Cholecystokinin Bioactivity in Human Plasma
`Molecular Forms, Responses to Feeding, and Relationship to Gallbladder Contraction
`
`Rodger A. Liddle, Ira D. Goldfine, Melvin S. Rosen, Randy A. Taplitz, and John A. Williama
`Cell Biology Laboratory and Departments of Medicine and Radiology, Mount Zion Hospital and Medical Center San Francisco
`'
`'
`California 94120; and Gastroenterology Unit, Departments of Medicine and Physiology,
`University of California, San Francisco, California 94143
`
`Abstract
`
`Introduction
`
`A sensitive and specific bioassay for the measurement of
`cholecystokinin (CCK) in human plasma was developed to
`determine the molecular forms of CCK in circulation, CCK
`responses to feeding, and the physiologic role of CCK in
`gallbladder contraction. First. plasma was quantitatively ex(cid:173)
`tracted and concentrated with octadecylsilylsilica, and the
`extracts were then assayed for their ability to stimulate amylase
`release from isolated rat pancreatic acini. Acini were highly
`sensitive to CCK whereas gastrin reacted only weakly in this
`system. With the assay, plasma levels of cholecystokinin
`octapeptide (CCK-8) bioactivity as low as 0.2 pM were detect(cid:173)
`able. CCK bioactivity in plasma was inhibited by the CCK
`antagonist. bibutyryl cyclic guanosine monophosphate, and was
`eliminated by immnnoadsorption with an antibody directed
`against the carboxyl terminus of CCK. Detection of fasting
`levels of CCK was possible in all individuals tested and
`averaged 1.0±0.2 pM (mean±SE, n = 22) CCK-8 equivalents.
`Plasma CCK biological activity was normal in patients with
`gastrin-secreting tumors. Mter being fed a mixed liquid meal,
`CCK levels rose within 15 min to 6.0±1.6 pM. The individual
`food components fat. protein, and amino acids were all potent
`stimulants of CCK secretion; in contrast. glucose caused a
`significant but smaller elevation in plasma CCK levels. Gel
`filtration studies identified three major forms of CCK bioactivity
`in human plasma: an abnndant form that eluted with CCK-33,
`a smaller form that eluted with CCK-8, and an intermediate
`form that eluted between CCK-33 and CCK-8. Ultrasonic
`measurements of gallbladder volume indicated that this organ
`decreased S1% in size 30 min after feeding a mixed liquid
`meal. This contraction occurred coincidentally with the increase
`in plasma CCK levels. Next CCK-8 was infused to obtain CCK
`levels similar to postprandial levels. This infusion caused a
`decrease in gallbladder volume, similar to that seen with a
`meal. The present studies indicate, therefore, that CCK can be
`bioassayed in fasting and postprandial human plasma. These
`studies also suggest that CCK may be an important regulator
`of gallbladder contraction.
`
`A preliminary report of this study was presented at the Annual Meeting
`of the American Gastroenterological Association in New Orleans, LA,
`1984, and has appeared in abstract form ( 1984. Gastroentero/ogy. 86:
`1163).
`Address correspondence to Dr. Liddle, Cell Biology Laboratory.
`Received for publication 17 July 1984 and in revised form 18
`December 1984.
`
`J. Oin. Invest.
`© The American Society for Oinical Investigation, Inc.
`0021-9738/85/04/ 1144/09 $ 1.00
`Volume 75, April 1985, 1144-1152
`
`Since its discovery by Ivy and Oldberg (1) in 1928, cholecys(cid:173)
`tokinin (CCK)1 has generally been accepted to be a major
`hormonal regulator of gallbladder contraction. Harper and
`Raper (2) in 1943 extracted from the small intestine a peptide
`that stimulated pancreatic enzyme secretion and named it
`pancreozymin. It is now known that CCK and pancreozymin
`are the same molecule that has both gallbladder contractile
`and pancreatic stimulatory properties (3). Cholecystokinin(cid:173)
`pancreozymin, now termed cholecystokinin, was originally
`identified in porcine intestine as a 33-amino acid peptide
`(CCK-33) that shares an identical carboxyl terminal pentapep(cid:173)
`tide sequence with gastrin. Other molecular forms of CCK
`have been identified in intestinal extracts, brain, and plasma
`of various species (4-11). CCK-39 was characterized from hog
`intestine as a hexapeptide extension on the amino terminus of
`CCK-33 (4, 5). A larger molecular form, CCK-58, has been
`extracted from dog intestine and partially characterized (6).
