THE JOURNAL OF BUXOGICAL CHEMISTRY Vol. 253, No. 11, Issue of June 10, pp. 4022-4030, 1978

Printed in U.S.A.

Immunochemical Studies on Choleeystokinin II. DISTRIBUTION AND MOLECULAR HETEROGENEITY IN THE CENTRAL NERVOUS SYSTEM AND

SMALL INTESTINE OF MAN AND HOG*

(Received for publication, September 13, 1977)

JENS F. REHFELD

From the Institute of Medical Biochemistry, University of Aarhus, UK-8000 Aarhus C, Denmark

Using sequence-specific radioimmunoassays, the distri- bution and heterogeneity of cholecystokinin (CCK) has been determined in extracts of tissues from the central nervous system and small intestine of adult man and hog. The CCK determinations were correlated with measurements with a radioimmunoassay specific for the structurally related pep- tide hormone gastrin. The amount and molecular forms of CCK extracted from brain as well as gut tissue were pH- dependent in that boiling at neutral pH released mainly small molecular forms, whereas acid released mainly large forms. Consequently, the tissue was first boiled at pH 6.6, after which acetic acid was added to pH 2.5. Large amounts of CCK were found in the central nervous system with a differential distribution: cerebral cortex, 200 to 2300 pmol/g with the highest concentrations in the neocortical regions; corpus striatum, 158 to 480 pmol/g; thalamus, 8 to 59 pmol/ g; hypothalamus, 24 to 220 pmol/g; mesencephalon, 13 to 93 pmol/g, and brain stem, 6 to 43 pmol/g. In the pineal body, the pituitary gland, cerebellum, pons, and spinal cord CCK was undetectable. In contrast with CCK proper, gastrins were measured in the pituitary gland (7 to 51 pmol/g), but gastrin was not detectable in other regions of the central nervous system of the two species. Immunoreactive CCK in the brain was found by gel chromatography to be present almost exclusively in three molecular forms: 1) a minor fraction corresponding to the triacontatriapeptide, CCK,, (5%); 2) a large fraction corresponding to the COOH-termi- nal octapeptide of CCK,, (80%); and 3) a small immunoreac- tive component corresponding to the COOH-terminal tetra- peptide of CCK,,. Determination of the true concentrations of the latter peptide, however, awaits purification and use of a proper standard. Duodenum and jejunum in both species contained CCK in concentrations similar’ to those measured in the cerebral cortex, varying from 550 to 2400 pmol/g of intestinal wall. The molecular heterogeneity of intestinal CCK was similar in both species studied. CCK in gut extracts eluted in five positions on gel chromatography: 1) immunoreactivity corresponding to peptides larger than CCK,, and CCK,,, was eluted with a constant, K,, , of 0.15 to 0.20 (10% of the immunoreactivity); 2) molecular forms

* This study was supported by grants from the Danish Medical Research Council, j. no. 512-5273, 512-5675, and 512-7230. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- tisernent” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

corresponding to CCK,, and CCK,,, (K,, - 0.50 to 0.55,8% of the immunoreactivity); 3) molecular forms presumably cor- responding to the COOH-terminal dodecapeptide of CCK,, Wa, - 0.63 to 0.74, 6% of the immunoreactivity); 4) a large fraction corresponding to the COOH-terminal octapeptide of CCK,, (K,,, = 1.10, 60% of the immunoreactivity); and finally 5) two peptides corresponding to each half of the COOH-terminal ociapeptide (K,, - 1.25 to 1.30, 16 to 20% of the apparent immunoreactivity). The existence and charac- ter of the different molecular forms of CCK were substanti- ated by rechromatography, chromatography with an 8 M urea gradient, affinity chromatography, and tryptic cleav- age. The results suggest (a) that CCK and gastrin are distributed in both brain and gut in a unique, differential manner; (b) that the amount of CCK in the brain by far exceeds the amount, of any other known hormonal brain peptide; (e) that CCK in both brain and gut is heteroge- neous, the predominating molecular forms corresponding to the small COOH-terminal octa- and tetrapeptides of CCK,,.

An important feature of peptide hormones is their molecular heterogeneity, i.e. the hormones are secreted as peptides of different size and biological activity within the same species. A fruitful way to study the heterogeneity of a known hormone has proved t,o be radioimmunochemical monitoring of tissue fractionation. Maximal information using such procedures is obtained by application of sequence-specific radioimunoas- says, which are assays that in a well defined manner recognize different regions of the hormonal sequence (I). In doing so it is possible to predict essential parts of the structure of the variant molecular forms of a hormone.

Cholecystokininl is a hormone which regulates enzyme secretion from the pancreas and gallbladder emptying. It has been purified as a triacontatriapeptide and a triacontanona- peptide (2-4) from hog intestine. Recent measurements with cross-reacting gastrin assays have suggested that CCK’-like peptides besides being located in the intestine are present also in brain tissue (5).

