followed by Southern blot hybridization. The autoradio-
graphic study of CCK-A and -B receptors in the human
duodenum and pancreas was examined in vitro. To
determine the subtypes to CCK receptors in the pancreas
or duodenum, we studied the abilities of CCK-A and -B
receptor agonists (CCK-8 and gastrin) and antagonists
(loxiglumide, L-364,718 and L-365,260) to inhibit binding
of 125I-CCK-8. CCK-A receptor mRNA was not expressed
in the human pancreas, but was expressed in the gall-
bladder and duodenum, although it was expressed in the
pancreas by RT-PCR. CCK-B receptor mRNA was ex-
pressed in the pancreas, but not in gallbladder and duo-
denum. Using autoradiography, high concentrations of
CCK-A receptors were detected in the duodenal mucosa,
although in the pancreas only CCK-B receptors were
detected by this method. These results suggest that
localization of CCK-A receptor in human duodenum pro-
vides a biochemical and morphological basis for some
physiological functions of CCK.
Introduction
CCK is not only a gastrointestinal hormone but also animportant neuropeptide [1]. CCK receptors in peripheraltissues and the central nervous system have been classi-fied into two subtypes, CCK-A and CCK-B, on the basisof their affinities for a structurally and functionally relat-ed family of peptides with identical COOH-terminal pen-tapeptide sequences, but differences in sulfations at thesixth (gastrin) and seventh (CCK) tyrosyl residues and inresponse to specific antagonists [2, 3]. In peripheral tis-sues, through binding to CCK-A receptors, CCK is amajor physiological mediator of gallbladder contractionand pancreatic enzyme secretion. In addition, CCK me-diates pancreatic enzyme secretion and satiety via theCCK receptors of vagal afferent nerves [4, 5]. Infusion ofCCK-A-receptor antagonists loxiglumide or L-364,718dose-dependently inhibited CCK-8-stimulated pancreaticenzyme responses in humans [6, 7]. However, by in vitroautoradiography human pancreas predominantly ex-
presses CCK-B receptors, whereas only CCK-A receptorswere localized in the human gallbladder [8]. Thus, thesedata suggested that secretion of human pancreas pro-duced by CCK is mediated via CCK-A receptors, al-though it probably resides on neurons that may even beextrapancreatic. In the previous rat study, we confirmedby reverse transcriptase-polymerase chain reaction (RT-PCR) in control rats that CCK-A- and CCK-B-receptorgene expression in the proximal intestine, which involvesvagal afferent nerve terminals, was substantial, whereasthese transcripts could not be detected by Northern trans-fer analysis [9]. Intraduodenal administration of capsa-icin, which stimulates pancreatic secretion via the vagalafferent nerve through CCK-A receptor significantly in-creased the protein output in normal rats [10], whereas noincrease with capsaicin stimulation was observed inOLETF rats, which lack CCK-A receptors because of agenetic abnormality [9, 11–13]. Thus, it is suggested thatthe lack of stimulatory effect of capsaicin is due to lack ofCCK-A receptors in the small intestine and/or in the pan-creas, although the functional involvement of CCK-Breceptors has not yet been clarified. Therefore, in thepresent study, we examined the gene expressions of CCK-A and CCK-B receptors and the localization of CCK-Aand CCK-B receptors in the human duodenum whichmay be another candidate of a receptor site of the intesti-nal phase of pancreatic secretion.
Materials and Methods
Materials and ChemicalsSynthetic CCK-8 and gastrin-17-I were purchased from the Pep-
tide Institute, Osaka, Japan. cDNA probes of the coding regions ofthe human CCK-A and CCK-B receptors were kindly provided byDr. S.A. Wank, NIH, Bethesda, Md., USA. Samples of healthy duo-denal tissue and gallbladder and pancreas with adenocarcinoma ofthe pancreas were obtained at surgery. Immediately after removingthe tumors, the healthy tissues were resected. One part was used forhistological examination, whereas the other part was rapidly frozenat –80°C. All samples were evaluated histopathologically, and nosigns of abnormalities or inflammation were detected.
