- Email: [email protected]
Immunoextracted calcitonin in human gastric secretion
Peptides, Vol. 9, pp. 295--299.©PergamonPress plc, 1988. Printed in the U.S.A.
0196-9781/88 $3.00 + .00
Immunoextracted Calcitonin in Human Gastric Secretion E. BUCHT,10. WIStSN,* B. G R A N B E R G A N D H. E. S J O B E R G Departments of Endocrinology and *Medicine, Karolinska Institute and Hospital Stockholm, Sweden R e c e i v e d 24 M a r c h 1987 BUCHT, E., O. WISI~N, B. GRANBERG AND H. E. SJI3BERG. lmmunoextracted calcitonin in human gastric secretion. PEPTIDES 9(2) 295-299, 1988.--Immunoreactive calcitonin (iCT) has been demonstrated in human gastric juice after immunoextraction with immobilized antibodies and subsequent radioimmunoassay. The basal levels were 4.5_+3.1 (mean_+SD) pg-Eq/ml gastric juice; range 1.2-9.1 pg-Eq/mi; n=7, and after stimulatory gastric secretion test with pentagastrin 0.3_+0.2 pg-Eq/ml; range 0.1-0.7 pg-Eq/ml; n=7 (p<0.01). The main fraction ofiCT from gastric juice eluted in the same region as synthetic human calcitonin (hCT) on Sephadex G-75 gel chromatography. Reverse phase chromatography in a fast protein liquid chromatography (FPLC) system revealed a slightly less hydrophobic character of the iCT from gastric juice compared to synthetic monomeric hCT. The results were further confirmed by using an additional antiserum. In plasma, the calcitonin (CT) levels were after immunoextraction at the basal state 6.6_+1.7 pg-Eq/ml (mean_+SD); range 5.1-10. l pg-Eq/ml; n=7 and after pentagastrin stimulation 9.4_+5.4 pg-Eq/ml; range 6.3-18.5 pg-Eq/ml; n=7. Calcitonin Gastric juice Affinity chromatography Fast protein liquid chromatography Reverse phase
Gel chromatography
C A L C I T O N I N (CT) was discovered by C o p p e t al. [7] in 1962 and one year later it was established that CT is secreted by the thyroid gland [14]. This gland has been regarded as the main source o f CT, but immunoreactive CT (iCT) has also been found in extra-thyroidal tissues such as lung, thymus, urinary bladder and the gastrointestinal tract [1]. We have demonstrated the presence of iCT in immunoextracted seminal fluid [3] and in milk and plasma from totally thyroidectomized mothers [4], which indicates extra-thyroidal production. Recently, CT has also been demonstrated in cells in the human gastric mucosa [16]. CT has been shown to interact with somatostatin and gastrin [5]. The latter two peptides are synthesized in the gastric mucosa and are, at least partly, released to the gastric lumen [20]. Thus it seemed of interest to examine whether CT could also be detected in gastric juice.
siduum removed. Ten mi of isotonic saline was instilled into the stomach and the easy recovery of the solution served as a check for the accurate positioning of the gastric tube. Gastric contents were collected continuously by gentle manual suction. Samples were immediately titrated to pH 8 with 1 M NaOH, boiled for one minute and then kept frozen at -20°C until analysis. After four basal 15 min collection periods, gastric acid secretion was stimulated by the subcutaneous injection of pentagastrin (Peptavlon, ICI), 6 ~g/kg and gastric contents were collected for a further four 15 min periods. The gastric juice from the last three basal periods was pooled, as were all of the samples collected during pentagastrin stimulation prior to analysis for iCT. Basal gastric acid output (BAO) was calculated from the acid output during the last two basal test periods and peak acid output (PAO) from the two highest consecutive responses to pentagastrin injection. Venous blood was withdrawn from an antecubital vein and after centrifugation the heparinized plasma was immediately frozen and kept at -20°C.
METHOD
Subjects
Radioimmunoassay
Gastric juice was obtained from seven patients who were referred to gastric secretory testing. Patient data and their gastric secretory responses are given in Table 1. Gastric juice from three healthy volunteers were used for further conformation tests.
iCT was determined in extracts of plasma and gastric secretion with a sensitive radioimmunoassay (RIA) with a detection limit of 0.8 pg/tube. The antiserum (No. 1) was directed against the carboxyterminal portion of CT. Synthetic human CT (hCT) (Peninsula Lab. Inc., San Carlos, CA) was used as a standard and for radiolabelling. F o r details see [2]. To further support that iCT was present, another antiserum (No. 2) was used for RIA in one series of experiments after im-
Procedure After an overnight fast, the subjects were intubated with a 14 F G gastric tube (Portex, England) and the gastric re-
1Requests for reprints should be addressed to Dr. Elisabet Bucht, Department of Endocrinology, Karolinska Hospital, Box 60500, S-104 01 Stockholm, Sweden.
295
296
BUCHT, WISI~N, GRANBERG AND SJOBERG TABLE 1 PATIENT CHARACTERISTICS 0-
Patient
Age
Sex
Diagnosis
1
42
F
2
50
F
3
56
F
Irritable bowel syndrome Irritable bowel syndrome Small-intestine bacterial overgrowth Pernicious anemia Macrocytic anemia Macrocytic anemia Gastric ulcer
4 5 6 7
35 78 42 42
M M F F
BAO PAO mmol/hr mmol/hr 8.2
30.4
1.8
28.5
1.8
23.8
30-
20.
O.
