Abbreviations used : AM, acetoxymethyl ester ; [Ca2+]i, intracellular Ca2+ concentration ; VMAT, vesicular monoamine transporter.
Amine Weak Bases Disrupt Vesicular Storage and Promote Exocytosis in Chromaffin Cells
Version of Record online: 5 MAY 2004
Journal of Neurochemistry
Volume 73, Issue 6, pages 2397–2405, December 1999
How to Cite
Mundorf, M. L., Hochstetler, S. E. and Wightman, R. M. (1999), Amine Weak Bases Disrupt Vesicular Storage and Promote Exocytosis in Chromaffin Cells. Journal of Neurochemistry, 73: 2397–2405. doi: 10.1046/j.1471-4159.1999.0732397.x
- Issue online: 5 MAY 2004
- Version of Record online: 5 MAY 2004
- Chromaffin cells.
Abstract : The vesicular contents in bovine chromaffin cells are maintained at high levels owing to the strong association of its contents, which is promoted by the low vesicular pH. The association is among the catecholamines, Ca2+, ATP, and vesicular proteins. It was found that transient application of a weak base, methylamine (30 mM), amphetamine (10 μM), or tyramine (10 μM), induced exocytotic release. Exposure to these agents was also found to increase both cytosolic catecholamine and intracellular Ca2+ concentration, as measured by amperometry and fura-2 fluorescence. Amphetamine, the most potent amine with respect to evoking exocytosis, was found to be effective even in buffer with out external Ca2+ ; however, the occurrence of spikes was suppressed when BAPTA-acetoxymethyl ester was used to complex intracellular Ca2+. Amphetamine-induced spikes in Ca2+-free medium were not suppressed by thapsigargin or ruthenium red, inhibitors of the sarco(endo)plasmic reticulum Ca2+-ATPase and mitochondrial Ca2+ stores. Atomic absorption measurements of amphetamine- and methylamine-treated vesicles reveal that intravesicular Ca2+ stores are decreased after a 15-min incubation. Taken together, these data indicate that amphetamine and methylamine can disrupt vesicular stores to a sufficient degree that Ca2+ can escape and trigger exocytosis.
Secretory vesicles in neurons and endocrine cells have the unique ability to store chemical messengers and release them rapidly on exocytosis. Chromaffin cells isolated from the adrenal medulla typify such cells (Wightman et al., 1991). In these cells, the intravesicular concentration of catecholamines and ATP are remarkably high, ~550 and 125 mM, respectively (Njus et al., 1986). In addition, Ca2+ has a total concentration of >20 mM within the vesicle (Winkler and Westhead, 1980). These high concentrations are partially maintained by the vesicular matrix, containing acidic, soluble proteins that associate with the smaller molecules and ions located within the vesicle (Yoo and Albanesi, 1990a ; Videen et al., 1992). The intravesicular association reduces the osmotic pressure and facilitates long-term storage within the small volume of the vesicle (Helle et al., 1990). The exocytotic release of the vesicular contents is triggered by an increase in intracellular Ca2+ concentrations ([Ca2+]i), which can arise from influx through voltagesensitive Ca2+ channels or liberation from intracellular pools (Llinas et al., 1992 ; Matthews, 1996). [Ca2+]i is highly regulated, returning to basal levels following stimulation via several routes, including the Na+/Ca2+ exchanger (Jan and Schneider, 1992), thapsigargin-sensitive Ca2+ uptake (Pan and Kao, 1997), and mitochondrial uptake (Herrington et al., 1996 ; Xu et al., 1997).
The Ca2+ -dependent fusion of the vesicle with the cellular membrane enables its releasable contents to be extruded into the extracellular space. At the normal intravesicular pH of 5.5, the vesicular contents have an optimal association that facilitates storage, but when exposed to the normal pH of the extracellular fluid following vesicle membrane fusion, the association weakens, and the intravesicular matrix unravels (Jankowski et al., 1993 ; Wightman et al., 1995 ; Yoo, 1996). Similar storage complexes are found in the vesicles of mast cells (Zimmerberg et al., 1987) and in goblet cells (Verdugo, 1990). Indeed, this type of association, which promotes storage at the intravesicular pH and release when exposed to the pH of the extracellular fluid, has been proposed for many secretory vesicles, including those of neurons (Rahamimoff and Fernandez, 1997).