Northern Transfer Analysis of CCK-A and CCK-B Receptors inHuman Duodenum, Pancreas and GallbladderTotal RNA was extracted by the acid-guanidium thiocyanate-
phenol-chloroform method. The quantity and purity of RNA weredetermined from the absorbances at 260 and 280 nm. RibosomalRNA bands were visualized under ultraviolet light after ethidiumbromide staining. Samples showing no signs of degradation wereused for Northern transfer analysis using cDNA probes of the codingregions of the human CCK-A- [14] and CCK-B-receptor [15] andß-actin cDNA (Wako Chemical Industries, Osaka, Japan).
Amplification of the Coding Regions of CCK-A- andCCK-B-Receptor cDNAs from Human Duodenum, Pancreas andGallbladderFor cDNA synthesis, RNA was incubated in reaction mixture
with RT using a First-Strand cDNA Synthesis kit (Pharmacia LKBBiotechnology, Uppsala, Sweden). For amplification of the codingregion of the CCK-A- and CCK-B-receptor precursor cDNAs, thePCR was carried out for 40 cycles using sense and antisense primerscorresponding to the nucleotide sequence of human CCK-A receptorprecursor cDNA [14] (nucleotides 823-842: GTGATGATGGTGG-CATATGG and 1613-1593: CCTTCATCTGATTCCAGAGCA),human CCK-B receptor precursor cDNA (nucleotides 215-235,ATATGCTCATCATCGTGGTCC and nucleotides 747-727,CGCTCGAGATGAATCCCGAA) [15] and human ß2-microglobu-lin precursor cDNA as an internal standard, respectively.
Southern Blot Analysis of PCR Amplified cDNA (PCR-Southern)The PCR reaction product was separated electrophoretically in
NuSieve agarose (3%) gel and blotted onto a nylon membrane. Theblot was hybridized with [32P]-labelled cDNA probes of the codingregions of human CCK-A [14] and CCK-B [15] receptors, and thenautoradiographed as described [12].
was obtained from New England Nuclear (Boston, Mass., USA). Thespecific activity was about 2,200 Ci/mmol. Human pancreas andduodenum tissue sections were cut at –20°C using a cryostat micro-tome, mounted onto APS-coated slides, and dried overnight at–80°C. Binding of 125I-BH-CCK-8 to tissue sections was performedaccording to the method of Tang et al. [8] with a slight modification.Sections mounted on slides were preincubated in 50 mM 2-(N-mor-pholino)ethanesulfonic acid buffer (MES) (pH 6.0), 5 g/l BSA for30 min at 22°C. The slides were washed wtih 50 mM MES. Theslides were incubated at 22 °C in 50 mM MES, 130 mM NaCl,7.7 mM KCl, 1 mM EGTA, 5 mM MgCl2, 4 Ìg/ml leupeptin, 2 Ìg/mlchymostatin, 0.25 mg/ml bacitracin and 200 pM 125I-BH-CCK-8 for180 min. To determine the subtypes of CCK receptors on the pan-crease or duodenum, we studied the abilities of CCK-A and CCK-Breceptor agonists (CCK-8 and gastrin-17-I) and antagonists (loxiglu-mide, L-364,780 and L-365,260) to inhibit binding of labelled CCKunder identical conditions. On completion of the incubation, theslides were washed 5 times for 5 min each in ice-cold 50 mM MES.The slides were then dried under a stream of cold air. The dried slideswere placed in a storage phosphor cassette for 24 h at 22°C tempera-ture. The latent image stored in the storage phosphor screen wasvisualized by laser scanning the screen in the Molecular Imager (GS-525, BIORAD).
