Pharmacology of Marine Toxins - American Chemical Society

Chapter 1

Pharmacology of Marine Toxins Effects on Membrane

Channels

Gary Strichartz and Neil Castle

Downloaded by 80.82.77.83 on October 26, 2017 | http://pubs.acs.org Publication Date: January 29, 1990 | doi: 10.1021/bk-1990-0418.ch001

Anesthesia Research Laboratories, Harvard Medical School, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115 I o n channels i n plasma membranes are primary targets o f marine toxins. These channels are important regulators o f a cell's physiology, and many o f the pathophysiological effects o f toxins arise from actions o n i o n channels. I n this chapter w e present the voltage-gated Na+ channel, as it exists i n excitable cells, as an example o f a receptor w i t h m u l t i p l e binding sites for different types o f toxins. T h e toxins are classified according to their physiological effects and described by their chemistry. Occluders, activators and stabilizers are considered as modes for toxins binding to and acting directly o n the i o n channel. Other, indirect actions o f toxins, mediated by cellular metabolism are also discussed to provide a broad overview o f the many possible modes for action o f marine toxins.

" C h a n n e l s " is the w o r d used to denote the entities that account for passive i o n transport across a l l membranes (1). Channels operate to produce very small, transient changes i n i o n concentrations w i t h i n cells. Ionic balances are important for the regulation o f cellular activities through c o n t r o l o f intracellular p H , i o n concentration, and membrane potential. F o r example, cytoplasmic C a levels critically determine the activities o f a variety o f enzymes, many o f which are elements i n extensive "cascades" that amplify the effects o f single "trigger" molecules, such as hormones (2). Internal Ca is often elevated by influx o f extracellular C a through C a - s e l e c t i v e channels (3). T h e example o f several i o n channels i n mediating some rapid physiological events i n neuromuscular systems is diagrammed i n Figure 1. A short primer o n electrophysiology w i l l benefit the subsequent description o f toxin actions. M e m b r a n e potential is the voltage o f the inside o f the cell (or organelle) relative to the outside and is most often dominated by the K diffusion potential; typical values range from -90 to -40 m V (negative inside) and result from the selective permeability to K o f the resting membrane. This permits a few K ions to flow passively o u t from the cell, leaving the interior with a residual net negative charge. This is an example o f net outward i o n i c current (charge flow) polarizing the membrane; conversely, net inward current depolarizes the membrane. Thus treatments that reduce outward K current at rest (e.g., K channel blockers) tend to depolarize cells, as do treatments that increase inward N a currents (e.g., N a channel activators). Cells are also depolarized by indirect dissipation o f the i o n ( K ) gradients through metabolic poisons o r by direct i n h i b i t i o n o f the active pumps, either the Na /K A T P a s e o r other exchange pumps that move o n e species o f i o n i n trade for another (4). F r o m a biochemical perspective, it is often helpful to view i o n channels as the enzymes that catalyze the reactions o f passive i o n transport. Indeed, these 2 +

2 +

2 +

2+

+

+

+

+

+

+

+

+

+

+

0097-6156/90/0418-0002$06.00/0 o 1990 American Chemical Society

Hall and Strichartz; Marine Toxins ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

1.

Marine Toxins and Membrane Channels

STRICHARTZ & CASTLE


spinal motorneuron

sympathetic n.s.

F i g u r e 1. A depiction o f the several different ionic currents necessary for the acute function o f neuromuscular transmission i n the skeletal m o t o r and the efferent autonomic nervous system. T h e boxed current designations are associated, by the arrows, with those cellular regions where their physiological role is most evident, although these currents often exist i n other regions o f the cell. I M T R i a t e d current; I = acetylcholine-activateacurrent; I = i o n selective current where X = N a , C a , K , C K , as noted. =