In addition to promoting intravesicular association in chromaffin cells, an acidic vesicular interior is necessary for catecholamine packaging by the vesicular monoamine transporter, VMAT (Njus et al., 1986 ; Henry et al., 1998). The weak base mechanism for psychostimulants proposes that compounds such as amphetamine decrease vesicular content as a result of effects on the VMAT (Johnson, 1988 ; Sulzer and Rayport, 1990). Consistent with both the storage and transporter mechanisms, evoked catecholamine release is diminished following incubation with methylamine (Kuijpers et al., 1989), a substance that raises the intravesicular pH from 5.5 to 6.3 (Holz et al., 1983). Whereas methylamine accesses the vesicular interior by passive diffusion of its unprotonated form through the cellular and vesicular membranes, amphetamine and related compounds are also transported by the cell (Scherman and Henry, 1980 ; Seiden et al., 1993) and vesicular catecholamine transporters (Fon et al., 1997). Their transport inhibits catecholamine transport (Scherman and Henry, 1980). The net result is a decrease in evoked catecholamine release (Sulzer et al., 1995) and an increased leakage of catecholamines from cells (Rubin and Jaanus, 1966).
In this work, carbon fiber microelectrodes have been used to analyze, at the level of single exocytotic events, the effects of methylamine, amphetamine, and tyramine (Mack and Bonisch, 1979 ; Knoth et al., 1984) on the release of catecholamines from bovine chromaffin cells. We found that all three weak base amines could trigger exocytotic release without addition of other secretagogues when used at doses previously shown to alkalize vesicles. Amphetamine and methylamine cause a significant amount of vesicular dissociation before exocytosis as evidenced by increased intracellular levels of Ca2+ and catecholamine. The resultant elevated [Ca2+]i triggers exocytosis. Thus, these data provide strong evidence that vesicles in chromaffin cells have the capacity to trigger their own release, a property demonstrated in several types of endocrine cells (van de Put and Elliot, 1996 ; Scheenen et al., 1998) and proposed for some neuronal systems (Stanley, 1993 ; Fossier et al., 1998).
MATERIALS AND METHODS
Cultured chromaffin cells
Primary cultures of bovine adrenal chromaffin cells were prepared as previously described (Leszczyszyn et al., 1991). Bovine adrenal glands were obtained from a local slaughterhouse and digested with collagenase, and the chromaffin cells were isolated using a Renografin density gradient. The epinephrine fraction was collected and plated at a density of 3 × 105 cells per 35-mm-diameter plate. The cultures were maintained in a controlled atmosphere with 5% CO2 in air at 37°C, and the medium was changed on days 3, 5, and 7 of the cell culture. Experiments were performed during days 3-7 of culture.
Before experiments, the cells were washed three times and then maintained in a pH 7.4 buffer containing 145 mM NaCl, 5 mM KCl, 11.2 mM glucose, 10 mM HEPES, 2 mM CaCl2, and 1 mM MgCl2·6H2o. Chemical stimulations and incubations of the cells were performed with the desired agents dissolved in this solution for the Ca2+-containing experiments.
Electrodes and electrochemical procedures
Microelectrodes were fabricated by sealing 6-μm-diameter carbon fibers (Thornel T-650 ; Amoco Corp., Greenville, SC, U.S.A.) in glass as previously described (Kawagoe et al., 1993). The electrodes were polished at a 45° angle on a diamond dust-embedded micropipette beveling wheel (Sutter Instrument Co., Novato, CA, U.S.A.) and soaked in isopropyl alcohol before use. Electrodes were calibrated using a flowinjection apparatus where epinephrine was the analyte detected.
Amperometry used an Axopatch 200B (Axon Instruments, Foster City, CA, U.S.A.) in the voltage-clamp mode with the whole-cell configuration (β=1). The low-pass filter was set to 10 kHz, and the gain was 10 mV/pA for Ca2+ -containing buffers and 20 mV/pA for others. The applied potential was 650 mV with repect to a sodium-saturated calomel electrode.