0 1.2 3.8 0
0 7.0 23.9 3.4
10
0 munoextraction with antiserum No. 1. This antiserum was directed against the midportion and carboxyterminal of CT [4]. RIA results are expressed as pg-equivalents (pg-Eq) of hCT, since the assay curves were parallel for iCT in extracts of plasma and gastric juice and synthetic hCT.
Immunoextraction To avoid nonspecific interference of substances present in gastric secretion, all of the samples were immunoextracted with antiserum No. 1 prior to RIA and gel chromatography. CT antibodies were coupled to CNBr-activated Sepharose 4B as earlier described [2]. The immunoextraction was performed at neutral pH at 4°C and CT was eluted with 1 M acetic acid. The extracts were lyophilized and reconstituted in assay buffer (0.05 M phosphate buffer, pH 7.4, with 0.02% BSA, 0.02% sodium azide, and one ampoule of Trasylol/1) before RIA and gel chromatography. The recovery was 80% when synthetic hCT was extracted. No nonspecific binding of gastric juice to Sepharose 4B was observed prior to coupling with CT-antibodies.
o
2'0
G'o Elution volume (mr)
FIG. 1. Gel chromatography on Sephadex G-75 (115×1 cm) of pooled immunoextracted calcitonin from gastric secretion collected initially and after pentagastrin stimulation. Markers: Vo=dextran blue, I=ovalbumin, II=cytochrome C, Ill=synthetic human calcitonin and V~=sodium chloride. The detection limit of the assay is indicated by the broken line.
enced by boiling for 5 min. Incubation for 24 hr at 4°C of ~25I-CT in gastric juice boiled for 1 rain showed no enzymatic degradation of the tracer regarding elution position on Sephadex G-50. The addition of 200 pg hCT to gastric secretion before alkalisation, boiling and freezing resulted in 77_+4% (n=4) recovery.
Statistical Methods Student's t-test [17] for paired data was used for evaluating differences between basal and stimulated levels. p-Levels <0.05 were considered significant.
Gel Chromatography CT extracts from gastric juice were gel chromatographed on a Sephadex G-75 column (115× 1 cm) developed with 0.1 M ammonium acetate, pH 6.8, at 4°C. The 2-3 ml fractions were lyophilized and reconstituted in 0.3 ml assay buffer. RIA with antiserum 1 was performed in duplicate for every fraction.
Reverse Phase FPLC Immunoextracted CT from gastric juice collected at basal conditions from three healthy females was chromatographed on a reverse phase Pep RPC, HR 10/10 column in a fast protein liquid chromatography (FPLC) system (Pharmacia, Sweden). The chromatogram was developed with a linear gradient of 0.1% trifluoracetic acid (Fluka AG) in water/acetonitril from 0% to 80% acetonitril (Merck), flow rate: 2 ml/min. Two ml fractions were lyophilized and reconstituted in assay buffer before RIA. In these experiments RIA was performed with both antiserum 1 and 2.
Recovery Experiments The immunoreactivity of synthetic hCT was not influ-
iGo
RESULTS
iCT in Gastric Secretion The levels at the basal state after immunoextraction were 4.5_+3.1 pg-Eq/ml (mean-+SD) range 1.2-9.1 pg-Eq/ml, n=7) and after pentagastrin stimulation of the gastric juice 0.3_+0.2 pg-Eq/ml (range 0.1-0.7 pg-Eq/ml, n=7) (Table 2). All patients bad lower concentrations of iCT in stimulated than in basal gastric secretion (p<0.01).
Gel Chromatography Three chromatograms of immunoextracted CT from gastric secretion were performed on Sephadex G-75. All chromatograms had the same pattern. Figure 1 shows the elution profile of a pool of 6 extracts from basal and stimulated gastric secretion. The main peak of the iCT eluted at the same position as synthetic hCT.
FPLC Reverse phase FPLC chromatography demonstrated that the iCT from gastric juice eluted as a slightly less hydrophobic material than monomeric hCT. The results
CALCITONIN IN GASTRIC JUICE
297
TABLE 2 DETERMINATIONOF IMMUNOEXTRACTEDCALCITONININ PLASMAAND GASTRICJUICE AT THE BASAL STATEAND AFTER PENTAGASTRINSTIMULATION iCT in Plasma pg-Eq/ml Patient 1 2 3 4 5 6 7
iCT in Gastric Juice pg-Eq/ml
iCT in Gastric Juice pg-Eq/hour
Basally
After Stimulation
Basally
After Stimulation
Basally
After Stimulation
7.1 6.6 5.7 5.4 10.1 6.2 5.1
4.5 15.5 6.3 6.2 18.5 8.1 6.6
8.6 3.8 1.8 1.2 4.0 9.1 2.9
0.2 0.3 0.1 0.5 0.7 0.2 0.4
366 116 86 85 106 424 77
26 50 18 33 36 28 22
were verified by use of two different antisera. As in earlier studies [4], antiserum No. 2 gave lower levels than No. 1, but the same pattern was achieved (Fig. 2). iCT in Plasma After immunoextraction the plasma levels of iCT were at the basal state 6.6_+ 1.7 pg-Eq/ml (mean_SD) (range 5.1-10.1 pg-Eq/ml, n=7), and after pentagastrin injection 9.4_+5.4 pg-Eq/ml (range 6.5-18.5, n=7) (Table 2). There was an increase of plasma iCT after pentagastrin stimulation in 6 out of 7 subjects (p>0.05). In this heterogenous group of patients, no correlation between plasma iCT and iCT from gastric juice was found.
