Analysis of regulated exocytosis in adrenal chromaffin cells: insights into NSF/SNAP/SNARE function

Many of the proteins that function in regulated exocytosis have now been identified. Several proteins form part of a conserved core machinery that acts in many intracellular vesicular fusion steps and their essential roles confirmed by molecular genetic analysis. In addition, studies with adrenal chromaffin and PC12 cells have demonstrated the function of various proteins in regulated exocytosis and have permitted dissection of the stages of exocytosis in which they act. N-Ethylmaleimide-sensitive fusion protein (NSF) and soluble NSF attachment proteins (SNAPs) are key proteins in exocytosis. Examination of their function has indicated that they have a predocking role most likely as molecular chaperones to prepare the docking/fusion machinery. The exact site and time of action in exocytosis of many of the other identified proteins are unknown. A major emphasis for the future will be analysis of the molecular physiology of regulated exocytosis to permit the assignment of functions to identified proteins in particular stages of the regulated exocytotic pathway. BioEssays 20 :328-335, 1998.© 1998 John Wiley & Sons, Inc.     

Loss of proteins from digitonin-permeabilized adrenal chromaffin cells essential for exocytosis

Cultured chromaffin cells can be permeabilized with digitonin; the cell interior is then accessible to the cytoplasm, and addition of calcium provokes release of catecholamines. Increasing the incubation time between the permeabilization step and calcium-induced stimulation resulted in a progressive inhibition of secretion reaching 60% after 20 min. Cytosoluble proteins which leak from detergent-permeabilized cells were collected, dialyzed, and concentrated. When these proteins were added back to permeabilized cells which were unable to secrete, catecholamine release was fully restored, suggesting that certain proteins necessary for exocytosis had been dialyzed from these cells. One of the released proteins was characterized as calmodulin. However, addition of calmodulin alone was ineffective in maintaining or restoring secretory activity in digitonin-permeabilized cells, excluding calmodulin as the sole factor responsible for the loss of release. Protein kinase C was also identified as one of the leaked proteins. This enzyme is known to be retained in cells in the presence of 12-O-tetradecanoylphorbol 13-acetate (TPA). However, under TPA-dependent conditions, there was also a loss of secretory activity. The present paper shows that among the proteins leaked from digitonin-permeabilized cells, there are specific proteins crucial to the exocytotic mechanism.     

Novel putative targets of N-ethylmaleimide sensitive fusion protein (NSF) and alpha/beta soluble NSF attachment proteins (SNAPs) include the Pak-binding nucleotide exchange factor betaPIX

N-ethylmaleimide sensitive fusion protein (NSF) is a chaperone that plays a crucial role in the fusion of vesicles with target membranes. NSF mediates the ATP-consuming dissociation of a core protein complex that assembles during vesicle fusion and it thereby recharges the fusion machinery to perform multiple rounds of fusion. The binding of NSF to the core complex is mediated by co-chaperones named soluble NSF attachment proteins (SNAPs), for which three isoforms (alpha, beta and gamma) are known. Here, we sought to identify novel targets of the NSF-SNAP complex. A yeast two-hybrid screen using the brain specific betaSNAP isoform as bait revealed, as expected, NSF and several isoforms of the SNARE protein syntaxin as interactors. In addition, three isoforms of the reticulon protein family and two isoforms of BNIP3 interacted with betaSNAP. A yeast two-hybrid screen using NSF as bait identified Rab11-FIP3 and the Pak-binding nucleotide exchange factor betaPIX as putative binding partners. betaPIX interacts with recombinant NSF in co-sedimentation assays and the two proteins may be co-immunoprecipitated. A leucine zipper (LZ) motif within the C-terminus of betaPIX mediates binding to NSF; however, this fragment of betaPIX does not exhibit dominant negative effects in a cellular assay. In summary, our results support the evolving view that NSF has numerous targets in addition to conventional SNARE complexes.     