Single chromaffin cell experiments
A culture plate was placed on the stage of an inverted microscope (Axiovert 35 ; Zeiss, Eastern Microscope, Raleigh, NC, U.S.A.) resting on a vibration isolation table (Stable Top TXA-2 ; Newport Corp., Fountain Valley, CA, U.S.A.). For experiments at 37°C, the stage of the microscope was heated. A pressure-ejection device (Picospritzer ; General Value Corp., Fairfield, NJ, U.S.A.) connected to a micropipette with a 10μ-diameter tip was used to apply the secretagogue onto the cell. The working electrode and micropipette were positioned using micromanipulators (Patch Clamp Driver PCS-250 ; Burleigh Instruments, Fishers, NY, U.S.A.). The micropipette was positioned 25-35 μm from the cell, and the carbon fiber microelectrode was positioned so that it firmly touched the cell surface. To determine viability, the cells were stimulated with 50 μM nicotine. Then, the agents of interest were pressureejected onto the cell. A 2-min interval was left between each ejection.
In experiments that involved incubation of the cells, they were first tested by pressure ejection of nicotine (typically four times for 4-5 s with 2 min between each ejection). Next, the amine of interest was added to the plate for 15 min. Cytosolic catecholamine levels were determined by incubating the cells with the amine and then permeabilizing it via the pressure ejection of 10 μM digitonin (Jankowski et al., 1992).
Amines were incubated with the cells in the buffer described above prepared without Ca2+, termed Ca2+ -free buffer in Results. In this buffer, pressure ejection of 10 mM caffeine was used to test the exocytotic viability of release in cells (Cheek et al., 1993). To chelate intracellular Ca2+, the cells were incubated in a Ca2+ -containing buffer containing 50 μM BAPTAacetoxymethyl ester (AM) for 10 min (procedure from Powis et al., 1996). Then, cells were rinsed with a Ca2+ -free buffer and maintained in that solution while agents were pressure-ejected onto the cells. To investigate thapsigargin-sensitive stores, cells were permeabilized by incubation with 1 μM digitonin in Ca2+ -free buffer for 2-5 min followed by a 5-min incubation with 100 nM thapsigargin in the same buffer. The cells were then rinsed, and release was tested in Ca2+ -free buffer (procedure modified from Poulsen et al., 1995). To investigate mitochondrial effects, cells were permeabilized by incubation with 1 μM digitonin for 4 min to allow entry of 50 μM ruthenium red. Approximately 5 min later, the cells were stimulated via pressure ejection (procedure modified from Troadec et al., 1998).
Fura-2 Ca2+ measurements
Chromaffin cells were incubated in a buffer solution containing 1 μM fura-2 AM and 0.1% bovine albumin for 20 min at 22°C. Fura-2 AM was dissolved in a mixture of 10 μl of 10% Pluronic F-127 and 40 μl of dry dimethyl sulfoxide. Following the incubation period, the cells were rinsed with fresh buffer solution, and the fura-2 AM was allowed to deesterify for 20 min. Next, the cells were rinsed twice with buffer solution.
Single-cell fluorescent measurements were made with a pinhole that allowed light from a 27-μm-diameter spot on the plate to be detected by a photomultiplier tube (Photometer System ; EMPIX, Mississauga, ON, Canada). Cells were alternately excited at 340 ± 7.5 and 380 ± 7.5 nm, whereas emission at 510 ± 10 nm was measured. Light was collected through a 40 × oil immersion objective (Fluar 40× ; Carl Zeiss, Thornwood, NY, U.S.A.). To reduce photobleaching, excitation light from the Xe arc lamp passed through a 0.5 neutral density filter (30% transmission) positioned between the arc lamp and the 40× objective. Ratios of F340/F380 were determined at 750-ms intervals.