Results
Northern Transfer Analysis of CCK-A- andCCK-B-Receptor mRNAs in the Duodenum, Pancreasand GallbladderThe 32P-labelled cDNA probe of the CCK-A receptor
coding region hybridized with 6 kb mRNA from thehuman gallbladder and duodenum. No mRNA of about
Mechanism of Cholecystokinin-A-ReceptorAntagonist
Digestion 1999;60(suppl 1):75–80 77
Fig. 1. Autoradiograms of the CCK-A andCCK-B receptor and ß-actin in mRNA sam-ples of human pancreas, duodenum and gall-bladder detected by Northern transfer analy-ses with CCK-A and -B receptors and ß-actinprobes. Autoradiograms of Northern blotsshowed a single band of ß-actin in all sam-ples, of CCK-A receptor in duodenum andgallbladder and of CCK-B receptor in pan-creas. Lane 1 = Human gallbladder; 2 = pan-creas; 3 = duodenum.
Fig. 2. Expression of the CCK-A receptor precursor gene in humanpancreas, duodenum and gallbladder by RT-PCR Southern transferanalyses. The coding region of cDNAs of the CCK-A receptor (791bp) were amplified by PCR. Autoradiograms showed a single band(0.8 kb) of CCK-A receptor in all samples. Lane = 1 Human duode-num; 2 = pancreas; 3 = gallbladder.
6 kb hybridizing with the probe was detected in thehuman pancreas, although ß-actin mRNA was detected.In the same blot, no hybridizing CCK-B receptor mRNAof about 3.3 kb was detected in blots of mRNAs of duode-num and gallbladder, although it was detected in blots ofmRNAs of pancreas (fig. 1).
Expressions of CCK-A and CCK-B Receptor PrcursorGenes in Human TissuesWe obtained two amplified DNAs of the expected sizes
(791 bp for DNA amplified from CCK-A receptor precur-sor mRNA and 109 bp for DNA amplified from ß-micro-globulin precursor mRNA used as an internal standard)by PCR from RNA of pancreas, duodenum and gallblad-der. These DNAs gave a single hybridizing fragment of0.8 kb by PCR-Southern blotting using a CCK-A-receptorcDNA probe (fig. 2).
We obtained one amplified DNA by PCR from RNAof the pancreas, duodenum and gallbladder, which corre-sponded to the expected size of 533 bp of DNA amplifiedfrom CCK-B receptor precursor mRNA. These DNAsgave single hybridizing fragments of 0.6 kb by the PCR-Southern blot method (data not shown).
AutoradiographyStorage phosphor autoradiographs showed the specific
binding of 125I-BH-CCK-8 on human tissue sections ofboth pancreas and duodenum. By comparison with thecorresponding HE-stained sections, we found the bindingof 125I-BH-CCK-8 diffusely distributed in the exocrinecomponent of the pancreatic tissue sections (fig. 3a). Add-ing 10–6 M CCK-8 and gastrin caused marked similarinhibitions of labelled CCK-binding, showing a specificbinding to CCK-B receptor (fig. 3b).
By comparison with the corresponding HE-stained sec-tions, we found the binding of 125I-BH-CCK-8 diffuselydistributed in the muscle layer and in part of the mucosallayer (fig. 4a) of the duodenal tissue sections. Adding10–6 M CCK-8, but not 10–6 M gastrin caused markedinhibition of labelled CCK binding, showing a specificbinding to CCK-A receptor. Dose-inhibition curvesshowed that marked competitive binding of 125I-BH-CCK-8 was found with CCK-8, but little competitivebinding of 125I-BH-CCK-8 was found with gastrin (notshown). Human duodenum mucosal layer had CCK-Areceptors labelled with 125I-BH-CCK-8, which was dis-placed mainly by CCK, loxiglumide and L-364,718 butnot gastrin and L-365,260 (fig. 4b).