Kinetics of exocytosis in adrenal chromaffin cells

Upon repetitive or maintained stimulation, chromaffin cells secrete catecholamines initially at a very high rate which then relaxes with multiple kinetic components. The complex kinetics are often modeled as resulting from the successive depletion of several functional pools of secretory granules which may reflect specific protein-mediated steps in granule maturation. The fastest component represents granules fully primed for exocytosis. This 'readily releasable pool' may, under some circumstances, consist of only about a dozen granules which can be released within tens of milliseconds. Modulating the size of this pool may be an important way for cells to regulate secretion.     

Intragranular pH rapidly modulates exocytosis in adrenal chromaffin cells

Several drugs produce rapid changes in the kinetics of exocytosis of catecholamines, as measured at the single event level with amperometry. This study is intended to unveil whether the mechanism(s) responsible for these effects involve changes in the intravesicular pH. Cell incubation with bafilomycin A1, a blocker of the vesicular proton pump, caused both a deceleration in the kinetics of exocytosis and a reduction in the catecholamine content of vesicle. These effects were also observed upon reduction of proton gradient by nigericin or NH4Cl. pH measurements using fluorescent probes (acridine orange, quinacrine or enhanced green fluorescent protein-synaptobrevin) showed a strong correlation between vesicular pH and the kinetics of exocytosis. Hence, all maneuvers tested that decelerated exocytosis also alkalinized secretory vesicles and vice versa. On the other hand, calcium entry caused a transient acidification of granules. We therefore propose that the regulation of vesicular pH is, at least partially, a necessary step in the modulation of the kinetics of exocytosis and quantal size operated by some cell signals.     

Calcium gradients and exocytosis in bovine adrenal chromaffin cells

The relationship between the localized Ca(2+) concentration and depolarization-induced exocytosis was studied in patch-clamped adrenal chromaffin cells using pulsed-laser Ca(2+) imaging and membrane capacitance measurements. Short depolarizing voltage steps induced Ca(2+) gradients and small "synchronous" increases in capacitance during the pulses. Longer pulses increased the capacitance changes, which saturated at 16 fF, suggesting the presence of a small immediately releasable pool of fusion-ready vesicles. A Hill plot of the capacitance changes versus the estimated Ca(2+) concentration in a thin (100 nm) shell beneath the membrane gave n = 2.3 and K(d) = 1.4 microM. Repetitive stimulation elicited a more complex pattern of exocytosis: early pulses induced synchronous capacitance increases, but after five or more pulses there was facilitation of the synchronous responses and gradual increases in capacitance continued between pulses (asynchronous exocytosis) as the steep submembrane Ca(2+) gradients collapsed. Raising the pipette Ca(2+) concentration led to early facilitation of the synchronous response and early appearance of asynchronous exocytosis. We used this data to develop a kinetic model of depolarization-induced exocytosis, where Ca(2+)-dependent fusion of vesicles occurs from a small immediately releasable pool with an affinity of 1-2 microM and vesicles are mobilized to this pool in a Ca(2+)-dependent manner.     

A role for synaptotagmin (p65) in regulated exocytosis

Proteins that are specifically localized to synaptic vesicles in the nervous system have been proposed to mediate aspects of synaptic transmission. Antibodies raised against the cytoplasmic domains of five of these proteins, vamp, rab3A, synaptophysin, synaptotagmin, and SV2, were used to investigate their function. Microinjection of monoclonal and polyclonal antibodies raised against synaptotagmin (p65), but not the other vesicle proteins, decreases K⁺/Ca²⁺-mediated dopamine β-hydroxylase surface staining, a measure of regulated secretion in PC12 cells. Microinjection of a soluble fragment of synaptotagmin encompassing one of the domains homologous to the C2 regulatory region of protein kinase C, but lacking the membrane anchor, also inhibits evoked dopamine β-hydroxylase surface staining. These results provide support for the hypothesis that synaptotagmin, a Ca²⁺- and phospholipid-binding protein, is important for regulated exocytosis in neurons.     

