Secretion of 5-hydroxytryptamine from electropermeabilised human platelets. Effects of GTP and cyclic 3',5'-AMP

Secretory granule exocytosis

Regulated exocytosis of secretory granules or dense-core granules has been examined in many well-characterized cell types including neurons, neuroendocrine, endocrine, exocrine, and hemopoietic cells and also in other less well-studied cell types. Secretory granule exocytosis occurs through mechanisms with many aspects in common with synaptic vesicle exocytosis and most likely uses the same basic protein components. Despite the widespread expression and conservation of a core exocytotic machinery, many variations occur in the control of secretory granule exocytosis that are related to the specialized physiological role of particular cell types. In this review we describe the wide range of cell types in which regulated secretory granule exocytosis occurs and assess the evidence for the expression of the conserved fusion machinery in these cells. The signals that trigger and regulate exocytosis are reviewed. Aspects of the control of exocytosis that are specific for secretory granules compared with synaptic vesicles or for particular cell types are described and compared to define the range of accessory control mechanisms that exert their effects on the core exocytotic machinery.

Characterization of small-molecular-mass guanine-nucleotide-binding regulatory proteins in insulin-secreting cells and PC12 cells

The distribution of ras-related small-molecular-mass guanine-nucleotide-binding regulatory proteins (SMG) of two insulin-secreting cell lines, RINm5F and HIT-T15, and of a catecholamine-secreting cell line, PC12, have been studied using different techniques. About ten such proteins were detected by [32P]GTP binding after two-dimensional gel electrophoresis and transfer to nitrocellulose membranes. In insulin-secreting cells, rho protein(s) that cannot be detected with the GTP-binding technique were identified by ADP ribosylation with Clostridium botulinum C3 exoenzyme. After subcellular fractionation, SMG displayed specific distributions. The insulin-secreting cell line RINm5F and the catecholamine-secreting cell line PC12 expressed a similar set of these proteins with analogous localization. [32P]GTP binding analysis revealed that at least seven SMG were associated with the secretory granule enriched fraction of RINm5F cells and with the fraction containing dense secretory granules from PC12 cells, proteins of 27 (pI 5.4), 23 (pI 6.8) and 25 kDa (pI 6.7) being the most abundant. These proteins were present in a highly purified granule fraction of a solid rat insulinoma. The 23 kDa (pI 6.8) and 25 kDa (pI 6.7) proteins, but not the protein migrating at 27 kDa (pI 5.4), were detected in the corresponding fraction from HIT-T15 cells. A monoclonal antibody directed against smg25A/rab3A recognized the SMG in secretory granules migrating at 25 kDa (pI 6.7) and 27 kDa (pI 5.4). This antibody also revealed the presence of such protein(s) in homogenates of rat pancreatic islets. During stimulation of insulin secretion of either intact or permeabilized cells, there was no detectable redistribution to the cytosol or to the plasma membrane of the major proteins located on secretory granules. In view of the invariable presence of at least two of the SMG in granules of secretory cells, these proteins are good candidates for regulation of hormone secretion.

Okadaic acid inhibits activation of phospholipase C in human platelets by mimicking the actions of protein kinases A and C

1. The effect of okadaic acid, a potent inhibitor of protein phosphatases 1 and 2A (PP1 and PP2A), on human platelets has been investigated. 2. Okadaic acid exerts a general increase in phosphorylation of platelet proteins but did not induce aggregation or secretion of 5-hydroxytryptamine (5-HT). Okadaic acid, however, did inhibit thrombin-induced functional responses. 3. Maximally effective concentrations of prostacyclin, to elevate adenosine 3'-5'-cyclic monophosphate (cyclic AMP), or phorbol dibutyrate, to activate protein kinase C, inhibited the formation of inositol phosphates by thrombin by approximately 60%. When used in combination, prostacyclin and phorbol dibutyrate reduced the levels of inositol phosphates induced by thrombin to 11%. 4. Okadaic acid (1 microM) decreased thrombin-induced formation of inositol phosphates by approximately 55% and increased the inhibitory action of prostacyclin or phorbol dibutyrate. Okadaic acid had no further effect when prostacyclin and phorbol dibutyrate were used in combination. 5. These results suggest that protein kinases A and C act to inhibit phospholipase C by distinct mechanisms and that their action is reversed by PP1 and/or PP2A.

