Pancreas Vol. 6, No. 3, pp. 260-265 0 1991 Raven Press, Ltd., New York
Plasma Concentrations of Neurotensin and CCK in Patients with Chronic Pancreatitis with and without Enzyme Substitution Rainer Nustede, Heinz Kohler, "Ulrich R. Folsch, and Anton Schafmayer Department of Surgery and *Department of Medicine (Division of Gastroenterology and Endocrinology), Georg-August-University of Gottingen, F.R.G.
Summary: The peptide hormones neurotensin (NT) and cholecystokinin (CCK) are commonly attributed with a physiological role in the stimulation of exocrine pancreatic secretion. However, on the other hand, little is known about the effect of diminished exocrine pancreatic function and of the resulting maldigestion on postprandial plasma levels of these two gastrointestinal peptides. We investigated, therefore, the effect of enzyme substitution therapy on the magnitude and time course of plasma concentrations of both hormones in patients suffering from severe chronic pancreatitis. Pancreatic insufficiency led to elevated NT-concentrations, in response to a standard meal, which could be reduced by enzyme replacement therapy. Prior to enzyme therapy, the mean integrated postprandial release of NT amounted to 2800 -t 250 pglml after 60 min in patients with severe chronic pancreatitis. This amount was significantly reduced to 1250 k 150 p g h l after 60 min after enzyme therapy, compared to 810 t 90 pglml after 60 min in healthy volunteers after the standard meal. The integrated postprandial CCK level in patients investigated was significantly lower (35 2 4.8 pmol/L after 60 min) without any substitution therapy, compared to the integrated peptide amount in healthy volunteers (145 13.5 pmoll L after 60 min). Enzyme therapy in patients suffering from chronic pancreatitis led to an increased postprandial CCK-level (80 9.6 pmoYL after 60 min). Elevated CCK-plasma concentrations have not been demonstrated in these patients with pancreatic insufficiency. We therefore suggest that CCK might not play a major role in feedback regulation in patients with chronic pancreatitis. However, in light of elevated NT plasma concentrations in patients with chronic pancreatitis, NT-mediated influence on the pancreas deserves further study. Key Words: Chronic pancreatitis-CCK; neurotensin-MaldigestionEnzyme therapy-Feedback-regulation.
*
*
NT stimulates exocrine pancreatic secretion in dogs (3-7), rats (8) and humans (9-1 1). On the other hand, controversy exists regarding interaction between exocrine pancreatic function and CCK and NT plasma concentrations. Varying plasma concentrations Of the peptides CCK and NT have been described in the postprandial state in patients with exocrine pancreatic insufficiency, resulting from chronic PanCreatitiS Or Cystic fibrosis. Numerous studies describe plasma concentrations of CCK,
The tridecapeptide neurotensin (NT) (1) has been increasingly attributed with a physiological role, along with CCK and secretin (2), in the regulation of pancreatic exocrine secretion. Manuscript received March 22, 1990; revised manuscript accepted August 8, 1990. Address correspondence and reprint requests to Dr. Rainer Nustede, at Department of General Surgery, Georg-AugustUniversity of Gottingen, Robert-Kocfl-Strasse 40, 3400 Gottingen, F.R.G.