`Forms of CCK similar in size to CCK-12, CCK-8, and CCK-
`4 have been characterized immunochemically in intestinal
`extracts ( 10, 11).
`The ability to study the circulating forms of CCK and the
`regulation of CCK secretion has been hampered by the lack
`of a rapid, sensitive, and specific assay for the hormone. In
`general, prior bioassays for CCK have not been sensitive
`enough to measure circulating levels of the hormone (12, 13),
`and radioimmunoassays have been hampered by crosm:activity
`with gastrinlike substances. Estimations of circulating CCK
`levels by radioimmunoassay have been useful, but to distinguish
`CCK from gastrin, either two antibodies that differ in their
`ability to recognize CCK and gastrin must be used, or the
`plasma must be chromatographed to separate CCK from
`gastrin (14-16). This need for processing may account for the
`wide variation in CCK levels that have been reported (14-24).
`In addition, CCK exists in multiple molecular forms, and
`antibodies directed against one portion of one molecular form
`may not recognize another molecular form. This immunovar(cid:173)
`iability may also account for some of the variability of molecular
`forms of CCK that has been reported.
`Recently, it has even been questioned whether CCK is a
`primary physiologic regulator of gallbladder contraction (25,
`26). Similarly, it has been suggested that physiologic postpran(cid:173)
`dial CCK levels alone cannot account for postprandial pan(cid:173)
`creatic enzyme secretion ( 15). Reliable measurements of CCK
`in plasma are necessary, therefore, to determine the hormonal
`role of CCK in normal and pathologic states.
`
`1. Abbreviations used in this paper: CBZ-L-tryptophan, N-carbobcnzoxy(cid:173)
`L-tryptophan; cGMP, guanosine 3'-5'-cyclic monophosphate; KHB,
`Krebs-Henseleit bicarbonate (bulfer); TR, Tris-Ringer (buffer); VIP,
`vasoactive intestinal polypeptide.
`
`1144
`
`R. A. Liddle, I. D. Goldfine, M. S. Rosen, R. A. Tap/itz, and J. A. Williams
`
`
`
`
`MAIA Exhibit 1042
`MAIA V. BRACCO
`IPR PETITION
`
`

`

`We have now developed a method for measuring human
`plasma CCK based on the ability of CCK in plasma extracts
`to stimulate amylase release from isolated rat pancreatic acini.
`These acini respond to CCK concentrations as low as 1 pM,
`and with the ability to concentrate plasma circulating CCK,
`levels as low as 0.2 pM can be measured. This assay has
`allowed us to measure both fasting and postprandial CCK
`levels, the relative contribution of various food components to
`CCK release, and the molecular forms of CCK in plasma.
`
`Methods
`
`The following substances were purchased: soybean trypsin inhibitor
`(types 1-S and 11-S), atropine sulfate, N 20 21-dibutyryl guanosine 3'-5'(cid:173)
`cyclic monophosphate (dibutyryl cGMP); N-carbobenzoxy-L-tryptophan
`(N-CBZ-L-tryptophan), carbamylcholine (carbachol), and Sephadex G-
`50 superfine from Sigma Chemical Co., St. Louis, MO; chromatograph(cid:173)
`ically purified collagenase from Worthington Biochemical Corp., Free(cid:173)
`hold, NJ; minimal Eagle's medium amino acid supplement from
`Gibco Laboratories, Grand Island, NY; bovine serum albumin, fraction
`V, from Miles Laboratories, Inc., Elkhart, IN; procion yellow dye from
`Polysciences, lnc., Warrington, PA; Staphylococcus aureus 10% cell
`suspension from New England Enzyme Center, Boston, MA; octade(cid:173)
`cylsilylsilica (SEP-PAK C-18) cartridges from Waters Associates, Mil(cid:173)
`lipore Corp., Milford, MA; instant breakfast supplement from Carnation
`Co., Los Angeles, CA; vasoactive intestinal polypeptide (VIP) and
`synthetic human gastrin-17 I from Bachem, Inc., Torrance, CA and
`Sincalide (CCK-8) from E. R. Squibb & Sons, Inc., Princeton, NJ.