Fig. 3. a Autoradiographs of the humanpancreas tissue section and adjacent HE-stained section (D). Autoradiograph of thetissue sections shows total binding of 125I-BH-CCK-8 (A), binding of 125I-BH-CCK-8with addition of 10–6 M CCK-8 (B) or gas-trin-I (C). b Ability of unlabelled CCK-8 andgastrin-I to inhibit binding of 125I-BH-CCK-8 to the human pancreas tissue section. Eachvalue is the mean (n = 3).
0
CCK-810 mol/l-6
Gastrin10 mol/l-6
20
40
60
80
100
%
0
Gastrin I10 mol/l-6
CR150510 mol/l-5
L-364, 71810 mol/l-7
L-365, 26010 mol/l-7
20
40
60
80
100
%
4b
3a
3b
Mechanism of Cholecystokinin-A-ReceptorAntagonist
Digestion 1999;60(suppl 1):75–80 79
Fig. 4. a Autoradiographs of the human duodenum tissue section and adjacent HE-stainedsection (G). Autoradiograph of the tissue sections shows total binding of 125-I-BH-CCK-8 (A),binding of 125I-BH-CCK-8 with addition of 10–6 M CCK-8 (B), 10–6M gastrin-I (C), 10–4 Mloxiglumide (D), 10–7 M L-364,718 (E) and 10–7 M L-365,260 (F). b Ability of unlabelledgastrin-I, loxiglumide, L-364,718 and L-365,260 to inhibit binding of 125I-BH-CCK-8 to thehuman duodenum mucosal layers, pancreas tissue section. Percentage of maximum inhibi-tion (100%) is defined as the inhibition by 10–6 M CCK-8. Each value is the mean (n = 3–5).
In this work we observed no expression of the CCK-A-receptor gene in the pancreas by the Northern transfermethod, but its expression was observed in the duodenumand gallbladder. However, we also confirmed by the RT-PCR method and PCR-Southern blotting that the CCK-A-receptor gene is expressed in appreciable amounts inthe pancreas. Although the CCK-A receptor is known tobe a major form in rat pancreas, Wank and co-workerscould not detect gene expression of CCK-A receptors inhuman pancreas by Northern transfer analysis, but de-tected CCK-B receptors [3, 14, 15]. Thus, in humans, themajor form in the pancreas is the B type, although thegallbladder has CCK-A receptors. Therefore, there is spe-cies difference in receptor expression in rats and hu-mans.
In comparison of corresponding human duodenal tis-sue sections stained by HE, the specific binding of 125I-BH-CCK-8 are found to be localized in the duodenalmucosa and muscle layer, and the specific CCK-A-recep-tor antagonists (loxiglumide and L-364,718) inhibited thebinding, indicating the major form of CCK receptor istype A. The mechanism responsible for CCK-inducedpancreatic enzyme secretion under physiological condi-tions remains controversial in humans [8, 16]. Both directand neural actions of CCK for mediating pancreatic secre-tion have been proposed. However, we have preliminary
data that dispersed human pancreatic acini did not re-spond to CCK stimulation, although acetylcholine stimu-lated amylase secretion [unpubl. observation]. Therefore,the presence of abundant CCK-A receptors in the mucosalsite of humans suggests a role of CCK in the intestinalneural phase of pancreatic secretion.
A recent electrophysiological study in the ferret indi-cated that the duodenum is endowed with a rich supply ofmucosal afferent fibers that are sensitive to luminal stim-uli such as hypertonic saline, HCl, and light stroking [17].These neurons initiate a reflex leading to inhibition offood intake (satiety), gastroduodenal motility, acid secre-tion, and stimulation of pancreatic secretion. Thus, it ispossible that CCK and duodenal stimuli may act on dis-tinct duodenal afferent fibers to potentiate pancreaticsecretion and that specific CCK-A-receptor antagonistsinhibited pancreatic secretion possible acting via thereceptor sites of terminal afferent fibers, although it isalmost impossible to record vagal afferent nerve excita-tions in humans.
