617
J. Physiol. (1975), 245, pp. 617-638 With 10 text-ft gure8 Printed in Great Britain
SECRETION OF CALCIUM IN PANCREATIC JUICE
By B. CECCARELLI, F. CLEMENTE* AND J. MELDOLESI From the C.N.R. Centre of Cytopharmacology, Department of Pharmacology, University of Milan, 20129 Milan, Italy
(Received 11 June 1974) SUMMARY
1. The origin of the calcium secreted by the pancreas has been investigated in vivo in the guinea-pig by a study carried out in parallel (a) in the juice secreted in response to the injection of either secretin or caerulein and (b) in the pancreatic tissue and in cell fractions isolated therefrom. 2. In agreement with previous findings we observed that the concentration of calcium is low in the secretin-stimulated and high in the caeruleinstimulated juice. In the latter calcium and protein are proportional (ca. 50 n-mole: mg). 3. After i.v. injection of 45Ca the radioactivity decreases rapidly and quasi-exponentially in the blood plasma. A roughly parallel time course is found in the secretin-stimulated juice: the evolution of the juice: plasma radioactivity ratio resembles that observed with the extracellular space marker [3H]D-sorbitol. In contrast, the time course of 45Ca in plasma and caerulein-stimulated juice are not proportional: the high levels characteristic of this juice are reached several minutes after the injection and maintained thereafter. This increase is followed ca. 50 min later by the appearance of the newly synthesized [3H]L-leucine-labelled proteins. 4. The pancreatic tissue is rich in calcium which is localized primarily in zymogen granules (ca. 36 n-mole: mg protein) and mitochondria; the soluble cytoplasm is low in calcium. 5. The injected "Ca accumulates in zymogen granules faster than [3H]L-leucine-labelled proteins. The 45Ca : protein ratio of these organelles is considerably lower than that of the caerulein-stimulated juice. 6. It is concluded (a) that calcium is secreted into the pancreatic juice in two fractions, one (possibly released by simple diffusion) associated with the electrolyte component, the other with the protein of the juice, * Present address: INSERM, Groupe de Recherche de Pathologie Digestive, Hopital Pourpan, Toulouse, France.
618 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI (b) that zymogen granules are the major, but not the only source of the latter fraction, and (c) that the zymogen granule-associated calcium joins the exportable proteins some time after their synthesis, possibly in the Golgi complex and/or in the condensing vacuoles. INTRODUCTION
It has been clearly established in several animal species that the calcium ions contained in the pancreatic juice fulfil a number of important functions. Thus, amylase, a juice enzyme, contains one atom of tightly bound calcium per molecule (Stein, Hsiu & Fischer, 1964). Furthermore Ca2+ is necessary for maintaining the molecular configuration of other secretary zymogens, such as trypsinogen, chymotrypsinogen (Delaage & Lazdunski, 1967), prephospholipase (Pieterson, Volwerk & De Haas, 1974) and lipase (Benzonana, 1968), and high concentration of the cation is required for the activation of trypsinogen (Desnuelle & Gabeloteau, 1957; Desnuelle, 1960) and for the activity of trypsin, lipase and phospholipase (Sipos & Merkel, 1970; Sarda, Marchis-Mouren, Constantin & Desnuelle, 1957; De Haas, Bonsen, Pieterson & Van Deenen, 1971). The correct understanding of the mechanisms by which an adequate concentration of calcium is achieved and maintained in the juice represents therefore a problem of great importance in pancreatic physiology. Previous studies carried out in intact animals as well as in in vitro preparations (Goebel, Steffen & Bode, 1972; Argent, Case & Scratcherd, 1973) suggest that calcium reaches the pancreatic juice from at least two different sources. One of these sources appears to be correlated with, and the other independent of the secretary function of the acinar cells of the gland. However, a definite identification of the origin of these two calcium fractions and of the mechanisms of their secretion has not been obtained so far. These problems are considered in the present investigation in which two types of experiments have been carried out in parallel: (a) the analysis of the pancreatic juice secreted in vivo in response to stimulation with either secretin or caerulein (a polypeptide of well known pancreozymin-like activity (Bertaccini, De Caro, Endean, Erspamer & Impicciatore, 1969)), and (b) the subcellular study of the pancreatic tissue, in relation to its calcium secretary activity. We used the guinea-pig, because in this animal species the pancreatic secretary process is well known and because subcellular fractions of high purity can be isolated from the pancreas tissue by means of established techniques (Jamieson & Palade, 1967a, b; Meldolesi, Jamieson & Palade, 1971).