`The following substances were gifts: CCK-8 from Dr. Miguel
`Ondetti of the Squibb Institute for Medical Research, Princeton, NJ;
`purified porcine CCK-33 from the Gastrointestinal Hormone Labora(cid:173)
`tory, Karolinska Institute, Stockholm, Sweden; gastrinoma-derived
`gastrin-17 II and gastrin antibody 1611 from Dr. John Walsh of
`University of California, Los Angeles and Center For Ulcer Research
`and Education, Los Angeles, CA; CCK/gastrin antibody RSB70 directed
`at the carboxyl terminus of CCK and gastrin from Dr. Margery
`Beinfeld, St. Louis University, St. Louis, MO; 1251-gastrin-17 from Dr.
`Steven Vigna, University of Oregon, Eugene, OR; Lipomul emulsion
`consisting of 71 % corn oil (88% oleic and linoleic acids) from the
`Upjohn Co., Kalamazoo, MI; Casec, casein powder from Mead Johnson
`and Co., Evansville, IN; and Nutrisource mixed L-amino acids consisting
`of 7% isoleucine, 14% leucine, 7% valine, 1.5% tryptophan, 8%
`phenylalanine, 3% methionine, 7% lysine, 5% threonine, 5.5% arginine,
`0. 7% tyrosine, I% cysteine, 7% alanine, 11.8% glutamic acid, 5.5%
`aspartic acid, 3% histidine, 2% serine, 6% glycine, and 5% proline from
`Sandoz, Minneapolis, MN.
`
`Bioassay of CCK
`Preparation of isolated pancreatic acini and measurement of amylase
`release. The buffer used to prepare isolated acini was modified Krebs(cid:173)
`Henseleit bicarbonate buffer (KHB), enriched with minimal Eagle's
`medium amino acid supplement and 0.1 mg/ml purified soybean
`trypsin inhibitor. KHB buffer was equilibrated to pH 7.4 with 95% 0 2
`and 5% CO2 (27).
`The incubation buffer was Tris-Ringer (TR) that contained 40
`mM Tris (hydroxymethyl)aminomethane, 103 mM NaCl, I mM NaH2
`PO., 4.7 mM KO, 1.28 mM Ca02, 0.56 mM MgC[i, II.I mM
`glucose, 0.1 mg/ml soybean trypsin inhibitor, minimal Eagle's medium
`amino acid supplement, and 5 mg/ml bovine serum albumin (BSA).
`TR buffer was equilibrated with 100% 0 2 and adjusted to pH 7.4
`at 37°C.
`Isolated pancreatic acini were prepared from 180-200-g female
`Sprague-Dawley rats by collagenase digestion of pancreas in KHB as
`previously described (27, 28). Acini were then incubated with plasma
`extracts or standard CCK-8 concentrations for 30 min at 37°C (28).
`Amylase released into the medium was assayed using procion yellow
`coupled starch as substrate (29). Amylase release expressed as percent
`of total amylase content, was compared with a dose-response curve
`
`for CCK-8 in order to calculate the CCK content of plasma expressed
`as CCK-8 equivalents.
`In this preparation of isolated pancreatic acini, CCK-8 is the most
`potent stimulus for amylase release (30). CCK-8 is threefold more
`potent than CCK-33. In contrast, gastrins are much less potent than
`either CCK-8 or CCK-33. Compared with CCK-8, the relative potencies
`of gastrins I and II are 0.00046 and 0.0025 to I, respectively. In
`addition, the C-terminal pentapeptide of CCK, CCK-5, is 5,000 times
`less active than CCK-8 on a molar basis (30).
`Feeding and collection of plasma. Human subjects for all studies
`were healthy volunteers between 21 and 43 yr of age. Subjects
`underwent an overnight 12-15-h fast prior to each study performed.
`Blood samples were drawn from an indwelling 19-gauge butterfly
`catheter in the antecubital fossa during the 2-3-h course of the study.
`Blood was collected into iced heparinized tubes and immediately
`centrifuged at 1,000 g to obtain plasma for CCK determinations.
`Blood for gastrin determination was collected into nonheparinized
`tubes, at room temperature, for recovery of serum.
`Subjects were fed orally liquid diets of either a mixed meal or
`various food components (fat, protein, amino acids, or glucose). The
`mixed liquid meal was made of Carnation instant breakfast supplement,
`one egg, and "half and hair' milk and cream, totaling 1.5 cal/ml and
`consisted of 40% fat, 20% protein, and 40% carbohydrate. This meal
`was given as 5.6 ml/kg body wt and was consumed over a 1-2-min
`period. Blood was drawn for CCK level determinations at various
`times up to 2 h after feeding.