The preceding paper described the development and char- acter of sensitive sequence-specific radioimmunoassays for

1 The abbreviations used are: CCK, cholecystokinin; CCK,!,, tria- contanonapeptide cholecystokinin; CCK,,, triacontatriapeptide cho- lecystokinin; CCK,, COOH-terminal octapeptide of CCK,,; gastrin,,, heptadecapeptide gastrin; gastrin,,, triacontatetrapeptide gast,rin.

4022

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Distribution and Heterogeneity of Clzolecystokinin in Brain and Gut 4023

CCK:,, (6). The present communication reports the amounts and molecular forms of CCK in extracts of different parts of the central nervous system and the small intestine as mea- sured with sequence-specific radioimmunoassays. The immu- nochemical and chromatographic properties of CCK in these extracts suggest that CCK exists in a number of different forms of which molecules resembling the COOH-terminal octa- and tetrapeptides of CCK,, predominate both in brain and intestinal mucosa of man and hog.

MATERIALS AND METHCJDS

Tissues - Porcine brains and small intestines were obtained at a local abattoir from freshly killed pigs. The interval between killing of the hogs and freezing of tissue on dry ice was 15 to 20 min. In addition, brains were resected from three anesthetized pigs (Depart- ment of Experimental Surgery, Rigshospitalet, Copenhagen). These brains were immediately dissected on an ice-bath, and each region then frozen on dry ice. Specimens of human brains, duodeni, and jejuni were obtained at surgery at the University Hospitals in Copenhagen and Aarhus and immediately frozen on dry ice. The human brain specimens were small pieces (0.86, 2.40, and 3.42 g) of morphologically normal tissue in the periphery of resections of tumors in the frontal and parietal lobes. The specimens of human duodenum and jejunum were morphologically normal from resec- tions due to pancreatic cancers (Whipple’s resection). The frozen tissue was stored at -80°C. They were then weighed and extracted on the same day.

Extraction of Tissues-In the frozen state the tissue was cut into small pieces weighing a few milligrams. In order to study the effect of pH on the extraction procedure the frozen tissue pieces were then dropped directly either into boiling water (10 ml/g of tissue, pH 6.6) or into boiling dilute acetic acid (3% 0.5 &r acetic acid, pH 2.5) and boiled for 25 min, or the tissue was boiled in water as described above, after which 0.5 M acetic acid was added, resulting in a pH of 2.5. Remaining in the boiled solution the tissue samples were then homogenized in a Potter S homogenizer (B. Braun-Melsungen, W. Germany) for 10 min at high speed. Aft,er homogenizing each sample, the grinder was washed either with 5 ml of 0.05 M sodium phosphate, pH 7.5, or by acid extraction with 3% 0.5 hl acetic acid, pH 2.5, which was added to the homogenate. The homogenates were centrifuged for 20 min at 15,000 x g (Sorvall). The supernatant fractions were decanted and frozen at -20°C until they were assayed in duplicate by sequence-specific radioimmunoassays in the dilu- tions ‘/lo, l/30, ‘/loo, ‘lnoo, ll~,ooo, l/a,ooo, and ‘/lo,ooo. The results of the measurements were calculated as the mean of those dilutions resulting in concentrations measured within the working range of t,he assays, which generally required two or three degrees of dilu- tion.

Validation of Extraction Procedures - Control extractions with and without addition of 1.25 nmol of purified CCK:,, and 150 pmol of the synthetic COOH-terminal octapeptide CCK,, respectively, were performed using both boiling water extractions at neutral pH (6.6) and acetic acid extractions (pH 2.5) of jejunal and cortical tissue. The extracts were assayed in duplicate in dilutions from ‘ilo to l/lo,ooo in order to measure recovery. 1.5-ml extracts were then taken for gel chromatography as described below in order to deter- mine the molecular forms of CCK in the extracts.

Fractionation of Tissue Extracts- 2- to 6-ml extracts were applied to Sephadex G-50 (superfine) columns (25 x 2000 mm) eluted with 0.25 M ammonium bicarbonate, pH 8.2, at 4°C at a flow rate of 20 ml/ h. Fractions of 3.0 ml were collected. All columns were calibrated with ‘“SI-albumin and ““NaCl (Amersham) for indication of void volume and total volume. Besides, the columns were calibrated with 99% pure porcine CCK,!, and CCK,, (generous gift from V. Mutt, Karolinska Institute, Stockholm), synthetic COOH-terminal octa- peptide (SQ 19.844, batch NNOllNB), and synthetic COOH-terminal tetrapeptide (ICI, Alderley Park, Cheshire, England). The existence and nature of the different molecular forms in an extract of porcine jejunum and brain were substantiated by the following procedures: 1) refiltration on the Sephadex G-50 (superfine) columns of the individual components after pooling of fractions, lyophilization, and reconstitution in 2.0 ml of ammonium bicarbonate, pH 8.2; 2) dilution of the individual component and comparison of the dilution curves with CCK,, and synthetic fragments of CCK:,:,; 3) gel tiltration in an 8 M urea (Aristar) gradient as previously described (7); 4) gel filtration of the larger molecular forms of CCK after incubation with

trypsin previously treated with L-l-tosylamido-2-phenylethyl chlo- romethyl ketone, 10 pg of trypsin (Lot 37D778, Worthington) per ml for 20 min at 20°C; 5) gel filtration of the extracts after incubation with an immunosorbent prepared as previously described (8) using antiserum 2609 raised against synthetic heptadecapeptide gastrin and reacting against the common COOH-terminal pentapeptide sequence of gastrin and CCK (9).