12s] iCT "80%
,-, 30. to O
DISCUSSION By the use of immunoextraction and a sensitive RIA we have demonstrated iCT in gastric juice in man. Immediate neutralization and boiling of the samples were performed to prevent degradation of CT. Direct RIA measurements of gastric fluids have been questioned. High values of hormones have been reported, probably due to extensive proteolytic damage to labelled antigen and antibody in the commonly used double-antibody and charcoal separation techniques [18]. In the present study, the immunoextraction with immobilized CT antibodies served both as a purification and a concentration step. Further purification and characterization by gel chromatography of iCT from immunoextracted gastric juice showed that the major peak of iCT eluted in the region of the monomer, which is regarded as the biologically active form [11]. On reverse phase FPLC the gastric iCT eluted somewhat earlier than the monomeric CT. Less hydrophobic iCT compounds have also been demonstrated in plasma [19]. However, further studies have to be performed before the character of the gastric iCT can be described. In plasma, the pattern is more heterogenous. Plasma iCT from healthy males seems to mainly consist of monomer-like CT, the dimer and a large form which may correspond to procalcitonin [2,19]. The presence of mainly monomer-like calcitonin in gastric juice might be explained by proteolytic degradation of larger forms of CT. Extra-thyroidal biosynthesis of CT has been demonstrated in lung carcinoma [8]. In the prostatic gland a large number of epithelial cells containing iCT have been found [9] and high concentrations of iCT are present in seminal fluid
o
t-
13 rl
O"
po.
o D
t 20'
o.
"40"/.
Io
10.
0.4--z 0
~ 25
,0% 50 75 E l u t i o n v o l u m e (mL)
100
FIG. 2. Reverse phase chromatography on a Pep RPC, HR I0/10 column in a FPLC system. Immunoextracted calcitonin from gastric secretion collected at basal conditions was lyophilized and reconstituted in A. The column was developed with a linear gradient from A to B. A=Water with 0.1% trifluoracetic acid (TFA), B=80% acetonitril with 0.1% TFA. Flow rate 2 ml/min. The elution positions of synthetic monomeric CT (A), immunoextracted CT from gastric juice measured with antiserum 1 (O) and antiserum 2 (@). The elution position of 1~5I-CTis indicated by the arrow. The detection limits of the assay with the two different antisera are indicated by broken lines.
298
BUCHT, WISt~N, GRANBERG AND SJOBERG
[3]. The demonstration of large amounts of iCT in milk from healthy as well as thyroidectomized mothers is further evidence for an extra-thyroidal production of CT [4]. Recently, the presence of iCT in a small number of cells in the gastric antral mucosa of man has been reported [16] and a role for CT as a paracrine messenger was suggested. Studies on paracrine interactions in the gastric mucosa have so far concentrated on gastrin and somatostatin, which are produced by antral G- and D-cells, respectively. These hormones are probably released into the surrounding interstitial tissue and have been shown to reach the antral lumen, as well as the blood [10,20]. At pH 1-2, a relative predominance in the release of somatostatin as compared to gastrin has been observed [20]. It has also been shown that the inhibitory action of somatostatin on gastric acid secretion is at least partly due to the inhibition of gastrin release [20]. CT is known to exhibit multiple actions on the gastrointestinal system and interestingly, seems to interact with both gastrin and somatostatin. The local application of CT in the stomach inhibits gastric secretion [21] and this effect could be mediated by somatostatin, since eel-calcitonin has been shown to stimulate gastric somatostatin release [5]. Our finding of decreased iCT in gastric juice after pentagastrin infusion, concomitantly with the same or at most occasions increased plasma iCT, is difficult to understand. However, it has been demonstrated that gastrin releases somatostatin in isolated perfused pancreas [15] and intravenous somatostatin has an inhibitory action on CT secretion in man [12]. In comparison with other tissues the gastric mucosa is rich in somatostatin. It can therefore be contemplated that, in analogy with the pancreas, an increased level of somatosta-
tin should lead to a decrease in the release of calcitonin. An increased proteolytic degradation due to the pentagastrin induced secretion of acid and pepsin could also explain the lower levels of iCT in stimulated gastric secretion. Gastrin is a well recognized releaser of calcitonin [6]. In the present study the levels of plasma iCT were increased after pentagastrin injection in all patients except one. The difference was not statistically significant (p>0.05) probably because of the low dose of pentagastrin used and the low number of patients. When doses which clearly stimulate plasma iCT release are used, adverse reactions have been observed in the patients [13]. The relatively low concentrations of iCT found in gastric juice correspond well with the small number of iCT containing cells found in gastric mucosa. The ratio between iCT containing cells and G-cells is about 1:50-100 [16]. The low concentration ofiCT in gastric juice does not rule out a physiological importance for the peptide, as local concentrations at a presumptive action site in gastric mucosa could be much higher. Even if it is tempting to suggest a physiologically important secretion of CT directly from the gastric mucosa, the plasma could also be the source. Further studies have to be done to establish the origin of the gastric CT. In conclusion, immunoreactive calcitonin with a molecular size similar to the monomer has been demonstrated in human gastric juice by means of immunoextraction and gel chromatography. Reverse phase FPLC reveals that the iCT is slightly less hydrophobic than synthetic human calcitonin. The physiological significance of the finding, as well as the decrease in iCT concentration in gastric juice after the pentagastrin test, remains to be evaluated.