SECRETION OF CALCIUM IN PANCREATIC JUICE 619 METHODS Male albino guinea-pigs weighing 450-500 g (gift of Sigurta Drug Co., Milan, Italy) were fasted 20-24 hr before the experiments. Pancreatic 8ecretion 8tudie8 In guinea-pigs anaesthetized with pentobarbitone the pancreatic duct was cannulated with a soft, plastic tube and the juice collected at various time intervals as indicated under Results. Tracers (45CaCl2 (40 Fuc/kg), [3H]L-leucine + CaCl2 (80 and 40 ,tc/kg, respectively) or [3H]D-sorbitol (50 ,uc/kg)) and secretagogue drugs (caerulein (500 nglkg) or secretin (2 u./kg)) were injected i.v. Experiments with pancreatic tissue and cell fraction Guinea-pigs anaesthetized with pentobarbitone were injected i.v. with either 45CaCl2 (40 /zcckg) + [3H]L-leucine (80 ftc/kg) or with [3H]D-sorbitol (50 ftc/kg) and sacrificed at various times after the injection by a blow over the head. The pancreases were quickly excised and immersed in ca. 50 ml. ice-cold 0-3 M sucrose. After 2 min they were removed, blotted with filter paper, weighed, minced with scissors and homogenized in 10 ml. 0 3 M sucrose as described previously. Fractionation of the homogenates by differential centrifugation was carried out as described by Meldolesi et al. (1971) to yield the following preparations: crude mitochondria, total microsomes, zymogen granules and post-microsomal supernatant. Isolated microsomes and mitochondria were washed once by resuspension in 0 3 M sucrose and recentrifugation; in the case of zymogen granules a washing step is included in the isolation procedure. An extensive characterization of these fractions has been reported previously (Meldolesi et al. 1971).
Radioactivity and chemical asaays Aliquots of the pancreatic homogenates, cell fractions, juice and blood plasma were usually mixed directly with 10 ml. of Packard Instagel and transferred to disposable plastic counting vials. Onlywhen a separate assay of the protein-incorporated 3H-radioactivity was needed, the samples were precipitated at 4° C with 10 % trichloroacetic acid (TCA) and centrifuged. The precipitates were washed as described previously (Meldolesi, 1970). Aliquots of the TCA-soluble and insoluble samples were dissolved in 10 ml. of Packard Instagel and counted. All samples were counted in an Intertechnique SL 30 liquid scintillation spectrometer. In single label counting, the 3H efficiency was over 40 % and "5Ca efficiency over 80 %. In double label counting the conditions were selected in order to have no spill of 3H counts into the 45Ca channel; efficiency of 3H counts was ca. 22 % and of 45Ca ca. 70 %. Spill of "Ca counts into the 3H channel was always less than 5 %. Corrections for quenching and spill were made by external standardization. Protein was determined according to Lowry, Rosebrough, Farr & Randall (1951) using bovine serum albumin as the standard. Calcium was assayed in a Hillger and Watts atomic absorption spectrometer, model Atomspek, after deproteinization with 10 % TCA and washing of the precipitate with 5 % TCA. In order to overcome interferences, all the samples contained 1 % La3+ and were adjusted at pH 1-5 before the assay. Furthermore, all the glassware and plasticware was carefully washed with either 2 N-HCl or 10 mm disodium ethylenediaminetetraacetate (EDTA), pH 7-5 and then rinsed repeatedly with distilled water. The calcium content of the solutions used was carefully determined, and the experimental results were corrected for the impurities when necessary.
620 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI MateriaUl L-[4,5-3H]leucine (sp.act. 32 mclpmole) and D-[1-3H]sorbitol (sp.act. 6-8 mcftmole) were purchased from New England Nuclear, Langen, Germany; 45CaCl2 (3 mc/ molel) from the Radiochemical Centre, Amersham, U.K.; sucrose (Special Enzyme Grade) from Schwarz-Mann, Orangeburg, New York, U.S.A. Secretin was obtained from the G.I.H. Research Unit, Chemistry Department, Karolinska Inst., Stockholm, Sweden; caerulein was the kind gift of the Farmitalia Laboratories for Basic Research, Milan, Italy. All other chemicals were reagent grade. RESULTS
In vivo stimulation of pancreatic secretion Fig. 1 illustrates the secretary response elicited by secretin and cerulein in guinea-pigs bearing an acutely cannulated pancreatic duct. The doses of the secretagogues used were able to provoke maximal secretary responses, as shown in a previous report (Meldolesi, 1970), as well as in preliminary experiments of this work. As expected, the i.v. injection of secretin produced a large increase of the juice output, which lasted for several minutes; the secretin-stimulated juice was very poor in protein. Slightly reduced responses were obtained by repeating the injection of the hormone. The effect of cerulein, on the other hand, was lower than that of secretin in terms of volume and much higher in terms of protein concentration of the juice. The response was more transient and showed a faster decrease on repeating the stimulation. Fig. 2 shows a representative example of the time course of calcium secretion induced by the two polypeptides. The data are plotted as a rate of release (upper panel) and concentration (lower panel). If one compares the results presented in this Figure it is clear that the cumulative calcium release is larger after stimulation with secretin, whereas the calcium concentration in the juice is always much higher with caerulein. The statistical analysis revealed that no correlation exists between calcium and protein concentration in the secretin-stimulated juice, whereas a linear correlation exists after stimulation with caerulein (Fig. 3). However, the slope of the regression line fitted to the data indicates that at 0 protein a small amount of calcium (ca. 07 ,umole: ml.) is still contained also in the caerulein-stimulated juice (Fig. 3). In order to obtain some information on the origin of the calcium recovered in the pancreatic juice, the kinetics of secretion of i.v. injected 45Ca was investigated. In Fig. 4 the time course of the radioactivity in the blood plasma (upper panel) is compared with that observed in the pancreatic juice secreted in response to repeated stimulation with either secretin or cerulein (lower panel). The curves are quite different. In the
SECRETION OF CALCIUM IN PANCREATIC JUICE 621
3 C
E E C
0 0. U
1
220 240 260 280
il1 C Z._
El 0
E 0 41 >
U
._% 11
10203040 60 80 100 120 140 160 180 200 220 240 260 280
$
t
I Time (min)
Fig. 1. Effect of secretin (0) and caerulein (0) on the protein output (upper panel) and on the volume (lower panel) of the pancreatic juice in guinea-pigs bearing an acutely cannulated pancreatic duct. The arrows represent single injections of either secretin (2 u./kg) or caerulein (0 5 ,jg/kg). In this, as well as in the following Figures which concern the pancreatic juice (Figs. 2, 4, 5, 6, 7 and 8), the indicated time points correspond to the times at which the juice samples were collected. Thus, the corresponding values pertain to the juice secreted during the interval preceding the time indicated in the Figure. Each curve shown is representative of the results obtained in seven consistent experiments.