Radioimmunoassay -- CCKs in the extracts were measured and characterized using the three different sequence-specific assays as described in the preceding paper (6) (Fig. 1). Antiserum 4478 was used in the assay specific for region 20 to 25 of CCK,,,,. This assay measured neither the COOH-terminal octapeptide nor tetrapeptide of CCK,,,. Neither does it measure any gastrins (type A assay). Antiserum 4698 was used in the assay specific for region 25 to 30 of CCK,,. This assay measures the COOH-terminal octapeptide and larger molecular forms of CCK. It measures neither the COOH- terminal tetrapeptide of CCK,, nor any gastrins. Antiserum 4698 binds only CCKs provided tyrosyl in position 27 of CCK,,, contains O-metbylsulfate (type B assay). Antiserum 2609 was used in the assay specific for region 29 to 33 of CCK,,. This antiserum, originally raised against heptadecapeptide gastrin (91, binds all cholecystoki- nins and gastrins containing the biologically active COOH-terminal tetrapeptide common to gastrin and CCK (type C assay). In all three CCK assays ‘“‘I-hydroxyphenylpropionic acid-succinimide ester cou- pled to CCK:,, was used as tracer. 99% pure porcine CCK,,,, was used as standard in type A assays and synthetic porcine CCK, as standard in type B and C assays.

In addition to the CCK radioimmunoassays, one gastrin radioim- munoassay was used for control and reference. Antiserum 2604 raised against human heptadecapeptide gastrin was used in this assay (9). It reacts with the COOH-terminal half of gastrin,,. It binds the three largest molecular forms of gastrin (10) with equimo- lar potency (111, whereas it binds <0.002 M CCK,:, and <0.008 M CCK, as compared to gastrin,, on a molar base at inhibition dose 50. Monoiodinated gastrin,, was used as tracer (12), and synthetic human gastrin,, was used as standard in this gastrin assay (type D assay).

The results of measurements on the extracts are expressed as picomoles of equivalent standard/g wet weight of tissue. The meas- urements on the fractionated extracts are expressed as nanomoles or picomoles of equivalent standard liter of fraction.

RESULTS”

Effect of Extraction Procedure on Cerebral CCK- At neu- tral as well as acid pH large amounts of immunoreactive CCK were extracted from porcine cerebrum (Table I). By boiling at neutral pH, however, 4 to 6 times more CCK was extracted than by acid extraction. Upon gel chromatography it appeared that at neutral pH 90% (mean, n = 6) of the cerebral CCK immunoreactivity was eluted in a peak at a position corre- sponding to that of CCK,. The remaining immunoreactivity was eluted later (K8, - 1.30) in a position similar to that for the COOH-terminal tetrapeptide of CCK,:, (Fig. 2).’ At acid pH the smaller amount of CCK extracted was eluted mainly in three positions. The first peak constituting 20% of the immunoreactivity eluted in a position corresponding to that of CCK,,. The second peak constituted approximately 10% of the immunoreactivity and eluted in a position corresponding to the COOH-terminal dodecapeptide of CCK,,,,. Finally, the majority oFthe immunoreactivity (70%) eluted like CCK, (Fig. 2). Addition of exogenous CCK, (Table I) or CCK,,,, (Table II) to boiling water or to acetic acid per se did not destroy the immunoreactivity nor degrade the added CCK to smaller molecular forms as revealed by gel chromatography. Of exog- enous CCK, added to frozen cerebral tissue pieces extracted

2 Portions of this paper (including Figs. 2 to 11 and Tables III to VI) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, Md. 20014. Request Document No. 77M-1476, cite author(s), and include a check or money order for $2.40 per set of photocopies.

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

4024 Distribution and Heterogeneity of Cholecystokinin in Brain and Gut

Ab 2609 Ab L698

FIG. 1. The amino acid sequence of porcine cholecystokinin (1 to 33). Shaded residues depict the additional NH,-terminal hexapeptide of the variant form of cholecystokinin. Brackets indicate the se- quences of cholecystokinin for which the immunoassays used are specific. The numbers indicate the antisera used in the sequence- specific assays.

TABLE I Effect of extraction procedure on concentration of endogenous

immunoreactive CCK in porcine cerebral cortex (pmollg of tissue (wet weight)) and on recovery of exogenous CCK, (150 pmol)

The concentrations were measured by type C assay using antise- rum 2609, which is specific for sequence 29 to 33 of CCK,,,,.

Acetic acid (PH 2.5)

Cerebral CCK Exogenous CCK, (recovery)

Cerebral CCK + exogenous CCK, (recovery)

925 250 141 152 (94%) (101%)

1225 350 (104%) (88%)

TABLE II Effect of extraction procedure on concentration of endogenous

immunoreactive CCK in porcinejejunum (pmolig of tissue (wet weight)) and on recovery of exogenous CCK,,,, (1250 pmol)

The concentrations were measured by type C assay using antise- rum 2609. which is sDecific for seauence 29 to 33 of CCK,,,.