ACKNOWLEDGEMENTS We thank Professor Hans Low for valuable discussions, Sylvie Karlsson and Katarina Breitholtz for typing the manuscript and UIrika Ernsten for drawing the figures. Financial support was obtained from the Swedish Medical Research Council (grant No. 5992), the Funds of the Karolinska Institute, the Magn. Bergvall Foundation, King Gustav V's and Queen Victoria's Research Foundation and the Royal Academy of Science (Amundsons Fund).
REFERENCES
1. Becker, K. L., R. H. Snider, C. F. Moore, K. G. Monaghan and O. L. Silva. Calcitonin in extrathyroidal tissues of man. Acta Endocrinol (Copenh) 92: 746-751, 1979. 2. Bucht, E., O. Tarring and H. E. Sjrberg. Gel chromatography of immunoextracted plasma calcitonin in response to the calcium clamp in healthy males. Acta Endocrinol (Copenh) 110: 421-428, 1985. 3. Bucht, E., S. Arver and H. E. Sjoberg. Immunoextracted calcitonin from human seminal plasma constitutes different forms. Acta Physiol Scand 126: 289-293, 1986. 4. Bucht, E., M. Telenius-Berg, G. Lundell and H. E. Sjoberg. Immunoextracted calcitonin in milk and plasma from totally thyroidectomized women. Evidence of monomeric calcitonin in plasma during pregnancy and lactation. Acta Endocrinol (Copenh) 113: 529-535, 1986. 5. Chiba, T., T. Taminato, S. Kadowaki, Y. Goto, K. Mori, Y. Seino, H. Abe, K. Chihara, S. Matsukura, T. Fujita and T. Kondo. Effects of [AsuLI-eel calcitonin on gastric somatostatin and gastrin release. Gut 21: 94-97, 1980. 6. Cooper, C. W., W. H. Schwesinger, A. M. Marhgoub and D. A. Ontjes. Thyrocalcitonin: stimulation of secretion by pentagastrin. Science 172: 1238-1240, 1971.
7. Copp, D. H., E. C. Cameron, B. A. Cheney, A. G. F. Davidson and K. G. Henze. Evidence for calcitonin--A new hormone from the parathyroid that lowers blood calcium. Endocrinology 70: 638--649, 1962. 8. Edbrooke, M. R., D. Parker, J. H. McVey, J. H. Riley, G. D. Sorenson, O. S. Pettengill and R. K. Craig. Expression of the human calcitonin/CGRP gene in lung and thyroid carcinoma. EMBO J 4: 715-724, 1985. 9. FetissoL F., G. Bertrand, D. Guilloteau, M. P. Dubois, Y. Lanson and B. Arbeille. Calcitonin immunoreactive cells in prostate gland and cloacal derived tissues. Virchows Arch 409: 523-533. 1986. I0. Fiddian-Green, R. G., J. Farrell, D. Havlichek, P. Kothary and G. Pittenger. A physiological role for luminal gastrin? Surgeo' 83: 663-668, 1978. 11. Goltzman, D. and A. S. Tischler. Characterization of the immunochemical forms of calcitonin released by a medullary thyroid carcinoma in tissue culture. J Clin Invest 61: 449-458, 1978. 12. Gordin, A., B-A. Lamberg, R. Pelkonen and S. Almqvist. Somatostatin inhibits the pentagastrin-induced release of serum calcitonin in medullary carcinoma of the thyroid. Clin Endocrinol (OxJ) 8: 289-293, 1978.
C A L C I T O N I N IN G A S T R I C J U I C E
13. Heynen, G., A. Brassine, J. C. Daubresse, G, Ligny, J. A. Kanis, S. Gaspar and P. Franchimont. Lack of clinical and physiological relationship between gastrin and calcitonin in man. Eur J Clin Invest 11: 331-335, 1981. 14. Hirsch, P. F., G. F. Gauthier and P. L. Munson. Thyroid hypocalcemic principle and recurrent laryngeal nerve injury as factors affecting the response to parathyroidectomy in rats. Endocrinology 73: 244--252, 1963. 15. Ipp, E., R. E. Dobbs, A. Arimura, W. Vale, V. Harris and R. H. Unger. The effect of gastrin, gastric inhibitory polypeptide, secretin and the octapeptide of cholecystokinin upon immunoreactive somatostatin release by the perfused pancreas. J Clin Invest 60: 1216--1219, 1977. 16. Ito, H., J. Hato, H. Yokozaki and E. Tahara. Calcitonin in human gastric mucosa and carcinoma. J Cancer Res Clin Oncol 112: 50-56, 1986.
299
17. Snedecor, G. W. and W. G. Cohran. Statistical Methods, 7th edition. Ames, IA: The Iowa State University Press, 1980. 18. Straus, E. and R. S. Yalow. Artifacts in the radioimmunoassay of peptide hormones in gastric and duodenal secretions. J Lab Clin Med 87: 292-298, 1976. 19. Tobler, P. H., F. A. Tschopp, M. A. Dambacher, W. Born and J. A. Fischer. Identification and characterization of calcitonin forms in plasma and urine of normal subjects and medullary carcinoma patients. J Clin Endocrinol Metab 57: 74%754, 1983. 20. Uvn~is-Wallensten, K., S. Efendic and R. Luft. Vagal release of somatostatin into the antral lumen of cats. Acta Physiol Scand 99: 126--128, 1977. 21. Ziegler, R., H. Minne, J. Hotz and H. Goebell. Inhibition of gastric secretion in man by oral administration of calcitonin. Digestion 11: 157-160, 1974.