260
NE UR0 TENSIN IN CHRONIC PANCREA TITIS
which were either increased (12-14), decreased (15,16) or unchanged (17-20). Increased (21), as well as decreased (22), postprandial NT plasma concentrations have been reported for patients with chronic pancreatitis. At present, under experimental conditions, it remains controversial (23,24) whether the level of intestinal pancreatic enzymes (particularly proteases) influences plasma concentrations of regulatory peptides (e.g., NT and CCK). Nevertheless, such experimental attempts are .still necessary on account of their potential clinical relevance: In pancreatic insufficiency, elevated peptide concentrations might contribute to the stimulation of this gland. Since there are some reports indicating that, in chronic pancreatitis, enzyme substitution therapy reduces pain (25,26), it is conceivable that this therapy might operate through a modified regulation of NT secretion. The present study was, therefore, designed to investigate the effect of reduced pancreatic exocrine secretion, with or without enzyme substitution, on the magnitude and time-course of basal and postprandial plasma concentrations of CCK and NT. METHODS Basal and postprandial plasma concentrations of CCK and NT were determined in a prospective study of six male patients (47-59 years of age), suffering from chronic pancreatitis; five of these patients also had insulin-dependent diabetes mellitus. All six patients were on the same enzyme therapy for more than six months. Without enzyme therapy (measured three days after terminating substitution therapy) daily fecal fat excretion averaged 36.8 g (22.3-72.8 g), with a mean stool mass of 329 g (165595 g) per day. After an overnight fast, patients were given a standard morning meal (1022 kcal), consisting of 152 g carbohydrates, 36 g protein, and 30 g fat, in a volume of 550 ml. Studies concerning postprandial plasma concentrations of CCK and neurotensin were conducted over a 60-min period postprandially, with enzyme replacement (lipase 56,000 u, amylase 44,000 u, and protease 3,000 u three times daily at main meals; Pankreon, KaliChemie, Hannover, F.R.G.), as well as 3 days after terminating substitution therapy. Therefore, each patient served as his own control. Blood samples for radioimmunological CCK and NT determinations were drawn from an indwelling venous cannula at 15-min intervals for 60 min. Aprotinin and heparin were added to the collecting tubes to give a
261
final concentration of 500 IE/ml and 20 IE/ml, respectively. Ten healthy volunteers (21-31 years of age) received the same standard meal and served as controls. Blood samples were drawn as described above. Laboratory determinations Neurotensin-radioimmunoassay Blood samples were centrifuged immediately (10 min at 2,876 X g and + 4°C) and stored at - 20°C. Plasma was extracted with absolute ethanol, 1:3, to remove unspecific plasma interference, centrifuged (2,876 g, +4"C), and freeze-dried. Samples were reconstituted in 0.02 M veronal buffer, pH 8.0, (with 0.2% bovine serum albumin (BSA) and 0.02% NaNJ immediately prior to the radioimmunological assay. The recovery rate was, on the average, 90%. Synthetic NT (NT 1-13) was used as standard and '25J-NT(NEN, Boston, MA, U.S.A.) as tracer. The assay employed our own antibody, GN21, in a final dilution of 1:420,000. This antibody has been raised in rabbits and requires the entire peptide (NT 1-13) for binding and displacement. Shorter fragments of peptide are not recognized. The antibody has no cross-reactivity with other peptides. A preincubation of 1 day was followed by a 2-day incubation in the presence of tracer. Sensitivity amounted to 0.78 pmol/L. The intraassay variance was 1.3% and the interassay variance was 4.0%, determined by ten consecutive measurements. A standard curve is shown in Fig. 1.
CCK-radioimmunoassay Plasma concentrations of CCK were determined by a newly-modified method (27), using the antibody G160 (13). The G160 antibody, when used in a final dilution of 1:90,000, binds to the sulfated tyrosine region of the hormone. This antibody does not cross-react with other gastrointestinal peptides. Sulfated CCK-8, in the concentration range of 0.3% 50 pmoVL, was used as a standard. Radioiodinelabelled, CCK-8 (2,200 mCi/mmoL) (NEN, Boston, MA, U.S.A.). The sensitivity of the method was 0.7 pmoYL. The intraassay variance was 5.8% (n = 10) with an interassay variance of 8.3%. Blood samples were collected and extracted, as described above, and dissolved in assay buffer (0.02 M veronal buffer, pH 8.4 with 0.1% BSA) for assaying. The recovery rate averaged 92%. Extracted and reconstituted plasma samples (0.1 ml aliquots) and standards (dissolved in hormone-free plasma and extracted in the same manner) were incubated with Pancreas, Vol. 6, No. 3. 1991
R . NUSTEDE ET AL.