Acetic acid (DH 2.5)

Jejunal CCK Exogenous CCK,:, (recovery)

Jejunal CCK + exogenous CCK,, (recovery)

1875 1001 1400 1447 (112%) (116%) 3000 2100

(96%) (93%)

either at neutral or acid pH, 88 to 104% was recovered (Table I) without change in molecular size. Based on these results, it was decided for further extractions to boil brain tissue first in water at pH 6.6 and then to add acetic acid to pH 2.5 in order to obtain maximal extraction of both large and small molecu- lar forms of CCK.

Effect of Extraction Procedure on Jejunal CCK - At both neutral and acid pH large amounts of CCK were extracted from porcine jejunum. At neutral pH, however, twice as much

CCK was extracted as at acid pH (Table II). Gel chromatog- raphy showed five different main components of CCK. The first fractions containing immunoreactive CCK were eluted in position corresponding to gastrin component I and proinsulin K, - 0.15 to 0.20). The second region of CCK immunoreac- tivity was in a position corresponding to CCK,,, (K,, = 0.50) and CCK,, (Ka, = 0.52). The third region of CCK immuno- reactivity eluted with an elution constant of 0.63 to 0.74. The fourth and major peak of CCK in jejunum eluted like CCK,

O-C,, - 1.10). Finally, extracted at pH 6.6 an immunoreactive peak was eluted in a position corresponding to the COOH- terminal tetrapeptide of CCK,, (K,, - 1.30). By extraction at neutral pH the tetra- and octapeptide-like component consti- tuted 92% of the total immunoreactivity (mean, n = 61, whereas the three components eluted earlier constituted from 1 to 8% of the immunoreactivity (Fig. 3). By extraction at acid pH the first peak comprised 20%, the third peak ll%, and finally the octapeptide-like peak the remaining 43% (Fig. 3). Based on these results it was decided for further extractions to boil intestinal tissue first in water at pH 6.6 and then to add acetic acid to pH 2.5 in order to obtain maximal extraction of both large and small molecular forms and CCK.

Distribution of CCK in Central Nervous System- The con- centration of immunoreactive CCK in the different regions of the central nervous system varied from immeasurable levels to 2.3 nmol/g of tissue (Table III). The highest concentrations were found in the telencephalic cortex, the frontal, parietal, temporal, and occipital lobes. Also the rhinencephalic regions, especially the olfactory area, contained large amounts of immunoreactive CCK. The concentrations of CCK in the diencephalon were of an order of 10 times below the telence- phalic concentrations. In the diencephalon, the hypothalamus contained the highest concentrations, 0.1 nmol/g of tissue. The more distal regions contained progressively less CCK. Some regions appeared devoid of CCK immunoreactivity. These regions were the epithalamus, including the pineal body, the mesencephalon including the cerebellum and pons, the spinal cord, and the pituitary. Also corpus callosum and the optic tract contained very little CCK.

Molecular Forms of CCK in Central Nervous System- As shown in Table III, the concentrations of CCK in the central nervous systems of hogs varied considerably depending on the assay used. The highest concentrations were, except for the pituitary gland, always measured with type B and type C assays and always in a similar order of magnitude. Since type A and type D assays measured much lower concentrations, the findings suggest that the predominating molecular form of CCK is the COOH-terminal octapeptide of CCK,,,,. This suggestion was confirmed by gel chromatography of extracts from both telencephalon (gray cortex and olfactory area), diencephalon (hypothalamus), mesencephalon (nigral sub- stance), and myelencephalon (vagal nucleus). Fig. 4 shows the elution pattern of an extract from the porcine parietal lobe of the cortex. The biopsies from human cerebral cortex contained immunoreactive CCK in concentrations of a magnitude simi- lar to those found in porcine cerebral cortex (Table IV); and the predominating molecular form in human cerebrum also resembled the COOH-terminal octapeptide of CCK,,, (Fig. 5). Both porcine and human brains contained, in addition to the octapeptide-like CCK, a larger molecular form of CCK with chromatographic and radioimmunochemical characteristics like CCK,,,. This form constituted 2 to 5% of the immunoreac- tivity. Finally, the brain tissue contained little immunoreac- tivity apparently corresponding to the COOH-terminal tetra-

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Distribution and Heterogeneity of Cholecystokinin in Brain and Gut 4025

peptide of CCK and gastrin (Figs. 4 and 5). Gastrin in Central Nervous System- While the gastrin

assay (type D assay) measured nothing or very low concentra- tions of immunoreactivity in most brain regions (Table III), and while these small amounts of immunoreactivity reflected only the weak cross-reactivity of the octapeptide-like CCK in the gastrin assay (Figs. 4 and 5), the pattern differed in the posterior lobe of the pituitary and the pituitary stalk (Table III). In the pituitary neither of the specific CCK assays (type A and B assays), devoid of cross-reactivity with gastrin, measured any immunoreactivity, while the cross-reacting CCK assay (type C) and the more specific gastrin assay measured substantial amounts of true gastrin immunoreactiv- ity.

