INTRODUCTION

Mastoparan, a cationic-amphiphilic tetradecapeptide isolated from wasp venom, and similar peptides have a wide variety of cellular effects including the activation of phospholipase A₂ and PLC¹ (1, 2, 3), the stimulation of glycogenolysis (4), and the induction of insulin, serotonin, and histamine secretion (5, 6, 7). All these mastoparan effects have been linked to the activation of heterotrimeric G_o and G_i proteins (8, 9, 10), small GTP-binding proteins (11), ATP- and GTP-degrading nucleotidases (12), guanylyl cyclase (13, 14), and nucleoside diphosphate kinase (15, 16). Mastoparan also induces the degranulation of mast cells (17, 18) and the mitochondrial permeability transition (19). However, mastoparan inhibits calmodulin-dependent enzymes (20, 21), PKC (22), Na⁺/K⁺-ATPase (23), PLC (24), catecholamine exocytosis (25), and ATP-sensitive potassium channels (26) in various cell types.

The mechanism by which mastoparan stimulates G-proteins appears to be by peptide binding into the phospholipid bilayer and the formation of an -helix that resembles the intracellular loops of G-protein-coupled receptors (27, 28). The peptides forming amphiphilic helices directly trigger G_i and G_o protein activation and induce production of inositol phosphate by activated PLC in a way similar to receptor activation (29, 30, 31). Therefore, the [Ca²⁺]_i rise induced by mastoparan appears to depend on PLC-catalyzed inositol phosphate production and Ca²⁺ mobilization, a pathway known to be regulated by G-protein in many cases. However, there are also several reports on mastoparan effecting membrane perturbation following alterations in the physical state of membranes (24, 29, 32). Perianin and Snyderman (33) reported that mastoparan increases intracellular Ca²⁺ in neutrophils via a pertussis toxin-insensitive pathway that requires extracellular Ca²⁺.

To further investigate the processes responsible for the increase of cytosolic Ca²⁺ during mastoparan stimulation and to examine the regulatory mechanism by which one could control the mastoparan actions, we used the mastoparan analogs Mas-7 and Mas-17 on SK-N-BE(2)C human neuroblastoma and other cells. The active mastoparan analog Mas-7 has been synthesized by substituting alanine for the positively charged lysine in position 12. This produced an enhancing effect on GTPase activity (34). In contrast, substitution with a neutral residue in the same place resulted in a loss of mastoparan activity, creating the inactive analog Mas-17 (30, 34). In the present study, we suggest that Mas-7 induces the formation of pores permeable to Ca²⁺, Mn²⁺, Na⁺, EtBr, and lucifer yellow, in addition to the phosphoinositide turnover resulting in IP₃ production and Ca²⁺ influx. Interestingly, the phosphoinositide turnover was regulated by PKC and Ca²⁺ entering from the extracellular environment.

EXPERIMENTAL PROCEDURES

Carbamylcholine (carbachol) chloride, thapsigargin, (±)-sulfinpyrazone, Triton X-100, EGTA, EDTA, Trizma base, trichloroacetic acid, bovine serum albumin, and IP₃ were obtained from Sigma. Phorbol 12-myristate 13-acetate (PMA) and 4--PMA were purchased from Research Biochemicals Inc. (Natick, MA). [³H]IP₃ was obtained from DuPont NEN, and fura-2/AM was purchased from Molecular Probes (Eugene, OR). Mas-7 and Mas-17 were purchased from Peninsula Laboratories, Inc. (Belmont, CA). Phospholipids were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL).

Human neuroblastoma clone SK-N-BE(2)C, human promyelocytic leukemia HL-60, and human astrocytoma 1321N1 cells were cultured in DMEM (Life Technologies, Inc.) supplemented with 10% (v/v) heat-inactivated bovine calf serum (Hyclone, Logan, UT) and 1% antibiotics (Life Technologies, Inc.) in a humidified atmosphere of 5% CO₂ at 37 °C. The medium was changed every 2 days, and cells were subcultured about once a week. Bovine chromaffin cells were isolated from fresh bovine adrenal glands according to the method of Kilpatrik et al., (35). The cells were maintained with DMEM/F-12 (Life Technologies, Inc.) containing 10% bovine calf serum and 1% antibiotics.

