Peptides, Vol. 13, pp. 125-128, 1992 Printed in the USA.
0196-9781/92 $5.00 + .00 Copyright© 1992 PergamonPressLtd.
Cholecystokinin, Gastrin and Stress Hormone Responses in Marathon Runners E L I S A B E T H P H I L I P P , * T H O M A S W I L C K E N S , f E L I S A B E T H FRIESS,i" P E T R A PLATTEI" A N D K A R L - M A R T I N P I R K E i-
*Klinikum rechts der Isar, Institute of Clinical Chemistry, W-8000 Munich 80, Ismaninger Strafle 22, Germany and -?Max-Planck-Institutefor Psychiatry, W-8000 Munich 40, Kraepelinstr. 10, Germany R e c e i v e d 16 J u l y 1991 PHILIPP, E., T. WILCKENS, E. FRIESS, P. PLATTE AND K.-M. PIRKE. Cholecystokinin, gastrin and stress hormone responses in marathon runners. PEPTIDES 13(1) 125-128, 1992.--The purpose of this investigation was to determine the influence of long-distance running on the secretion of the gastrointestinal peptide hormones cholecystokinin (CCK) and gastrin. Several known stress hormones, ACTH, cortisol and norepinephrine, were also measured. The hormones were estimated before and after a competitive marathon run of 46.5 km and under control conditions a few weeks later. Except gastrin, all hormones were significantly higher under prerun conditions than under control conditions and were highest after the run. The most marked prerun elevation was in CCK. Therefore, CCK seems to be an important regulation factor in response to anticipatory stress. Cholecystokinin (CCK)
Competitive marathon run
after a competitive marathon run and under control conditions. Control conditions were defined as a different day, independent of the marathon competition state. This was so that we would have a basis for detecting effects of any anticipatory stress.
NUMEROUS investigations have been undertaken to study changes in circulating hormones after severe exercise, such as a marathon run (2,11). The investigations have been attempts to establish the physiological roles of these hormones and the stimuli responsible for their release. Many of these efforts have been made to explain the interrelationship among the anterior pituitary hormones in response to the physiological demands of exercise (5,21,32). It has been shown that adrenocorticotropic hormone (ACTH) and cortisol plasma values are elevated after exercise at maximal and submaximal (12) intensities. In contrast, little is known about the effects of prolonged exercise on gastrointestinal function, which is under hormonal control. Sullivan et al. (31) studied gastrointestinal regulatory peptide responses in long-distance runners and found an increased concentration of gastrin, motilin, somatostatin, pancreatic glucagon, pancreatic polypeptide and vasoactive peptide (VIP). Cholecystokinin (CCK) is a gut hormone that contracts gallbladder, stimulates pancreatic enzyme release (19), delays gastric emptying (13), enhances small bowel motility and potentially mediates satiety (5,18). It plays a crucial role both in the regulation of gastrointestinal digestion and in the control of human feeding behavior, but in addition it seems to influence various facets of pituitary function. In particular, CCK stimulates pituitary growth hormone (GH) and vasopressin secretion (16,17) in vitro and basal pituitary ACTH output and adrenal corticosterone secretion in vivo (24). The objective of the present work was to study the release of CCK before and after a marathon run, which to our knowledge has not yet been done, and to compare the findings with those for the secretion of other hormones already known to be increased after strenuous exercise. Therefore, we measured CCK, gastrin, cortisol, ACTH and norepinephrine (NE) before and
Eleven male and 8 female marathon runners participated in the study. Their ages, body mass indices (BMI = kg/m 2) and exercise times are given in Table 1. One female runner took oral contraceptives.
Design Blood samples were obtained by venipuncture about 1 hour before the competitive run (between 7:00 and 8:00 a.m.), immediately after the run and under control conditions. Control venipuncture was several weeks later without any physical or mental stress (8:00 a.m.). Composition and size of breakfast and time interval to venipuncture were the same on control and competition days. This was to exclude nutrition-dependent changes in CCK and gastrin release. All subjects came directly to venipuncture, without physical exercise and warming up. After collection, the blood was divided into two parts; about two-thirds were placed in chilled ethylendiaminetetraacetic acid (EDTA)-containing tubes and immediately centrifuged. The other third was centrifuged immediately after clotting. After centrifugation, serum and plasma were stored at - 70°C until assayed. Analyses
Plasma CCK concentrations were measured by a specific and sensitive radioimmunoassay (RIA). Synthetic sulfated and un-
PHILIPP ET AL.