Vesicle isolation experiments
Epinephrine cells were plated on 100-mm-diameter plates at a density of 2.5 × 106 cells per plate. After 3 days in culture, the contents of six plates were removed from the plate with trypsin. The cells were centrifuged at 360 g for 5 min, resuspended in 30 ml of a hypoosmotic medium (20 mM MOPS, 5 mM EDTA, and 0.235 M sucrose, pH 7.2), and homogenized in a cold room (Gratz et al., 1981). The homogenate was diluted to 70 ml with the hypoosmotic medium and centrifuged at 3,500 g for 10 min. The supernantant was removed and centrifuged again at 13,100 g for 20 min. The uppermost layers of the resultant pellet are brown and contain mitochondria and lysosomes, whereas the lower portion contains the desired vesicles and is pink. The upper layer was removed by decanting off the supernatant and using a small aliquot of the isolation medium to loosen it. This process significantly reduces the vesicle yield but is efficient at removing nonvesicular organelles (Wills, 1996). The vesicle pellet was resuspended in 70 ml of hypoosmotic medium, and the centrifugation step was repeated twice to purify the vesicles further. Finally, the pellet was resuspended with 70 ml of Ca2+ -free isotonic buffer containing no chelating agent. This fraction was centrifuged at 3,500 g for 5 min to precipitate clumped vesicles.
The desired concentrations of amines were added to the supernatant. Following a 10-min incubation period, the solutions were centrifuged at 1,200 g for 5 min. The pellet was resuspended in Ca2+-free isotonic buffer, and LaCl3 (lanthanum oxide dissolved in concentrated HCl) was added to obtain a final concentration of 1% (wt/vol). Calcium levels were determined by atomic absorption spectroscopy (Instrumental Laboratory S-12 AA/AE spectrohotometer ; λ = 422.7 nm). Lanthanum is used to minimize interference in the flame from anions such as phosphates. Standard additions of Ca2+ were used to determine the moles of Ca2+ per milligram of protein.
The amperometric signal was digitized (PCM-2 ; Medical Systems Corp., Greenvale, NY, U.S.A.) and stored on VCR tape. The data were played back into a PC with commercially available hardware and software (Cyberamp 320 and Axotape ; Axon Instruments). The current records were digitally filtered at 400 Hz before analysis. Individual spikes (signal-to-noise ratio = 5) were located by locally written software. The average root mean square current noise level after filtering was 0.13 pA. For each spike, its area (Q), amplitude (A), and width at half-height (t1/2) were determined. Spikes whose shapes were poorly correlated with an exponentially modified Gaussian were not included in the data set (Wightman et al., 1991). Data were analyzed by single-factor ANOVA for statistical significance.
Nicotine and BAPTA-AM were obtained from Research Biochemicals International (Natick, MA, U.S.A.) Fura-2 AM and Pluronic F-127 were purchased from Molecular Probes (Eugene, OR, U.S.A.). Pen/Strep was obtained from the Lineberger Cancer Research Center (Chapel Hill, NC, U.S.A.). The chromaffin cell culture medium was Dulbecco's modified Eagle's medium/Ham's F12 medium from GIBCO Laboratories (Grand Island, NY, U.S.A.). Collagenase was acquired from Worthington Chemicals (Freehold, NJ, U.S.A.). Renografin-60 was from Squibb Diagnostics (New Brunswick, NJ, U.S.A.). Dimethyl sulfoxide was purchased from Mallinckrodt Chemical (Paris, KY, U.S.A.). All other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). All solutions were prepared in distilled, deionized water (Mega Pure System MP-3A ; Corning Glass Works, Corning, NY, U.S.A.).
Amine-induced concentration spikes
In preliminary experiments, it was found that incubation of chromaffin cells with methylamine (30 mM), amphetamine (10 μM), or tyramine (10 μM) evoked concentration spikes of catecholamine at room temperature that could be detected with a carbon fiber microelectrode. To investigate this further, each amine was individually introduced by pressure ejection onto a single cell. Before exposure to the amines, exocytosis was evoked with 50 μM nicotine pressure-ejected onto each cell for 4-5 s to establish cell viability.