The intracellular free Ca²⁺ concentration was determined using the fluorescent Ca²⁺ indicator fura-2, as previously reported (36). Briefly, SK-N-BE(2)C cells were loaded with fura-2 acetoxymethylester (fura-2/AM) to a final concentration of 3 µM in complete medium and incubated at 37 °C for 50 min with stirring. The final concentration of dimethyl sulfoxide in the incubation medium was 0.3%. After the loading, the cells were pelleted and washed twice with Locke's solution (154 mM NaCl, 5.6 mM KCl, 1.2 mM MgCl₂, 2.2 mM CaCl₂, 5.0 mM HEPES, 10 mM glucose, pH 7.4) to remove the extracellular dye. Sulfinpyrazone was added to both the loading medium and the washing solution to a final concentration of 250 µM to prevent dye leakage (37). For the fluorimetric measurement of the [Ca²⁺]_i, 1 × 10⁶ cells were placed into a quartz cuvette in a thermostatically controlled cell holder at 37 °C with continuous stirring. Fluorescence ratios were taken by dual excitation at 340 and 380 nm and emission at 500 nm by an alternative wavelength time scanning method. The concentration of intracellular Ca²⁺ was calculated using the following equation where R_max and R_min are the ratios obtained when fura-2 is saturated with Ca²⁺ or when EGTA is used to remove Ca²⁺, respectively. S_b2 and S_f2 are the proportionality coefficients of Ca²⁺-saturated fura-2 and free fura-2, respectively. Calibration of the fluorescence signal in terms of [Ca²⁺]_i was performed according to Grynkiewicz et al. (38). In extracellular Ca²⁺-free experiments, the Locke's solution did not contain Ca²⁺ ion but instead included 200 µM EGTA.

IP₃ mobilization in the cells was determined by competition assay using [³H]IP₃ in binding to IP₃ binding protein (39). To determine the IP₃ production induced by Mas-7, SK-N-BE(2)C cells were grown in 6-well culture plates to about 95% confluency. The cells were then stimulated with the agonists for various periods of time. The reaction was terminated by removal of the medium followed by the addition of 0.3 ml of ice-cold 15% (w/v) trichloroacetic acid (TCA) containing 10 mM EGTA. The plates were left on ice for 30 min to extract the water-soluble inositol phosphates. The extract was transferred to an Eppendorf tube, and the TCA was removed by four extractions with diethyl ether. The final extract was neutralized with 200 mM Trizma base (Tris base) and its pH adjusted to about 7.4. 20 µl of the cell extract was added to 20 µl of assay buffer (0.1 M Tris buffer containing 4 mM EDTA, 4 mg/ml bovine serum albumin), and 20 µl of [³H]IP₃ (0.1 µCi/ml). Then 20 µl of a solution containing the binding protein was added. The mixture was incubated for 15 min on ice and then centrifuged at 2,000 × g for 10 min. 100 µl of water and 1 ml of scintillation mixture were added to the pellet to measure radioactivity. IP₃ concentration in the sample was determined by comparison with a standard curve and expressed as pmol/µg of protein. The IP₃ binding protein has been prepared from bovine adrenal cortex according to the method of Challiss et al. (40).

The level of intracellular Na⁺ was assessed using SBFI/AM, the fluorescent sodium indicator (41). Cells were harvested and incubated in culture medium with 15 µM SBFI/AM, 0.2% pluronic acid, and 250 µM sulfinpyrazone at 37 °C for 2 h with continuous stirring. The increase in cytosolic Na⁺ was measured as the increase in the fluorescence ratio determined at the dual excitation wavelength of 340 and 380 nm and the emission wavelength of 530 nm at 37 °C. In the Na⁺-free Locke's solution, 154 mM choline chloride was added instead of NaCl.

