Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Plants belonging to the genus Delphinium have been recognized for their toxic effects on insects and mammals for centuries (Dioscroides, The Greek Herbal of Dioscorides; Gerard, 1975; Mitton and Mitton, 1982). To this day, cattle are frequently poisoned in the western regions of North America by Delphinium spp. ingestion (Benn and Jacyno, 1983). Methyllycaconitine (MLA), one of the many norditerpenoid alkaloids found in these plants, has been reported to be a competitive antagonist for nicotinic acetylcholine receptors (nAChRs) at mammalian neuromuscular junctions (Dozortseva, 1959; Nambi-Aiyar et al., 1979). In addition to MLA, North American Delphinium spp. contain numerous toxic alkaloids, including deltaline, nudicauline, 14-deacetylnudicauline (14-DN), and barbinine. The effects of these alkaloids on neuromuscular transmission are largely unknown (Pelletier, 1983; Manners et al., 1993, 1995).

Alkaloids commonly found in Delphinium spp. are derivatives of the norditerpenoid lycoctonine. Among these alkaloids, deltaline is a 7,8-methylenedioxylycoctonine-type (MDL) norditerpenoid alkaloid, whereas MLA, nudicauline, 14-DN, and barbinine are lycoctonine derivatives esterified with N-(methylsuccinyl)anthranilic acid at C18 and designated collectively as N-(methylsuccinimido)anthranoyllycacotonine (MSAL)-type alkaloids (Manners et al., 1993, 1995). The N-(methylsuccinyl)anthranilic acid moiety appears to affect alkaloid toxicity and affinity for nAChR types because norditerpenoid alkaloids lacking this group are at least 100 times less lethal/potent than the MSAL-type alkaloids (Manners et al., 1993, 1995; Hardick et al., 1995, 1996). The only structural difference among these four MSAL-type alkaloids is the chemical substitution at C14 (Fig. 1). Deltaline, which is far less toxic than the MSAL-type alkaloids, may contribute substantially to plant toxicity because it is the most abundant alkaloid in Delphinium spp. (Manners et al., 1993).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Norditerpenoid alkaloid structures. Top, skeletal structure of the MSAL-type alkaloids. The R group on C14 provides the only structural difference among the MSAL-type alkaloids tested, and may be O, barbinine; OH, 14-DN; OMe, MLA; or OAc, nudicauline. Bottom, skeletal structure of the MDL-type alkaloid deltaline. Deltaline is hydroxylated at C10 and lacks the N-(methylsuccinimido)anthranilic acid at C-18 that is characteristic of the MSAL-type alkaloids.

All of the alkaloids used in the study described here bind to nicotinic receptors in the central nervous system (Alkondon et al., 1992; Kukel and Jennings, 1994; Hardick et al., 1996; Yum et al., 1996). However, paresis and paralysis are principal signs of Delphinium poisoning in animals (Olsen and Sisson, 1991a,b), suggesting that alkaloid-induced morbidity and mortality could result from the effects of these alkaloids on nicotinic receptors at the neuromuscular junction. Because MLA is the only Delphinium alkaloid whose effects on muscle-type nAChRs have been investigated functionally at the cellular level (Dozortseva, 1959; Nambi-Aiyar et al., 1979; Benn and Jacyno, 1983), it was used as a reference alkaloid for comparing the effects of nudicauline, 14-DN, barbinine, and deltaline on synaptic transmission between nerve and muscle in a lizard preparation. A preliminary account of some of these findings has been presented previously (Dobelis et al., 1993).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Neuromuscular Preparations and Solutions. Adult lizards (Anolis carolinensis) were deeply anesthetized with halothane and then decapitated. For extracellular recording experiments, the hind limbs were removed, the sciatic nerve was isolated, and all hind limb muscles except the m. extensor digitorum longus (EDL) were removed. The resulting neuromuscular preparation was pinned to a Sylgard-coated 35-mm culture dish and continuously perfused by gravity feed at 2 to 3 ml/min with physiological saline solution (PSS; 132 mM NaCl, 3 mM KCl, 2 mM CaCl₂, 0.5 mM MgCl₂, 5 mM glucose, 5 mM HEPES, pH 7.2) at 20-22°C. To record miniature end-plate potentials (MEPPs), intercostal muscles and their attached ribs were removed, pinned to a Sylgard-coated 35-mm culture dish, and perfused with PSS as described above. The free-base form of each MSAL-type alkaloid was solubilized in PSS at pH 4.0 as a 100 µM stock solution. Deltaline was solubilized similarly as a 1 mM stock solution. All stock solutions were frozen at 70°C as 1-ml aliquots. Working alkaloid solutions were made fresh before each experiment by serial dilution and adjustment to pH 7.2.