262 ryeo(x)
an average of 17.2 g (3.5-27.4) and the mean stool mass decreased from 329 to 194 g (87-304). Basal and postprandial plasma concentrations of NT were increased in all patients with chronic pancreatitis, compared to healthy volunteers. Whereas the mean neurotensin plasma concentrations increased from 8 ? 2 pg/ml to 14.9 ? 2.8 pg/ml after 15 min and to 36.9 7 p g / d after 60 rnin in healthy subjects, they increased from 30.2 ? 3.5 pg/ml to 80.1 5 pg/ml after 15 rnin and 62 It 4.5 pg/ml after 60 min, respectively in patients without enzyme substitution therapy (Fig. 2). Enzyme replacement caused a partial reduction of postprandial neurotensin concentrations to 40.8 pg/ml at 15 rnin and 63 k 4 pg/ml after 60 min, respectively. Basal values remained elevated at 31 ? 2 pg/ml, compared to healthy volunteers. The mean integrated postprandial release of neurotensin in patients prior to enzyme therapy, amounted to 2,800 ? 250 pg/ml after 60 min and was significantly higher (p < 0.001) compared to the peptide's release in healthy volunteers or in enzyme-treated patients (p < 0.02). The integrated postprandial neurotensin release amounted to 1,250 150 pg/ml after 60 rnin in patients after enzyme therapy, compared to 810 90 pg/ml after 60 rnin in healthy volunteers. This difference was also statistically significant (p < 0.02) (Fig. 3a). Plasma concentrations of CCK in normal subjects increased from mean basal values of 1.2 -+ 0.3 pmol/ L to an average of 5.3 ? 0.8 pmoYL after 15 min, and to 4.1 ? 0.8 pmol/L after 60 min. In untreated patients with chronic pancreatitis CCK concentrations increased from a mean basal concentration of 0.79 0.3 pmoVL to 1.8 rt 0.5 pmoVL after 15 min
*
*
*
1 2.0
1 3.6
1 1 1 1 1 1 1 I I 6.8 12.7 23.7 44.2 82.4 153.8 287.0535.8 1000
CONC. (pg/ml)
FIG. 1. Neurotensin (NT) radioimmunoassay-standardcurve. The assay employed antibody GN21 in a final dilution of 1:420,000, which requires the entire peptide (NT 1-13) for binding. Synthetic NT (NT 1-13) was used as a standard and '*'Jneurotensin as a tracer. The intraassay variance was 1.3% and the interassay variance was 4%, determined by 10 concecutive measurements.
0.4 ml of diluted antibody (dilution 1:90,000)for 72 h before adding labelled tracer for further 48 h. The bound fraction was separated on charcoal (6% charcoal with 20% hormone-free plasma in 0.02 M barbital buffer, pH 8.4) at the end of the total incubation time of 5 days.
*
t 8070
Statistical evaluation Results are expressed as mean ? SEM and compared, using the Student t test for paired and unpaired samples. A p value
Plasma Concentrations of Neurotensin and CCK in Patients with Chronic Pancreatitis with and without Enzyme Substitution Rainer Nustede, Heinz Kohler, "Ulrich R. Folsch, and Anton Schafmayer Department of Surgery and *Department of Medicine (Division of Gastroenterology and Endocrinology), Georg-August-University of Gottingen, F.R.G.
Summary: The peptide hormones neurotensin (NT) and cholecystokinin (CCK) are commonly attributed with a physiological role in the stimulation of exocrine pancreatic secretion. However, on the other hand, little is known about the effect of diminished exocrine pancreatic function and of the resulting maldigestion on postprandial plasma levels of these two gastrointestinal peptides. We investigated, therefore, the effect of enzyme substitution therapy on the magnitude and time course of plasma concentrations of both hormones in patients suffering from severe chronic pancreatitis. Pancreatic insufficiency led to elevated NT-concentrations, in response to a standard meal, which could be reduced by enzyme replacement therapy. Prior to enzyme therapy, the mean integrated postprandial release of NT amounted to 2800 -t 250 pglml after 60 min in patients with severe chronic pancreatitis. This amount was significantly reduced to 1250 k 150 p g h l after 60 min after enzyme therapy, compared to 810 t 90 pglml after 60 min in healthy volunteers after the standard meal. The integrated postprandial CCK level in patients investigated was significantly lower (35 2 4.8 pmol/L after 60 min) without any substitution therapy, compared to the integrated peptide amount in healthy volunteers (145 13.5 pmoll L after 60 min). Enzyme therapy in patients suffering from chronic pancreatitis led to an increased postprandial CCK-level (80 9.6 pmoYL after 60 min). Elevated CCK-plasma concentrations have not been demonstrated in these patients with pancreatic insufficiency. We therefore suggest that CCK might not play a major role in feedback regulation in patients with chronic pancreatitis. However, in light of elevated NT plasma concentrations in patients with chronic pancreatitis, NT-mediated influence on the pancreas deserves further study. Key Words: Chronic pancreatitis-CCK; neurotensin-MaldigestionEnzyme therapy-Feedback-regulation.