Dual effects of guanosine 5'-[gamma-thio]triphosphate on secretion by electroporated human neutrophils

It is generally believed that G-proteins play stimulatory roles on cell activation. In contrast, we found that guanosine 5'-[gamma-thio]triphosphate (GTP[S]) was a potent inhibitor of Ca(2+)-induced secretion from specific granules (as monitored by vitamin B-12-binding protein). GTP[S] inhibition of specific-granule release occurred in the presence or absence of adenine nucleotides, required Mg2+ (1-3 mM), and was half-maximal at 30 microM-GTP[S]. The dual stimulatory and inhibitory effects of GTP[S] could be readily observed and differentiated when degranulation was monitored over a range of Ca2+ concentrations. Inhibition of specific-granule release by GTP[S] was observed at low Ca2+ concentrations and resulted from shifting the Ca2+ dose-response curves to the right. In contrast, GTP[S] promoted azurophil-granule secretion at relatively high concentrations of Ca2+ and appeared to be due to a general enhancement at all Ca2+ concentrations. A series of hydrolysable and non-hydrolysable nucleotides did not mimic GTP[S] or block its action. Inhibition by GTP[S] occurred in cells which were sensitized with a protein kinase C agonist, suggesting that inhibition of secretion took place distal to this enzyme. However, the inhibitory effects of GTP[S] on specific-granule secretion were reversed by cytochalasin D, which prevents new microfilament formation; this compound also enhanced the stimulation of azurophil-granule release by GTP[S]. We also found that GTP[S] greatly increased the F-actin content of permeabilized neutrophils, whereas Ca2+ (to a lesser extent) decreased F-actin. These data are consistent with the hypothesis that at least two G-proteins are involved in regulating secretion: one which has been previously described as stimulating Ca(2+)-induced secretion (particularly from azurophil granules) and a second, possibly involved in promoting microfilament assembly, which inhibits the discharge of specific granules.

Stimulation of Ca2(+)-independent catecholamine secretion from digitonin-permeabilized bovine adrenal chromaffin cells by guanine nucleotide analogues. Relationship to arachidonate release

The effect of GTP analogues on catecholamine secretion and [3H]arachidonic acid release from digitonin-permeabilized adrenal chromaffin cells was examined. Several GTP analogues stimulated Ca2(+)-independent exocytosis, with the order of efficacy being XTP greater than ITP greater than guanosine 5'-[beta gamma-imido]triphosphate (p[NH]ppG) greater than guanosine 5'-[gamma-thio]triphosphate (GTP[S]). The stimulatory effect of the GTP analogues appeared to be due to activation of a conventional GTP-binding protein, as it was inhibited by guanosine 5'-[beta-thio]diphosphate (GDP[S]). In contrast, Ca2(+)-dependent exocytosis was only partially inhibited by high doses of GDP[S]. GTP did not stimulate Ca2(+)-independent exocytosis, but instead was found to inhibit secretion caused by micromolar Ca2+. Arachidonic acid (100 microM) also stimulated Ca2(+)-independent catecholamine secretion. Determination of the effect of GTP analogues on release of free [3H]arachidonic acid into the medium showed that it was stimulated by GTP[S] but inhibited by GTP, p[NH]ppG, ITP and XTP. The inhibition of [3H]arachidonic acid release by XTP was not prevented by GDP[S]. These results demonstrate that activation of a GTP-binding protein by certain GTP analogues can induce Ca2(+)-independent secretion in adrenal chromaffin cells and that the effect of GTP analogues on Ca2(+)-independent secretion can be dissociated from generation of arachidonic acid.

Guanine nucleotides and Ca2(+)-dependent lysosomal secretion in electropermeabilised human platelets

1. Metabolically stable analogues of GTP, e.g. guanosine 5'-[gamma-thio]triphosphate (GTP[S]) and guanosine 5'-[beta,gamma-imido]triphosphate (pp[NH]pG), enhance the extent of Ca2(+)-dependent secretion of beta-N-acetylglucosaminidase and beta-galactosidase from electropermeabilised human platelets in the presence of less than 5 microM Ca2+. A similar effect is observed on addition either of 1,2-dioctanoin or of GTP in in the presence or absence of thrombin. 2. In the presence of higher Ca2+ concentrations the extent of enhancement of lysosomal secretion declines and little, or no, enhancement is observed at a [Ca2+] of 30-40 microM. Addition of leupeptin or antipain prevents this decrease in lysosomal secretion and enhances the extent of Ca2(+)-dependent lysosomal secretion obtained in the presence or absence of guanine nucleotides, thrombin or 1,2-dioctanoin. 3. The concentration of GTP[S] or pp[NH]pG required to obtain half-maximal enhancement of lysosomal secretion is dependent on [Ca2+] for secretion of 5-hydroxytryptamine, beta-N-acetylglucosaminidase and beta-galactosidase. At two fixed [Ca2+] the median effective concentration (EC50) values for GTP[S] and pp[NH]pG which characterise enhancement of 5-hydroxytryptamine secretion are significantly different from those characterising enhancement of the secretion of beta-N-acetylglucosaminidase and beta-galactosidase. 4. In the presence of a saturating concentration of GTP[S] marked 5-hydroxytryptamine and beta-N-acetylglucosaminidase secretion is observed at nanomolar [Ca2+] and these responses show little dependence on [Ca2+] over the attainable range. Secretion of beta-N-acetylglucosaminidase is also induced at nanomolar Ca2+ concentrations by addition of activators of protein kinase C. 5. Guanosine 5'-[beta-thio]diphosphate inhibits enhancement of beta-N-acetylglucosaminidase secretion induced by GTP[S] but has no effect on secretion of this enzyme induced by Ca2+ when added alone. 6. Our data provide some support for a model in which addition of metabolically stable guanine nucleotides enhances Ca2(+)-dependent platelet lysosomal secretion by activating a guanine-nucleotide-binding protein (GE) located close to the exocytotic site. However, not all the data are consistent with this postulate.