Methylamine (30 mM), amphetamine (10 μM), and tyramine (10 μM) also evoked concentration spikes when administered by pressure ejection (Fig. 1). The frequency of spikes, evaluated in the first minute after drug introduction, was lower than with 50 μM nicotine for all three amines (Table 1.) but much higher than the spontaneous release rate. Normally spikes are never seen in the absence of a secretagogue (Wightman et al., 1991). Spikes induced by methylamine tended to be clustered near the time of application onto the cell (Fig. 1B). whereas spikes induced by amphetamine tended to occur over a longer interval (Fig. 1C). Tyramine was the least effective at inducing spikes (Fig. 1D). The amine-induced spikes were similar in shape to those induced by 50 μM nicotine but had a lower molecular content per event (Table 1) at room temperature. However, at 37°C, amphetamine-induced spikes increased in frequency, whereas nicotine-induced spikes did not significantly change in frequency (Table 2).
|Secretagogue||Spike frequency (spikes/min)||Quantal size (pC)||Amplitude (nA)||t1/2 (ms)||No. of cells tested|
|Nicotine (50 μM)||33.5±2.1||1.42±0.03||0.012±0.002||22.13±4.31||21|
|Methylamine (30 mM)||8.6±0.2||0.77±0.02||0.009±0.001||47.28±5.36||9|
|Amphetamine (10 μM)||16.8±0.5||1.14±0.12||0.010±0.001||43.45±2.11||6|
|Tyramine (10 μM)||4.5±0.5||0.42±0.04||0.007±0.001||32.64±3.16||16|
|Condition at 37°C||Spike frequency (spikes/min)|
|2 mM Ca2+ buffer||Nicotine (50 μM)||Amphetamine (10 μM)|
|Ca2+-free buffer||Caffeine (10 mM)||Amphetamine (10 μM)|
|Control||11.9±1.2 (9)||11.6±1.7 (9)a|
|BAPTA-AM (50 μM)||3.5±0.4 (6) b||2.7±0.3 (6) c|
|Thapsigargin (100 nM)||1.8±0.8 (6) d||9.4±1.4(6)|
|Ruthenium red (50 μM)||12.2±1.8 (6)||9.5±0.7(6)|
Amine-induced increase in cytosolic catecholamines
To evaluate the effects of the amines on cytosolic catecholamines, cells were incubated with the drug of interest in Ca2+ -containing buffer for 15 min. The cells were transferred to Ca2+ -free buffer at room temperature and permeabilized with 10 μM digitonin delivered from a pressure-ejection pipette for 5 s. In some cases, multiple exposures to digitonin were required. Under these conditions, the cytosolic contents of the cell can leak out, and the electroactive species can be detected with an adjacent carbon fiber electrode (Leszczyszyn et al., 1991). No response was observed from untreated cells permeabilized with digitonin ; however, a broad efflux was observed after incubation with 30 mM methylamine, 10 μM amphetamine, or 10 μM tyramine (Fig. 2). The prolonged time response of the efflux is quite different from the sharp concentration spikes observed during exocytosis, indicating a cytosolic as opposed to a vesicular origin (Cahill and Wightman, 1995). For methylamine, the detected amount of cytosolic catecholamine was 1.51 ± 0.35 × 108 molecules (n = 6 cells). Prior incubation with amphetamine evoked efflux of 1.78 ±0.16 × 107 molecules (n = 4 cells), and incubation with tyramine evoked 1.46 ± 0.12 ± 107 molecules (n = 4 cells).
The effect of these amines on [Ca2+]i at room temperature was examined by monitoring the fluorescence of intracellular fura-2 alternately at two excitation wavelengths (340 and 380 nm). Exposure of the cell to 60 mM K+ in Ca2+ -containing medium caused an increase in the ratio of the fluorescence, indicating elevated [Ca2+]i (Finnegan and Wightman, 1995). Subsequent exposure to amphetamine or methylamine also caused a transient increase in the ratio (Fig. 3). The fractional response of amphetamine-induced [Ca2+]i with respect to that induced by K+ was 23.3 ± 3.5%. For methylamine, the fractional response with respect to K+ was 20.3 ± 2.9% (average ± SEM). Ca2+ effects induced by tyramine were not measurable. At least five cells were analyzed for each amine.