Distribution of CCK in Small Intestine - As shown in Table V the concentration of CCK varied considerably in the upper intestinal tract of hogs when measured with the four different types of assays. A similar distribution and variation were found in the intestinal tract of man (Table VI). In the antral mucosa the specific CCK assays, type A and B, did not measure significant amounts of immunoreactivity, whereas the cross-reacting type C assay and the more specific gastrin type D assay measured high concentrations. The midportion of the duodenum contained comparatively low concentrations of gastrin, whereas high concentrations of immunoreactive CCK were found. The jejunum contained similar amounts of CCK as found in duodenum, whereas the content of gastrin as measured with type D assay was minute.

Molecular Forms of CCK in Proximal Small Intestine- Since type B and C assays measured the highest concentra- tions and concentrations in a similar range (1 to 2 nmol/g of tissue) in duodenum and jejunum, whereas type A assays measured concentrations between 0.2 and 0.9 nmol/g of tissue, the findings indicate that the predominating molecular form of CCK corresponds to the COOH-terminal octapeptide of CCK,,. By gel chromatography the interpretation of measure- ments on extracts was confirmed. Almost identical patterns of immunoreactive CCK components were found in duodenum and jejunum, without significant species differences between man and hog (Figs. 6 and 7). Immunoreactive CCK was eluted mainly in five positions as described above in the comments on the extraction procedure. The gastrin assay (type D) showed that small amounts of gastrin immunoreactivity also eluted in four positions: component I, component II (gastrin,,) with a Ka,. of 0.42, component III (gastrin,,) with a Ka, of 0.76, and finally a peak corresponding to the COOH-terminal octapeptide of CCK,,,,, which cross-reacts weakly with antise- rum 2604.

Validation of Existence of Different Molecular Forms of CCK- By refiltration of the various molecular forms of CCK in extracts of porcine jejunum, they eluted in positions corre- sponding to those found by gel filtration of the crude extract (Fig. 8). In an 8 M urea gradient the position of the individual peaks remained unchanged (Fig. 9). Immunosorption using a gastrin antiserum with COOH-terminal reactivity removed all CCK components measured with an assay specific for sequence 29 to 33 of CCK,, (type C assay), which suggests that none of the peaks measured with this assay were due to unspecific interference. Also, the peaks measured by type A assay, presumably specific for sequence 20 to 25 of CCKZ3, were removed. However, type B assay, specific for sequence 25 to 30 of CCK,,,,, still revealed one peak after immunosorption with the gastrin antiserum. This peak was eluted after CCKII, with an elution constant of 1.30 (Fig. 9). Incubation with

trypsin displaced immunoreactive CCK to the position corre- sponding to CCK, (Fig. 10). The CCK,,-like peptide in porcine jejunal extracts showed a dilution curve parallel to that of pure porcine CCK,,,, (Fig. 1lA). The CCK,-like materials in porcine jejunal and cerebral extracts diluted in parallel with synthetic porcine CCK, (Fig. llB), and material from the peak corresponding to the COOH-terminal tetrapeptide in extracts from porcine jejunum and cerebrum diluted in paral- lel with synthetic COOH-terminal tetrapeptide (Fig. IlC).

DISCUSSION

The results presented in this paper argue that CCK, al- though first isolated from the small intestine of hogs, is also widely distributed throughout the central nervous system in a unique differential manner and in amounts beyond those known for other hormonal peptides in the brain. Second, the results indicate that CCK in tissues is more heterogeneous than previously known, without significant species differences between man and hog. The predominating molecular forms in brain and gut from both species are peptides resembling the COOH-terminal octa- and tetrapeptides of CCK:,:,. Finally, the results demonstrate that the central nervous system, in addi- tion to its content of different molecular forms of CCK, also contains the related gastrointestinal hormone gastrin differ- entially located in the pituitary.

Extraction of CCK from Tissue - The study emphasizes the importance of evaluation of the extraction procedure for hor- mones present in different molcular forms. While the different gastrin components are all extracted in maximal amounts using boiling water at neutral pH (13-151, the structurally related CCK components differ. The larger molecular forms of CCK in intestines and nervous tissue are, thus, only partly extracted at neutral pH, while the octa- and tetrapeptide-like CCK are extracted in maximal amounts. Conversely, the large molecular forms of CCK are extracted more efficiently at acid pH, at which the small forms are only partly released (Figs. 1 and 2). The rational procedure is, consequently, first to boil the tissue at neutral pH to release the octa- and tetrapeptide-like CCK and then to add acid to release the large molecular forms. Since these, like CCK:,:, and CCK:,!,, are readily dissolved in water or buffers at neutral pH (Tables I and II), the necessity for acid probably reflects ionic binding in the tissue of the large molecular forms of CCK. The pH- dependent extraction explains in part why the small molecu- lar forms of CCK, in particular the highly bioactive COOH- terminal dodecapeptide- and octapeptide-like CCK (161, were not previously discovered in the intestines since acid extrac- tion was used to isolate CCK (2-4, 17). Moreover, failure of alginic acid to absorb the octa- and tetrapeptides, late elution of these peptides on Sephadex G-50 and G-25 columns, and finally disposal of water used for initial boiling of the intes- tines (2-4, 17) may further explain the discrepancy between the present and earlier extraction results with regard to the molecular nature of CCK in the intestinal tissue.