622 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI plasma the concentration of the 45Ca declines rapidly within a few minutes after the injection and then subsides more gradually. In the secretinstimulated juice an initial decrease is also evident; however, it is smaller
g 50 0
E
E R 25
U
IV
0
E
E
*2: U
3-
20 40
t
60 80 100 120 140 160 180 200
I
I
Time (min) Fig. 2. Effect of secretin (e) (2 u./kg) and caerulein (0) (0-5 /zg/kg) on the rate of secretion (upper panel) and the concentration (lower panel) of calcium in the pancreatic juice. The arrows represent single injection of either secretagogue. Each curve shown is representative of the results of six consistent experiments.
than that observed in the plasma. In the caerulein-stimulated juice the curve is completely different: during the first 30 min the concentration of the tracer remains approximately at the level of the first sample of the secretin-stimulated juice. Between the 30th and the 50th min the radioactivity approximately doubles and remains at this high level (even if with distinct fluctuations) for at least 3 hr.
SECRETION OF CALCIUM IN PANCREATIC JUICE 623 In another set of experiments (Fig. 5) the evolution of the radioactivity was followed in the juice several hours after the i.v. injection of 45Ca, i.e. at a time when the concentration of the tracer remains relatively constant in the plasma. The results of these experiments corroborate our previous findings in the sense that the concentration of the tracer in the secretinstimulated juice remains relatively stable, at a level one third to one v=54
0 0
0
r=087 7.5
0 0
.2
~~ ~~00 0
E
205 _ E
2*5
OdM
0
o
00 0
0000
50
0
0
o
0~
0
00
0
100 mg protein/mi. juice
150
Fig. 3. Correlation between the concentration of calcium and of protein in the pancreatic juice secreted in response to the i.v. injection of caerulein (0.5 jug/kg). The solid line is the calculated regression line (P < 0.001). v = degrees of freedom; r = coefficient of correlation.
quarter of that of the plasma, whereas in the caerulein-stimulated juice the concentration is four- to eightfold higher. In these animals, the switching from one type of secretion stimulation to the other results in a rapid shift in the level of the juice radioactivity (Fig. 5, right-hand side). Since it was important to establish whether or not the secretion of calcium is correlated with the secretion of other components of the pancreatic juice, experiments were carried out in this direction. Thus, the kinetics of 45Ca
624 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI pancreatic secretion was compared with that of other tracers: [3H]D-sorbitol, a molecule which in different tissues, including the pancreas, is known to penetrate very slowly within the cells and therefore can be considered as an extracellular tracer (Goodford & Leach, 1966; Van Venroij, Poort,
15p-
10
C
5p
4)
0 U L.
I-
0 f ._ I
EE a.
4p
:0 EM
3
xa
2
1
10 I
5
a
50 30
f
210
150 I
.
I
*
*
70 90 110 130
*
f
I
I
170 190
I
I
t
230
Time (min)
Fig. 4. Short interval kinetics of pancreatic secretion of i.v.-injected 45Ca. The Figure shows the time course of the radioactivity in the blood plasma (m) (upper panel) and in the juice secreted in response to either secretin (0) (2 u./kg) or caerulein (0) (05/jg/kg) (lower panel). The large arrow marks the injection of the tracer (40,uc/kg) together with either one of the secretagogue drugs; the small arrows mark the injection of the secretagogue drugs alone. The curve shown is representative of the results obtained in four consistent experiments.