Extracellular Recording. Compound muscle action potentials (CMAPs) were recorded extracellularly using a bipolar platinum wire recording electrode. Recordings were made by placing the uninsulated tips of the platinum wires on the surface of the EDL muscle and stimulating the sciatic nerve supramaximally with a suction electrode. Recordings were digitized using Super Scope acquisition software (GW Instruments, Somerville, MA), stored on a Macintosh II ci computer, and later analyzed using Microsoft Excel spreadsheet software.

Intracellular Recording. The effects of 14-DN, MLA, barbinine, and deltaline on MEPPs were studied in lizard (A. carolinensis) intercostal muscle preparations. Each intercostal muscle fiber was used as an internal control. MEPPs were recorded intracellularly using a Warner IE-201 Intracellular Electrometer (Warner Instrument Corporation, Hamden, CT). Signals were filtered at 10 kHz and digitized at 3 kHz using the Super Scope acquisition software. Digitized data were stored on a Macintosh II ci computer and later analyzed using the Microsoft Excel spreadsheet software. MEPPs were gathered in event triggered mode by setting a threshold to twice the amplitude of the background noise, which was typically about 200 µV. Data points acquired from 5 ms before to 20 ms after the threshold setpoint were stored temporarily in a memory buffer, inspected visually, and either accepted and saved to disk or rejected and dumped from the memory buffer. Criteria for acceptance were a rapid rise time (<2 ms) and an apparently exponential decay. This procedure reduced the requisite digital storage space and facilitated analysis of MEPP characteristics but did not permit an analysis of alkaloid effects on MEPP frequency. Manual selection of MEPPs introduces the possibility of bias toward selection of larger-amplitude events. To check this possibility, MEPP amplitude distributions were tested for normalcy. All MEPP amplitudes were normally distributed about the mean, suggesting that our selection method was not biased toward larger-amplitude events.

Acetylcholine-Induced Muscle Contraction. To investigate the ability of MLA, 14-DN, and deltaline to prevent acetylcholine-induced muscle contraction, intercostal muscles were pinned out as described above in a bath volume of about 250 µl. Acetylcholine (100 µM) was manually applied directly above the muscle preparation. The volume of acetylcholine was adjusted to obtain reliable muscle contraction and was typically 5 µl. The interval between acetylcholine applications was 5 min to avoid receptor desensitization. After reliable muscle contraction was achieved, 14-DN (5 µM), MLA (10 µM), or deltaline (500 µM) was perfused into the bath. Alkaloid concentrations were chosen to ensure complete blockade of neuromuscular transmission. After a 20- to 45-min incubation in alkaloid, acetylcholine accompanied by alkaloid was again applied, and muscle contraction was monitored visually through a dissecting microscope. To ensure that none of the alkaloids affected direct stimulation of muscle contraction, 25 µl of osmotically adjusted PSS containing 50 mM K⁺ was manually added to the bath in the presence of alkaloid after alkaloid blockade was achieved. Alkaloids were removed by bath perfusion with normal PSS for 30 min, and acetylcholine-induced contractions were again monitored to ensure preparation viability and reversibility of the alkaloid effect.