0196-9781/88 $3.00 + .00
Immunoextracted Calcitonin in Human Gastric Secretion E. BUCHT,10. WIStSN,* B. G R A N B E R G A N D H. E. S J O B E R G Departments of Endocrinology and *Medicine, Karolinska Institute and Hospital Stockholm, Sweden R e c e i v e d 24 M a r c h 1987 BUCHT, E., O. WISI~N, B. GRANBERG AND H. E. SJI3BERG. lmmunoextracted calcitonin in human gastric secretion. PEPTIDES 9(2) 295-299, 1988.--Immunoreactive calcitonin (iCT) has been demonstrated in human gastric juice after immunoextraction with immobilized antibodies and subsequent radioimmunoassay. The basal levels were 4.5_+3.1 (mean_+SD) pg-Eq/ml gastric juice; range 1.2-9.1 pg-Eq/mi; n=7, and after stimulatory gastric secretion test with pentagastrin 0.3_+0.2 pg-Eq/ml; range 0.1-0.7 pg-Eq/ml; n=7 (p<0.01). The main fraction ofiCT from gastric juice eluted in the same region as synthetic human calcitonin (hCT) on Sephadex G-75 gel chromatography. Reverse phase chromatography in a fast protein liquid chromatography (FPLC) system revealed a slightly less hydrophobic character of the iCT from gastric juice compared to synthetic monomeric hCT. The results were further confirmed by using an additional antiserum. In plasma, the calcitonin (CT) levels were after immunoextraction at the basal state 6.6_+1.7 pg-Eq/ml (mean_+SD); range 5.1-10. l pg-Eq/ml; n=7 and after pentagastrin stimulation 9.4_+5.4 pg-Eq/ml; range 6.3-18.5 pg-Eq/ml; n=7. Calcitonin Gastric juice Affinity chromatography Fast protein liquid chromatography Reverse phase
Gel chromatography
C A L C I T O N I N (CT) was discovered by C o p p e t al. [7] in 1962 and one year later it was established that CT is secreted by the thyroid gland [14]. This gland has been regarded as the main source o f CT, but immunoreactive CT (iCT) has also been found in extra-thyroidal tissues such as lung, thymus, urinary bladder and the gastrointestinal tract [1]. We have demonstrated the presence of iCT in immunoextracted seminal fluid [3] and in milk and plasma from totally thyroidectomized mothers [4], which indicates extra-thyroidal production. Recently, CT has also been demonstrated in cells in the human gastric mucosa [16]. CT has been shown to interact with somatostatin and gastrin [5]. The latter two peptides are synthesized in the gastric mucosa and are, at least partly, released to the gastric lumen [20]. Thus it seemed of interest to examine whether CT could also be detected in gastric juice.
siduum removed. Ten mi of isotonic saline was instilled into the stomach and the easy recovery of the solution served as a check for the accurate positioning of the gastric tube. Gastric contents were collected continuously by gentle manual suction. Samples were immediately titrated to pH 8 with 1 M NaOH, boiled for one minute and then kept frozen at -20°C until analysis. After four basal 15 min collection periods, gastric acid secretion was stimulated by the subcutaneous injection of pentagastrin (Peptavlon, ICI), 6 ~g/kg and gastric contents were collected for a further four 15 min periods. The gastric juice from the last three basal periods was pooled, as were all of the samples collected during pentagastrin stimulation prior to analysis for iCT. Basal gastric acid output (BAO) was calculated from the acid output during the last two basal test periods and peak acid output (PAO) from the two highest consecutive responses to pentagastrin injection. Venous blood was withdrawn from an antecubital vein and after centrifugation the heparinized plasma was immediately frozen and kept at -20°C.
METHOD
Subjects
Radioimmunoassay
Gastric juice was obtained from seven patients who were referred to gastric secretory testing. Patient data and their gastric secretory responses are given in Table 1. Gastric juice from three healthy volunteers were used for further conformation tests.
iCT was determined in extracts of plasma and gastric secretion with a sensitive radioimmunoassay (RIA) with a detection limit of 0.8 pg/tube. The antiserum (No. 1) was directed against the carboxyterminal portion of CT. Synthetic human CT (hCT) (Peninsula Lab. Inc., San Carlos, CA) was used as a standard and for radiolabelling. F o r details see [2]. To further support that iCT was present, another antiserum (No. 2) was used for RIA in one series of experiments after im-
Procedure After an overnight fast, the subjects were intubated with a 14 F G gastric tube (Portex, England) and the gastric re-
1Requests for reprints should be addressed to Dr. Elisabet Bucht, Department of Endocrinology, Karolinska Hospital, Box 60500, S-104 01 Stockholm, Sweden.
295
296
BUCHT, WISI~N, GRANBERG AND SJOBERG TABLE 1 PATIENT CHARACTERISTICS 0-
Patient
Age
Sex
Diagnosis
1
42
F
2
50
F
3
56
F
Irritable bowel syndrome Irritable bowel syndrome Small-intestine bacterial overgrowth Pernicious anemia Macrocytic anemia Macrocytic anemia Gastric ulcer
4 5 6 7
35 78 42 42
M M F F
BAO PAO mmol/hr mmol/hr 8.2
30.4
1.8
28.5
1.8
23.8
30-
20.
O.