SNAPs, a family of NSF attachment proteins involved in intracellular membrane fusion in animals and yeast

Three new and likely related components of the cellular fusion machinery have been purified from bovine brain cytosol, termed alpha-SNAP (35 kd), beta-SNAP (36 kd), and gamma-SNAP (39 kd). Transport between cisternae of the Golgi complex measured in vitro requires SNAP activity during the membrane fusion stage, and each SNAP is capable of binding the general cellular fusion protein NSF to Golgi membranes. The SNAP-NSF-membrane complex may be an early stage in the assembly of a proposed multisubunit "fusion machine" on the target membrane. SNAP transport factor activity is also found in yeast. Yeast cytosol prepared from a secretion mutant defective in export from the endoplasmic reticulum (sec17) lacks SNAP activity, which can be restored in vitro by the addition of pure alpha-SNAP, but not beta- or gamma-SNAPs. These data suggest that the mechanism of action of SNAPs in membrane fusion is conserved in evolution.     

MgATP-independent and MgATP-dependent exocytosis. Evidence that MgATP primes adrenal chromaffin cells to undergo exocytosis

The MgATP dependency of secretion was investigated in digitonin-permeabilized adrenal chromaffin cells. Shortly after permeabilization there is a component of Ca2+-dependent secretion that occurs in the absence of MgATP in the medium. This secretion occurs from cells which are permeable to Ca2+/[ethylene-bis(oxyethylenenitrilo)]tetraacetic acid buffers, to nucleotides, and to proteins. It is prevented by treatment of cells with metabolic inhibitors to reduce cellular ATP prior to permeabilization. The rate of MgATP-independent secretion is rapid and terminates by approximately 2 min after introduction of Ca2+. MgATP-independent secretion is labile and is lost unless Ca2+ is introduced within 8 min of permeabilization. MgATP-dependent secretion occurs at a slower rate than MgATP-independent secretion and continues at a constant rate for 12 min. Preincubation of permeabilized cells with MgATP enhances Ca2+-dependent secretion during a subsequent incubation in the absence of MgATP. Similar MgATP sensitivities are observed when MgATP is present only prior to or only during stimulation with Ca2+ with half-maximal stimulation occurring at 0.4-0.5 and 0.6 mM MgATP, respectively. The data indicate that intact cells are primed by intracellular ATP so that immediately upon permeabilization, there is a component of secretion which is independent of medium MgATP. MgATP partially maintains the primed state after permeabilization by acting before Ca2+ in the secretory pathway.     

Sodium-dependent calcium efflux from adrenal chromaffin cells following exocytosis. Possible role of secretory vesicle membranes.

Although cytosolic Ca2+ transients are known to influence the magnitude and duration of hormone and neurotransmitter release, the processes regulating the decay of such transients after cell stimulation are not well understood. Na(+)-dependent Ca2+ efflux across the secretory vesicle membrane, following its incorporation into the plasma membrane, may play a significant role in Ca2+ efflux after stimulation of secretion. We have measured an enhanced 45Ca2+ efflux from cultured bovine adrenal chromaffin cells following cell stimulation with depolarizing medium (75 mM K+) or nicotine (10 microM). Such stimulation also causes Ca2+ uptake via voltage-gated Ca2+ channels and secretion of catecholamines. Na+ replacement with any of several substitutes (N-methyl-glucamine, Li+, choline, or sucrose) during cell stimulation inhibited the enhanced 45Ca2+ efflux, indicating and Na(+)-dependent Ca2+ efflux process. Na+ deprivation did not inhibit 45Ca2+ uptake or catecholamine secretion evoked by elevated K+. Suppression of exocytotic incorporation of secretory vesicle membranes into the plasma membrane with hypertonic medium (620 mOsm) or by lowering temperature to 12 degrees C inhibited K(+)-stimulated 45Ca2+ efflux in Na(+)-containing medium but did not inhibit the stimulated 45Ca2+ uptake. Enhancement of exocytotic secretion with pertussis toxin resulted in an enhanced 45Ca2+ efflux without affecting calcium uptake. The combined results suggest that Na(+)-dependent Ca2+ efflux across secretory vesicle membranes, following their incorporation into the plasma membrane during exocytosis, plays a significant role in regulating calcium efflux and the decay of cytosolic Ca2+ in adrenal chromaffin cells and possibly in related secretory cells.     