Data Analysis. Measurements of CMAP amplitudes were taken as peak-to-peak values. Concentration-dependent inhibition curves were fit to the CMAP data using the equation
where R is the fractional response, R_max is the response recorded in the absence of alkaloid, [alkaloid] is the alkaloid concentration, n_H is the slope of the curve, and IC₅₀ is the alkaloid concentration that produces a 50% inhibition of the maximal response. Curves were iteratively fit by allowing the IC₅₀ value and slope of the curve to float. Slopes and IC₅₀ values were obtained from the best-fit curves using the least-squares method.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The use of extracellular and intracellular recording techniques in lizard muscles permitted an analysis of alkaloid effects on action potential generation and nAChR function at neuromuscular synapses. Because the lizard EDL is a small, nearly cylindrical muscle that yields a high current density, extracellular wire electrodes could be used to investigate alkaloid effects on action potential generation in a population of muscle fibers. Lizard intercostal muscles were used to investigate alkaloid effects on nAChRs. This preparation is especially well suited for intracellular recording and pharmacological studies because end-plate regions are easily visualized and diffusion barriers are minimized in this one-muscle-fiber-layer-thick muscle. However, the arrangement of muscle fibers in a thin sheet precluded the use of extracellular recording techniques to measure the nearly simultaneous generation of action potentials in populations of muscle fibers. Alkaloid effects on nAChR function are comparable among skeletal muscles because postsynaptic nAChRs are similar regardless of the muscle type (Salpeter, 1987). Cross-species comparisons of alkaloid effects on neuromuscular transmission are also practical because nAChR function at the neuromuscular junction is similar regardless of the vertebrate species (Salpeter, 1987).

Alkaloid Blockade of CMAPs. Compound muscle action potentials were elicited in an isolated lizard EDL through sciatic nerve stimulation (Fig. 2). For each experiment, the stimulus strength was increased until the CMAP peak-to-peak amplitude reached a maximum, indicating that the number of muscle fibers reaching threshold had been maximized. Reductions in maximal CMAP amplitude after exposure to alkaloid indicated a decrease in the number of muscle fibers brought to threshold by nerve stimulation and provided a measure of the ability and potency of each alkaloid to block neuromuscular transmission (Fig. 2).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Representative CMAPs. Three superimposed traces (control, in the presence of alkaloid, and after alkaloid washout) are shown for each alkaloid. Each alkaloid reduced the CMAP amplitude, which was restored by alkaloid washout. The alkaloid concentrations that produced the illustrated responses were 0.3 µM nudicauline (a), 1 µM 14-DN (b), 2 µM MLA (c), 10 µM barbinine (d), and 160 µM deltaline (e). Scale: vertical, 10 µV, horizontal, 2.5 ms.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Concentration-dependent reduction in CMAP amplitudes. The effect of each Delphinium alkaloid is expressed as a percent of the control response. Amplitudes were measured as peak-to-peak. The IC50 values and the corresponding slopes are listed in Table 1. The effects of two or three alkaloid concentrations are presented for four preparations for nudicauline, 14-DN, barbinine, and deltaline and for five preparations for MLA.

Alkaloid Blockade of MEPPs. Alkaloid-induced changes in MEPP amplitude were determined for 14-DN, MLA, barbinine, and deltaline (Fig. 4 and Table 1). Nudicauline was not studied because insufficient quantities of the alkaloid were available at the time. Measurements of MEPP amplitudes provided a more direct determination of alkaloid effects on nicotinic receptors than measurements of CMAP amplitude because a MEPP occurs when the contents of a single synaptic vesicle opens nicotinic receptor channels in the postsynaptic membrane (Anderson and Stevens, 1973; Kuffler and Yoshikami, 1975; Salpeter, 1987). All of the alkaloids tested significantly reduced (p < .05 to p < .001) the mean MEPP amplitude compared with the control state (Tables 1 and 2). The relative potency for this reduction was the same as that for the reduction of CMAP amplitudes (Table 1 and Fig. 4). After alkaloid washout, the mean MEPP amplitude was not different compared with controls for any of the alkaloids (Table 2). For 14-DN, barbinine, and deltaline, the alkaloid effect appeared to be completely reversible because the mean MEPP amplitude after alkaloid washout was greater than that in the presence of alkaloid (p < .05) and not different from controls (p > .19). However, for MLA, the alkaloid effect appeared to be incompletely reversible because mean MEPP amplitudes after washout, although larger than in the presence of alkaloid, were not significantly different from one another (p = .09). When the amplitudes of all of the MEPPs collected in the MLA studies were compared rather than comparing the mean values for each treatment, the amplitudes were significantly larger before and after alkaloid administration.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Representative signal averaged MEPPs. Three superimposed traces (control, in the presence of alkaloid, and after alkaloid washout) are shown for each alkaloid. Each alkaloid reduced the MEPP amplitude. Averages were assembled as described in Materials and Methods; each trace is an average of 15 MEPPs, except for the deltaline-treated trace, which is an average of 11. The alkaloid concentrations that produced the illustrated responses were 0.1 µM MLA (a), 0.05 µM 14-DN (b), 0.5 µM barbinine (c), and 200.0 µM deltaline (d). Scale: vertical, 100 µV, horizontal, 10 ms.