Amine-induced spikes in Ca2+ -free buffer
In preliminary experiments, it was found that concentration spikes were also evoked during incubation with 30 mM methylamine, 10 μM tyramine, and 10 μM amphetamine in extracellular medium without Ca2+, albeit at a lower rate than in the presence of Ca2+. Such release from chromaffin cells is similar to that evoked by agents such as caffeine that promote release of internal Ca2+ stores. Since release was most robust with amphetamine at physiological temperature (37°C), its actions were tested further in external buffer containing no extracellular Ca2+. Pressure ejection of amphetamine (10 μM) elicted a similar number of spikes as caffeine (10 mM) in this medium (Table 2 and Fig. 4A). However, the frequency of spikes elicited by amphetamine was significantly less than in Ca2+ -containing medium. Measurements with fura-2 showed that the caffeine exposure increased [Ca2+]i as reported earlier (Finnegan and Wightman, 1995), but no change was detected with amphetamine.
To examine further whether the concentration spikes observed in Ca2+ -free medium were evoked by intracellular Ca2+, the cells were loaded with the Ca2+ chelator BAPTA-AM. Caffeine was used for comparison because its stores should be negligibly depleted by the 10-min, 50 μM BAPTA-AM treatment. Intracellular chelation by BAPTA-AM decreased significantly the frequency of concentration spikes evoked by both 10 mM caffeine and 10 μM amphetamine at 37°C (Table 2).
Origin of the intracellular Ca2+ store evoked by amphetamine
Ca2+-ATPases account for Ca2+ uptake into inositol 1,4,5-trisphosphate- as well as caffeine-sensitive Ca2+ stores in bovine chromaffin cells (Poulsen et al., 1995). Thapsigargin is a highly selective inhibitor of the sarco-(endo)plasmic reticulum Ca2+ -ATPase pump. To evaluate the effect of these stores on the observed concentration spikes, cells were permeabilized with 1 μM digitonin in Ca2+-free buffer and exposed to 100 nM thapsigargin for 5 min at 37°C. Under these conditions 10 mM caffeine-evoked responses were significantly diminished—typically two or three vesicles per caffeine exposure (Table 2 and Fig. 4B, right). In contrast, the frequency of spikes evoked by exposure to 10 μM amphetamine was not significantly altered (an example is shown in Fig. 4B, left).
The mitochondrial store of Ca2+ was investigated with ruthenium red, an inhibitor of Ca2+ transport in mitochondria (Troadec et al., 1998). Cells were again permeabilized with 1 μM digitonin and then incubated with 50 μM ruthenium red for 4 min. Subsequent pressure of 10 μM amphetamine and 10 mM caffeine evoked concentration spikes at a frequency that was not significantly altered for either drug (Table 1 ; an example is shown in Fig. 4C).
Vesicular [Ca2+] in isolated vesicles
To examine further the effects of these amines on intravesicular Ca2+ in Ca2+ in Ca2+ -free buffer, the Ca2+ content of isolated vesicles was examined by atomic absorption spectroscopy. Vesicles were maintained in Ca2+-free buffer or in Ca2+ -free buffer with the amines for 15 min. Three samples were examined for each condition, and a minimum of two measurements per sample was made. The values (Table 3) obtained in the absence of amines were quite similar to those found previously for vesicles : 125 ± 15 nmol of Ca2+ /mg of protein (Serck-Hanssen and Christiansen, 1973) and 112 ± 6.3 nmol of Ca2+ /mg of protein (Yoo and Albanesi, 1990b). With the exception of tyramine, the amounts were lower in vesicles incubated with the amines (Table 3).
|Condition (no. of samples)||nmol of Ca2+/mg of protein|
|Tyramine (10 μM) (3)||117.7±1.9|
|Amphetamine (10 μM) (3)||99.2±2.3a|
|Methylamine (30 mM) (3)||101.3±1.0b|
The results of this work reveal that the application of various amines onto bovine chromaffin cells causes release of catecholamines by an exocytotic process. All of the amines selected for study have been documented in previous investigations to cause alkalization of the vesicular interior (Holz et al., 1983) ; Knoth et al., 1984Sulzer and Rayport, 1990). This process should relax the tight association of the vesicular contents that maintains them in storage for prolonged periods and also inhibit inward transport by the VMAT. Our findings reveal that the amine-induced alkalization causes an increase in the cytosolic concentration of both catecholamines and Ca2+. On crossing the plasma membrane, the first vesicles that the amines would encounter and perturb are those close to the protein machinery that mediates exocytosis, perhaps those vesicles that are already docked. Thus, vesicular Ca2+ liberation would occur in the intracellular space adjacent to the plasma membrane where it can trigger exocytosis.