n e u r o t r a n s m

t t e r

a c t r v

X

ACh

x

+

2 +

+


3

4

MARINE TOXINS: ORIGIN, STRUCTURE, AND MOLECULAR PHARMACOLOGY

enzymes have turned out, when analyzed, to be proteins, and they express the classical enzyme attributes o f substrate specificity (ion selectivity), i n h i b i t i o n , and regulation. That many o f the toxins to be described here act as inhibitors o r regulators o f i o n channel activities further consolidates this analogy. Consequences o f channel activity often feed back o n the channel itself, through reciprocally antagonistic actions that may be biochemical o r physiological. A good electrophysiological example is posed by certain neurons that act as "bursting pacemakers", firing impulses i n "bursts" for brief periods interrupted by impulse-free intervals (5). T h e initial burst results from an underlying, l o w frequency "pacemaker depolarization", due to the membrane potential-dependent opening o f C a channels restricted to the soma; the subsequent activation o f similarly "voltage-gated" N a channels produces the superimposed spiking burst discharge which propagates along the nerve's axon. T h e accompanying entry o f C a ions o f itself activates "calcium-dependent" K channels (6), which catalyze K efflux and thus repolarize the soma membrane, shutting off the C a channels and the pacemaker potential. Restoration o f " i n i t i a l " conditions, by uptake and transport o f C a (7), witnesses a reappearance o f the pacemaker C a current and re-initiation o f the bursting output. Intracellular C a may also trigger the catalyzed phosphorylation o f proteins (8), release o f internally sequestered C a (9), and itself may be elevated through the activation o f other "second messenger" systems (10). T h e possibilities o f reciprocal, complex interactions are numerous, and a single response to i o n channel activation is probably a rare physiological event. +

2 +

+

+


2 +

2 +

2 +

2 +

2 +

A l t h o u g h most examples o f i o n channel actions are drawn from neurobiology, their importance goes well beyond cells specifically designated as neuronal. I o n channels exist i n every animal and plant cell membrane, plasmalemma and organelle included. T h e voltage-gated variety, previously considered hallmarks o f "excitable cells" o f nerve, heart, and muscle, are widespread among plasma membranes. Fibroblasts ( i i ) , endocrine cells (12-14), and those o f the i m m u n e system (15) all possess and utilize i o n channels i n their repertoire o f physiological responses. Therefore, systemically distributed toxins may directly affect many types o f tissues not merely those traditionally termed "excitable". O u r goal i n this chapter is to describe mechanisms through w h i c h marine toxins may act o n i o n channels. Examples often cite the actions o f specific toxins, but sometimes a potential target w i l l be noted o r the effects o f a non-marine toxin described. Since new toxins are reported regularly, a broad basis for the actions o f future discoveries seems to us appropriate here. It is not o u r intention to c o m p i l e an exhaustive survey, but rather to categorize and to organize toxins into mechanistically identifiable classes. A t first this seems a simple, direct task, but we w i l l quickly demonstrate the complexity o f actions. F r o m the present understanding o f i o n channel function, no single toxin appears to exert only o n e simple action. O n e o f the reasons for this is that i o n channels are complex proteins whose functions involve subtle, concerted conformational changes, and, at present, o u r understanding o f the structural basis for these functions is inadequate for a proper, molecular description o f toxin-induced phenomena. Nevertheless, the observed effects are important to describe and understand, for o u r comprehension o f channel mechanics is paralleled by our evolving knowledge o f their molecular toxinologyClassification of Toxin Actions T h e actions o f toxins may be classified according to the o u r current perspective o f i o n channel function. Channels open, o r "gate", i n response to a range o f " s t i m u l i " , variables that perturb the p o p u l a t i o n distribution among a set o f possible channel


1.