View larger version (30K): [in this window] [in a new window]   Fig. 5.   a-d, relative cumulative frequency distribution of MEPP amplitudes. MEPP amplitudes were plotted as relative cumulative frequency distributions for MLA (a), 14-DN (b), barbinine (c), and deltaline (d). The number of MEPPs was normalized to unity for each alkaloid to facilitate comparing the pre- (solid line) and post- (broken line) alkaloid effects. In each case, the alkaloid caused a left shift in the distribution of MEPP amplitudes but had no effect on the distribution of MEPP amplitudes around the mean. The number of preparations and the number of MEPPs sampled were different for each alkaloid (Table 2). For each preparation, MEPPs were recorded in the same muscle fiber before, during, and after alkaloid administration.

View this table: [in this window] [in a new window]   TABLE 3 Comparison of alkaloid effects on MEPP decay constants Each averaged MEPP is an average of 15 individual MEPPs for MLA; 11, 15, and 11 for 14-DN controls, treated and rinse, respectively; 10 for barbinine; and 14, 11, and 11 for deltaline controls, treated and rinse, respectively. Measurements of  were made from a subset of MEPPs sampled from all of the animals (Table 2).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Correlation of alkaloid concentrations that produced an equivalent reduction in CMAP and MEPP amplitudes. For each alkaloid, a concentration (micromolar) was chosen that reduced MEPP and CMAP amplitudes by an equal amount; the concentration of each alkaloid producing this effect was then plotted. Data were obtained from the theoretical curve fits in Fig. 3 and the MEPP data in Table 1. Analysis by linear regression yielded a correlation coefficient of 0.99.

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Correlation of alkaloid LD50 and IC50 values for CMAP blockade. The IC50 values obtained in the present study (Table 1) were plotted against previously published LD50 values in mammals (Manners et al., 1995). Linear regression analysis yielded a correlation coefficient of 0.98.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Electrophysiological measurements were used to characterize the effects of the Delphinium norditerpenoid alkaloids MLA, nudicauline, 14-DN, barbinine, and deltaline on synaptic transmission at lizard neuromuscular junctions. Neuromuscular synapses were selected for these studies because a single type of nAChR mediates neuromuscular transmission and because the clinical signs of Delphinium poisoning are consistent with a curariform block of this receptor. In addition, MLA is known to interact with muscle-type nAChRs (Dozortseva, 1959; Nambi-Aiyar et al., 1979; Ward et al., 1990; Garcha et al., 1993; Yum et al., 1996; Tian et al., 1997). The IC₅₀ value obtained for MLA-induced CMAP blockade in lizards (1.5 µM) is in good agreement with the previously reported IC₅₀ value (2.3 µM) for blockade of nerve-evoked twitch in rat diaphragm (Nambi-Aiyar et al., 1979). All five alkaloids blocked neuromuscular transmission in a concentration-dependent manner, and the relative potencies among the alkaloids were similar to those reported for in vivo toxicities in mammals (Manners et al., 1995).

Nudicauline, 14-DN, MLA, and deltaline reduced CMAP amplitudes with similarly steep concentration-dependent relationships (Table 1). The similarity in these slopes and the similar characteristics for MEPP amplitude reduction suggest that these four alkaloids block neuromuscular transmission via the same mechanism. The finding that MLA and 14-DN block acetylcholine-induced muscle contraction indicates that these alkaloids block nAChRs at the neuromuscular junction. The results of our studies combined with the well-established competitive interaction of MLA with muscle-type receptors (Dozortseva, 1959; Nambi-Aiyar et al., 1979; Ward et al., 1990; Garcha et al., 1993; Yum et al., 1996; Tian et al., 1997) suggest that nudicauline, 14-DN, and deltaline are likely competitive nAChR antagonists at the neuromuscular junction.

