With the availability of modern biochemical and molecular techniques developed in the 1980s and 1990s, it was possible to characterize and clone the two cholecystokinin receptors known today. They are now identified as the cholecystokinin-1 (CCK-1) and the cholecystokinin-2 (CCK-2) receptor subtypes.
In this review entry, we will summarize what we now know about the CCK-2 receptor. More specifically, we will discuss its biochemical characterization, its organ localization, its functions in normal organs, its role in cancer along with the intracellular signaling routes used to relay its messages. References to the other cholecystokinin receptor are given in the preceding chapter.
Initially, the CCK-1 receptor was known as the CCKA receptor (A for alimentary) and the CCK-2 receptor as the CCKB receptor (B for brain). This reclassification of the CCK receptor subtypes occurred recently based on recommendations from the International Union of Pharmacology Committee regarding receptor nomenclature and drug classification.
These two receptors can be differentiated by their respective affinity for CCK and gastrin, their respective natural agonist. Indeed, the CCK-1R binds and responds to sulfated CCK with a 500–1,000-fold higher affinity or potency than sulfated gastrin or nonsulfated CCK. On the contrary, the CCK-2R binds and responds to gastrin and CCK with almost the same affinity or potency, and discriminates poorly between sulfated and nonsulfated agonists (Dufresne et al. 2006). The CCK-2R gene has been cloned in five different species (human, rat, mouse, dog, and rabbit). According to published data, the human CCK-2R has 90% identity to the rat and canine receptor. The predicted mouse CCK-2R shares 87% and 92% amino acid identity with the human and rat receptor, respectively. The rabbit CCK-2R protein exhibits 93–97% amino acid similarly with corresponding cDNA identified in human, canine and rat brain or stomach. In general, the CCK-2R cloned from different species exhibits a comparable high affinity for the agonists CCK-8 and gastrin-17-1 with IC50 in the nM range. On the contrary, it has a higher affinity for its specific antagonist, L-365,260 (IC50 around 10 nM) than for the CCK-1R antagonist L-364,718 (IC50 around 1 μM). Finally, as opposed to the CCK-1R, the CCK-2R binds to a single class of high affinity sites with a KD in the range of 0.2–2.0 nM (Morisset 2005).
The presence and cellular localization of a receptor in a specific organ can be detected and determined by at least five different techniques: agonist- or antagonist- binding studies, RT-PCR, Western blot, immunological staining, and autoradiography. Of these five, the Western blot and the immunological staining techniques are the most reliable and informative on the presence of the receptor and its specific localization if the specificity of the antibodies used has been carefully evaluated. However, so far, the nonquantitative RT-PCR technique has been the major method of analysis. This technique is indeed very sensitive but there is always this possibility that the message levels detected could be extremely low and therefore, may not translate into the expression of functional levels of receptors. The autoradiography technique with a radioactive ligand seems the least sensitive while the binding technique mostly used cell membrane preparations. All five techniques can be used for receptor detection but cellular localization can be most efficiently and accurately revealed by immunological staining and confirmation with Western blot using preferably the same antibody. Therefore, to relate a physiological response to a specific receptor’s occupation, we need to know that the receptor is there and that it is really localized on the cell where the physiological response originates from in the target organ.
The CCK-2R was mainly found throughout the central nervous system in areas that parallel the distribution of CCK and gastrin immunoreactivity and mRNAs. Samples from rat brain tissue allowed the detection of CCK-2R mRNA. Mature forms of the receptor mRNA were identified in cerebral cortex, hypothalamus, and hippocampus. Moreover, a truncated form of the receptor was identified in all regions of the brain studied (Jagerschmidt et al. 1994). Using the PCR amplification technique, the CCK-2R was also identified in the brain, but also in the stomach, pancreas, small intestine, liver, colon, spleen, lung, thymus, ovary, breast, prostate, testes, adrenals, and kidneys, as opposed to what was observed previously (Morisset 2005).