SECRETION OF CALCIUM IN PANCREATIC JUICE 625 Kramer & Jansen, 1972), and [3H]L-leucine, used here as a precursor of the juice proteins. The i.v.-injected rH]D-sorbitol was found to appear after a short lag of time in the juice elicited by the injection of both caerulein (Fig. 6) and secretin (not shown). The time courses of the decline of the r3H]D-sorbitol concentrations in juice and plasma run almost in parallel, with a ratio of ca. 0 5, irrespective of the stimulant used (Fig. 6). The juice: plasma radioactivity ratios observed after injection of either 6
.40
E
an' 2
-
\
I
40
80
f
120
160
200
tf
240
f
280
f
320
360
Time (min)
Fig. 5. Kinetics of pancreatic secretion of 45Ca injected several hours before the experiment. Animals received one i.v. injection of the tracer (40 fc/kg). After 3 hr their pancreatic duct was cannulated and 1 hr later the stimulation of secretion with either secretin (e) (2 u./kg) or caerulein (0.5 ,tg/kg) (Q) was started. The arrows indicate single injections of the secretagogues. Each curve shown is representative of the results obtained in two consistent experiments.
[3H]D-sorbitol or 45Ca and stimulation of secretion with secretin and caerulein are plotted in Fig. 7 as a function of time. It is clear that the time course of the ratios concerning the 45Ca in the secretin-stimulated juice is almost parallel to the VH]D-sorbitol curve (especially at the early time points of the experiment), whereas no correlation is evident between the latter curve and the 45Ca curve concerning the caerulein-stimulated juice. In this case the 45Ca ratios are lower than the corresponding PH]Dsorbitol ratios at the beginning of the experiment; after the 30th min they rise progressively to very high values (ca. 3) indicating that the bulk of the calcium secreted in this juice is not in equilibrium with the plasma. Finally, experiments were carried out to investigate the correlation between the appearance in the juice of 45Ca and that of newly synthesized 26
P HY 245
626 B. CECCARELLI, P. CLEMENTE AND J. MELDOLESI exportable protein. Animals were injected simultaneously with 45Ca, [H]L-leucine and with either one of the stimulatory drugs. In both cases the 3H-labelled proteins appeared in the juice after a lag of about 50 mmi, and reached the maximum concentration after 130 min. No correlation was found between the 3H-protein curve and the 45Ca curve in the secretinstimulated juice (not shown in Figures), whereas in the cerulein-stimulated juice the 3H-protein data fall along a line which seems to follow the increase in "5Ca radioactivity but with a delay of ca. 80 min (Fig. 8). 8 co (U
*6
.:
6E
A
1
4E 2_
Xi
III
510 20
40
I
60
80
I
100
I
120
I
150
I
170
Time (min) Fig. 6. Kinetics of pancreatic secretion of i.v.-injected [3H]D-sorbitol. The Figure shows the time course of the radioactivity in the blood plasma (n) and in the pancreatic juice secreted in response to either secretin (2 u./kg) (0) or caerulein (0-5 /zg/kg) (0). The large arrow indicates the injection of the tracer together with cerulein, the small arrows the injection of the secretagogues alone. Each curve is representative of the results obtained in two consistent experiments.
Experiments with pancreas tissue and cell fractions In a previous paper we have reported the distribution of calcium in cell fractions isolated from the pancreas of the guinea-pig (Clemente & Meldolesi, 1975). These results were also corrected for the adsorption artifacts occurring during and after tissue homogenization. The corrected
SECRETION OF CALCIUM IN PANCREATIC JUICE 627 data concerning the fractions which might be involved in calcium secretion are reported in Table 1. It is worth noting that very little calcium is found in the post-microsomal supernatant, whereas a much larger amount is associated with the zymogen granules, the storage organelles containing the secretion products available for discharge. This amount, however (36-0 n-mole/mg protein), is considerably lower than that recovered in the cerulein-stimulated juice (ca. 50 n-mole/mg protein, see Fig. 3). On the other hand, it is likely that not all the calcium found in the zymogen granules is discharged upon exocytosis since ca. 20 % of it appears associated 4
43
.2 E U,
10
30
so
70
90 110 130 150 170 190 210 Time (min) .
Fig. 7. Time course of the juice: blood plasma radioactivity ratios in guineapigs injected with either [3H]D-sorbitol (e-e---) or 45Ca ( ~) and with secretagogue drugs secretin (2 u.lkg) (-) or cerulein (0-5 ,/tglkg) (0). The Figure was drawn on the basis of the results reported in Figs. 4 and 6.