To calibrate the intracellular Na⁺ concentration, SBFI-loaded cells were bound to the experimental chamber and illuminated with ultraviolet light applied through the rear epifluorescence port of the Nikon Diaphot microscope (Nikon Corp., Tokyo, Japan). At the end of each experiment, cells were treated with calibration solutions consisting of specific concentrations of Na⁺, Na⁺ ionophores gramicidin D (1 µM) and monensin (40 µM), and Na⁺/K⁺ pump inhibitor strophanthidin (100 µM). The final concentration of Na⁺ was obtained by mixing two solutions containing either NaCl or KCl (130 mM NaCl, or 130 mM KCl, 2 mM EGTA, 10 mM HEPES, pH 7.2) in an appropriate proportion.

The membrane permeabilization was determined with the help of fluorescent dyes. The EtBr influx was measured as described previously (42). Briefly, cells were harvested in serum-free DMEM medium, then one-half million cells were aliquoted and centrifuged, and the pellets were resuspended in Locke's buffer solution containing 25 µM ethidium bromide. The membrane permeability to EtBr after the addition of the indicated reagents was determined by measuring the change in fluorescence intensity at an excitation wavelength of 310 nm and the emission wavelength of 580 nm. The effect of Mas-7 on uptake of lucifer yellow, fura-2, and fluorescein-conjugated phalloidin was assessed as previously reported (43) with some modifications. Aliquoted 3 × 10⁵ harvested cells were transferred to 24-well plates and incubated for 5 h to let the cells attach to the bottom. Then the medium was removed, and the cells were incubated for 5 min in Locke's solution with or without Mas-7 and fluorescent dyes at the following concentrations: 0.5 mg/ml lucifer yellow, 15 µM fura-2, 1 mg/ml Evans blue, and 0.04 µg/ml fluorescein-conjugated phalloidin. At the end of each treatment, the cells were washed in their wells twice with Ca²⁺-free Locke's solution and once with Ca²⁺ containing Locke's solution. Then, 300 µl of 3% (w/v) TCA containing Locke's solution was added to each well, stopping the reaction, and the plates were incubated at 4 °C for 30 min. The extract was then transferred to a quartz cuvette, and the fluorescence was measured. Lucifer yellow fluorescence was measured at an excitation wavelength of 450 nm and the emission wavelength of 521 nm; for fura-2, the wavelengths used were 340 and 510 nm; for fluorescein-conjugated phalloidin, the wavelengths used were 495 and 520 nm. The uptake of Evans blue was measured by absorbance at 600 nm using a spectrophotometer.

Cells loaded with fura-2/AM as described above were stimulated with Mas-7 in the presence of 25 µM Mn²⁺, and fluorescence quenching was measured. Fluorescence was excited at 360 nm, i.e. the isosbestic wavelength at which Ca²⁺ does not affect fura-2 fluorescence and at which, therefore, changes were caused by Mn²⁺ quenching (44, 45). Emission was recorded at 500 nm. Values for maximal Mn²⁺ quenching in each preparation were determined immediately following the recording of the permeabilization of the cells with 0.1% Triton X-100.

The formation of channels by Mas-7 was measured by the method of Kim et al. (46). Briefly, planar lipid bilayers were composed of phosphatidylethanolamine and phosphatidylserine dissolved in n-decane (20 mg/ml) in a 1:1 ratio. Bilayers were formed on a hole of 0.3 mm diameter in a Delrin cup. Mas-7 (10 µM) was added to the cis chamber containing Locke's solution without glucose. The trans solution was the modified Locke's solution (5.6 mM NaCl, 154 mM KCl, 1.2 mM MgSO₄, 5.0 mM HEPES, pH 7.4). The cis solution was connected via an Ag/AgCl electrode and an agar/KCl bridge to the bilayer recording head-stage input of an Axopatch 1D amplifier (Axon Instruments, Foster City, CA). The trans solution was held at ground potential using the same electrode arrangement. Recordings were filtered through a low pass Bessel Filter (Dagan Corporation, Minneapolis, MN) at 0.5 kHz. Data were analyzed by pClamp 6.02 software (Axon Instrument).