The concentration spikes evoked by the amines have the form of exocytotic events evoked by exposure to depolarizing agents. They are sharp, transient events, as anticipated for vesicular extrusion, but have smaller amplitudes than normal exocytotic events. Evidently, the vesicles that exocytose have already experienced the relaxation in association and subsequent vesicular displacement of catecholamines caused by alkalization leading to the smaller amounts released. For the concentrations examined, amphetamine is much more effective at evoking exocytosis than tyramine. Indeed, at physiological temperature amphetamine (10 μM) evokes exocytotic events at a frequency almost as great as that evoked by 50 μM nicotine, an agent that causes membrane depolarization and activation of voltage-dependent Ca2+ channels. Both tyramine and amphetamine have access to the vesicle interior because they are substrates for the cell (Banerjee et al., 1987) and vesicular transporters (Knoth et al., 1984 ; Romanenko et al., 1998). However, because of the nonpolar nature of amphetamine, its access is more efficient because it can also cross both membranes by passive diffusion (Fischer and Cho, 1979). In contrast, methylamine, which can only gain access to the vesicle interior by diffusion of its unprotonated form, evokes exocytosis only at a high concentration.
All three amines elevate levels of cytosolic catecholamines as a result of vesicle displacement, with tryamine being the least effective in this regard. For amphetamine and methylamine in external medium containing Ca2+, this displacement is accompained by a measureable increase in [Ca2+]i. A previous report also documents increases in [Ca2+]i following incubation of chromaffin cells with another weak base, ammonium chloride, although its effects on release were not examined (Kuijpers et al., 1989). Although the increases in [Ca2+]i are lower than those caused by 60 mM K+, the amine-induced [Ca2+] increases are sufficient to generate exocytosis. Thus, like classical, depolarizationevoked exocytosis, elevated [Ca2+]i accompanies the amine-generated release events.
To investigate further the Ca2+ dependence of amineinduced release, it was evaluated in Ca2+ -free media. In isolated vesicles methylamine and amphetamine were able to displace ~20% of stored Ca2+, and incubation of intact cells with those amines induced exocytosis. Although we have not investigated the mechanism of release of Ca2+ from the vesicles, we note there are Ca2+ - activated Ca2+ channels (Woodbury, 1995) on vesicles as well as Ca2+ exchangers.
Since amphetamine was most robust with respect to Ca2+ mobilization in Ca2+ -free conditions, it was studied in some detail. At physiological temperature, the rate of exocytotic spikes induced by amphetamine was reduced by half in Ca2+-free media, but the rate was similar to that evoked by 10 mM caffeine, an agent that causes release of Ca2+ from internal stores. Loading of the cells with BAPTA-AM, a rapid Ca2+-chelating agent, suppressed the exocytotic rates evoked by both caffeine and amphetamine, further suggesting they are Ca2+-dependent. However, because amphetamine-induced increases in [Ca2+]i in Ca2+-free medium were not observable in whole-cell measurements with fura-2, they must be quite small or localized to a microregion of the cell interior. To evoke exocytosis, Ca2+ levels would only need to be elevated in the very small region between the plasma membrane and an adjacent vesicle. The readily releasable pool of vesicles (Neher, 1998) occupies this region, and it would be the first to encounter the amine and the one most susceptible to release as a result of displacement of its Ca2+.
In many types of cells, including bovine chromaffin cells, mobilization of intracellular Ca2+ triggers the opening of Ca2+ channels in the plasma membrane that can further increase [Ca2+]i. It has been postulated that this capacitative Ca2+ entry can enhance exocytotic rates (Cheek and Thastrup, 1989 ; Powis et al., 1996 ; Fomina and Nowycky, 1999). Because vesicles tend to dock near Ca2+ channels, this mechanism would provide a potent way to potentiate amine-induced release in Ca2+-containing medium. Our results are consistent with this hypothesis as release rates are greater in medium containing Ca2+.