STRICHARTZ & CASTLE

5

conformations. Recognized perturbations include changes i n membrane potential, ions themselves, neurotransmitters binding directly to the channel and a range o f "regulatory" substances which modify the response to other perturbations through covalent or non-covalent association with the channel itself (1). Thus, a channel performs the operations o f i o n conductance, when it is "open", o f gating between o p e n and closed states (of which there are often several), and o f regulating between "activatable" (gateable = openable) states and other, non-activatable (non-gateable o r "inactivated") states. Ions may carry current through the channel's " p o r e " (16), b i n d i n the channel's pore and occlude it (17), b i n d to o r associate w i t h structural features beyond the pore and thereby modify gating (18), and alter the actions o f other ligands (19). There are many different types o f i o n channels, exhibiting different i o n selectivities (thus, " N a " channels, " K " channels, etc.), and a variety o f gating and regulatory mechanisms. R a t h e r than describe all o f them here, we w i l l use as an example the voltage-gated Na channels. A l t h o u g h certain quantitative physiological and pharmacological properties o f N a channels differ among neuronal, skeletal muscular, and myocardial N a channels, we w i l l treat them here as a single entity, with exceptions as noted. The advantage o f using the N a channel as an example rests w i t h its elaborate and welldocumented toxinology. M a n y animals have evolved toxins as offensive o r defensive weapons that are directed, respectively, against their prey's o r predator's N a channels, and these have been characterized extensively (20). T h e disadvantage o f exemplifying the N a channel lies i n its apparent insensitivity to endogenous regulatory mechanisms, either reversibly binding ligands (hormones, peptides) o r covalent modification (e.g., phosphorylation), which are c o m m o n l y seen with potassium and calcium channels (see below). These activities may o r do exist, respectively, o n N a channels but their physiological importance remains u n k n o w n . +

+


+

+

+

+

+

+

+

+

T h e toxinology o f N a channels is best understood i n the context o f their n o r m a l physiology. S o d i u m channels exist i n several different conformations o r states, most o f which do not conduct ions across the membrane (i.e., are "closed") and some which do conduct ions (i.e., are "open"). Transitions among these different states are termed "gating" reactions. A very simple scheme for N a channel gating and its resultant kinetic output is diagrammed i n Figure 2. T h e channel has two non-conducting states — one existing i n resting membranes (R, resting) and one i n membranes held depolarized (/, inactivated) - as well as one conducting state (O, open). Numbers at the arrows denote the relative rate constants for the transitions among different states. In normal, toxin-free channels responding to membrane depolarization, the R —• O process, called "activation", is markedly faster than inactivation (O - » 7) and is moderately reversible; i f inactivation did not occur, 8 0 % o f the channels w o u l d populate O at equilibrium. However, inactivation reactions do occur and are almost totally irreversible, 99.9% o f the channels populating the I state at the end o f l o n g depolarizations. T h e transient p o p u l a t i o n o f O is shown by the heavy dashed curve i n F i g u r e 2, which graphs the dynamics o f the fraction o f N a channels i n state O after a depolarizing step i n membrane potential. In contrast to this "macroscopic" p o p u l a t i o n function, the " m i c r o s c o p i c " behavior o f a single N a channel is depicted by the horizontal traces displayed o n the same time scale. E a c h o f these traces shows the probability o f o n e sodium channel being open. F o r the toxin-free channel, the first o p e n i n g occurs shortly after depolarization; temporary reversal o f opening ("deactivation") accounts for many o f the early closings, but eventually the virtually irreversible O —> I reaction traps the channel i n the inactivated state. Inactivation must also be occurring directly from the resting state (i.e., R —> I) i n order to assure microscopic reversibility, but normally that process is slow and competes ineffectively with the rapid activation gating. +

+

+

At

a

different

membrane

potential

the

gating

rate

constants

would


differ.



6

+

F i g u r e 2. K i n e t i c schemes for N a channel gating (right) and the graphed time-course for single channels (solid lines, the higher position is "open") and for the p o p u l a t i o n o f many channels (broken line, the fraction " o p e n " increases upwardly). Numbers at the arrows o f the kinetic scheme are the rate constants, i n 1 0 sec" . T h e period o f simulation is 5 msec. C o m p u t e r i z e d m o d e l courtesy o f D r . D a n i e l Chernoff. 3

1


1.