`Other food components included 25% solutions (100 g/400 ml of
`water) of either glucose, amino acids, fat (in the form of corn oil,
`Lipomul), or protein (given as casein). Separate control diets tested
`included: 400 ml of either water, 0.9% sodium chloride, or 2% sodium
`chloride. The osmolality of these solutions ranged from 0 to 600
`mosmol.
`
`Extraction of CCK from plasma
`CCK was extracted from plasma by adsorption onto SEP-PAK cartridges
`previously washed with 5 ml of methanol and 20 ml of water. The
`cartridges were then washed again with 20 ml of water and the CCK
`was eluted with I ml of 80% ethanol and 0.2% tritluoroacetic acid.
`The eluants were collected in 30-ml flat-bottomed incubation vials and
`dried under a nitrogen stream at 45°C. These vials were subsequently
`used for incubation with I ml of acini suspended in TR. CCK was
`concentrated up to sixfold by adsorbing up to 6 ml of plasma through
`a single cartridge and eluting the CCK into a single vial.
`Recoveries of CCK standards were measured by adding known
`amounts of CCK-8 or CCK-33 (dissolved in either saline or 50 mM
`acetic acid containing 5 mg/ml BSA) to plasma from fasting subjects
`or charcoal-stripped plasma. These plasma samples were then processed
`through SEP-PAK cartridges as described above and assayed for CCK(cid:173)
`like activity by comparing the bioactivity of plasma samples with those
`of standard curves of CCK-8 and CCK-33. Concentrations of CCK-8
`and CCK-33 ranging from 10 to 100 fmol, were added to plasma and
`yielded recoveries of 92±8% (mean±SD, n = 14) for CCK-8 and
`85±10% (n = 10) for CCK-33. Samples were usually assayed on the
`day of collection, however, it was found that recovery of plasma CCK
`bioactivity was unchanged if plasma was stored at -20°C in SEP-PAK
`cartridges for up to 10 d.
`Gastrin radioimmunoassays were kindly performed by Dr. Clifford
`Deveney by the method of Stadil and Rehfeld (31}. Antibody 1611,
`previously characterized ( 15), was used in a concentration of I :500,000.
`Gastrin 17 was used as a standard, and the assay had a sensitivity of
`0.1 fmol/ml. The normal range of basal serum gastrin values with this
`assay is 0-50 pM.
`
`lmmunoprecipitation of CCK
`Staphylococcus aureus (0.5 ml of a 10% suspension) was washed and
`resuspended in I ml of incubation buffer. IO µI of either saline, normal
`rabbit serum, or anti-CCK antibody RSB70 were added and incubated
`2 hat 4°C. The mixture was then washed twice, resuspended in TR
`
`Cho/ecystokinin Bioactivity in Plasma
`
`1145
`
`

`

`buffer, incubated 2 hat 4°C with either CCK-8 or plasma extract and
`then centrifuged (32). The supernates were bioassayed for CCK activity.
`In addition, to determine if an inhibitor of CCK was present in
`plasma, these supernates were added to various concentrations of
`CCK-8 or carbamylcholine and incubated with 1 ml of pancreatic
`acini and assayed as described above. Final concentrations of these
`plasma extracts in the incubation were 1-2 ml of plasma/ml of acini.
`
`Column chromatography
`Plasma samples were collected and extracted. The extracted material
`was resuspended in a buffer of 0.25 M ammonium carbonate, pH 8.2,
`with 0.2% BSA at 4°C and chromatographed on a Sephadex G-50
`superfine column, 0.9 X 58 cm. I-ml fractions from the column were
`then passed through separate SEP-PAK cartridges. These samples were
`eluted and bioassayed for CCK activity. Recovery of biological activity
`from plasma extracts subjected to column chromatography averaged
`65% (n = 5). To determine if CCK-8 aggregated in plasma, 200 fmol
`CCK-8 was added to 15 ml of plasma from a fasted subject. This
`enriched plasma was incubated for 10 min at 37°C, extracted through
`SEP-PAK cartridges, chromatographed, and bioassayed.
`
`CCK infusion and gallbladder ultrasonography
`After an overnight (12-15 h) fast five male subjects underwent intra(cid:173)
`venous infusion ofCCK-8.2 CCK-8 (Sincalide) was diluted to appropriate
`concentrations in a total volume of 20 cm3 normal saline. By use of a
`Harvard pump (Harvard Apparatus Co., Inc., S. Natick, MA), CCK
`was infused through an indwelling butterfly catheter in the antecubital
`vein at a rate of 14 pmol/kg per h. The actual infusion rate was
`determined by measuring the CCK concentration of the infusate taken
`from the delivery system. This measurement corrects for losses of
`CCK by adsorption to syringes and intravenous tubing. Blood was
`collected before and during the infusion through another indwelling
`intravenous catheter in the opposite antecubital vein.