with the membrane and not with the segregated content of the organelle (Clemente & Meldolesi, 1975). Hence, these results might be consistent with the idea that the zymogen granules represent the major, but not the only source of the calcium recovered in the caerulein-stimulated juice. In order to obtain further information on this question we have studied the kinetics of in vivo labelling of the whole pancreas and of pancreatic cell fractions after i.V. injection of 45Ca. Fig. 9 compares the time course of the 45Ca concentration in the pancreatic tissue with that observed in the blood plasma, seen in Fig. 4. It is clear that the amount of radioactivity declines in the tissue almost in parallel with the plasma and reaches a 26-2
628 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI 4
89
T
3 Io
0_1I
6 T_
I
UE 2
o
l
U I-
I ,,
4
J. Physiol. (1975), 245, pp. 617-638 With 10 text-ft gure8 Printed in Great Britain
SECRETION OF CALCIUM IN PANCREATIC JUICE
By B. CECCARELLI, F. CLEMENTE* AND J. MELDOLESI From the C.N.R. Centre of Cytopharmacology, Department of Pharmacology, University of Milan, 20129 Milan, Italy
(Received 11 June 1974) SUMMARY
1. The origin of the calcium secreted by the pancreas has been investigated in vivo in the guinea-pig by a study carried out in parallel (a) in the juice secreted in response to the injection of either secretin or caerulein and (b) in the pancreatic tissue and in cell fractions isolated therefrom. 2. In agreement with previous findings we observed that the concentration of calcium is low in the secretin-stimulated and high in the caeruleinstimulated juice. In the latter calcium and protein are proportional (ca. 50 n-mole: mg). 3. After i.v. injection of 45Ca the radioactivity decreases rapidly and quasi-exponentially in the blood plasma. A roughly parallel time course is found in the secretin-stimulated juice: the evolution of the juice: plasma radioactivity ratio resembles that observed with the extracellular space marker [3H]D-sorbitol. In contrast, the time course of 45Ca in plasma and caerulein-stimulated juice are not proportional: the high levels characteristic of this juice are reached several minutes after the injection and maintained thereafter. This increase is followed ca. 50 min later by the appearance of the newly synthesized [3H]L-leucine-labelled proteins. 4. The pancreatic tissue is rich in calcium which is localized primarily in zymogen granules (ca. 36 n-mole: mg protein) and mitochondria; the soluble cytoplasm is low in calcium. 5. The injected "Ca accumulates in zymogen granules faster than [3H]L-leucine-labelled proteins. The 45Ca : protein ratio of these organelles is considerably lower than that of the caerulein-stimulated juice. 6. It is concluded (a) that calcium is secreted into the pancreatic juice in two fractions, one (possibly released by simple diffusion) associated with the electrolyte component, the other with the protein of the juice, * Present address: INSERM, Groupe de Recherche de Pathologie Digestive, Hopital Pourpan, Toulouse, France.
618 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI (b) that zymogen granules are the major, but not the only source of the latter fraction, and (c) that the zymogen granule-associated calcium joins the exportable proteins some time after their synthesis, possibly in the Golgi complex and/or in the condensing vacuoles. INTRODUCTION
It has been clearly established in several animal species that the calcium ions contained in the pancreatic juice fulfil a number of important functions. Thus, amylase, a juice enzyme, contains one atom of tightly bound calcium per molecule (Stein, Hsiu & Fischer, 1964). Furthermore Ca2+ is necessary for maintaining the molecular configuration of other secretary zymogens, such as trypsinogen, chymotrypsinogen (Delaage & Lazdunski, 1967), prephospholipase (Pieterson, Volwerk & De Haas, 1974) and lipase (Benzonana, 1968), and high concentration of the cation is required for the activation of trypsinogen (Desnuelle & Gabeloteau, 1957; Desnuelle, 1960) and for the activity of trypsin, lipase and phospholipase (Sipos & Merkel, 1970; Sarda, Marchis-Mouren, Constantin & Desnuelle, 1957; De Haas, Bonsen, Pieterson & Van Deenen, 1971). The correct understanding of the mechanisms by which an adequate concentration of calcium is achieved and maintained in the juice represents therefore a problem of great importance in pancreatic physiology. Previous studies carried out in intact animals as well as in in vitro preparations (Goebel, Steffen & Bode, 1972; Argent, Case & Scratcherd, 1973) suggest that calcium reaches the pancreatic juice from at least two different sources. One of these sources appears to be correlated with, and the other independent of the secretary function of the acinar cells of the gland. However, a definite identification of the origin of these two calcium fractions and of the mechanisms of their secretion has not been obtained so far. These problems are considered in the present investigation in which two types of experiments have been carried out in parallel: (a) the analysis of the pancreatic juice secreted in vivo in response to stimulation with either secretin or caerulein (a polypeptide of well known pancreozymin-like activity (Bertaccini, De Caro, Endean, Erspamer & Impicciatore, 1969)), and (b) the subcellular study of the pancreatic tissue, in relation to its calcium secretary activity. We used the guinea-pig, because in this animal species the pancreatic secretary process is well known and because subcellular fractions of high purity can be isolated from the pancreas tissue by means of established techniques (Jamieson & Palade, 1967a, b; Meldolesi, Jamieson & Palade, 1971).