0 1.2 3.8 0
0 7.0 23.9 3.4
10
0 munoextraction with antiserum No. 1. This antiserum was directed against the midportion and carboxyterminal of CT [4]. RIA results are expressed as pg-equivalents (pg-Eq) of hCT, since the assay curves were parallel for iCT in extracts of plasma and gastric juice and synthetic hCT.
Immunoextraction To avoid nonspecific interference of substances present in gastric secretion, all of the samples were immunoextracted with antiserum No. 1 prior to RIA and gel chromatography. CT antibodies were coupled to CNBr-activated Sepharose 4B as earlier described [2]. The immunoextraction was performed at neutral pH at 4°C and CT was eluted with 1 M acetic acid. The extracts were lyophilized and reconstituted in assay buffer (0.05 M phosphate buffer, pH 7.4, with 0.02% BSA, 0.02% sodium azide, and one ampoule of Trasylol/1) before RIA and gel chromatography. The recovery was 80% when synthetic hCT was extracted. No nonspecific binding of gastric juice to Sepharose 4B was observed prior to coupling with CT-antibodies.
o
2'0
G'o Elution volume (mr)
FIG. 1. Gel chromatography on Sephadex G-75 (115×1 cm) of pooled immunoextracted calcitonin from gastric secretion collected initially and after pentagastrin stimulation. Markers: Vo=dextran blue, I=ovalbumin, II=cytochrome C, Ill=synthetic human calcitonin and V~=sodium chloride. The detection limit of the assay is indicated by the broken line.
enced by boiling for 5 min. Incubation for 24 hr at 4°C of ~25I-CT in gastric juice boiled for 1 rain showed no enzymatic degradation of the tracer regarding elution position on Sephadex G-50. The addition of 200 pg hCT to gastric secretion before alkalisation, boiling and freezing resulted in 77_+4% (n=4) recovery.
Statistical Methods Student's t-test [17] for paired data was used for evaluating differences between basal and stimulated levels. p-Levels <0.05 were considered significant.
Gel Chromatography CT extracts from gastric juice were gel chromatographed on a Sephadex G-75 column (115× 1 cm) developed with 0.1 M ammonium acetate, pH 6.8, at 4°C. The 2-3 ml fractions were lyophilized and reconstituted in 0.3 ml assay buffer. RIA with antiserum 1 was performed in duplicate for every fraction.
Reverse Phase FPLC Immunoextracted CT from gastric juice collected at basal conditions from three healthy females was chromatographed on a reverse phase Pep RPC, HR 10/10 column in a fast protein liquid chromatography (FPLC) system (Pharmacia, Sweden). The chromatogram was developed with a linear gradient of 0.1% trifluoracetic acid (Fluka AG) in water/acetonitril from 0% to 80% acetonitril (Merck), flow rate: 2 ml/min. Two ml fractions were lyophilized and reconstituted in assay buffer before RIA. In these experiments RIA was performed with both antiserum 1 and 2.
Recovery Experiments The immunoreactivity of synthetic hCT was not influ-
iGo
RESULTS
iCT in Gastric Secretion The levels at the basal state after immunoextraction were 4.5_+3.1 pg-Eq/ml (mean-+SD) range 1.2-9.1 pg-Eq/ml, n=7) and after pentagastrin stimulation of the gastric juice 0.3_+0.2 pg-Eq/ml (range 0.1-0.7 pg-Eq/ml, n=7) (Table 2). All patients bad lower concentrations of iCT in stimulated than in basal gastric secretion (p<0.01).
Gel Chromatography Three chromatograms of immunoextracted CT from gastric secretion were performed on Sephadex G-75. All chromatograms had the same pattern. Figure 1 shows the elution profile of a pool of 6 extracts from basal and stimulated gastric secretion. The main peak of the iCT eluted at the same position as synthetic hCT.
FPLC Reverse phase FPLC chromatography demonstrated that the iCT from gastric juice eluted as a slightly less hydrophobic material than monomeric hCT. The results
CALCITONIN IN GASTRIC JUICE
297
TABLE 2 DETERMINATIONOF IMMUNOEXTRACTEDCALCITONININ PLASMAAND GASTRICJUICE AT THE BASAL STATEAND AFTER PENTAGASTRINSTIMULATION iCT in Plasma pg-Eq/ml Patient 1 2 3 4 5 6 7
iCT in Gastric Juice pg-Eq/ml
iCT in Gastric Juice pg-Eq/hour
Basally
After Stimulation
Basally
After Stimulation
Basally
After Stimulation
7.1 6.6 5.7 5.4 10.1 6.2 5.1
4.5 15.5 6.3 6.2 18.5 8.1 6.6
8.6 3.8 1.8 1.2 4.0 9.1 2.9
0.2 0.3 0.1 0.5 0.7 0.2 0.4
366 116 86 85 106 424 77
26 50 18 33 36 28 22
were verified by use of two different antisera. As in earlier studies [4], antiserum No. 2 gave lower levels than No. 1, but the same pattern was achieved (Fig. 2). iCT in Plasma After immunoextraction the plasma levels of iCT were at the basal state 6.6_+ 1.7 pg-Eq/ml (mean_SD) (range 5.1-10.1 pg-Eq/ml, n=7), and after pentagastrin injection 9.4_+5.4 pg-Eq/ml (range 6.5-18.5, n=7) (Table 2). There was an increase of plasma iCT after pentagastrin stimulation in 6 out of 7 subjects (p>0.05). In this heterogenous group of patients, no correlation between plasma iCT and iCT from gastric juice was found.