RESULTS

The ability of Mas-7 to alter [Ca²⁺]_i in SK-N-BE(2)C neuroblastoma cells was examined using the fluorescent Ca²⁺ indicator fura-2. We found that, in the presence of extracellular Ca²⁺, Mas-7 caused a sustained elevation of the cytosolic Ca²⁺ concentration, as illustrated in Fig. 1A. A high concentration of Mas-7 (50 µM) increased the [Ca²⁺]_i rapidly (Fig. 1A), while treatment with a submaximal concentration (20 µM) of Mas-7 led to a slow increase in [Ca²⁺]_i (data not shown). Interestingly, the Mas-7-induced response was partially inhibited by pretreatment with PMA, the PKC activator. Addition of Mas-17, an analog of mastoparan that does not activate G-proteins (34), was ineffective in raising the [Ca²⁺]_i. In the absence of extracellular Ca²⁺, exposure of the cells to Mas-7 resulted in a transient increase in [Ca²⁺]_i, which was completely abolished by PMA pretreatment (Fig. 1B). The concentration dependence of the [Ca²⁺]_i rise induced by Mas-7 with or without PMA treatment is presented in Fig. 1C. In the presence of extracellular Ca²⁺, maximal effective concentrations of Mas-7 raised the [Ca²⁺]_i to over 1600 nM from a resting level of ~80 nM. The maximal and half-maximal (EC₅₀) effective concentrations were 80 ± 15 µM and 28 ± 6 µM, respectively. It is noteworthy that, under these conditions, the elevated [Ca²⁺]_i was sustained at all concentrations of Mas-7 tested. 4--PMA, the inactive analog of PMA, did not inhibit the Mas-7-induced [Ca²⁺]_i rise, indicating that the effect is mediated by PKC (data not shown).

We performed a similar analysis with regard to the effect on IP₃ generation. We found that the peak IP₃ generation is obtained 1 min after the addition of Mas-7 and that the effect is maintained for more than 20 min (Fig. 2A). Mas-17 was ineffective in eliciting IP₃ production. Fig. 2B shows that Mas-7 produced IP₃ in a concentration-dependent manner, the maximal response and half-maximal response being obtained at 78 ± 25 and 28 ± 7 µM, respectively. The IP₃ generation in response to variations in Mas-7 concentration occurred in a pattern similar to that observed for the [Ca²⁺]_i rise. However, removal of extracellular Ca²⁺ decreased the Mas-7-induced IP₃ production, and PMA treatment completely blocked the response. These data suggest that Mas-7 induces [Ca²⁺]_i rise and IP₃ production and that PMA treatment inhibits both responses. However, PMA treatment inhibits the [Ca²⁺]_i rise and the IP₃ production only partially in the presence of extracellular Ca²⁺ (Figs. 1C and 2B), while in the absence of extracellular Ca²⁺, it does so completely (Fig. 1B and 2B).

To look into the mechanism of the [Ca²⁺]_i rise during Mas-7 stimulation, we depleted the internal Ca²⁺ stores by pretreating the cells with thapsigargin, an inhibitor of endomembrane Ca²⁺-ATPases. Incubation of the cells with 1 µM thapsigargin for 10 min increased [Ca²⁺]_i and then the [Ca²⁺]_i was slowly decreased and sustained to a slightly elevated level as we previously demonstrated (47). The above thapsigargin treatment abolished a subsequent [Ca²⁺]_i rise induction by carbachol treatment (Fig. 3A), indicating that the IP₃-sensitive Ca²⁺ stores were depleted by the thapsigargin treatment. However, addition of Mas-7 after the thapsigargin treatment produced another [Ca²⁺]_i rise although slightly lower than without the pretreatment (Fig. 3B). This result suggests that Mas-7 can evoke Ca²⁺ influx from the extracellular space by a second pathway, one that is separate from Ca²⁺-release activated channel. Furthermore, PMA treatment now could not prevent the Mas-7-induced [Ca²⁺]_i rise. Mas-17 was not able to promote [Ca²⁺]_i rise in thapsigargin-pretreated cells.

