Molecular mechanism of magnesium transport in epithelial cells

Abstract

Gastrointestinal secretions are controlled mainly by the gut hormones and by the neurotransmitters released by the autonomic nervous system. The hormones and neurotransmitters (secretagogues) utilise different intracellular mediators to elicit enzyme and fluid secretion. One particular mediator is the second messenger calcium (Ca2+) and there is now much evidence that the abundant divalent cation magnesium (Mg 25 may play an important physiological role in regulating the mobilisation of cellular Ca 2t Thus, the present study was designed to investigate (a) the effect of a modification of extracellular Mg2 on secretagogue-evoked enzyme and hydrochloric acid (HCI) secretion and Ca 2 homeostasis in the pancreatic acinar cells and parietal cells and (b) to characterise the molecular mechanism of Mg 2 transport using fluorimetric and spectroscopic studies.
The results have shown that application of either cholecystokinin-octapeptide (CCK-8) or acetylcholine (ACh) to rat pancreatic segments can result in marked increases in amylase and trypsinogen output in normal (1.1 mlvi) extracellular magnesium [Mg 2 0. When [Mg2 0 was elevated to 10 mlvi, there was a significant (P C 0.05) decrease in secretagogue-evoked pancreatic enzyme secretion. On the other hand, in low (0 mM) [Mg2 0 both CCK-8 and ACh elicited marked increases in enzyme secretion similar to the responses obtained in the presence of normal [Mg 2 o . The effects closely correlate with the concurrent reductions and increases in intracellular free calcium concentrations 2 +-.
[Ca j ifrom studies performed with fljra-2-AM loaded pancreatic acmi dunng perturbation of [Mg2 0 These findings indicate that Mg 24 can influence enzyme secretion by regulating Ca24 mobilisation. The possible site of action of Mg2 in controlling Ca24 appears to be at the level of Ca 24 influx, since experiments, using thapsigargin or ionomycin, agents which release Ca2+ from intracellular Ca2+ stores, were not affected by a variation in [Mg 2 0 .
This study also employed the technique of microspectrofluoritnetry to fi.irther characterise the mechanism of Mg2 transport using mouse pancreatic acinar cells loaded with magfisra-2-AM. Stimulation of acini with CCK-8 evoked an initial sharp rise in [Mg 2 1 followed by a decrease to a new steady-state level (Mg 24 efflux). On removal of CCK-8 [Mg2 1 returned to the pre-stimulated basal level (Mg 24 reuptake). In contrast, CCK-8 gave rise to Ca24 oscillations. When acinar cells were co-loaded with 1 ,2-bis (2- aminophenoxy) ethane-N,N,N',N'-tetraacetic acid (BAPTA) (10 piM) and either magflira-2-AM or fijra-2-AM, CCK-8 evoked only the normal decrease in [Mg 2 j and a slight [Ca2 j+ ,e levation. (10 nM). This . results suggests that the in. i.t ial ns. e in. [Mg 2+-. ji seen with CCK-8 in normal conditions may be due to the Ca 24 interfering with the magfura-2-AM signal. Both thapsigargin (0.5 pM) and ionomycin (5 piM) evoked a marked decrease in [Mg 2+,. . . . 2-i-,jj in magfiira-2-AM loaded acmar cells and an elevation in [Ca j' from fi.ira-2- VAM loaded acinar cells. The results indicate that cytosolic Ca 2 is associated with secretagogue-induced decrease in [Mg 24]1. When acinar cells were pre-treated with either forskolin (10 pM), N Nitro-L-arginine (N-NLA; 2 mM), 8 -Bromo guanosine guanosine cyclic mono-phosphate (Br cGMP; 100 pM) or staurosporine (1 pM), CCK-8 elicited only a small decrease in [Mg24]j compared to a much larger response with CCK-8 alone. In contrast, genistein (10 pM) and 12-0-tetradecanoyl phorbol 13 a acetate (TPA; 1 pM) augmented the decrease in [Mg 24]1 evoked by CCK-8. The results indicate that Ca2 mobilising secretagogues can stimulate Mg 2 efflux which is also mediated by a number of intracellular mediators.
This study also investigated the mechanism of Mg 2 transport from the rat pancreas. Permeabilised pancreatic acini were loaded with Mg 2 ' by employing a high Mg2 (12 mM) buffer containing the tonophore A23 187 (6 .dv1). Net Mg 2+ efflux was measured by using the technique of atomic absorption spectrophotometry. Incubation of pre-loaded acini in a buffer deficient in Mg2 resulted in a large and time-dependent release of Mg 2 with maximal efflux occuning within 40 min. Pre-treatment of loaded acini with either bumetadine, SITS or ouabain had no significant effect on Mg2 efflux. In contrast, when acini were pre-treated with either 10 mM dinitrophenol (DNP), io M amiloride, 1 mM lidocaine or I mM quinidine there were significant (P <0.05) decreases in net Mg 2 efflux. Replacement of extracellular sodium [Na4]0 with either N -methyl-D-glucamine (NMDCI), or choline chloride resulted in a significant (P C 0.05) inhibition of Mg2 effiux. The results of this study indicate that Mg 2 transport (efflux) in rat pancreatic acinar cells may not be associated with the N atKtATPase, the NatKtCl cotransporter or the anion exchanger, but with a Na+ -sensitive Mg2+ transport system.
However, this Na+ sensitivity was found to be species dependent as in mouse pancreatic acinar cells Mg2 transport occurred in the absence of [NC] 0 . Studies performed on rat gastric panetal cells have demonstrated that elevated [Mg 2 10 has the same inhibitory effects on secretagogue-evoked acid secretion and cellular Ca 2 transport as that observed with pancreatic acinar cells whereas low [Mg 2 0 had the opposite effect.
In conclusion, the results of this study have demonstrated marked interactions between the two divalent cations Mg2 and Ca2 in epithelial secretory cells of the gastrointestinal tract. Mg2+ seems to regulate secretagogue-evoked secretion by controlling cellular Ca2+ mobilisation.