Several lines of evidence support the contention that some of the radioimmunoassayable molecular forms of CCK in brain and in gut are closely related and probably identical with the known triacontatriapeptide CCK and its COOH- terminal octa- and tetrapeptide. (a) Parts of the radioimmu- noassayable CCK in brain and gut extracts elute by gel chromatography on Sephadex G-50 (and G-25 using different buffer systems) in positions identical with those of CCKRB, CCKI(, and the COOH-terminal tetrapeptide (Figs. 3 and 5).

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

4026 Distribution and Heterogeneity of Cholecystokinin in Brain and Gut

(b) The substances in cerebral and intestinal extracts chro- matographically similar to CCKss, CCKN, and the COOH- terminal tetrapeptide react with the four different sequence- specific antisera in proportions identical with the Y-CCK displacements of the pure peptides (Figs. 3 to 6, Fig. 11, and Table III in Ref. 6). (c) Trypsin known to cleave the peptide bond between the arginine and aspartic acid residues in positions 25 and 26 cleaves all the larger molecular forms to the octapeptide-like CCK (Fig. 10). The same lines of evidence as discussed above support the contention that porcine and human CCK are very closely related, at least in the COOH- terminal half of their sequence.

Distribution and Heterogeneity of CCK in Central Nervous System - In recent years an increasing number of hormonal peptides have been isolated or demonstrated by radioimmu- noassays with a dual localization in both the central nervous system and the intestinal tract. Some of these peptides, substance P (18, 19), somatostatin (20, 21), neurotensin (22, 23), and enkefalin (24, 25) were first isolated from brain tissue. Of peptides primarily isolated from the gut, vasoactive intes- tinal polypeptide (VIP) was recently located in the central nervous system (26, 27). The present study extends the list of gut peptides located in the central nervous system with two classical hormones, CCK and gastrin. It has previously been suspected that a small peptide related to CCK was present in the brain, since gastrin radioimmunoassays that cross-reacted with CCK indicated that the gastrin-like peptide earlier reported (28) rather resembled the COOH-terminal region of CCK,, (5). However, the present results based on specific CCK radioimmunoassays give direct evidence for the existence of peptides similar to native CCK octapeptide in the brain. Moreover, the results argue that molecules similar to the large CCK,,, and the small COOH-terminal tetrapeptide also are present in brain tissue. Finally, the simultaneous use of assays specific for both CCK and gastrin demonstrated that not only CCK but also proper gastrins are located in the central nervous system (Table III). Further studies have shown that pituitary gastrin in the pituitary almost exclu- sively is present as gastrin,, (“big” gastrin) and gastrin,7 (29).

The distribution pattern for CCK in the central nervous system is unique and differs markedly from that of other hormonal brain peptides like thyrotropin-releasing hormone (TRH) (30, 31), somatostatin (32), substance P (33), vasoactive intestinal polypeptide (26, 27), neurotensin (21), and luteiniz- ing hormone-releasing hormone (LHRH) (34, 35). While the latter peptides are all located mainly in the hypothalamus and present only sporadically and in low concentrations in the cerebral cortex, CCK is predominantly located to the cerebral cortex, particularly in the neocortical part of the telence- phalon, the pallium (Table III), which dominates the central nervous system in man. CCK is, however, still present in substantial amounts in the diencephalon, mesencephalon, and the brain stem (Table III). In all regions CCK appears like the other peptides located in neurons (36, 37).

Another unique feature of cerebral CCK is the amount present in the brain. While the other hormonal peptides mentioned above mainly are found in the hypothalamus in concentrations of the order 10 to 100 pmol/g of tissue (21, 26, 30-35), hypothalamic CCK is present in similar concentra- tions, 25 to 100 pmol/g (Table III). However, in addition, CCK is present in concentrations 10 to 20 times higher in cortex. Since adult porcine and human hypothalamus constitute only 3 to 4 g, whereas the weight of cortex in man for instance is 400 g (38), the brain in pig and man thus contains on an

estimate approximately lo3 times more CCK than any of the other peptides on a molar basis. Thus, human brains contain totally 1 to 2 mg of CCKR, whereas other hormonal peptides are present in microgram amounts. The central nervous system also contains on a molar basis- more CCK than is present in the small intestine in man.