*
*
NT stimulates exocrine pancreatic secretion in dogs (3-7), rats (8) and humans (9-1 1). On the other hand, controversy exists regarding interaction between exocrine pancreatic function and CCK and NT plasma concentrations. Varying plasma concentrations Of the peptides CCK and NT have been described in the postprandial state in patients with exocrine pancreatic insufficiency, resulting from chronic PanCreatitiS Or Cystic fibrosis. Numerous studies describe plasma concentrations of CCK,
The tridecapeptide neurotensin (NT) (1) has been increasingly attributed with a physiological role, along with CCK and secretin (2), in the regulation of pancreatic exocrine secretion. Manuscript received March 22, 1990; revised manuscript accepted August 8, 1990. Address correspondence and reprint requests to Dr. Rainer Nustede, at Department of General Surgery, Georg-AugustUniversity of Gottingen, Robert-Kocfl-Strasse 40, 3400 Gottingen, F.R.G.
260
NE UR0 TENSIN IN CHRONIC PANCREA TITIS
which were either increased (12-14), decreased (15,16) or unchanged (17-20). Increased (21), as well as decreased (22), postprandial NT plasma concentrations have been reported for patients with chronic pancreatitis. At present, under experimental conditions, it remains controversial (23,24) whether the level of intestinal pancreatic enzymes (particularly proteases) influences plasma concentrations of regulatory peptides (e.g., NT and CCK). Nevertheless, such experimental attempts are .still necessary on account of their potential clinical relevance: In pancreatic insufficiency, elevated peptide concentrations might contribute to the stimulation of this gland. Since there are some reports indicating that, in chronic pancreatitis, enzyme substitution therapy reduces pain (25,26), it is conceivable that this therapy might operate through a modified regulation of NT secretion. The present study was, therefore, designed to investigate the effect of reduced pancreatic exocrine secretion, with or without enzyme substitution, on the magnitude and time-course of basal and postprandial plasma concentrations of CCK and NT. METHODS Basal and postprandial plasma concentrations of CCK and NT were determined in a prospective study of six male patients (47-59 years of age), suffering from chronic pancreatitis; five of these patients also had insulin-dependent diabetes mellitus. All six patients were on the same enzyme therapy for more than six months. Without enzyme therapy (measured three days after terminating substitution therapy) daily fecal fat excretion averaged 36.8 g (22.3-72.8 g), with a mean stool mass of 329 g (165595 g) per day. After an overnight fast, patients were given a standard morning meal (1022 kcal), consisting of 152 g carbohydrates, 36 g protein, and 30 g fat, in a volume of 550 ml. Studies concerning postprandial plasma concentrations of CCK and neurotensin were conducted over a 60-min period postprandially, with enzyme replacement (lipase 56,000 u, amylase 44,000 u, and protease 3,000 u three times daily at main meals; Pankreon, KaliChemie, Hannover, F.R.G.), as well as 3 days after terminating substitution therapy. Therefore, each patient served as his own control. Blood samples for radioimmunological CCK and NT determinations were drawn from an indwelling venous cannula at 15-min intervals for 60 min. Aprotinin and heparin were added to the collecting tubes to give a
261
final concentration of 500 IE/ml and 20 IE/ml, respectively. Ten healthy volunteers (21-31 years of age) received the same standard meal and served as controls. Blood samples were drawn as described above. Laboratory determinations Neurotensin-radioimmunoassay Blood samples were centrifuged immediately (10 min at 2,876 X g and + 4°C) and stored at - 20°C. Plasma was extracted with absolute ethanol, 1:3, to remove unspecific plasma interference, centrifuged (2,876 g, +4"C), and freeze-dried. Samples were reconstituted in 0.02 M veronal buffer, pH 8.0, (with 0.2% bovine serum albumin (BSA) and 0.02% NaNJ immediately prior to the radioimmunological assay. The recovery rate was, on the average, 90%. Synthetic NT (NT 1-13) was used as standard and '25J-NT(NEN, Boston, MA, U.S.A.) as tracer. The assay employed our own antibody, GN21, in a final dilution of 1:420,000. This antibody has been raised in rabbits and requires the entire peptide (NT 1-13) for binding and displacement. Shorter fragments of peptide are not recognized. The antibody has no cross-reactivity with other peptides. A preincubation of 1 day was followed by a 2-day incubation in the presence of tracer. Sensitivity amounted to 0.78 pmol/L. The intraassay variance was 1.3% and the interassay variance was 4.0%, determined by ten consecutive measurements. A standard curve is shown in Fig. 1.