`Determinations of gallbladder volumes were made by abdominal
`ultrasonography as described by Everson et al. (38). Longitudinal
`sonograms of the gallbladder were recorded on film with a commercially
`available Advanced Technological Laboratories (Bellvue, WA) real(cid:173)
`time scanner utilizing a 3.5-MH or 5-MH transducer. Scans were
`obtained with subjects in the upright position. Long axis views were
`obtained by manipulating the transducer so that it followed the
`appropriate long axis of the gallbladder and the largest gallbladder
`dimensions at each time were recorded. All of the subjects' gallbladders
`were found in the usual subhepatic position and orientation. No
`gallstones, wall thickening, or other pathology was identified. After
`base-line blood samples were collected and base-line gallbladder sono(cid:173)
`grams obtained, subjects either drank a mixed liquid meal (as described
`above) or were infused with CCK-8. Blood samples for CCK level
`determinations and simultaneous gallbladder sonograms were obtained
`at various times over a 2-h period. Gallbladder volumes were calculated
`by the sum of cylinders method (38).
`This study was approved by the Committee on the Protection of
`Human Subjects of Mt. Zion Hospital and the Committee on Human
`Research of the University of California, San Francisco. Informed
`consent was obtained from each subject.
`
`Statistical analysis
`All values from a single experiment are expressed as the mean± 1 SD
`and values from pooled experiments as the mean± I SEM. Comparison
`
`2. For several reasons CCK-8 was chosen for the infusion studies
`rather than CCK-33. First, CCK-8 is present in human plasma. Second,
`CCK-8 has all of the biological activities ofCCK-33 in gallbladder and
`other tissues; moreover, in the gallbladder CCK-8 has been reported
`to be either equipotent (33, 34) or more potent (35, 36) than CCK-
`33. Third, human CCK-8 and porcine CCK-8 are identical; in contrast,
`human CCK-33 and porcine CCK-33 are different by two amino acids
`(37). Fourth, CCK-8 is available for human use whereas human CCK-
`33 is not available.
`
`of responses were made by analysis of variance with repeated measures
`(39). Post-hoc analysis of the difference between points was carried
`out by means of the Newman-Keuls test. Differences with a P value
`of <0.05 were considered significant.
`
`Results
`
`Specificity, sensitivity, and validation of the pancreatic acini
`bioassay. In the bioassay system employed, a detectable effect
`of CCK-8 was seen at 1 pM and maximal effects were seen at
`100 pM (Fig. 1). Assay sensitivity was defined as the amount
`of CCK-8 that produced a statistically different response in
`amylase release (a response that differed by 2 SD from that
`observed with no hormone). In all assays perf~ed, incubation
`of acini with 1 pM CCK-8 resulted in higher amylase release
`values than acini incubated in the absence of hormone, and
`in 26 of 30 consecutive experiments this difference was statis(cid:173)
`tically significant. In no individual experiment was a given
`plasma CCK concentration calculated unless the value was
`statistically different from basal.
`In this system plasma extracts stimulated amylase release
`from pancreatic acini. The dose-response curve of stimulated
`amylase release with postprandial plasma from subjects fed a
`mixed liquid meal paralleled that of the CCK-8 standards (Fig.
`1). Similar parallelism was seen for standards of CCK-33.
`Plasma extracts were then incubated with dibutyryl cGMP,
`a known antagonist of CCK action ( 40). Dibutyryl cGMP
`suppressed plasma-stimulated amylase release in a dose-depen(cid:173)
`dent manner similar to that which was shown in rat plasma
`(30). The inhibition curves of amylase release by dibutyryl
`cGMP for both the hormones and plasma were parallel. The
`concentrations of dibutyryl cGMP that completely suppressed
`stimulated amylase release were 0.3-1 mM. In addition, a
`different antagonist of CCK, CBZ-tryptophan (41), was incu(cid:173)
`bated with plasma extracts. Similarly to dibutyryl cGMP, CBZ(cid:173)
`tryptophan also inhibited plasma activity and CCK-stimulated
`amylase release in a parallel dose-dependent manner.
`Tritluoroacetic acid used in eluting CCK from SEP-PAK
`cartridges did not interfere with the bioassay. There was no
`difference in either basal or CCK-stimulated amylase release
`
`35
`
`30
`
`25
`
`20
`
`15
`
`10
`
`5
`
`!