CCK in the brain is heterogenous. A minor fraction corre- sponds to CCKZ3, but the predominating form appears similar to CCK,. However, based on chromatographic behavior and immunoreactivity it can be assumed that the brain also contains a peptide closely related to the COOH-terminal fragment of CCKZ3, Trp-Met-Asp-Phe-NH, (Figs. 1, 4, and 5). Since the potency of this tetrapeptide is much smaller than that of the octapeptide, used as standard, with respect to displacement of ““I-CCK,, from antiserum 2609 (Fig. 11 and Table III in Ref. 6), the true molar concentrations of the tetrapeptide-like substance are 30 times above those indicated in Figs. 4 and 5. Thus, the tetrapeptide-like fragment of CCKSJ, which quantitatively appears to predominate in corti- cal tissue, may be the principal chemical messenger in the family of CCK- and gastrin-like peptides, and the larger molecular forms may merely be biosynthetic precursors. This hypothesis is partly supported by the observation that the tetrapeptide on an equimolar basis releases insulin more potently than the larger heptadecapeptide gastrin (39). It is not likely that the COOH-terminal octa- and tetrapeptide-like fragments of CCK,, are artifacts due to degradation of CCK,, during the extraction, since pure CCK,, and CCK, added to the tissue were not degraded during extraction. Furthermore, CCK immunoreactivity in spinal fluid, which is not subjected to extraction appears like CCKs and the COOH-terminal tetrapeptide (40).

Distribution and Heterogeneity of CCK in Small Intestine - In the duodenum and the jejunum of both man and hog CCK is present in concentrations (Tables V and VI) of the same order of magnitude as measured for other gastrointestinal hormones (41). Also, in the gut CCK is highly heterogeneous (Figs. 6 and 7), and more different molecular forms of CCK are present than previously described (2-4). The predominat- ing forms are, as in brain tissue, small peptides resembling the COOH-terminal octa- and tetrapeptide of CCK,,. In addi- tion, type B assay (antiserum 4698) measured a small peptide eluted after CCK,, (Figs. 6B and 7B). This peptide was not removed by immunosorption using the gastrin antiserum binding sequence 29 to 33 of CCK,,. Since antiserum 4698 binds sequence 25 to 30 of CCK,,,, (6), the small peptide discovered in the intestinal extract presumably corresponds to the NH,-terminal part of CCKs. Since a peptide resembling the COOH-terminal tetrapeptide of CCK, is present in the extracts as measured by type C assay, the small peptide bound by antiserum 4698 conceivably constitutes the remaining NH,- terminal tetrapeptide of CCK8, Asp-Tyr (SO,)-Met-Gly. The true amount of this peptide in gut is unknown since a useful standard has not been available. It is possible that brain tissue also contains this peptide, although in amounts consid- erably smaller than in the gut (Fig. 5B). Using arguments similar to those discussed for the small peptides in extracts of brain tissue, the small CCK-related peptides in the gut are not likely to be artifactually produced during extraction of the larger molecular forms of CCK. In addition to the small peptides, the gut also contained large molecular forms, larger than CCK,, and CCK3!,, in substantial amounts. These large forms eluted after the void volume of Sephadex G-50 columns and were not detectable in brain tissue. Further studies are

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Distribution and Heterogeneity of Cholecystokinin in Brain and Gut

needed to clarify their structure and role as putative biosyn- thetic precursors.

tiers in Gastrointestinal Hormone Research (Andersson. S.,

Acknowledgments -1 thank Professor V. Mutt, Department of Biochemistry, Karolinska Institute, Stockholm, for his generous gifts of various ‘cholecystokinin preparations. I am grateful to Dr. G. J. Dockray, Physiological Laboratories, University of Liverpool, for valuable discussion of extraction procedures. The expert technical assistance of Bodil Basse and Gitte Hollase is gratefully acknowledged.

Addendum - Since the present article was submitted, Muller, Straus, and Yalow ((1977) Proc. Natl. Acad. Sci. U. S. A. 74, 3035-3037) have reported the presence of radioim- munoassayable CCK in pig cerebral cortex, and Dockray ((1977) Nature 270, 359-361) the presence of octapeptide-like CCK in porcine intestines.

REFERENCES

1. Segre, G. V., Tregear, G. W., and Potts, J. T., Jr. (1975) in Methods in Enzymology, Hormone Action, Part B: Peptide Hormones (O’Malley, B. W., and Hardman, J. G., eds) Vol. 37, pp. 38-66, Academic Press, New York

2. Mutt, V., and Jorpes, J. E. (1967) Biochem. Biophys. Res. Commun. 26, 392-397

3. Mutt, V., and Jorpes, E. (1971) Biochem. J. 125,57P-58P 4. Mutt, V. (1976) Clin. Endocrinol. 5, 175s-183s 5. Dockray, G. J. (1976) Nature 264, 568-570 6. Rehfeld, J. F. (1978) J. Biol. Chem. 253. 4016-4021 7. Rehfeld, J. F. (1972) Biochim. Biophys. kcta 285, 364-372 8. Rehfeld. J. F.. and Stadil. F. (1973) Stand. J. Clin. Lab. Invest.

31, 455-464 9. Rehfeld, J. F., Stadil, F., and Rubin, B. (1972) &and. J. Clin.

Lab. Invest. 30, 221-232 10. Rehfeld, J. F., Stadil, F., and Vikelsoe, J. (1974) Gut 15, 102-111 11. Rehfeld, J. F. (1976) J. Clin. Invest. 58, 41-49 12. Stadil, F., and Rehfeld, J. F. (1972) Stand. J. Clin. Lab. Invest.