Acknowledgments
This study was supported in part by a Grant-in-Aid for ScientificResearch in Japan and a grant from the Pancreatic Research Founda-tion of Japan. We thank Dr. G. Green at the University of TexasHealth Science Center in San Antonio for reviewing this manu-script.
References
1 Dockray GJ: Immunochemical evidence ofcholecystokinin-like peptides in brain. Nature1976;264:568–570.
2 Jensen RT, Wank SA, Rowley WH, Sato S,Gardener JD: Interaction of CCK with pan-creatic acinar cells. Trends Pharmacol Sci1989;10:418–423.
3 Wank SA: Cholecystokinin receptors (editorialreview). Am J Physiol 1995;269:G628–646.
4 Li Y, Owyang C: Vagal afferent pathway me-diates physiological action of cholecystokininon pancreatic enzyme secretion. J Clin Invest1993;92:418–424.
5 Silvente-Poirot S, Dufresne M, Vaysse N,Fourmy D: The peripheral cholecystokinin re-ceptors. Eur J Biochem 1993;215:513–529.
6 Adler G, Beglinger C, Braun U, Reinshagen M,Koop I, Schafmayer A, Rovati L, Arnold R:Interaction of the cholinergic system and chole-cystokinin in the regulation of endogenous andexogenous stimulation of pancreatic secretionin humans. Gastroenterology 1991;100:537–543.
7 Cantor P, Olsen O, Gertz BJ, Gjorup T, Worn-ing H: Inhibition of cholecystokinin-stimulatedpancreaticobiliary output in man by the chole-cystokinin receptor antagonist MK-329. ScandJ Gastroenterol 1991;26:627–637.
8 Tang C, Biemond I, Lamers CBHW: Cholecys-tokinin receptors in human pancreas and gall-bladder muscle: A comparative study. Gastro-enterology 1996;111:1621–1626.
9 Miyasaka K, Masuda M, Kawanami T, Funa-koshi A: Neurohormonal regulation of pan-creatic exocrine function in rats without geneexpression of the cholecystokinin-A receptor.Pancreas 1996;12:272–279.
10 Gicquel N, Nagain C, Chariot J, Tsocas A, Lev-enez F, Corring T, Roze C: Modulation of pan-creatic secretion by capsaicin-sensitive sensoryneurons in the rat. Pancreas 1994;9:203–211.
11 Miyasaka K, Kanai S, Ohta M, Kawanami T,Kono A, Funakoshi A: Lack of satiety effect ofcholecystokinin (CCK) in a new rat model notexpressing the CCK A receptor gene. NeurosciLett 1994;180:143–146.
12 Funakoshi A, Miyasaka K, Shinozaki H, Masu-da M, Kawanami T, Takata Y, Kono A: An
animal model of congenital defect of gene ex-pression of cholecystokinin (CCK)-A receptor.Biochem Biophys Res Commun 1995;210:787–796.
13 Takiguchi S, Takata Y, Funakoshi A, Miyasa-ka, Kataoka K, Fujimura Y, Goto T, Kono A:Disrupted cholecystokinin type-A receptor(CCK-AR) gene in OLETF rats. Gene 1997;197:169–175.
14 De Weerth A, Pisegna JR, Huppi K, Wank SA:Molecular cloning, functional expression andchromosomal localization of the human chole-cystokinin type A receptor. Biochem BiophysRes Commun 1993;194:811–818.
15 Wank SA, Pisegna JR, De Weerth A: Brain andgastrointestinal cholecystokinin receptor fami-ly: Structure and functional expression. ProcNatl Acad Sci USA 1992;89:8691–8695.
16 Adler G, Nelson DK, Katschinski M, BeglingerC: Neurohormonal control of human pancreat-ic exocrine secretion. Pancreas 1995;10:1–13.
17 Blackshaw LA, Grundy D: Effect of cholecysto-kinin on two classes of gastroduodenal vagalafferent fibers. J Auton Nerve Syst 1990;31:191–202.