Ca2(+)-induced secretion by electropermeabilized human neutrophils. The roles of Ca2+, nucleotides and protein kinase C

Studies of stimulus-response coupling have benefitted from the availability of permeabilization techniques, whereby putative second messengers and intracellular modulators can be introduced into the cell interior. Electropermeabilization, which uses high-intensity electric fields to breach the plasma membrane, creates small pores, permitting access of solutes with molecular masses below 700 KDa. Neutrophils permeabilized by this technique, but not intact cells, discharged lysosomal constituents when exposed to micromolar levels of Ca2+. Secretion by electroporated neutrophils was significantly enhanced by the presence of Mg-ATP (0.3-1.0 mM). Contrary to expectations, it was determined that ATP was not the only nucleotide which enhanced Ca2(+)-induced secretion in the presence of Mg2+. Not only could GTP, XTP, ITP, UTP or ADP partially or completely replace ATP, but even non-hydrolyzable nucleotides such as ADP beta S ATP gamma S, and App[NH]p were effective. GTP gamma S and GDP beta S were inhibitory, while Gpp[NH]p was inactive. None of these nucleotides induced secretion on its own. In contrast, neutrophils which were permeabilized and then washed, were only slightly activated by Mg-ATP and other nucleotides; even the response to Ca2+ alone was less. This hyporesponsiveness of washed cells proved to be due to a time-dependent deactivation of the permeabilized neutrophils taking place at 4 degrees C. In an effort to assess the role for protein kinase C (PKC) in secretion in this system, we examined the effects of phorbol myristate acetate (PMA), a PKC agonist. PMA enhanced degranulation induced by Ca2+ by lowering the requirement for this divalent cation; enhancement by PMA was not dependent upon exogenous ATP. Three inhibitors of PKC with varying specificity, namely H-7, K-252a, and staurosporine, all abrogated PMA-enhanced secretion. These agents also inhibited secretion stimulated by Ca2+ plus ATP in parallel with that induced by Ca2+ plus PMA, strongly suggesting a role for PKC in modulation of degranulation by ATP. Our results show that electropermeabilized neutrophils provide a convenient, useful model for stimulus-secretion coupling. These data also suggest that the 'requirement' for Mg-ATP, which has been observed in other permeabilized cell systems, is not simply for metabolic energy or as a substrate for kinases. It is possible that these nucleotides all interact with a recently described neutrophil receptor for adenine nucleotides or with a recently postulated exocytosis-linked G-protein.

Calcium-dependent and calcium-independent exocytosis

A large body of evidence supports the concept that calcium (Ca2+) plays a pivotal role in the control of exocytosis. However, recent experiments suggest that a rise in intracellular Ca2+ does not necessarily trigger secretion, and also that secretion can occur independently of cytosolic free calcium levels. This article briefly summarizes the early evidence that has formulated the role of Ca2+ in secretion, and then examines some of the recent evidence suggesting a Ca2+-independent mechanism of exocytosis.

The control of cytoskeletal actin and exocytosis in intact and permeabilized adrenal chromaffin cells: role of calcium and protein kinase C

The control of cytoskeletal actin and exocytosis was examined in intact and digitonin-permeabilized chromaffin cells. Cytoskeletal actin was assayed by determining the actin content of Triton-insoluble cytoskeletons. The secretagogues nicotine, high K+ and Ba2+ resulted in a rapid reduction in the amount of actin associated with the cytoskeleton. The effect of nicotine but not high K+ on cytoskeletal actin was independent of external Ca2+ and the reduction in cytoskeletal actin was mimicked by the phorbol ester 12-O-tetradecanoylphorbol-13-acetate suggesting a role for protein kinase C. In digitonin-permeabilized cells micromolar calcium produced both catecholamine secretion and a reduction in cytoskeletal actin. The reduction in cytoskeletal actin was transient. Secretion was enhanced by the GTP analogue guanosine 5'-(3-O-thio)triphosphate and the analogue also reduced cytoskeletal actin at low calcium levels. The effects of guanosine 5'-(3-O-thio)triphosphate were inhibited by the phospholipase C inhibitor neomycin and were mimicked by 12-O-tetradecanoylphorbol-13-acetate. An additional GTP analogue, guanyl-5'-yl imidodiphosphate, had no effect on cytoskeletal actin. These results provide further evidence for a requirement for reorganisation of cortical actin in the secretory processes and suggest that the reduction in actin associated with the cytoskeleton may be mediated by protein kinase C and/or calcium in intact and permeabilized chromaffin cells.