30, 361-368 13. Gregory, R. A., and Tracy, H. J. (1964) Gut 5, 103-117 14. Gregory, R. A., and Tracy, H. J. (1972) Lancet 2, 797-799 15. Rehfel& J. F., Stadil, F.,-MalmstrBm, J., and Miyata, M. (1975)

in Gastrointestinal Hormones (Thompson, J. C., ed) pp. 43-58, University of Texas Press, Austin

16. Rubin, B., and Engel, S. L. (1973) Nobel Symposium 16: Fron-

17. 18.

19.

20.

21.

22.

23.

24.

25.

26.

27. 28.

A. 6. G. (1977)Lancet 1,972-974 ’ ’ ’ Larsson, L. I., Fahrenkrug, J., Schaffalitzky de Muckadell, 0.

B., HBkanson, R., Sundler, F., and Rehfeld, J. F. (1976) Proc. Natl. Acad. Sci. U. S. A. 73, 3197-3200

Said, S. I., and Rosenberg, R. N. (1976) Science 192,907-908 Vanderhaeghen, J. J., Signeau, J. C., and Gepts, W. (1975)

Nature 257, 604-605 29. Rehfeld, J. F. (1978) Nature 271, 771-773 30. Oliver, C., Eskay, R. L., Ben-Jonathan, N., and Porter, J. C.

31. 32.

(1974) Endocrinology 95,540-546 Winokur, A., and Utiger, R. D. (1974) Science 185, 265-267 Brownstein, M., Arimura, A., Sato, H., Schallv, A. V., and

33. Kizer, J. S. (1975) Endocrinology 96, 1456-1461 -

Brownstein. M. J.. Mroz. E. A.. Kizer. S.. Palkovits. M.. and

34.

35.

36. 37.

38.

39. 40. 41.

ed) pp. 41-55, Almqvist & Wiksell, Stockholm Mutt. V.. and Jornes. J. E. (1968) Eur. J. Biochem. 6. 156-162 Chang k. M., and ieeman, S. E. (1970) J. Biol. Chem. 245,

4784-4790 Studer, R. O., Trzekiak, A., and Lergier, W. (1973) Helv. Chim.

Acta 56, 860-866 Brazeau, P., Vale, W., Burgus, R., Ling, N., Butcher, M.,

Rivier, J.. and Guillemin, R. (1973) Science 179, 77-79 Arimura, A., Sato, H., Dupont, A., Nishi, N., and Schally, A.

V. (1975) Science 189, 1007-1009 Carraway, R., and Leeman, S. E. (1976) J. Biol. Chem. 251,

7045-7052 Kitabgi, P., Carraway, R., and Leeman, S. E. (1976) J. Biol.

Chem. 251, 7053-7058 Hughes, J., Smith, T. W., Kosterlitz, H. W., Fothergill, L. A.,

Morgan, B. A., and Morris, H. R. (1975) Nature 258, 577-579 Polak. J.. Sullivan. S. N.. Bloom. S. R.. Facer. P.. and Pearse.

Leeman, k. E. (1976) Brain Res. 116, i99-305 ’ ’ Winters, A. J., Eskay, R. L., and Porter, J. C. (1974) J. Clin.

Endocrinol. Metab. 39, 960-963 Palkovitz, M., Arimura, A., Brownstein, M., Schally, A. V., and

Saavedrs, J. M. (1974) Endocrinology 95, 554-558 Larsson, L. I., and Rehfeld, J. F. (1977) Nature 269, 335-338 Hiikfelt, T., Elde, R., Fuxe, K., Johansson, O., Ljungdahl, A.,

Goldstein, M., Luft, R., Nilsson, G., Terenius, L., Ganten, D., Jeffcoate, S. L., Rehfeld, J. F., and Said, S. (1977) in The Hypothalamus (Reichlin, S., Baldessarini, R., and Martin, J. B., eds) pp. 69-135, Raven Press, New York

Ranson. S. W., and Clark, S. L. (1959) The Anatomy of the Nervous System, W. B. Saunders Co., Philadelphia ”

Rehfeld. J. F. (1971)Acta Endocrinol. 66. 169-176 Rehfeld; J. F., and Kruse-Larsen, C. (19iS) Brain Res., in press Bloom, S. R. (1974)Br. Med. Bull. 30, 62-67

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

4028 Distribution and Heterogeneity of Cholecystokinin in Brain and Gut

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Distribution and Heterogeneity of Cholecystokinin in Brain and Gut 4029

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

4030 Distribution and Heterogeneity of Cholecystokinin in Brain and Gut

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from

J F Rehfeldhog.heterogeneity in the central nervous system and small intestine of man and

Immunochemical studies on cholecystokinin. II. Distribution and molecular

1978, 253:4022-4030.J. Biol. Chem. 

  http://www.jbc.org/content/253/11/4022.citation

Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/253/11/4022.citation.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on October 16, 2020

http://ww

w.jbc.org/

Dow

nloaded from