SECRETION OF CALCIUM IN PANCREATIC JUICE 619 METHODS Male albino guinea-pigs weighing 450-500 g (gift of Sigurta Drug Co., Milan, Italy) were fasted 20-24 hr before the experiments. Pancreatic 8ecretion 8tudie8 In guinea-pigs anaesthetized with pentobarbitone the pancreatic duct was cannulated with a soft, plastic tube and the juice collected at various time intervals as indicated under Results. Tracers (45CaCl2 (40 Fuc/kg), [3H]L-leucine + CaCl2 (80 and 40 ,tc/kg, respectively) or [3H]D-sorbitol (50 ,uc/kg)) and secretagogue drugs (caerulein (500 nglkg) or secretin (2 u./kg)) were injected i.v. Experiments with pancreatic tissue and cell fraction Guinea-pigs anaesthetized with pentobarbitone were injected i.v. with either 45CaCl2 (40 /zcckg) + [3H]L-leucine (80 ftc/kg) or with [3H]D-sorbitol (50 ftc/kg) and sacrificed at various times after the injection by a blow over the head. The pancreases were quickly excised and immersed in ca. 50 ml. ice-cold 0-3 M sucrose. After 2 min they were removed, blotted with filter paper, weighed, minced with scissors and homogenized in 10 ml. 0 3 M sucrose as described previously. Fractionation of the homogenates by differential centrifugation was carried out as described by Meldolesi et al. (1971) to yield the following preparations: crude mitochondria, total microsomes, zymogen granules and post-microsomal supernatant. Isolated microsomes and mitochondria were washed once by resuspension in 0 3 M sucrose and recentrifugation; in the case of zymogen granules a washing step is included in the isolation procedure. An extensive characterization of these fractions has been reported previously (Meldolesi et al. 1971).
Radioactivity and chemical asaays Aliquots of the pancreatic homogenates, cell fractions, juice and blood plasma were usually mixed directly with 10 ml. of Packard Instagel and transferred to disposable plastic counting vials. Onlywhen a separate assay of the protein-incorporated 3H-radioactivity was needed, the samples were precipitated at 4° C with 10 % trichloroacetic acid (TCA) and centrifuged. The precipitates were washed as described previously (Meldolesi, 1970). Aliquots of the TCA-soluble and insoluble samples were dissolved in 10 ml. of Packard Instagel and counted. All samples were counted in an Intertechnique SL 30 liquid scintillation spectrometer. In single label counting, the 3H efficiency was over 40 % and "5Ca efficiency over 80 %. In double label counting the conditions were selected in order to have no spill of 3H counts into the 45Ca channel; efficiency of 3H counts was ca. 22 % and of 45Ca ca. 70 %. Spill of "Ca counts into the 3H channel was always less than 5 %. Corrections for quenching and spill were made by external standardization. Protein was determined according to Lowry, Rosebrough, Farr & Randall (1951) using bovine serum albumin as the standard. Calcium was assayed in a Hillger and Watts atomic absorption spectrometer, model Atomspek, after deproteinization with 10 % TCA and washing of the precipitate with 5 % TCA. In order to overcome interferences, all the samples contained 1 % La3+ and were adjusted at pH 1-5 before the assay. Furthermore, all the glassware and plasticware was carefully washed with either 2 N-HCl or 10 mm disodium ethylenediaminetetraacetate (EDTA), pH 7-5 and then rinsed repeatedly with distilled water. The calcium content of the solutions used was carefully determined, and the experimental results were corrected for the impurities when necessary.
620 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI MateriaUl L-[4,5-3H]leucine (sp.act. 32 mclpmole) and D-[1-3H]sorbitol (sp.act. 6-8 mcftmole) were purchased from New England Nuclear, Langen, Germany; 45CaCl2 (3 mc/ molel) from the Radiochemical Centre, Amersham, U.K.; sucrose (Special Enzyme Grade) from Schwarz-Mann, Orangeburg, New York, U.S.A. Secretin was obtained from the G.I.H. Research Unit, Chemistry Department, Karolinska Inst., Stockholm, Sweden; caerulein was the kind gift of the Farmitalia Laboratories for Basic Research, Milan, Italy. All other chemicals were reagent grade. RESULTS
In vivo stimulation of pancreatic secretion Fig. 1 illustrates the secretary response elicited by secretin and cerulein in guinea-pigs bearing an acutely cannulated pancreatic duct. The doses of the secretagogues used were able to provoke maximal secretary responses, as shown in a previous report (Meldolesi, 1970), as well as in preliminary experiments of this work. As expected, the i.v. injection of secretin produced a large increase of the juice output, which lasted for several minutes; the secretin-stimulated juice was very poor in protein. Slightly reduced responses were obtained by repeating the injection of the hormone. The effect of cerulein, on the other hand, was lower than that of secretin in terms of volume and much higher in terms of protein concentration of the juice. The response was more transient and showed a faster decrease on repeating the stimulation. Fig. 2 shows a representative example of the time course of calcium secretion induced by the two polypeptides. The data are plotted as a rate of release (upper panel) and concentration (lower panel). If one compares the results presented in this Figure it is clear that the cumulative calcium release is larger after stimulation with secretin, whereas the calcium concentration in the juice is always much higher with caerulein. The statistical analysis revealed that no correlation exists between calcium and protein concentration in the secretin-stimulated juice, whereas a linear correlation exists after stimulation with caerulein (Fig. 3). However, the slope of the regression line fitted to the data indicates that at 0 protein a small amount of calcium (ca. 07 ,umole: ml.) is still contained also in the caerulein-stimulated juice (Fig. 3). In order to obtain some information on the origin of the calcium recovered in the pancreatic juice, the kinetics of secretion of i.v. injected 45Ca was investigated. In Fig. 4 the time course of the radioactivity in the blood plasma (upper panel) is compared with that observed in the pancreatic juice secreted in response to repeated stimulation with either secretin or cerulein (lower panel). The curves are quite different. In the
SECRETION OF CALCIUM IN PANCREATIC JUICE 621
3 C
E E C
0 0. U
1
220 240 260 280
il1 C Z._
El 0
E 0 41 >
U
._% 11
10203040 60 80 100 120 140 160 180 200 220 240 260 280
$
t
I Time (min)
Fig. 1. Effect of secretin (0) and caerulein (0) on the protein output (upper panel) and on the volume (lower panel) of the pancreatic juice in guinea-pigs bearing an acutely cannulated pancreatic duct. The arrows represent single injections of either secretin (2 u./kg) or caerulein (0 5 ,jg/kg). In this, as well as in the following Figures which concern the pancreatic juice (Figs. 2, 4, 5, 6, 7 and 8), the indicated time points correspond to the times at which the juice samples were collected. Thus, the corresponding values pertain to the juice secreted during the interval preceding the time indicated in the Figure. Each curve shown is representative of the results obtained in seven consistent experiments.