12s] iCT "80%
,-, 30. to O
DISCUSSION By the use of immunoextraction and a sensitive RIA we have demonstrated iCT in gastric juice in man. Immediate neutralization and boiling of the samples were performed to prevent degradation of CT. Direct RIA measurements of gastric fluids have been questioned. High values of hormones have been reported, probably due to extensive proteolytic damage to labelled antigen and antibody in the commonly used double-antibody and charcoal separation techniques [18]. In the present study, the immunoextraction with immobilized CT antibodies served both as a purification and a concentration step. Further purification and characterization by gel chromatography of iCT from immunoextracted gastric juice showed that the major peak of iCT eluted in the region of the monomer, which is regarded as the biologically active form [11]. On reverse phase FPLC the gastric iCT eluted somewhat earlier than the monomeric CT. Less hydrophobic iCT compounds have also been demonstrated in plasma [19]. However, further studies have to be performed before the character of the gastric iCT can be described. In plasma, the pattern is more heterogenous. Plasma iCT from healthy males seems to mainly consist of monomer-like CT, the dimer and a large form which may correspond to procalcitonin [2,19]. The presence of mainly monomer-like calcitonin in gastric juice might be explained by proteolytic degradation of larger forms of CT. Extra-thyroidal biosynthesis of CT has been demonstrated in lung carcinoma [8]. In the prostatic gland a large number of epithelial cells containing iCT have been found [9] and high concentrations of iCT are present in seminal fluid
o
t-
13 rl
O"
po.
o D
t 20'
o.
"40"/.
Io
10.
0.4--z 0
~ 25
,0% 50 75 E l u t i o n v o l u m e (mL)
100
FIG. 2. Reverse phase chromatography on a Pep RPC, HR I0/10 column in a FPLC system. Immunoextracted calcitonin from gastric secretion collected at basal conditions was lyophilized and reconstituted in A. The column was developed with a linear gradient from A to B. A=Water with 0.1% trifluoracetic acid (TFA), B=80% acetonitril with 0.1% TFA. Flow rate 2 ml/min. The elution positions of synthetic monomeric CT (A), immunoextracted CT from gastric juice measured with antiserum 1 (O) and antiserum 2 (@). The elution position of 1~5I-CTis indicated by the arrow. The detection limits of the assay with the two different antisera are indicated by broken lines.
298
BUCHT, WISt~N, GRANBERG AND SJOBERG
[3]. The demonstration of large amounts of iCT in milk from healthy as well as thyroidectomized mothers is further evidence for an extra-thyroidal production of CT [4]. Recently, the presence of iCT in a small number of cells in the gastric antral mucosa of man has been reported [16] and a role for CT as a paracrine messenger was suggested. Studies on paracrine interactions in the gastric mucosa have so far concentrated on gastrin and somatostatin, which are produced by antral G- and D-cells, respectively. These hormones are probably released into the surrounding interstitial tissue and have been shown to reach the antral lumen, as well as the blood [10,20]. At pH 1-2, a relative predominance in the release of somatostatin as compared to gastrin has been observed [20]. It has also been shown that the inhibitory action of somatostatin on gastric acid secretion is at least partly due to the inhibition of gastrin release [20]. CT is known to exhibit multiple actions on the gastrointestinal system and interestingly, seems to interact with both gastrin and somatostatin. The local application of CT in the stomach inhibits gastric secretion [21] and this effect could be mediated by somatostatin, since eel-calcitonin has been shown to stimulate gastric somatostatin release [5]. Our finding of decreased iCT in gastric juice after pentagastrin infusion, concomitantly with the same or at most occasions increased plasma iCT, is difficult to understand. However, it has been demonstrated that gastrin releases somatostatin in isolated perfused pancreas [15] and intravenous somatostatin has an inhibitory action on CT secretion in man [12]. In comparison with other tissues the gastric mucosa is rich in somatostatin. It can therefore be contemplated that, in analogy with the pancreas, an increased level of somatosta-
tin should lead to a decrease in the release of calcitonin. An increased proteolytic degradation due to the pentagastrin induced secretion of acid and pepsin could also explain the lower levels of iCT in stimulated gastric secretion. Gastrin is a well recognized releaser of calcitonin [6]. In the present study the levels of plasma iCT were increased after pentagastrin injection in all patients except one. The difference was not statistically significant (p>0.05) probably because of the low dose of pentagastrin used and the low number of patients. When doses which clearly stimulate plasma iCT release are used, adverse reactions have been observed in the patients [13]. The relatively low concentrations of iCT found in gastric juice correspond well with the small number of iCT containing cells found in gastric mucosa. The ratio between iCT containing cells and G-cells is about 1:50-100 [16]. The low concentration ofiCT in gastric juice does not rule out a physiological importance for the peptide, as local concentrations at a presumptive action site in gastric mucosa could be much higher. Even if it is tempting to suggest a physiologically important secretion of CT directly from the gastric mucosa, the plasma could also be the source. Further studies have to be done to establish the origin of the gastric CT. In conclusion, immunoreactive calcitonin with a molecular size similar to the monomer has been demonstrated in human gastric juice by means of immunoextraction and gel chromatography. Reverse phase FPLC reveals that the iCT is slightly less hydrophobic than synthetic human calcitonin. The physiological significance of the finding, as well as the decrease in iCT concentration in gastric juice after the pentagastrin test, remains to be evaluated.