CCK binding sites were also identified on rat adipocyte membranes and their occupation lead to the control of leptin release (Attoub et al. 1999). In the adrenal glands, CCK-stimulated aldosterone release was related to occupation of both CCK subtypes in rats, but exclusively through CCK-2R occupation in humans (Mazzocchi et al. 2004). In normal esophagus, the CCK-2R has been revealed by RT-PCR and localized to columnar epithelia within the glandular and lower crypt regions of the mucosa by microautoradiography (Haigh et al. 2003).
Even though the CCK-2R was first identified in the brain, historically it was originally cloned by Kopin through the expression of a canine parietal cell cDNA library and isolation of a cDNA clone encoding a 453-amino-acid protein.
Although there is no doubt that the CCK-2R is present in the stomach, its precise location at the cellular level still presents some controversies depending on the technique used for detection. By RT-PCR, the receptor was identified in rat and mice antrum mucosae, and in the fundus by Northern blot analysis. By in situ hybridization, the CCK-2R was present only on parietal cells during the perinatal period in rats and mainly in ECL cells in adult rodents (Waki et al. 1998). This absence of CCK-2Rs on rat adult parietal cells was confirmed by Fluo-CCK-8 binding studies which located the receptor only on ECL cells co-localized with histidine decarboxylase (Bakke et al. 2001). These data contradict the cloning data of Kopin performed on a 95% pure isolated canine parietal cell preparation, but contamination with only a few ECL cells could be enough to give one CCK-2R clone.
Adding to the confusion about CCK-2R localization in the stomach, immunohistochemistry and electron microscopy studies performed with polyclonal antibodies to human gastrin indicate, on one hand, that gastrin binding sites were found in the guinea pig stomach on parietal and chief cells but much less on ECL and somatostatin cells (Tarasova et al. 1996). On the other hand, using an antibody raised against the C-terminal decapeptide of the CCK-2R, it was found that undifferentiated rat gastric epithelial cells express CCK-2R. Furthermore, other epithelial cells in the progenitor zone of the adult gastric gland also expressed the receptor and among them were the parietal, chief, and ECL cells (Tommeras et al. 2002).
In the pancreas, we encountered the same location problems as in the other organs with the usual techniques used. Indeed, from whole human pancreas and isolated acini, RT-PCR amplified both receptor subtypes mRNA but the receptor protein could not be identified. On the contrary, this CCK-2R has been detected in the pig pancreas by Northern blot analysis as well as in the whole human and rat fetal and adult pancreas and isolated islets; this finding was confirmed by Western blot analysis and by immunohistochemistry with a precise location on islet somatostatin cells (Julien et al. 2004). Binding studies with labeled G-17 confirmed the presence of high affinity sites on dog pancreatic acini (Fourmy et al. 1984). A similar situation was found in guinea pig pancreas with the same technique (Yu et al. 1987) and on rat acini by immunofluorescence (Morisset 2005). Such a location on pancreatic delta cells was also confirmed in six different species, the calf, pig, horse, rat, human, and dog. However, with a different antibody, the receptor was also localized on human islet glucagon cells. The only other location observed for the CCK-2R in the pancreas by immunofluorescence besides the somatostatin cells was on the ductal cells in the young calf and adult cow (Morisset 2005).
By comparing and analyzing all these data using different techniques, we can easily ascertain that technology has not yet been successful in helping us localize this CCK-2R in any tissues at the cellular level. We therefore need to produce more specific receptor antibodies which will have to be certified for their specificity by different laboratories using the same technique of immunohistochemistry for cellular localization and Western blot for protein detection. When this is elucidated, it will be much easier to definitively associate this receptor occupation with a cellular physiological response.