We further tested whether the Mas-7 effect was reversible. When the cells were washed with Locke's buffer after 30 µM Mas-7 stimulation, they were again normally responsive to carbachol (Fig. 3C). Furthermore, the washing out of Mas-7 did not affect the fluorescence intensity of the intracellular fura-2 (data not shown). These results indicate that Mas-7 acts on the plasma membrane, triggering a reversible Ca²⁺ response, and that the Mas-7-induced passages are not permeable to fura-2.

Mas-7-mediated influx of extracellular Ca²⁺ across the plasma membrane was verified by measuring the unidirectional uptake of Mn²⁺ into fura-2-loaded SK-N-BE(2)C cells. The rate of fluorescence decrease, measured at the Ca²⁺-insensitive isosbestic point, provides a relative measure of the divalent cation permeability. Mn²⁺ slowly and steadily enters unstimulated cells. Fig. 4A shows that the rate of fura-2 quenching produced by the Mn²⁺ entry accelerates with the addition of 1 µM thapsigargin, revealing the activation of a divalent cation permeability pathway in the plasma membrane. The subsequent addition of Mas-7 markedly increased the rate of fluorescence quenching. The total fluorescence quenching possible was obtained after the addition of 0.1% Triton X-100 and is represented by a dotted line in Fig. 4A. The rate of fluorescence quenching is dependent upon the amount of Mas-7, while Mas-17 does not trigger the Mn²⁺ quenching effect at all (Fig. 4B). The concentration dependence of the Mas-7 effect seen in Fig. 4C shows that maximal and half-maximal stimulation of the Mn²⁺ influx occurs at 72 ± 12 and 24 ± 5 µM Mas-7, respectively.

To test the selectivity for Ca²⁺ in the Mas-7-induced ion current, Na⁺ influx upon Mas-7 treatment was measured in cells loaded with SBFI/AM, a Na⁺-specific fluorescent dye. The treatment of cells with Mas-7 resulted in an increase of the [Na⁺]_i in the SBFI-loaded cells (Fig. 5), indicating that Mas-7 induces the influx of Na⁺ in addition to that of Ca²⁺. The Mas-7-induced influx of Na⁺ was also sustained, similar to the [Ca²⁺]_i rise. Mas-17 did not change the [Na⁺]_i (data not shown). Also, the Mas-7-induced influx of Na⁺ was not obtained when extracellular Na⁺ was substituted with choline (dotted trace in Fig. 5). Monensin, a Na⁺-specific ionophore, was used as a positive control in the test of the [Na⁺]_i elevation. These results reflect the ability of Mas-7 to regulate membrane permeability for a variety of ions.

To test whether Mas-7 makes membrane pores, we examined the effect of Mas-7 on the uptake of EtBr. Fig. 6A shows that submaximal concentration (30 µM) of Mas-7 evoked rapid uptake of EtBr (314 Da). Digitonin, as a positive control, also induced EtBr uptake. The maximal effective concentration (80 µM) increased the fluorescence intensity to the digitonin-induced level (data not shown). The concentration dependence, as seen in Fig. 6B, strongly supports the conclusion that membrane pores are formed upon Mas-7 stimulation.

To assess the pore size formed by Mas-7, we measured the permeability of various fluorescent dyes with different molecular weights. Mas-7 induced the uptake of lucifer yellow (457 Da) in a concentration-dependent manner with maximal and half-maximal concentrations seen at 75 ± 18 and 22 ± 5 µM, respectively (data not shown). However, Mas-7 did not induce the uptake of fura-2 (831 Da), Evans blue (961 Da), and fluorescein-conjugated phalloidin (1,175 Da) even at the maximal effective concentration of 80 µM Mas-7 (data not shown). These results indicate that Mas-7 forms specific pores permeable to molecules at least up to 457 Da in size.