Pharmacological intervention was used to investigate two other well-established internal Ca2+ stores. Thapsigargin, an inhibitor of Ca2+-ATPase in the endoplasmic reticulum, was unable to reduce the amphetamine-induced spikes. Interference of mitochondrial stores with ruthenium red also did not alter the amphetamine-induced release events. In both cases, the incubation times were kept short to avoid depletion of all internal Ca2+ stores (Guo et al., 1996). Thus, release evoked by amphetamine appears independent of the two best-characterized internal stores. Rather, the intracellular Ca2+ appears to arise from vesicles that are clearly disrupted by the amines as evidenced by the liberated intracellular catecholamine. This store contains 60% of the total cell content of Ca2+ (Haigh et al., 1989) and thus could be a potent source of Ca2+.
The ability of vesicles to trigger their own exocytosis as a result of liberation of their Ca2+ has received considerable attention recently. Internal release of Ca2+ from an acidic compartment has been shown to trigger exocytosis in insulin-secreting cells (Scheenen et al., 1998). In pancreatic acinar cells, vesicular Ca2+ has been shown to be mobilized by inositol 1,4,5-trisphosphate and ADP-ribose (Gerasimenko et al., 1996 ; van de Put and Elliott, 1996). A similar mechanism has also been previously proposed for chromaffin vesicles (Yoo and Albanesi, 1990a). In rat neurohypophysial nerves, uptake and efflux of Ca2+ have been demonstrated from secretory vesicles. The process is rapid and has been shown to be inhibited by the weak base ammonium chloride (Troadec et al., 1998). Cholinergic synaptic vesicles have been proposed to control the presynaptic Ca2+ concentration and trigger the release of acetylcholine (Fossier et al., 1998). It is interesting that elevated [Ca2+]i in PC12 cells causes an increase in vesicular pH, consistent with this hypothesis (Han et al., 1999). Thus, the observations reported here are not necessarily restricted to chromaffin cells but may be important at several cell types, including neurons.
The mechanism set forth in this work to explain exocytotic release induced by amphetamine likely operates in parellel with the classical mechanism accepted for the releasing actions of phenethylamines. Indeed, for many catecholamine systems such as in the striatum, release is Ca2+-independent (Carboni et al., 1989 ; Hurd and Ungerstedt, 1989). Rather, it depends on reversal of catecholamine transport by the plasma membrane transporter following vesicular displacement (Fischer and Cho, 1979 ; Jones et al., 1998). This nonexocytotic process results in slower release rates to which the amperometric electrode at isolated cells is less sensitive. Consistent with concurrent release by reversal of transport, we see increased baseline fluctuations at cells incubated with amphetamine (data not shown). Our results may explain why amphetamine release is found to be Ca2+-dependent in some neuronal systems (Crespi et al., 1997 ; Pierce and Kalivas, 1997 ; Hoffman and Gerhardt, 1999). The newly proposed mechanism, however, will only be operant in cells whose vesicles contain Ca2+.
Taken together, the results of this study demonstrate the importance of the strong association of the vesicular contents. The association allows the multiple vesicular contents to be maintained in an osmotically stable form. The contents include intravesicular Ca2+ that was packaged during vesicle formation at the Golgi apparatus (Orci et al., 1986 ; Yoo, 1996). Disruption of the association by alkalization with weak bases allows the vesicular contents to leak into the intracellular space. This action decreases the amount of catecholamine secreted per exocytotic event and increases [Ca2+]i, which can trigger exocytosis due to its close proximity to release sites. We are currently investigating the hypothesis that other perturbations of the vesicular matrix can also displace its contents.
Discussions with Marc Caron and Stephanie Cragg are gratefully acknowledged. This research was supported by the National Institutes of Health.
- 531996) Imaging of intracellular calcium stores in individual permeabilized pancreatic acinar cells. Apparent homogeneous cellular distribution of inositol 1,4,5-trisphosphate-sensitive stores in permeabilized pancreatic acinar cells.J. Biol. Chem. 271,4999–5006.& (
- 581996) Measurement of the size and catecholamine contents of individual adrenomedullary vesicles by tandem capillary hydrodynamic chromatography-open tubular liquid chromatography [Ph.D. Thesis]. University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.(