CCK-radioimmunoassay Plasma concentrations of CCK were determined by a newly-modified method (27), using the antibody G160 (13). The G160 antibody, when used in a final dilution of 1:90,000, binds to the sulfated tyrosine region of the hormone. This antibody does not cross-react with other gastrointestinal peptides. Sulfated CCK-8, in the concentration range of 0.3% 50 pmoVL, was used as a standard. Radioiodinelabelled, CCK-8 (2,200 mCi/mmoL) (NEN, Boston, MA, U.S.A.). The sensitivity of the method was 0.7 pmoYL. The intraassay variance was 5.8% (n = 10) with an interassay variance of 8.3%. Blood samples were collected and extracted, as described above, and dissolved in assay buffer (0.02 M veronal buffer, pH 8.4 with 0.1% BSA) for assaying. The recovery rate averaged 92%. Extracted and reconstituted plasma samples (0.1 ml aliquots) and standards (dissolved in hormone-free plasma and extracted in the same manner) were incubated with Pancreas, Vol. 6, No. 3. 1991
R . NUSTEDE ET AL.
262 ryeo(x)
an average of 17.2 g (3.5-27.4) and the mean stool mass decreased from 329 to 194 g (87-304). Basal and postprandial plasma concentrations of NT were increased in all patients with chronic pancreatitis, compared to healthy volunteers. Whereas the mean neurotensin plasma concentrations increased from 8 ? 2 pg/ml to 14.9 ? 2.8 pg/ml after 15 min and to 36.9 7 p g / d after 60 rnin in healthy subjects, they increased from 30.2 ? 3.5 pg/ml to 80.1 5 pg/ml after 15 rnin and 62 It 4.5 pg/ml after 60 min, respectively in patients without enzyme substitution therapy (Fig. 2). Enzyme replacement caused a partial reduction of postprandial neurotensin concentrations to 40.8 pg/ml at 15 rnin and 63 k 4 pg/ml after 60 min, respectively. Basal values remained elevated at 31 ? 2 pg/ml, compared to healthy volunteers. The mean integrated postprandial release of neurotensin in patients prior to enzyme therapy, amounted to 2,800 ? 250 pg/ml after 60 min and was significantly higher (p < 0.001) compared to the peptide's release in healthy volunteers or in enzyme-treated patients (p < 0.02). The integrated postprandial neurotensin release amounted to 1,250 150 pg/ml after 60 rnin in patients after enzyme therapy, compared to 810 90 pg/ml after 60 rnin in healthy volunteers. This difference was also statistically significant (p < 0.02) (Fig. 3a). Plasma concentrations of CCK in normal subjects increased from mean basal values of 1.2 -+ 0.3 pmol/ L to an average of 5.3 ? 0.8 pmoYL after 15 min, and to 4.1 ? 0.8 pmol/L after 60 min. In untreated patients with chronic pancreatitis CCK concentrations increased from a mean basal concentration of 0.79 0.3 pmoVL to 1.8 rt 0.5 pmoVL after 15 min
*
*
*
1 2.0
1 3.6
1 1 1 1 1 1 1 I I 6.8 12.7 23.7 44.2 82.4 153.8 287.0535.8 1000
CONC. (pg/ml)
FIG. 1. Neurotensin (NT) radioimmunoassay-standardcurve. The assay employed antibody GN21 in a final dilution of 1:420,000, which requires the entire peptide (NT 1-13) for binding. Synthetic NT (NT 1-13) was used as a standard and '*'Jneurotensin as a tracer. The intraassay variance was 1.3% and the interassay variance was 4%, determined by 10 concecutive measurements.
0.4 ml of diluted antibody (dilution 1:90,000)for 72 h before adding labelled tracer for further 48 h. The bound fraction was separated on charcoal (6% charcoal with 20% hormone-free plasma in 0.02 M barbital buffer, pH 8.4) at the end of the total incubation time of 5 days.
*
t 8070
Statistical evaluation Results are expressed as mean ? SEM and compared, using the Student t test for paired and unpaired samples. A p value