TABLE 1 SAMPLECHARACTERISTICS
Age (year) Range Mean BM1 (kg/m2) Range Mean Exercise/week (h) Range Mean
Women (n = 8)
Men (n = tl)
sulfated CCK-8, CCK-21, CCK-5 and CCK-4 were obtained from Bachem, and sulfated and unsulfated gastrin- 17 and gastrin34 were from Sigma. Porcine CCK-33 was obtained from Peninsula Laboratories. Sulfated CCK-8, J25iodine labeled with Bolton and Hunter reagent and the antiserum to CCK were from Amersham Corporation. The antiserum was raised in rabbits using CCK-8-S covalently conjugated to bovine serum albumin. The antiserum is specific for the C-terminal. We found no crossreactivity with nonsulfated compounds. The antiserum binds CCK-33 less well than CCK-8-S (0.59 versus 1.0) and has a cross-reactivity of up to 10% to sulfated gastrins. No cross-reactivity was found to other gastrointestinal peptides (e.g., VIP, somtostatin). Plasma was extracted and concentrated by adsorption onto octadecylsilica cartridges (C-18 Sep-Pak, Waters Associates). CCK was then eluted with 3 ml methanol and the extracts dried down under vacuum using a speed-vac sample concentrator. These plasma extracts were directly dissolved in RIA buffer, consisting of 50 mM sodium phosphate, pH 7.4, containing 0.2% gelatin and 10 mM EDTA. The standard curve ranged from 0.1 fmol CCK-8-S/tube to 50 fmol/tube. The sample, the label solution (3000 cpm) and the antiserum solution were incubated in this buffer at 4°C for at least 24 hours. Free from bound tracer was separated by dextran-charcoal. Normally, a blank of < 5% and a zero binding of 45-55% were obtained. This Sep-Pak procedure gives the total CCK immunoreactivity in plasma obtainable with this RIA system. As far as possible, all procedures were performed at 5°C in siliconized glass tubes. The recovery of CCK-8-S added to hormone-free plasma prior to extraction was 83% at a concentration of 0.6 fmol/ml, 80% at a concentration of 6 fmol/ml, and 77% at a concentration of 10 pmol/ml. The intraassay variability was 8.2% and the interassay variability 12.1% at a concentration of 10 pmol/1. The lowest standard concentration of CCK that could be detected in the solution containing no peptide with a 95% confidence limit was 0.4 pmol/1 (n = 6). In a previous study, plasma extracts were separated by high performance liquid chromatography (HPLC) and collected in 1 ml fractions. These fractions were dried and, subsequently, CCK immunoreactivity was measured in the different fractions, the most important being that of CCK-8.S. We found a strong correlation between CCK immunoreactivity after the Sep-Pak procedure and CCK-8-S immunoreactivity after the HPLC separation (r = .893) (22). Therefore, in the present study we decided to measure CCK immunoreactivity only after the Sep-Pak procedure. Gastrin was measured in serum using a commercially available RIA kit from Becton-Dickinson. The intraasay variability
was 3.7% at a concentration of 102 pg/ml, and the interassay variability was 4.3% at a concentration of 97 pg/ml (n = 8). ACTH was determined in EDTA-plasma by a commercially available RIA kit from Nichols Institute, Diagnostic GmbH. The intraassay variability was 3.0% at a concentration of 35, and the interassay variability was 7.8% at a concentration of 36 pg/ml (n = 10). The cortisol MAIA kit was supplied by Serono Diagnostics. Intraassay variability was 4.6% at a concentration of 98.8 ng/ ml, and the interassay variability was 7.1% at a concentration of 109 ng/ml (n = 15). NE was measured by HPLC as reported earlier (23). Intraassay variability was 4.9% at a concentration of 1.82 pmol/ml, and day to day precision was 5.6 at mean 1.97 pmol/ml (n = 6). Albumin was measured by lasernephelometry (Behring). Within-run variability was 2.4% and between-run variability was 3.2% at a concentration of 4.2 g/dl (n = 6).
Evaluation All results shown in the figures are means _+ SEM. Comparisons of values for a given hormone were made with the Friedman test. When this test revealed significant differences among situations, the Wilcoxon test was used. Spearman's coefficients of rank correlation were calculated. The Mann-Whitney U-test was used for comparisons of independent samples. RESULTS The mean running time for the women was 221 (+ 14) min and for the men 177 (+ 7.5) min. The plasma concentrations of the 5 hormones before (b) and after (a) the run and under control conditions (c) are shown in Fig. l and Fig. 2. The plasma CCK concentration was 4.46 (___0.63) fmol/ml under control conditions and 8.16 (+ 1.1) before the run (p