The formation of membrane pores by Mas-7 was further investigated in a planar lipid bilayer in an effort to explain the entry of various ions and molecules into Mas-7-treated cells. Fig. 7A shows the planar bilayer recordings of Mas-7-induced channels in the cis solution containing Locke's solution and in the trans solution containing the modified Locke's solution with high K⁺. In three separate experiments, we observed two types of channels with different conductances. The current-voltage relationships of these two channels appeared linear with little voltage dependence, and the slope conductances were 290 and 94 pS (Fig. 7B). The arrows in Fig. 7A represent the current level contributed by both the closed state of the large conductance channel and the open state of the small conductance channel. The open probability of the 290 pS channel was 0.25 ± 0.12 (n = 3), however, the small conductance channel remained open most of the time and its closure is marked with an asterisk.

We also looked at the effect of Mas-7 on [Ca²⁺]_i rise and EtBr uptake in other cell types beside SK-N-BE(2)C. We tested the Mas-7 effect also on HL-60, 1321N1 human astrocytoma, and bovine chromaffin cells. As Fig. 8 shows, Mas-7 increases the [Ca²⁺]_i in the presence of extracellular Ca²⁺ and EtBr uptake in all the tested cells. This suggests that Mas-7 regulates membrane permeability by a mechanism common to a variety of cell types.

DISCUSSION

Mastoparan has been widely used to prime G-protein activation in a receptor-independent manner with various cell types (9, 17, 30, 48). Although the priming mechanism is unclear, this agent appears to activate phosphoinositide-specific PLC via G-protein and to produce inositol phosphate (49, 50). To further characterize cytosolic calcium elevation induced by mastoparan, we applied Mas-7, a highly active form of mastoparan (34), to human neuroblastoma SK-N-BE(2)C and other cell types. The treatment of SK-N-BE(2)C cells with Mas-7 evoked IP₃ production and [Ca²⁺]_i rise in a concentration-dependent manner. Both responses increased gradually and stayed steady for a long time. In the [Ca²⁺]_i rise induced by Mas-7, the major event was a prolonged Ca²⁺ influx through membrane pores from the extracellular space while a minor contribution came from a transient IP₃-sensitive Ca²⁺ release from intracellular stores.

It has been suggested that IP₃ can directly activate plasmalemmal Ca²⁺-conductive channels in lymphocytes (51). However, in the cells we tested, such channels do not appear to contribute significantly to the Mas-7-generated [Ca²⁺]_i rise since the inhibition of IP₃ production by treatment with PMA had little effect on blocking the [Ca²⁺]_i rise. In addition, thapsigargin, the agent used to deplete internal Ca²⁺ stores, has also little effect on the Mas-7-induced [Ca²⁺]_i rise. These results indicate that depletion of Ca²⁺ stores is not the main cause for the Ca²⁺ permeability and that most of the Ca²⁺ entry is not from a capacitative entry pathway through Ca²⁺-release activated channel (52). Instead, our experiments with EtBr and fluorescent dyes suggest that Ca²⁺ influx from the extracellular space occurs through membrane pores. These pores are permeable to Na⁺, as well as Ca²⁺ and Mn²⁺, and also to EtBr and lucifer yellow. Previous studies suggested that the enhancement of membrane permeability by mastoparan might possibly be evoked by a cytotoxic mechanism independent of G-protein activation (53). Our findings indicate that this permeability is not a nonspecific membrane perturbation since the pores are not permeable to bigger molecules such as fura-2 (831 Da). In addition, removal of Mas-7 allowed the cells to restore their responsiveness to a subsequent muscarinic stimulation, indicating that the action site of Mas-7 is the plasma membrane rather than the internal membrane. Mas-17, which has an amino acid composition almost identical to Mas-7 except for two amino acids, has no effect in the formation of membrane pores. The data, therefore, suggest that Mas-7 forms membrane pores by a specific process and that a specific sequence in Mas-7 is essential for the interaction with the membrane.