ACKNOWLEDGEMENTS We thank Professor Hans Low for valuable discussions, Sylvie Karlsson and Katarina Breitholtz for typing the manuscript and UIrika Ernsten for drawing the figures. Financial support was obtained from the Swedish Medical Research Council (grant No. 5992), the Funds of the Karolinska Institute, the Magn. Bergvall Foundation, King Gustav V's and Queen Victoria's Research Foundation and the Royal Academy of Science (Amundsons Fund).
REFERENCES
1. Becker, K. L., R. H. Snider, C. F. Moore, K. G. Monaghan and O. L. Silva. Calcitonin in extrathyroidal tissues of man. Acta Endocrinol (Copenh) 92: 746-751, 1979. 2. Bucht, E., O. Tarring and H. E. Sjrberg. Gel chromatography of immunoextracted plasma calcitonin in response to the calcium clamp in healthy males. Acta Endocrinol (Copenh) 110: 421-428, 1985. 3. Bucht, E., S. Arver and H. E. Sjoberg. Immunoextracted calcitonin from human seminal plasma constitutes different forms. Acta Physiol Scand 126: 289-293, 1986. 4. Bucht, E., M. Telenius-Berg, G. Lundell and H. E. Sjoberg. Immunoextracted calcitonin in milk and plasma from totally thyroidectomized women. Evidence of monomeric calcitonin in plasma during pregnancy and lactation. Acta Endocrinol (Copenh) 113: 529-535, 1986. 5. Chiba, T., T. Taminato, S. Kadowaki, Y. Goto, K. Mori, Y. Seino, H. Abe, K. Chihara, S. Matsukura, T. Fujita and T. Kondo. Effects of [AsuLI-eel calcitonin on gastric somatostatin and gastrin release. Gut 21: 94-97, 1980. 6. Cooper, C. W., W. H. Schwesinger, A. M. Marhgoub and D. A. Ontjes. Thyrocalcitonin: stimulation of secretion by pentagastrin. Science 172: 1238-1240, 1971.
7. Copp, D. H., E. C. Cameron, B. A. Cheney, A. G. F. Davidson and K. G. Henze. Evidence for calcitonin--A new hormone from the parathyroid that lowers blood calcium. Endocrinology 70: 638--649, 1962. 8. Edbrooke, M. R., D. Parker, J. H. McVey, J. H. Riley, G. D. Sorenson, O. S. Pettengill and R. K. Craig. Expression of the human calcitonin/CGRP gene in lung and thyroid carcinoma. EMBO J 4: 715-724, 1985. 9. FetissoL F., G. Bertrand, D. Guilloteau, M. P. Dubois, Y. Lanson and B. Arbeille. Calcitonin immunoreactive cells in prostate gland and cloacal derived tissues. Virchows Arch 409: 523-533. 1986. I0. Fiddian-Green, R. G., J. Farrell, D. Havlichek, P. Kothary and G. Pittenger. A physiological role for luminal gastrin? Surgeo' 83: 663-668, 1978. 11. Goltzman, D. and A. S. Tischler. Characterization of the immunochemical forms of calcitonin released by a medullary thyroid carcinoma in tissue culture. J Clin Invest 61: 449-458, 1978. 12. Gordin, A., B-A. Lamberg, R. Pelkonen and S. Almqvist. Somatostatin inhibits the pentagastrin-induced release of serum calcitonin in medullary carcinoma of the thyroid. Clin Endocrinol (OxJ) 8: 289-293, 1978.
C A L C I T O N I N IN G A S T R I C J U I C E
13. Heynen, G., A. Brassine, J. C. Daubresse, G, Ligny, J. A. Kanis, S. Gaspar and P. Franchimont. Lack of clinical and physiological relationship between gastrin and calcitonin in man. Eur J Clin Invest 11: 331-335, 1981. 14. Hirsch, P. F., G. F. Gauthier and P. L. Munson. Thyroid hypocalcemic principle and recurrent laryngeal nerve injury as factors affecting the response to parathyroidectomy in rats. Endocrinology 73: 244--252, 1963. 15. Ipp, E., R. E. Dobbs, A. Arimura, W. Vale, V. Harris and R. H. Unger. The effect of gastrin, gastric inhibitory polypeptide, secretin and the octapeptide of cholecystokinin upon immunoreactive somatostatin release by the perfused pancreas. J Clin Invest 60: 1216--1219, 1977. 16. Ito, H., J. Hato, H. Yokozaki and E. Tahara. Calcitonin in human gastric mucosa and carcinoma. J Cancer Res Clin Oncol 112: 50-56, 1986.
299
17. Snedecor, G. W. and W. G. Cohran. Statistical Methods, 7th edition. Ames, IA: The Iowa State University Press, 1980. 18. Straus, E. and R. S. Yalow. Artifacts in the radioimmunoassay of peptide hormones in gastric and duodenal secretions. J Lab Clin Med 87: 292-298, 1976. 19. Tobler, P. H., F. A. Tschopp, M. A. Dambacher, W. Born and J. A. Fischer. Identification and characterization of calcitonin forms in plasma and urine of normal subjects and medullary carcinoma patients. J Clin Endocrinol Metab 57: 74%754, 1983. 20. Uvn~is-Wallensten, K., S. Efendic and R. Luft. Vagal release of somatostatin into the antral lumen of cats. Acta Physiol Scand 99: 126--128, 1977. 21. Ziegler, R., H. Minne, J. Hotz and H. Goebell. Inhibition of gastric secretion in man by oral administration of calcitonin. Digestion 11: 157-160, 1974.