Physiological Response to the CCK-2 Receptor Occupation
The stomach remains one of the main targets of gastrin, and the CCK-2R expressed in this organ is an important part of the system that regulates functions of the gastric epithelium. Two main pathways of activation of stomach secretion by gastrin have been proposed: direct and indirect. Direct implies the gastrin action on parietal and chief cells while indirect means that the gastrin effect is mediated by histamine stored and released by the enterochromaffin-like (ECL) cells of the mucosa. According to Bakke (2001), the indirect effect of gastrin via the ECL cells would prevail for the control of acid secretion since they could not locate the CCK-2R on parietal cells. Since these data were obtained by fluorescence CCK-8 binding, they will have to be confirmed by immunofluorescence or immunocytochemistry. Studies performed in gastrin-CCK double KO mice to study control of acid secretion via the CCK-2R are of limited use since acid secretion in such a case becomes only controlled by cholinergic vagal stimulation (Tarasova et al. 1996).
Besides being involved in the control of acid secretion, the occupation of the CCK-2R was also associated with proliferation of the gastric mucosa, especially during lactation in rats (Takeuchi and Johnson 1987). Gastrin was also described as a growth factor for the small intestine and colon. However, an elaborate study performed in fed rats using relatively high doses of pentagastrin for 5 days did not have any major effect on DNA synthesis in the stomach, duodenum, and colon, nor on the total organ weight of these three organs and their total DNA contents (Solomon et al. 1987). Furthermore, more recently, it was shown that gastrin elicited increased DNA synthesis in ECL cells but failed to do so in parietal cells. This increased DNA synthesis was preceded by phosphorylation and activation of MAP kinase as well as c-fos and c-jun gene expression only in the ECL cells (Kinoshita et al. 1998). These data clearly indicate that occupation of the CCK-2R can induce growth in this specific subpopulation of gastric cells and suggest that gastrin could act to promote commitment or differentiation of stomach precursor cells. The presence of the CCK-2R on the proliferating progenitor cell population remains to be confirmed.
As indicated earlier in this review, the location of the CCK-2R in the pancreas of mammals as well as humans remains an unsolved situation which should be addressed soon. Indeed, in recent literature, we can observe that the human pancreatic acinar cells lack functional responses to CCK and gastrin (Ji et al. 2001), while a more recent study (Murphy et al. 2008) clearly claimed a secretory response of purified human pancreatic acinar cells to CCK. However, when the human CCK-2R receptor is either transfected in purified human acinar cells or present in the murine pancreas via transfection using the elastase promoter, both models exhibited amylase release from their respective acini in response to CCK and gastrin (Ji et al. 2001; Saillan-Barreau et al. 1998). This indicates that the acini of these preparations can respond to CCK-2R activation when the receptor is present and that it can use the cell’s intracellular machinery leading to enzyme release.
The implication of the CCK-2R in the control of pancreas growth and regeneration via its agonist gastrin is far from being established. However, pentagastrin, given to pregnant rats, was found to be the most potent factor, with hydrocortisone, responsible for fetal pancreas growth as its effect was blocked by the specific CCK-2R antagonist L-365,260. The fetal rat pancreas also expressed gastrin whose expression disappeared early after birth (Morisset et al. 2004). Furthermore, after pancreatitis induction in rats, significant overexpression of the CCK-2R mRNA was observed quite early during the destruction period as well as during early regeneration; the presence of caerulein was however constantly needed for prolonged expression of the receptor (Morisset and Calvo 1998). Recent data obtained in transgenic CCK-2R mice with CCK-2R overexpression in pancreatic acinar cells supported the possibility of this receptor being involved in pancreas growth as significant increases were observed in pancreas weight and areas occupied by the exocrine cells. Further increases in pancreatic weight were observed when these rats were bred with insulin-gastrin transgenic mice to achieve continuous stimulation of the acinar cells overexpressing the CCK-2R (Morisset et al. 2004). With regard to the human pancreas, even though Tang claimed that the CCK-2R are distributed all over the exocrine pancreas as established by phosphorimaging detection, it remains that no one has yet established first, that the CCK-2Rs are present at any other location than on the endocrine delta cells and second, that the human pancreas did not show any sign of regeneration after partial pancreatectomy. Since we and others have shown that the pig pancreas can regenerate after partial pancreatectomy, perhaps we should use this species as a model to study pancreas regeneration (Morisset et al. 2004). However, prior to initiating such studies, it is mandatory that we clearly establish the location of this CCK-2R in human and pig pancreas to secure the pig model.