The formation of membrane pores by Mas-7 was demonstrated in a lipid bilayer. The Mas-7-induced channels appeared to be of two types with slope conductances of 290 and 94 pS when we used the Locke's solution in cis and modified Locke's solution containing high K⁺ and low Na⁺ in trans to mimic the extracellular and intracellular ion composition. These two channels seem to behave independently because we were able to observe only small conductance channels in four experiments. The reversal potentials of these two types of channels were observed at near 0 mV, suggesting that the conductance was not contributed by the passage of a single kind of cation. These results support the hypothesis that the Mas-7-induced channels are nonselective for cations. Although we still have to explain the permeability of the Mas-7-induced channels to large molecules, such as EtBr and lucifer yellow, the large conductance channel is likely to form a large pore that could allow these molecules to enter. The situation is reminiscent of the ryanodine receptor of muscle sarcoplasmic reticulum, which has a comparable conductance under similar ionic conditions and can let a molecule of the size of glucose pass through (54). The results clearly show that the partitioning of Mas-7 into the lipids is sufficient to make membrane pores without involvement of membrane protein and cytosolic components.

Previous studies suggested that mastoparan forms -helical structures that are highly amphiphilic in a lipid environment, with hydrophobic residues within the lipid bilayer, and with hydrophilic, positively charged residues facing outward (27, 28, 55). Our results indicate that the alignment of mastoparan makes pores in the membrane permeable to molecules of a specific molecular size although the mechanism of pore formation remains unclear. Recently, maganinin, an antimicrobial peptide isolated from frog skin, has been found to form -helical structures and pores in membranes, thus altering the permeability of the pores by the aggregation of a number of transmembrane peptide subunits within the bilayer (56). The effect of small peptide fragments on the lipid bilayer conductance was also explored with certain antibiotics, such as alamethicin and monazomycin, derived from fungi. Here water-filled pores spanning the bilayer were formed via the aggregation of variable numbers of individual molecules (57, 58). Those reports suggest the possibility that mastoparan too may form pores by a mechanism similar to that of the peptide fragments.

The cation permeability of the Mas-7-induced pores indicates that this bilayer conductance does not lend itself to the simple speculation that anions are favored due to the high density of positive charges lining the pore. Rather, the phenomenon is similar to that of maganinins, peptide fragments of the sodium channel, and other peptide antibiotics (56, 59). All these peptides are positively charged but generate cation-selective channels when incorporated into artificial lipid bilayers. It has also been reported that the presence of the negatively charged lipid phosphatidylserine is important to the peptides forming the pores in the bilayer (60). Interestingly, mastoparan is also known to penetrate into membranes that contain mainly acidic phospholipids (32). The maximal responsive concentration of Mas-7 differs from cell type to cell type, indicating cell-specific variations in the Mas-7-induced permeabilization. The difference may reflect variations in the affinity of Mas-7 for each particular cell membrane and/or the lipid composition of the plasma membranes.

The maximal and half-maximal responsive Mas-7 concentrations for the [Ca²⁺]_i rise, the Mn²⁺ influx, the IP₃ generation, and the uptake of EtBr and lucifer yellow are almost the same. The results indicate that Mas-7 is of similar potency with regard to G-protein activation as well as pore formation. However, the Mas-7-induced IP₃ production and [Ca²⁺]_i rise are dramatically decreased in the absence of extracellular Ca²⁺. This finding may reflect a Ca²⁺ requirement for the interaction of Mas-7 with the plasma membrane or for the activation of PLC. Removal of external Ca²⁺ is expected to result in a stoppage of Ca²⁺ influx, which then would in turn block the phosphoinositide turnover. This interpretation does not, however, rule out an additional direct modulatory effect of Ca²⁺ on Mas-7 mediated G-protein activation. On the contrary, in human neutrophils the G-protein modulating and membrane disrupting activities of mastoparan appear to be separate and not to effect each other (53).

The experiments with PMA suggest that PKC activation reduces the effectiveness of Mas-7. This inhibitory effect of PKC may be a result of a reduction in Mas-7-induced IP₃ generation. Since the depletion of the internal Ca²⁺ stores after treatment with thapsigargin abolishes the inhibitory effect of PMA on the Mas-7-induced [Ca²⁺]_i rise and since the Mas-7-induced Ca²⁺ release is blocked by PMA treatment, the regulation of Mas-7-induced [Ca²⁺]_i rise by PKC may be limited to PLC signaling. This is even more strongly supported by the fact that the Ca²⁺ influx from the extracellular space via membrane pores formed by Mas-7 is not effected by PMA treatment.