STRICHARTZ & CASTLE

7


A c t i v a t i o n is slower i n less depolarized membranes and inactivation drains the open (and resting) state more effectively. I n fact, real N a channels gate by m o r e complex pathways, including several closed states intermediate between R a n d O, as well as m u l t i p l e inactivated states. Inactivation from these intermediate states is probably faster than from R, and the entire activation process, i n its fully branched entirety, is rich w i t h kinetic possibilities. However, the effects o f toxins may be understood i n general by the simpler scheme presented i n Figure 2. A real protein underlies these kinetic abstractions. S o d i u m channels are large (ca. 200 k daltons) glyco-peptides (27), heavily acylated w i t h fatty acids (22) and carrying a net charge o r d i p o l e moment (23). A n c h o r e d within the plasma membrane by hydrophobic, electrostatic, and covalent bonding, the channel protein's electrically changed regions w i l l respond to changes o f the membrane potential by conformational changes to the most stable, energetically accessible disposition. These changes constitute the gating reactions w e have simplified for Figure 2. B i n d i n g o f toxins to the channels modifies these conformational changes, altering the energetics and changing the apparent gating kinetics.


+

+

T h e k n o w n toxinology o f the N a channel classifies compounds into three categories: activators, stabilizers, and occluders. A l t h o u g h these terms describe the apparent major effect o f the toxin, it should be realized that m u l t i p l e effects are more often the rule than the exception.

Activators +

These toxic agents increase the probability that N a channels w i l l o p e n at membrane potentials where such openings are usually quite rare. A s noted above, the normal, voltage-gated process o f opening is called "activation", thus the title o f this class o f drugs. M o s t o f the k n o w n activators are not derived from marine organisms, but instead are alkaloids extracted from terrestrial plants o r amphibians, e.g., aconitine, veratridine ( V T D : veratrum spp.) o r batrachotoxin ( B T X ) , respectively (Figure 3; ref 24). These relatively lipophilic, organic molecules, typically o f mass less than 1 0 daltons, produce their full pharmacological effects whether applied intracellularly o r from without the cell, and probably act from within the membrane. T h e effects are i n h i bited noncompetitively by occluder toxins (e.g., tetrodotoxin, see below), a n d result from binding at a specific site o n the channel (25, 26). Activators alter the opening and the closing probability o f N a channels, tending to activate channels at voltages near the cell's resting potential and to prevent their normal, complete inactivation. T h e result is a long-lasting increase i n N a permeability and a corresponding depolarization from the steady inward N a current (27). B u t if the membrane potential is held constant by laboratory manipulations (voltageclamp), then the drug-modified channel's activation and inactivation kinetics may be examined under more controlled conditions (28-30). 3

+

+

+

O n e example o f channel activation by a marine toxin is shown i n F i g u r e 4. H e r e the effects o f a brevetoxin ( B v T X ) activator o n N a currents under voltage-clamp are characterized by three aspects, according to the traditional description o f such phenomena; the amplitude o f currents is reduced, the voltage range for activation is "shifted", and the kinetics o f inactivation, shown as the declining phase o f the currents, are slowed (31). It should be noted that B v T X and a related organic marine toxin, ciguatoxin (32), act at o n e site o n the channel which is separate from the other activator (e.g., BTX) binding site (33-35), yet the physiological effects are largely indistinguishable (see chapter by Baden et al., this volume). A t first these effects o f activators may appear complicated, requiring m u l t i p l e factors to explain the total response, but reference to the kinetic model for channel gating (Figure 2) clarifies the complexity. Activators may produce spontaneous opening +