622 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI plasma the concentration of the 45Ca declines rapidly within a few minutes after the injection and then subsides more gradually. In the secretinstimulated juice an initial decrease is also evident; however, it is smaller
g 50 0
E
E R 25
U
IV
0
E
E
*2: U
3-
20 40
t
60 80 100 120 140 160 180 200
I
I
Time (min) Fig. 2. Effect of secretin (e) (2 u./kg) and caerulein (0) (0-5 /zg/kg) on the rate of secretion (upper panel) and the concentration (lower panel) of calcium in the pancreatic juice. The arrows represent single injection of either secretagogue. Each curve shown is representative of the results of six consistent experiments.
than that observed in the plasma. In the caerulein-stimulated juice the curve is completely different: during the first 30 min the concentration of the tracer remains approximately at the level of the first sample of the secretin-stimulated juice. Between the 30th and the 50th min the radioactivity approximately doubles and remains at this high level (even if with distinct fluctuations) for at least 3 hr.
SECRETION OF CALCIUM IN PANCREATIC JUICE 623 In another set of experiments (Fig. 5) the evolution of the radioactivity was followed in the juice several hours after the i.v. injection of 45Ca, i.e. at a time when the concentration of the tracer remains relatively constant in the plasma. The results of these experiments corroborate our previous findings in the sense that the concentration of the tracer in the secretinstimulated juice remains relatively stable, at a level one third to one v=54
0 0
0
r=087 7.5
0 0
.2
~~ ~~00 0
E
205 _ E
2*5
OdM
0
o
00 0
0000
50
0
0
o
0~
0
00
0
100 mg protein/mi. juice
150
Fig. 3. Correlation between the concentration of calcium and of protein in the pancreatic juice secreted in response to the i.v. injection of caerulein (0.5 jug/kg). The solid line is the calculated regression line (P < 0.001). v = degrees of freedom; r = coefficient of correlation.
quarter of that of the plasma, whereas in the caerulein-stimulated juice the concentration is four- to eightfold higher. In these animals, the switching from one type of secretion stimulation to the other results in a rapid shift in the level of the juice radioactivity (Fig. 5, right-hand side). Since it was important to establish whether or not the secretion of calcium is correlated with the secretion of other components of the pancreatic juice, experiments were carried out in this direction. Thus, the kinetics of 45Ca
624 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI pancreatic secretion was compared with that of other tracers: [3H]D-sorbitol, a molecule which in different tissues, including the pancreas, is known to penetrate very slowly within the cells and therefore can be considered as an extracellular tracer (Goodford & Leach, 1966; Van Venroij, Poort,
15p-
10
C
5p
4)
0 U L.
I-
0 f ._ I
EE a.
4p
:0 EM
3
xa
2
1
10 I
5
a
50 30
f
210
150 I
.
I
*
*
70 90 110 130
*
f
I
I
170 190
I
I
t
230
Time (min)
Fig. 4. Short interval kinetics of pancreatic secretion of i.v.-injected 45Ca. The Figure shows the time course of the radioactivity in the blood plasma (m) (upper panel) and in the juice secreted in response to either secretin (0) (2 u./kg) or caerulein (0) (05/jg/kg) (lower panel). The large arrow marks the injection of the tracer (40,uc/kg) together with either one of the secretagogue drugs; the small arrows mark the injection of the secretagogue drugs alone. The curve shown is representative of the results obtained in four consistent experiments.