Requirement for metalloendoprotease in exocytosis: evidence in mast cells and adrenal chromaffin cells

Exocytosis is initiated by the receptor-mediated influx of calcium that results in fusion of the secretory vesicle with the plasma membrane. We examined the possibility that calcium-dependent exocytosis in mast cells and adrenal chromaffin cells requires metalloendoprotease activity. Metalloendoprotease inhibitors and dipeptide substrates block exocytosis in these cells with the same specificity and dose dependency as that with which they interact with metalloendoproteases. Metalloendoprotease activity is identified in these cells with fluorogenic synthetic substrates, which also blocked exocytosis. Metalloendoprotease activity is highest in the plasma membrane of chromaffin cells. The metalloendoprotease appears to be required in exocytosis at a step dependent on or after calcium entry, since exocytosis initiated by direct calcium introduction in both mast cells and chromaffin cells is blocked by metalloendoprotease inhibitors.     

Soluble N-ethylmaleimide-sensitive fusion attachment proteins (SNAPs) bind to a multi-SNAP receptor complex in Golgi membranes.

Soluble N-ethylmaleimide-sensitive fusion attachment proteins (SNAPs) are required for the binding of N-ethylmaleimide-sensitive fusion protein (NSF) to Golgi membranes and are, therefore, required for intra-Golgi transport. We report the existence of distinct alpha/beta-SNAP and gamma-SNAP-binding sites in Golgi membranes that appear to be part of the same receptor complex. Cross-linking studies with alpha-SNAP demonstrate that an integral membrane protein of between 30-40 kDa is the alpha-SNAP binding component of the multi-SNAP receptor complex. These data suggest that SNAPs function by independently binding to a multi-SNAP membrane-receptor complex, thereby activating them to serve as adaptors for the targeting of NSF.     

Visualization of the exocytosis/endocytosis secretory cycle in cultured adrenal chromaffin cells

Cultured bovine adrenal medullary chromaffin cells were stimulated to secrete catecholamines by addition of veratridine or nicotine. The formation of an exocytotic pit exposes a major secretory granule membrane antigen, the enzyme dopamine beta-hydroxylase, to the external medium. By including antiserum to this enzyme in the medium, we were able to visualize sites of exocytosis by decoration of bound antibody using a fluorescent second antibody. Internalization of this antibody- antigen complex was then followed in chase experiments: approximately half the surface complex was internalized in 15-30 min. In other experiments, secretion was triggered in the absence of antiserum, and surface enzyme was revealed by binding antibodies at various times after secretion had been halted by an antagonist. Surface patches of antigen remained discrete from the bulk of the plasma membrane for at least 30 min, although a substantial proportion of the antigen was internalized within this time. Cell surface concanavalin A receptors were internalized at a roughly similar rate, suggesting that mechanisms may be similar. After internalization, chromaffin granule membranes fused to larger structures, possibly lysosomes, and were transported over a few hours to the perinuclear region of the cell.     

The chromaffin granule proton pump and calcium-dependent exocytosis in bovine adrenal medullary cells

Summary Calcium-dependent exocytosis in leaky bovine adrenal medullary cells has a requirement for Mg-ATP. One possibility is that exocytosis depends in some way on the operation of the ATP-dependent proton pump that serves to maintain the core of the secretory vesicles both acid and at a positive potential with respect to the cytosol. This possibility has been tested in leaky cells by monitoring exocytosis under conditions where the secretory vesicle pH and potential gradients are measuredin situ. The results show rather clearly that exocytosis can persist, with unchanged Ca-activation kinetics, in the virtual absence both of a difference in pH between the cytosol and secretory vesicle core and also of a difference in potential across the vesicle membrane. The results do not, however, exclude a small modulating effect of vesicle pH or potential on exocytosis and shed no light on whether or not the plasma membrane potential, which is maintained close to zero in these experiments, influences exocytosis.     