8

Brevetoxin-B F i g u r e 3. T h e chemical structures o f two "classical" alkaloid activators (batrachotoxin and veratridine) and o f a recently characterized marine toxin [brevetoxin B ( B v T X - B ) ] , that acts at a different site o n the N a channel. +



p

+

+

§

m

2

+

2

N a

+

N a

Figure 4. Effects o f dihydro-brevetoxin B ( H B v T X - B ) o n N a currents ( I ) i n crayfish axon under voltage-clamp. (A) A family o f N a currents i n c o n t r o l solution; each trace shows the current kinetics responding to a step depolarization (ranging from -90 to +100 m V i n 10 m V increments). Incomplete inactivation at large depolarizations is n o r m a l i n this preparation. (B) N a currents after internal perfusion w i t h H B v T X - B (1.2 / * M ) . I inactivation is slower and less complete than i n the c o n t r o l , and the current amplitudes are reduced. ( C ) A plot o f current amplitudes at their peak value ( I ; o, o ) and at steady-state (I ; A , A for l o n g depolarizations) shows that toxin-modified channels (filled symbols) activate at m o r e negative membrane potentials ( E ) and correspond to a reduced peak N a conductance o f the axon (Reproduced w i t h permission from Ref. 31. Copyright 1984 A m e r i c a n Society for Pharmacology and Experimental Therapeutics).

A CONTROL


10


by shifting channels at rest from R to intermediate closed (C) and even open (O) states. A c t i v a t i o n o f activator-modified channels therefore requires smaller membrane depolarization (leading to increased spontaneous openings) and occurs without the kinetic delay seen i n unmodified channels (56), consistent with few o r no intermediate states to pass through before opening. A c t i v a t o r binding stabilizes the open o r preo p e n states and thus inhibits their inactivation reaction, slowing the net decline o f the Na current. T h e shift i n activator-bound channels from resting to intermediate closed forms is equivalent to having a higher affinity o f these closed forms for the activators. This behavior is characteristic o f "modulated receptor" models, described later i n this chapter. Channels react faster with most activators i f the membrane is repetitively depolarized, again consistent with the drug's binding to activated channels (28, 29, 37). W h e n open, the drug-modified channels have a lower conductance and a different i o n selectivity compared to unmodified channels, i.e., the detailed energetics o f i o n transport are altered (30, 36, 38). Finally, modified channels may close and re-open many times i n a depolarized membrane, i n contrast to n o r m a l channels which rapidly inactivate (38). O n e activator molecule binding to a N a channel influences the total function, including all the gating and the i o n permeation. It is pharmacologically noteworthy that these l i p o p h i l i c ligands, o f a modest size compared to the large channel protein, exert such extensive influence. E i t h e r the structural features subserving channel functions are localized i n a small, central domain o f the protein, o r concerted, moleculewide structural changes attending these functions can be affected through o n e integrative site, o r the activator molecule can move i n milliseconds among m u l t i p l e points o n the N a channel. T h e summed effects o f activators are to increase the p o p u l a t i o n o f open channels (albeit o f reduced N a conductance) over a broad range o f membrane potentials. U n d e r n o r m a l conditions this results i n slow depolarization o f the membrane attended by a period o f rapid, spontaneous impulse firing. Nerve, muscle, and heart all appear to be sensitive to activators, which probably also penetrate the b l o o d brain barrier, affecting the central nervous system; the pathophysiology o f activator intoxication is extensive and interactive. A c t i v a t o r effects are antagonized by local anesthetics and thus, ironically, an antidote might be found i n the intentional, transitory administrat i o n o f systemic local anesthetic i n conjunction with artificial respiration, at least u n t i l the activators can be cleared from the body. B y partially blocking activator-modified N a currents, the often toxic systemic anesthetics can reverse the excited responses o f w h o l e organs and prevent the persistent depolarization that may otherwise result i n cell death.


+

+

+

+

+

Stabilizers A second category o f toxins has an apparently simple action. These drugs p r o l o n g and inhibit the inactivation phase o f N a currents during depolarization, thereby stabilizing channels i n the o p e n state without affecting the activation process (Figure 5). T h e best examples o f such toxins from marine sources are small peptides isolated from nematocysts o f a variety o f anemones (39) and larger proteins found i n venoms o f C o n i d a e (40). W h i l e some o f these toxins at high concentrations do reduce the amplitude o f N a currents, the more typical effect is prolongation o f the p e r i o d o f high N a conductance. In non-voltage clamped membranes, the effect o f such stabilizers is to greatly p r o l o n g the action potential, generating a "plateau" depolarization lasting from tenths o f seconds to seconds (41). This l o n g depolarization greatly increases the amount o f C a that w i l l enter a cell through voltage-gated C a channels, and also serves to generate "local circuit" currents that spread passively to +