`0
`c a,
`!:!
`i
`
`a, = I a, .,
`
`i
`.i
`
`• CCK-8
`o CCK-8 + VIP
`• Plasma (i ml)
`a Plasma I 1 ml) + VIP
`• Plasma 12 ml)
`A Plasma (2 ml) + VIP
`
`•
`
`6
`
`,f
`
`~
`
`Figure 1. CCK bioactivity
`of plasma extracts and the
`effect of exogenous VIP.
`Postprandial plasma ex(cid:173)
`tracts equivalent to I ml
`(o) or 2 ml (t.) of plasma
`were incubated with iso(cid:173)
`lated pancreatic acini and
`the resultant amylase re(cid:173)
`lease was compared with a
`standard curve of CCK-8
`(lower curve). Similar
`amounts of plasma extracts
`or CCK-8 standards were
`incubated with acini to
`which a maximally-stimu(cid:173)
`lating dose of VIP ( I nM)
`had been added (upper
`curve). This resulted in a shift upward of the standard curve and
`plasma samples but did not change the calculated amount of CCK in
`plasma.
`
`or ........ ---~---~
`
`0
`
`1
`
`10
`CCK-8 (pM)
`
`100
`
`1148
`
`R. A. Liddle. 1. D. Goldfine, M. S. Rosen, R. A. Tap/itz, and J. A. Williams
`
`
`
`
`

`

`between control vials and vials to which 1 ml of ethanol:
`trifluoroacetic acid had been added and dried under nitrogen.
`To test the immunoreactivity of the plasma CCK bioactivity,
`both plasma extracts and a CCK-8 standard were incubated
`with either a suspension of S. aureus to which normal rabbit
`serum was added, or a suspension of S. aureus to which
`antibody RSB70 (directed against the carboxyl terminus of
`CCK) had been added. After centrifugation the supernates
`were assayed for CCK biological activity. Incubation of the
`plasma extracts with S. aureus plus anti-CCK antibody (but
`not normal rabbit serum) completely removed CCK bioactivity.
`This result is similar to that which has been previously
`demonstrated for bioassayable CCK in rat plasma (30).
`To investigate the possibility that an inhibitor of CCK(cid:173)
`stimulated amylase release might be present in plasma extracts
`and thereby interfere in the bioassay, plasma extracts were
`treated with anti-CCK antibody RSB70 bound to S. aureus as
`described above. This process removed all CCK bioactivity
`from plasma extracts. These "CCK-free" plasma extracts were
`then incubated with acini to which either CCK or carbamyl(cid:173)
`choline standards were added (Fig. 2). Plasma extracts stripped
`of CCK had no inhibitory effect on either CCK- or carbamyl(cid:173)
`choline-stimulated amylase release.
`Like CCK, muscarinic cholinergic analogues stimulate
`amylase release via the mobiliz.ation of intracellular calcium
`(42). To exclude the possibility that acetylcholine or a similar
`agent in plasma was interfering in this assay, acini were
`incubated with plasma extracts in the presence and absence of
`5 µM atropine. No change in the CCK biological activity of
`the plasma was seen with atropine (Table I).
`In addition, VIP and secretin are two secretagogues that
`either stimulate or potentiate amylase release from the pancreas
`via the generation of cAMP (43). To test for the possibility
`that these hormones may have been present in significant
`concentrations to influence the bioassay of CCK, l nM VIP
`(a maximally stimulating concentration) was added to both
`the plasma extracts and the CCK standards. VIP increased
`amylase release to an equal degree in both plasma extracts
`and CCK-8 standards (Fig. 1). The net effect was a shift in the
`standard curve upward. The slope of the curve was unchanged.
`
`a, -
`
`G)
`
`-
`
`j~
`... -:B 15
`1~
`~ .e,
`
`20
`
`15
`
`10
`
`5
`
`o
`
`A
`
`/ l.! ................
`
`•
`
`.
`
`,,tr
`.K ,•
`~-
`
`/
`/2
`ff'
`
`.-·
`
`B
`
`.