All of the Delphinium alkaloids reduced the mean MEPP amplitude without affecting the distribution of amplitudes about the mean, suggesting that these alkaloids act postsynaptically to reduce synaptic efficacy. The amplitude of the postsynaptic potential is proportional to the number of open channels when its amplitude is less than 10% of the resting membrane potential (McLachlan and Martin, 1981). MEPP amplitudes were usually between 0.5 and 0.75 mV, whereas resting membrane potentials were between 75 and 85 mV. Alkaloid concentrations insufficient to block neuromuscular transmission decreased MEPP amplitudes by about 30 to 40%, suggesting that the alkaloids reduced the number of open nAChRs by a similar amount. The leftward shift of the curves in the cumulative frequency histograms (Fig. 5, a-d) is consistent with this interpretation.

The values for the coefficient of variation in MEPP amplitudes reported here are similar to those reported for mammalian neuromuscular junctions (Boyd and Martin, 1956a). Alkaloid-induced changes in the coefficient of variation would be consistent with variability in synaptic vesicle loading, implying a presynaptic effect. None of the alkaloids affected the coefficient of variation (Table 2), indicating that quantum size remained uniform in the presence of alkaloids. These results are consistent with postsynaptic nAChR blockade.

The alkaloids 14-DN, MLA, and deltaline reduced MEPP amplitudes without affecting the resting membrane potential, the rate of MEPP decay, or the coefficient of variation of MEPP amplitude. These results suggest that all three alkaloids act by blocking nAChRs and not by altering either muscle fiber passive membrane properties or neurotransmitter release. This interpretation is consistent with recent findings showing that MLA reduces spontaneous miniature end-plate currents without altering the time course of their decay in rat diaphragm (Tian et al., 1997). The similarity in the effects of MLA, 14-DN, and deltaline on MEPP amplitude, MEPP amplitude distribution, and MEPP decay constant is consistent with these alkaloids competing with acetylcholine for a common binding site on the nAChRs at the neuromuscular junction.

The high correlation coefficient obtained by comparing alkaloid concentrations equipotent for reducing CMAP and MEPP amplitudes supports the hypothesis that the MSAL and MDL alkaloids impair neuromuscular transmission by blocking postsynaptic nAChRs. An MEPP results when a quantum of neurotransmitter opens closely packed postsynaptic nAChRs (Stiles et al., 1996), and CMAP blockade at a site or sites other than nAChRs would yield little or no correlation between receptor and action potential blockade. For example, blockade of voltage-gated Ca²⁺ channels that govern transmitter release would block evoked transmitter release but be independent of effects on the amplitude of spontaneous MEPPs. Similarly, blockade of voltage-gated Na⁺ channels would block neuromuscular transmission by preventing action potential propagation without affecting MEPP amplitude.

As a more direct test of the ability of alkaloids to block nAChRs, acetylcholine-induced muscle contraction was assayed in the presence of MLA, 14-DN, and deltaline. Both MLA (10 µM) and 14-DN (5 µM) blocked muscle contractions induced by exogenously applied acetylcholine but had no effect on contractions induced by high K⁺ PSS. This result indicates that these two alkaloids block muscle contraction by blocking nAChRs. Deltaline (500 µM) failed to block exogenous acetylcholine-induced muscle contraction. This result appears to be inconsistent with deltaline-induced nAChR blockade; however, this assay may be inadequate to measure the effects of deltaline on exogenous acetylcholine-induced muscle contraction because of the low potency of deltaline for blocking CMAPs and reducing MEPP amplitudes.