Role of CCK-2R in Cancer
The expression of both CCK receptor subtypes in human gastrointestinal cancers remains poorly documented and is still of controversial nature. From a potential therapeutic point of view, it is very important to ascertain which CCK receptors are present on gastrointestinal tumors. Indeed, these receptors might act in concert with oncogenes to promote neoplastic transformation or contribute to tumor invasiveness. The presence of a known receptor may offer the opportunity to use receptor antagonists or a cytotoxic toxin linked to a stable ligand of this receptor as an adjunct in the treatment of a target cancer.
As an example of the confusion existing on the presence of CCK-2R on specific tumors, one study indicated that this receptor, evaluated by RT-PCR, was not detected in esophageal cancers (Clerc et al. 1997), while its presence was confirmed with gastrin in 100% of patients with Barrett’s metaplasia and in 70% of patients with esophageal adenocarcinomas. This suggests that an autocrine signal could be involved in the pathogenesis of Barrett’s metaplasia before development of dysplasia and cancer. Does this mean that this receptor disappears during cancer development?
Controversy also exists on the presence of the CCK-2R on gastric cancer cells. The receptor and gastrin were detected in some human gastric adenocarcinoma by immunohistochemistry while detected in only 7% of gastric cancer samples by RT-PCR or by receptor autoradiography. To amplify the doubt, another study, using the RT-PCR technique, claimed that seven out of eight specimens of gastric adenocarcinomas express the receptor (Clerc et al. 1997).
Data on the presence of the CCK-2R in colon cancer are less controversial. Indeed, by autoradiography the CCK-2R was either rarely or not at all expressed in colorectal cancer in one study, while found in only two out of 12 colon adenocarcinoma samples by RT-PCR in another study. In this organ, however, isoforms of the CCK-2R were detected, the short and longer forms. The short form, also called CCK-C, was detected in 100% of the tumors tested along with gastrin mRNA. This truncated form does not discriminate between amidated and glycine-extended forms of gastrin, and since the colonic adenocarcinomas synthesize progastrin but fail to process the prohormone, this prohormone could be the specific agonist of this receptor and thus be the intracrine growth factor in human colorectal cancer (Biagini et al. 1997). It thus becomes important to search for this receptor type in any other type of gastrointestinal cancers, and the presence of this new subtype may be the reason why so few CCK-2Rs were detected in colon cancer, we were not looking for the appropriate one.
The pancreatic tumors are no different than the others as some authors have failed to find transcripts of the CCK-2R at the mRNA level whereas others reported expression. A CCK-C receptor was also characterized in human pancreatic cancer (Smith et al. 2002). However, it is not yet clear if this newly reported pancreatic CCK-CR is comparable to the one reported in the colon. Although some authors claimed that the CCK-2Rs were present in all pancreatic carcinomas, others reported their occasional presence in gastroenteropancreatic tumors with rare expression in ductal pancreatic carcinoma (Morisset et al. 2004). However, when the CCK-2R is transfected in pancreatic acinar cells of the murine pancreas along with the already present CCK1R, its new presence seems to play a key role in the development of pre- and neoplastic lesions in the pancreas of these mice (Clerc et al. 2002). The new feature of this experimental approach is the simultaneous expression of CCK-1 and CCK-2Rs in the same cell. Could this result in novel signaling secondary to receptor cross talk and/or heterodimerization of these receptors? We could expect potentiation of signal transduction, and heterodimerization may represent a new mechanism for modulation of these receptors which may lead to cancer development.
The presence of both CCK receptor subtypes in cancer cell lines is not a negligible phenomenon, as we will see. Furthermore, the expression of CCK receptors in cancer cell lines has mostly given rise to controversial reports with regard to their type and density.