`
`/
`~
`
`I ~~---.----.----,
`f ~ I
`1.0
`0 3
`10
`30
`100
`0.3
`0
`0.1
`Carbamylcholine
`CCK-8 (pM)
`(JJM)
`
`Figure 2. Influence of CCK-free plasma on amylase release from
`pancreatic acini. Plasma extracts, collected 15 min (o) and 60 min
`(a) after feeding, were incubated with 10 µI of antibody RSB70 for 2
`h. S. aureus was then added to precipitate the antibody-antigen
`complex, and supernates were incubated with pancreatic acini con(cid:173)
`taining various concentrations of (A) carbamylcholine (B) CCK-8 or
`standards <• ). Values are the means±SD of triplicate determinations.
`A representative of two experiments is shown.
`
`Table I. Lack of Effects of Atropine and
`VIP on the CCK-like Bioactivity of Plasma
`
`Plasma
`
`Plasma
`plus S µM
`atropine
`
`Plasma
`plus I nM
`VIP
`
`CCK-8 equivalents
`(pM)
`
`12.0±0.8
`
`10.8±1.3
`
`12.6±1.7
`
`Pancreatic acini were incubated with 2 ml of plasma extracts in the
`presence or absence of either 5 µM atropine or I nM VIP. In the
`case of VIP, I nM VIP was added to each of the CCK.-8 standards as
`well as the plasma extracts. The CCK concentration (mean±SD,
`n = 6) in plasma was then calculated from this standard curve. Data
`are compiled from two separate experiments, with triplicate determi(cid:173)
`nations.
`
`Therefore, VIP did not change the calculated amounts of CCK
`present in plasma samples (Table I).
`Because gastrin, in very high concentrations, can stimulate
`amylase release from pancreatic acini, it was important to
`ascertain the possible interference of gastrin in this assay
`system. Fasting blood samples were collected from four patients
`with documented gastrinomas from which serum gastrin and
`plasma CCK levels were determined (Table II). Fasting gastrin
`levels were consistent with the diagnosis of Zollinger-Ellison
`syndrome and ranged from 205 to 724 pM gastrin 17 equiv(cid:173)
`alents. 3 In contrast, fasting CCK levels were normal ranging
`from 0.7 to 1.5 pM CCK-8 equivalents.
`Plasma CCK response to feeding. The plasma CCK re(cid:173)
`sponses to feeding a mixed liquid meal were then studied (Fig.
`3). 12 men and 10 women were fed a liquid meal of 5.6
`ml/kg and plasma CCK levels measured for up to 2 h. Fasting
`levels of CCK averaged 0.9±0.2 pM for the men and were
`slightly greater for the women, averaging 1.2±0.2 pM, but this
`difference was not statistically significant. There was a prompt
`rise in plasma CCK levels to 7.3±1.9 pM in male subjects and
`6.2±0.8 in female subjects within 7 .5-15 min after feeding.
`CCK levels fell within 60 min but remained significantly
`elevated for up to 2 h after feeding (P < 0.05). To determine
`the effect of both volume and osmolality on CCK release,
`control male subjects were fed either water, normal saline, or
`2% Naa (600 mosmol, the same osmolality as the mixed
`meal). There was a slight but not statistically significant
`e~yation in plasma CCK levels 7 .5 min after saline ingestion
`(Fig. 3). Peak postprandial CCK levels after water and 2%
`Naa were 1.6±0.6 pM (n = 3) and 1.5±0.4 pM (n = 3) CCK-
`8 equivalents, respectively. These levels were not statistically
`different from basal (P > 0.2).
`To determine the relative contributions of protein, fat,
`carbohydrate, and amino acids to CCK secretion, plasma levels
`of CCK were measured after the ingestion of 100 g of either
`casein, com oil, glucose, or mixed amino acids (Fig. 4). Plasma
`CCK levels were measured in response to feeding each of the
`separate food components in the same five male subjects on
`different days. Each of the food components stimulated CCK
`release (P < 0.05). Fat, protein, and amino acids were the
`
`3. Gastrin concentrations are often expressed as picograms per milliliter,
`because gastrin-17 has a molecular weight of ~ 2, I 00, I 00 pg/ml of
`gastrin is 48 pM.
`
`Cho/ecystokinin Bioactivitv in Plasma
`
`1147
`
`

`

`8
`
`.l!?
`5j
`1
`e-
`
`2
`
`0
`
`-.
`
`Amino acids
`
`o
`I
`"--6--<>---b
`
`&
`
`1'-Q
`
`..
`6 A.
`1·~,.
`I ~.~l~,.