The highly significant correlation between the LD₅₀ and the IC₅₀ for neuromuscular blockade is consistent with the hypothesis that Delphinium alkaloids exert their lethal effects by blocking nAChRs at neuromuscular junctions on skeletal muscle fibers. However, synaptic transmission in autonomic ganglia requires functional nAChRs. MLA exhibits nanomolar affinity for 7 nAChRs found in autonomic ganglia, but blockade of the 7 nAChRs alone fails to block synaptic transmission in these ganglia (Zhang et al., 1996). This is consistent with early studies on MLA suggesting that the acute, lethal effects of MLA do not result from blocking autonomic function (Dozortseva, 1959). The ability of the other alkaloids used in this study to block autonomic synaptic transmission is unknown. Therefore, these alkaloids might exert their lethal effects by blocking synaptic transmission in autonomic ganglia. Low concentrations of barbinine (0.5 µM) reduced MEPP amplitudes in a manner similar to the other alkaloids. In contrast to 14-DN, MLA, and deltaline, which had no effect on the time constant for MEPP decay (_MEPP), barbinine reversibly reduced _MEPP by about 13%. In principle, the effects of barbinine could result from decreasing the membrane resistance (R_m), but reductions in R_m can account for only a portion of the approximately 40% decrease in MEPP amplitude. If barbinine shortens _MEPP by reducing R_m, it exerts this effect without affecting the resting membrane potential.

The effects of barbinine on _MEPP could result entirely from the effects of barbinine on nAChR function. The value of _MEPP is affected by both the time constant of the membrane (_m) and the mean nAChR channel open time (_ch), which can be defined as 1/_MEPC, where _MEPC is the time constant of the miniature end-plate current decay. The influence of _m dictates that the mean of _MEPP exceeds the mean of _MEPC. However, _ch would influence _MEPP if _MEPC were an appreciable fraction of _MEPP (Sieb et al., 1996). For lizard intercostal muscles, _MEPC is 1.16 ms (Land et al., 1984), about 30% of the _MEPP reported here (Table 1). About 15% of the _MEPC values collected by Land et al. (1984), overlap with our mean _MEPP value, suggesting that barbinine-induced reductions in _ch might be observed as decreases in _MEPP. Allosteric modulation of nAChRs by barbinine could elicit such reductions. Open-channel block would also decrease _ch, but this explanation is inconsistent with the effects of barbinine on CMAPs.

Barbinine also affected CMAPs differently than the other alkaloids. Because of the safety factor for neuromuscular transmission, steep concentration-response relationships are expected for agents that block synaptic transmission at the neuromuscular junction (Paton and Waud, 1967; Waud and Waud, 1972; Lingle and Steinbach, 1988). The concentration-response relationship for barbinine, which exhibited a slope of 1.45, was considerably less steep than that for the other alkaloids (Fig. 2 and Table 1). These results suggest that compared with the other alkaloids, barbinine decreases CMAP amplitudes via alternative or additional mechanisms. Open channel block is a well known mechanism to reduce receptor function, but it seems unlikely that barbinine works through this mechanism at the neuromuscular junction. For an open channel block, the susceptibility to neuromuscular blockade would be expected to increase with increasing alkaloid concentrations. In contrast, a given reduction in safety factor required a proportionally greater increase in alkaloid concentration for barbinine than for any of the other alkaloids (Fig. 3). The mechanisms by which barbinine effects neuromuscular blockade have not been completely elucidated. Barbinine is a minor constituent of North American Delphinium, limiting available amounts and preventing additional experiments to investigate the effects of barbinine on membrane resistance and nAChR function.

Functional Implications of Alkaloid Structure. Measurements of alkaloid effects on CMAP and MEPP amplitude provided a functional assay for investigating the relationships between alkaloid structure and alkaloid function. Earlier binding studies on nAChRs from the central nervous system related the chemical substitution at C14 of the alkaloid to alkaloid affinity for these nicotinic receptors (Kukel and Jennings, 1994; Hardick et al., 1996). The rank order of potency for blockade of CMAPs and MEPPs at the neuromuscular junction is consistent with binding studies on neuronal nAChRs (Hardick et al., 1995, 1996; Dobelis et al., 1997). However, the relationship between the moiety substituted at C14 and alkaloid potency in the functional assays is not immediately clear (Fig. 1 and Table 1). If differences in alkaloid potency result simply from steric interactions of the C14 substitutions, the predicted order of potency would be either nudicauline > MLA > 14-DN > barbinine, or the reverse. The order of potency for diminishing CMAP amplitudes was nudicauline > 14-DN > MLA > barbinine. Except for nudicauline, which was not tested against MEPPs, this rank order of potency was the same for reducing MEPP amplitudes.