SECRETION OF CALCIUM IN PANCREATIC JUICE 625 Kramer & Jansen, 1972), and [3H]L-leucine, used here as a precursor of the juice proteins. The i.v.-injected rH]D-sorbitol was found to appear after a short lag of time in the juice elicited by the injection of both caerulein (Fig. 6) and secretin (not shown). The time courses of the decline of the r3H]D-sorbitol concentrations in juice and plasma run almost in parallel, with a ratio of ca. 0 5, irrespective of the stimulant used (Fig. 6). The juice: plasma radioactivity ratios observed after injection of either 6
.40
E
an' 2
-
\
I
40
80
f
120
160
200
tf
240
f
280
f
320
360
Time (min)
Fig. 5. Kinetics of pancreatic secretion of 45Ca injected several hours before the experiment. Animals received one i.v. injection of the tracer (40 fc/kg). After 3 hr their pancreatic duct was cannulated and 1 hr later the stimulation of secretion with either secretin (e) (2 u./kg) or caerulein (0.5 ,tg/kg) (Q) was started. The arrows indicate single injections of the secretagogues. Each curve shown is representative of the results obtained in two consistent experiments.
[3H]D-sorbitol or 45Ca and stimulation of secretion with secretin and caerulein are plotted in Fig. 7 as a function of time. It is clear that the time course of the ratios concerning the 45Ca in the secretin-stimulated juice is almost parallel to the VH]D-sorbitol curve (especially at the early time points of the experiment), whereas no correlation is evident between the latter curve and the 45Ca curve concerning the caerulein-stimulated juice. In this case the 45Ca ratios are lower than the corresponding PH]Dsorbitol ratios at the beginning of the experiment; after the 30th min they rise progressively to very high values (ca. 3) indicating that the bulk of the calcium secreted in this juice is not in equilibrium with the plasma. Finally, experiments were carried out to investigate the correlation between the appearance in the juice of 45Ca and that of newly synthesized 26
P HY 245
626 B. CECCARELLI, P. CLEMENTE AND J. MELDOLESI exportable protein. Animals were injected simultaneously with 45Ca, [H]L-leucine and with either one of the stimulatory drugs. In both cases the 3H-labelled proteins appeared in the juice after a lag of about 50 mmi, and reached the maximum concentration after 130 min. No correlation was found between the 3H-protein curve and the 45Ca curve in the secretinstimulated juice (not shown in Figures), whereas in the cerulein-stimulated juice the 3H-protein data fall along a line which seems to follow the increase in "5Ca radioactivity but with a delay of ca. 80 min (Fig. 8). 8 co (U
*6
.:
6E
A
1
4E 2_
Xi
III
510 20
40
I
60
80
I
100
I
120
I
150
I
170
Time (min) Fig. 6. Kinetics of pancreatic secretion of i.v.-injected [3H]D-sorbitol. The Figure shows the time course of the radioactivity in the blood plasma (n) and in the pancreatic juice secreted in response to either secretin (2 u./kg) (0) or caerulein (0-5 /zg/kg) (0). The large arrow indicates the injection of the tracer together with cerulein, the small arrows the injection of the secretagogues alone. Each curve is representative of the results obtained in two consistent experiments.
Experiments with pancreas tissue and cell fractions In a previous paper we have reported the distribution of calcium in cell fractions isolated from the pancreas of the guinea-pig (Clemente & Meldolesi, 1975). These results were also corrected for the adsorption artifacts occurring during and after tissue homogenization. The corrected
SECRETION OF CALCIUM IN PANCREATIC JUICE 627 data concerning the fractions which might be involved in calcium secretion are reported in Table 1. It is worth noting that very little calcium is found in the post-microsomal supernatant, whereas a much larger amount is associated with the zymogen granules, the storage organelles containing the secretion products available for discharge. This amount, however (36-0 n-mole/mg protein), is considerably lower than that recovered in the cerulein-stimulated juice (ca. 50 n-mole/mg protein, see Fig. 3). On the other hand, it is likely that not all the calcium found in the zymogen granules is discharged upon exocytosis since ca. 20 % of it appears associated 4
43
.2 E U,
10
30
so
70
90 110 130 150 170 190 210 Time (min) .
Fig. 7. Time course of the juice: blood plasma radioactivity ratios in guineapigs injected with either [3H]D-sorbitol (e-e---) or 45Ca ( ~) and with secretagogue drugs secretin (2 u.lkg) (-) or cerulein (0-5 ,/tglkg) (0). The Figure was drawn on the basis of the results reported in Figs. 4 and 6.
with the membrane and not with the segregated content of the organelle (Clemente & Meldolesi, 1975). Hence, these results might be consistent with the idea that the zymogen granules represent the major, but not the only source of the calcium recovered in the caerulein-stimulated juice. In order to obtain further information on this question we have studied the kinetics of in vivo labelling of the whole pancreas and of pancreatic cell fractions after i.V. injection of 45Ca. Fig. 9 compares the time course of the 45Ca concentration in the pancreatic tissue with that observed in the blood plasma, seen in Fig. 4. It is clear that the amount of radioactivity declines in the tissue almost in parallel with the plasma and reaches a 26-2
628 B. CECCARELLI, F. CLEMENTE AND J. MELDOLESI 4
89
T
3 Io
0_1I
6 T_
I
UE 2
o
l
U I-
I ,,
4