`Protein l 1 J
`i 6L•~J•
`
`8 4
`2
`
`' ~~~ . . - - : •
`
`Table /I. Plasma CCK Bioactivity in Patients with Gastrinomas
`
`Patient
`
`Serum pstrin-17 equivalents
`
`Plasma CCK-8 equivalents
`
`A
`B
`C
`D
`
`pM
`
`724
`252
`714
`205
`
`pM
`
`0.8
`0.7
`1.0
`1.5
`
`Immunoreactive serum gastrin and bioactive plasma CCK concentra(cid:173)
`tions were determined iii fasting blo<id samples from four subjects
`(A-D) with documented gastrinotnas. Serum gastrin was measured
`with antibody 1611 directed against the midportion of gastriil as de(cid:173)
`scribed in Methods.
`
`most potent stimulants of CCK secretion causing a four- to
`sevenfold increase above fasting CCK concentrations. The
`integrated response for 2 h after feeding each food is shown
`(Table III). Fat, protein, and amino acids caused a greater
`increase in plasma CCK than did glucose, which resulted in a
`small and transient elevation of CCK levels.
`Molecular forms of plasma CCK. To determine the molec(cid:173)
`ular forms of CCK in plasma, postprandial sampies were
`collected from four male and four female subjects 15 min
`after ingesting a mixed liquid meal. 30 ml of plasma were
`concentrated onto SEP-PAK cartridges, chromatographed over
`a Sephadex G-50 superfine column, and compared with CCK-
`33 and CCK-8 standards (Fig. 5). Three peaks of CCK
`bioactivity were detected in all subjects. The first and most
`prominent peak of biologically active material (peak a) had a
`molecular size similar to that ot CCK-33. The smallest and
`least prominent peak (peak c) eluted in a position similar to
`that of CCK-8, while a peak intermediate in size between
`CCK-33 and CCK-8 (peak b) was also identified. The percentage
`of total CCK bioactivity present in each peak is shown (Table
`IV). In all subjects, peak a contained the greatest amount of
`CCK bioactivity. A similar profile but with smaller peaks was
`also observed when I 00 ml of plasma from two fasting subjects
`was chromatographed.
`
`10
`
`o--.---~--~-~-~
`60
`90
`30
`·15
`0
`120
`Minutes
`
`Figure 3. Plasma CCK response to feeding. After an overnight fast 12
`male (e) and 10 female (o) subjects were fed 5.6 ml/kg of a mixed
`liquid meal or normal saline (ll). At the times indicated, plasma was
`collected and extracted. These extracts were then amyed for CCK-8
`bioactivity, expressed as CCK-8 equivalents. Each value is the
`mean±SE (n = 12 men, mixed meal; n = 10 women, mixed meal;
`and n = 4 mean; saline). Postprandial values for the mixed meal in
`both males and females were statistically different from basal
`(P < 0.05). There was no significant elevation in CCK after the
`ingestion of saline.
`
`Glucose
`
`-............A___.!
`
`1· j. ~
`"-•
`II
`I
`I
`90 120 -15 0
`Minutes
`
`o I I
`-15 0
`
`I
`30
`
`I
`30
`
`I
`60
`
`I
`90
`
`I
`120
`
`Figure 4. CCK responses to feeding fat, protein, amino acids, or
`glucose. After an overnight fast, five maie subjects were fed 100 g of
`protein (e, as casein), fat <•, as Lipomul), mixed amino acids (o), or
`
`glucose (ll). Plasma was collected at the times indicated and extracts
`were assayed for CCK-8 bioactivity. On separate days the same five
`subjects drank each of the four food components studied. Each food
`was given as a 25% solution (5-5.7 ml/kg) in a volume of 400 ml.
`Values are the CCK levels of the five subjects (mean±SE).
`
`To determine if the large molecular forms ofCCK resulted
`from aggregation of smaller forms, CCK-8 was incubated with
`fasting plasma and chtomatographed, A single peak of CCK
`bioactivity, eluting in the position of CCK-8 was recovered,
`indicating that the larger molecular forms are not aggregated
`CCK-8 fragments.
`Relation of plasma CCK to gallbladder contraction. To
`establish whether CCK has a physiologic role in regulating
`galll?Iadder contraction, two sets of experiments were performed.
`First, the relationship between plasma CCK levels and gall(cid:173)
`bladder volume, measured ultrasonographically was exa,mined
`(Fig. 6). Simultaneous measurements of plasma CCK and
`gallbladder volumes were obtained in five male subjects before
`and after feeding a mixed liquid meal. Plasma CCK levels rose
`to 5.0±1.2 pM within 15 min, declined by 60 min but
`remained elevated for up to 2 h. Coincidental