In SEG-1 cells, a human esophageal adenocarcinoma cell line, RT-PCR has established the presence of the CCK-2R receptors responsive to gastrin for their growth with growth inhibition by L-365,260, the specific CCK-2R antagonist. However, one study on esophageal cancers reported expression of low levels of CCK-2R mRNA with the CCK-1Rs being overexpressed (Morisset 2005).
In AGS-B cells, a human gastric carcinoma cell line expressing the CCK-2R, amidated gastrin G-17 was associated with increased DNA synthesis which was inhibited by the L-365,260 antagonist. This cell line could be derived from the 7% of the gastric cancer samples expressing the CCK-2R (Morisset 2005).
Contrary to the rare incidence of CCK receptors in colon cancer as seen above, it seems that some colorectal carcinoma cell lines Colo 320, HCT 116, LoVo, and the immortalized mouse colon cell line YAMC express the CCK-2R as they respond to gastrin for their growth and tumorigenicity. However, such cell lines are believed to be the exception because, as for tumors, most colorectal carcinoma cell lines do not express gastrin/CCK-2 receptors or the CCK-1Rs (Baldwin and Shulkes 1998). In general, however, expression of the CCK-2R isoform is always coupled with co-expression of the gastrin gene in GI tumor cell lines, implying the presence of a gastrin/CCK-2R autocrine loop. If there is any involvement of peptides of the CCK and gastrin family in the development of colon cancer, it could be through glycine-extended gastrin and progastrin as high concentrations of these two peptides have been observed in colon tumors and in blood of patients with colorectal cancer. The receptor for these peptides was described but never cloned (Dufresne et al. 2006).
In non-GI tract tumor cell lines, co-expression of gastrin and CCK-2R was not found, except in the lymphoblastic leukemia cell line Molt 4 which also expresses the receptor’s isoform. High CCK-2R density and incidence was reported in medullary thyroid carcinomas, small cell lung cancers, and stromal ovarian cancers; leiomyo-carcomas express both CCK-R subtypes (Dufresne et al. 2006).
In some pancreatic cell lines of ductal origin, a gastrin autocrine loop involving the CCK-2R exists as in the MIA PaCa-2 cells but such is not the case in the PANC-1 cells possessing also the CCK-2R (Morisset et al. 2004). In these same two cell lines, transfection of both CCK-R subtypes led to inhibition of their growth by CCK-8, a result which contradicts what was previously observed in these same cells. Would it be possible that overexpression of these CCK-Rs leads to inhibition of their growth? In the rat AR4-2J cells, an acinar cell line which expresses both CCK-R subtypes, a trophic response to gastrin has been shown which is coupled to CCK-2R occupation (Scemama et al. 1989). In BxPC-3 human cancer cells, the expression of the recently characterized CCK-CR was confirmed as its specific antibody resulted in growth inhibition (Smith et al. 2002). In the RIN-14B cells, a pancreatic somatostatin endocrine cell line, the presence of both CCK-Rs was revealed by Western blot and immunofluorescence; in these cells, occupation of both receptors by caerulein and gastrin led to somatostatin secretion, while cell growth was inhibited during stimulation of the CCK-1R (El-Kouhen and Morisset 2010). The message which can be drawn from these studies on cancer cell lines driven from any type of cancer is that it is very important to establish what kind of CCK-Rs are present and if there is any type of autocrine loop existing involving both specific agonists, CCK and gastrin. The other major observation remains that cancers in situ or cancer cell lines are unique in their growth control and that each typical cancer cell could respond to a specific chimiotherapeutic treatment depending on the intracellular signaling systems activated by the agonist.
Intracellular Signaling Pathways
Besides mutations of specific genes often causing cancer development, activation of intracellular signaling pathways by occupation of hormones and growth factor receptors in an unregulated way can be part of the tumor growth and invasiveness processes.
