CaCO3 is a major constituent of pancreatic stones, salivary stones, and many pigment gallstones. Elucidation of the physicochemical state of calcium is necessary for definition of calcium solubility in these systems. Pancreatic stones are observed in both humans and cattle, and are approximately 95% CaCO3 (calcite) in both species. Despite its importance, little is known about the physicochemical state of calcium in pancreatic juice. This paper presents an a priori model, based on established physicochemical principles, for the state of calcium in the juice at all levels of secretion. Two postulates of the model are the following: the limiting free [Ca2+] in the juice is governed by the solubility product (K' sp) for CaCO3; if K' sp is exceeded, the juice is supersaturated and precipitation of CaCO3 is thermodynamically possible; total calcium, [Ca], in the juice is the sum of four distinct species: free ionized calcium, Ca2+; calcium-bicarbonate complex, CaHCO3+; calcium carbonate ion-pair, CaCO3(0); and protein-bound calcium, CaProt. Overall equations of the model and graphical corollaries are presented. The model predicts an inverse hyperbolic relationship between [Ca2+] or [Ca] and [HCO3-]. Calcium solubility is maximal at low [HCO3-]; as [HCO3-] increases, both [Ca2+] and [Ca] decline to respective limiting values of approximately 0.015 and 0.15 mM. At low [HCO3-], most of [Ca] is present as Ca2+ and CaProt, whereas at high [HCO3-], most [Ca] is CaHCO3+ and CaCO3(0). Protein, HCO3-, and CO3(2-) ions are thus important buffers for Ca2+ in the juice. The model provides a quantitative framework for further elucidation of calcium lithogenicity in the pancreas, salivary gland, and biliary tract.

Pancreatic calcification is a frequent complication of chronic pancreatitis, and pancreatic stones (95% CaCO3 as calcite) are observed in both humans and cattle, but little is known about the complex equilibriums governing calcium solubility. Using the Ca2+ electrode and equilibrium dialysis of NaHCO3-CaCl2-NaCl solutions (24 +/- 2 degrees C; total ionic strength = 0.153-0.161), studies were made at variable pH and total calcium and carbonate concentrations to determine the formation constants of the three possible calcium complexes: CaHCO3+, CaC03(0), and Ca(HCO3)20. If the first two complexes were present, a plot of ([Ca]/[Ca2+]-1)/[HCO3-] against the ionic ratio [C03(2)-]/[HCO3-] should be linear with intercept of the CaHCO3+ formation constant (K'aB) and slope of the CaC03(0) formation constant (K'aC). This was found to be the case in both dialysands and dialysates, using two different methods ("pH" and "K'sp") for estimation of [CO32-] values. Two other methods of data analysis were also used: simultaneous equations and multiple linear regression by matrix analysis. For all three methods, mean K'aB = 12.48 and mean K'aC = 1,870. There was no evidence for a Ca(HCO3)2(0) complex. We conclude that CaHCO3+ and CaC03(0) complexes may account for a substantial fraction of total soluble calcium is pancreatic juice. These studies provide a necessary step toward the construction of a quantitative physicochemical model of pancreatic calcium solubility.

The effects of repeated intravenous calcium administration on pancreatic juice secretion were investigated in four Thomas fistula dogs. During stimulation by 1.0 U kg-1h-1 GIH secretin, three Ca doses were administered: 2, 4 and 8 mu mol kg-1 min-1 during 1 h, saline being used in control tests; one dose only was tested per day. It was found that Ca administration induced both acute and long-lasting effects. Acute effects were characterized by an increased response to secretin stimulation. Fluid, HCO3(-), protein and Ca outputs increased significantly in a dose-dependent manner, the increase of protein output being the most dramatic. Long-lasting effects, until now unrecognized, were characterized by a progressive increase of protein secretion during the first hour of secretin stimulation. This increase kept going during the 3 months of repeated calcium injections. Although protein plugs were observed in the juice, sometimes stopping the flow of juice, no pancreatic lesion was found. A second protocol showed that, after discontinuing calcium injections, the long-lasting effects decreased progressively, but protein hypersecretion was still significant 3.5 months later. The importance of these findings regarding chronic pancreatitis due to hyperparathyroidism is discussed.

To determine the solubility of pancreatic stones, in vitro experiments were performed by using the anaerobic percolation method. During steady-state conditions an inverse relationship between size and solubility in 150 mM NaCl was observed. The effects of several solvents on pancreatic stones solubility were studied: albumin induced a concentration-dependent increase, while bicarbonate induced a concentration-dependent decrease. Citrate dramatically increased the solubility of pancreatic stones, an increase which was not antagonized by adding bicarbonate. Total in vitro dissolution of 50 mg pancreatic stones was attempted by percolating 150 mM NaCl or 3.9 mM citrate, a concentration which can be reached in pancreatic juice during intraduodenal infusion of citrate. Extrapolated time of total dissolution was 60 days with NaCl percolation, while actual time with citrate percolation was 25 days. The potential usefulness of citrate treatment of chronic calcified pancreatitis is discussed.

Two major proteins have been identified in sodium citrate extracts of bovine pancreatic stones from 15 glands with lithiasis. They were found to have a molecular weight of about 24 000 and were further characterized by a variety of methods, including polyacrylamide-gel electrophoresis in the presence of sodium dodecyl sulphate, isoelectric focusing, two-dimensional electrophoresis, immunodiffusion, immunoelectrophoresis and determination of N-terminal residues. These two immunologically and electrophoretically different proteins were definitely shown to be immunoreactive forms of anionic and cationic trypsinogens, which are normal components of pancreatic juice. However, in contrast with both secretory trypsinogens, the stone proteins displayed an important charge heterogeneity under isoelectric-focusing conditions. A possible role for both secretory trypsinogens in pancreatic lithogenesis is suggested by the reproducibility of the data. Finally, two minor proteins with a lower molecular weight (about 11 000-13 000) have also been found to be present in all extracts, but have not yet been identified.