Proenkephalin A gene expression in bovine adrenal chromaffin cells is regulated by changes in electrical activity.

Concentrations of mRNA coding for the opioid peptide precursor proenkephalin A (mRNAENK) were measured in primary cultures of bovine adrenal chromaffin cells maintained in serum-free medium. Using a sensitive solution hybridization assay, an increase in mRNAENK levels from 45 to 300% above control with K+ (10-20 mM), Ba2+ (1 mM) and veratridine (5 microM) was found. The highest increase (300% above control) was obtained with the Na+ channel agonist veratridine. This effect was nearly abolished in the presence of the Na+ channel antagonist tetrodotoxin (TTX) (1 microM). Moreover, TTX partially inhibited the increase in mRNAENK levels caused by K+ (20 mM) depolarization (from 185 to 130% of control), but had no effect on the stimulation by Ba2+ (1 mM). The Ca2+ channel antagonists D600 (50 microM) verapamil (50 microM) and Co2+ (1 mM) inhibited the responses to either K+, Ba2+ or veratridine, whereas the Ca2+ channel agonist Bay K 8644 (0.1 microM) potentiated the effect of 20 mM K+ from 185 to 230% of control. The K+-induced increase in the mRNAENK levels was associated with an increase of immunoreactive proenkephalin A-derived peptides in both tissue and medium, indicating an enhanced production of opioid peptides. These results suggest that membrane depolarization may play an important role in the regulation of proenkephalin A gene expression in bovine adrenal chromaffin cells. It may represent a mode by which substances acting directly on Na+ or Ca2+ channels may modulate the regulation of proenkephalin A mRNA biosynthesis and opioid peptide production.     

Evaluation of the electrostatic field strength at the site of exocytosis in adrenal chromaffin cells.

Exocytosis in secretory cells consists of release from intracellular storage granules directly into the extracellular space via fusion of the granule membrane with the plasma membrane of the cell. It is considered here as comprising two distinct processes. One is the close apposition of granule and plasma membranes. The other arises from interactions between the two membranes during the process of apposition, leading to the formation of a fusion pore. In the following it is shown for the case of the adrenal medullary chromaffin cell that the fusion pore can be ascribed to electroporation of the granule membrane, triggered by the strong electric field existing at the site of exocytosis. Based on an electric surface charge model of the cytoplasmic side of the plasma membrane, resulting from the negatively charged phosphatidylserine groups, it is found that the electrostatic field strength at the site of exocytosis reaches values on the order of 10(8) V/m at small intermembrane distances of 3 nm and lower. The field strength increases with the size of the disc-shaped plasma membrane region generating the electric field, reaching an approximate limit for a radius of 10 nm, at a surface charge density of 5.4 x 10(-2) C/m2. According to previous experimental evaluations of threshold field strength, this field is sufficiently strong to cause membrane electroporation. This step is a precondition for the subsequent membrane fusion during the ongoing process of apposition, leading to secretion.     

Common mechanisms for regulated exocytosis in the chromaffin cell and the synapse

Chromaffin granule exocytosis differs in many physiological respects from neuronal synaptic vesicle exocytosis, which has led to the assumption that the two processes occur by distinct mechanisms. While different mechanisms are certainly in operation for the biogenesis of granules and synaptic vesicles, it is now becoming clear that similar mechanisms are used by both beyond this stage. The similarities extend to various aspects of regulated exocytosis, including regulation of the number of vesicles released in response to cell stimulation. Most strikingly, it now appears that the same proteins mediate the docking and fusion of both chromaffin granules and synaptic vesicles, and that homologues of these proteins act similarly in constitutive membrane traffic throughout evolution.