Occupation of the CCK-2R by gastrin is no exception and results in activation of a variety of cell-type–specific transduction pathways involved in proliferation. Among them are phospholipase C, c-src-like tyrosine kinases, p125 FAK, phosphatidylinositol 3-kinase and the MAPK, ERK and p38 kinases (Rozengurt and Walsh 2001).
Besides proliferation, malignant transformation also results in loss of epithelial differentiation, acquirement of mesenchymal characteristics, and increasing invasive and metastatic potential. Gastrin and the CCK-2R seem to be involved in epithelial-mesenchymal transition, and this process could involve p60-src. PI-3-kinase and ERK1/2 activation for periods of up to 4 h, leading to reduced association between E-cadherin, p120CTN, and β-catenin resulting in loss of cell adhesion and scattering (Bierkamp et al. 2002).
How can we now reconcile the growth-promoting effect of gastrin via occupation of its CCK-2R on different cancer cells with recent observations indicating that gastrin through this same CCK-2R can prevent rather than promote colorectal carcinogenesis via activation of the MAPK/ERK/AP-1 pathway and inhibition of NF-κB (Muerkoster et al. 2005)? Furthermore, how to explain this other finding that occupation of the CCK-2R in CHO and Swiss 3T3 cells activated similar intracellular signals with opposite growth effects. The reason given was that it depends on the cell model (Detjen et al. 1997).
With so many conflicting data on the effects of gastrin through occupation of its CCK-2R regarding growth and intracellular signaling pathways activated, it seems that all of these controversies come from the choice of the cell line selected to perform these studies and whether they possess the CCK-2 receptor or that it has been transiently or stably transfected. Indeed, we can list these cell lines: the MDCK cells, the CHO cells, the Swiss 3T3 cells, the COS-7 cells, the AR42J cells, the Colo 320 cells, the Lovo cells, the SW707 cells, and the HCT-115 cells, among others. It is believed that in order to have a clear image of the effects of gastrin on cancer cell growth and transformation through CCK-2R occupation, all the different signaling pathways involved in these processes should be studied in one selected cell line preferably expressing the CCK-2R. In doing so, it will then be possible to assign pathways to different responses happening in the cell leading to cancer development, growth control, and invasiveness.
The message to be stretched out of this review is that a lot of uncertainty still exists on the location and roles played by this CCK-2R on normal and tumoral cells. In order to clarify the existing dilemmas, we need, as scientists, to set up standard protocols to be followed using well-accepted techniques and tools which everybody agrees on. In this way, it will then be possible to clearly establish which organ or tumor expresses the receptor and on which specific cell it is operating. With all this new knowledge, we shall be able to clearly establish the physiological roles of this receptor on normal cells and possibly target or use the receptor for chemotherapy on tumor cells, if applicable.
- Clerc P, Leung-Theung-Long S, Wang TC, Dockray GJ, Bonisson M, Delisle MB, Vaysse N, Pradayrol L, Fourmy D, Dufresne M. Expression of CCK2 receptors in the murine pancreas: proliferation, transdifferentiation of acinar cells and neoplasia. Gastroenterology. 2002;122:428–37.CrossRefPubMedGoogle Scholar
- Mazzocchi G, Malendowicz LK, Aragona F, Spinazzi R, Nussdorfer GG. Cholecystokinin (CCK) stimulates aldosterone secretion from human adrenocortical cells via CCK2 receptors coupled to the adenylate cyclase/protein kinase A signaling cascade. J Clin Endocrinol Metab. 2004;89:1277–84.CrossRefPubMedGoogle Scholar
- Morisset J. The gastrointestinal cholecystokinin receptors in health and diseases. Ann Acad Med Bialostocensis. 2005;50:21–36.Google Scholar
- Morisset J, Calvo E. Trophicity of the pancreas: from rodents to man. J Physiol Pharmacol. 1998;49(Suppl 2):25–37.Google Scholar
- Morisset J, Lainé J, Biernat M, Julien S. What are the pancreatic target cells for gastrin and its CCKB receptor? Is this a couple for cancerous cells? Med Sci Monit. 2004;10:RA 242–6.Google Scholar