+

+

2 +


2 +



STRICHARTZ & CASTLE

I

I 1ms

a

I

1

5 ms Figure 5. M u l t i p l e actions o f toxin II from Anemonia sulcata ( A T X II) o n voltage-clamped N a currents ( I ) from amphibian myelinated nerve. This "stabilizer" toxin works i n a dose-dependent manner to inhibit channel inactivation (see b o t t o m panel) and, as a consequence, delay the time o f peak current (see top panel). T h e reduction o f peak current amplitude does n o t result directly from these kinetic alterations and is not observed w i t h a l l stabilizers (Reproduced w i t h permission from R e f . 39. Copyright 1981 SPPIF). +

N a


12


adjacent, t o x i n - f r e e regions o f a cell (e.g. along an axon) and initiate repeated impulses there. T h e net result i n either case is an increase i n depolarization-related activities such as secretion, contraction, etc. Stabilizers b i n d at a site separate from those o f traditional activators and o f ciguatoxin-brevetoxin, but they exert a synergistic action o n b o t h types o f activators (35, 42). This action potentiates the activators and generally increases their efficacy, yielding larger depolarizations at lower doses (42); it occurs uniquely w i t h the peptide stabilizers and not w i t h ions or oxidants that also slow the inactivation o f N a current (37). T h e actions o f stabilizers present one o f the classic voltage-dependent toxin effects. Depolarizations o f the membrane sufficient to energetically drive toxin-bound channels to an inactivated conformation also lead to dissociation o f the stabilizer (4345). T h e energetics o f this process are related to its kinetics i n an interesting way. F o r a tightly binding stabilizer toxin, o f K ~ l O ^ M , small depolarizations (e.g., to -20 m V ) result only i n slow dissociations that require minutes to complete, whilst large depolarizations (e.g., to + 50 m V ) result i n complete dissociation w i t h i n 10 msec. Stabilizers that b i n d less strongly, with K ~ l O ^ M , reverse m o r e rapidly at lower depolarizations. T o x i n dissociation results not from a direct effect o f the membrane potential o n the binding reaction per se, but o n a voltage-dependent m o d u l a t i o n o f the toxin binding site from a high to a l o w affinity form. A p p a r e n t l y the inactivated state has an affinity for stabilizers that is at least an order o f magnitude less than that o f the resting or open states. Sufficiently positive potentials force the stabilizer-bound channel into an inactivated conformation, leading to rapid relaxation o f toxin binding to the new equilibrium corresponding to the l o w affinity, inactivated conformation. +


D

D

Occluders Tetrodotoxin ( T T X ) and the saxitoxins ( S T X ) are classic examples o f marine toxins acting as the third class o f agents, occluders. These small (300-450 daltons), organic cations act o n l y from the external surface o f N a channels to produce a readily reversible, yet usually high affinity (K ~ 1 0 M ; block o f the channel (see Figure 6). This block is antagonized by N a , C a , and H (19, 47) and was originally thought to correspond to a "plugging" o f the channel's pore (48). T h e rapid kinetics o f activat i o n and inactivation o f conducting channels are unaffected by T T X / S T X , and the movement o f charge associated w i t h activation remains unchanged (23). T h e overall effect appears equivalent to reducing, reversibly, the number o f channels that have patent conductance pathways (49; see Figure 6). B u t more recent evidence questions this simple plugging model; the structural dependence o f toxin activity is more c o m p l i cated than necessary for simple pore blockade (50), and the physiological behavior extends beyond simple occlusion (47). In mammalian cardiac N a channels, an unusually l o w affinity for T T X (TC l O ^ M j is accompanied by an apparent "voltage-dependent" action - the toxin block is altered by the membrane potential, probably through some conformational change o f the channel (51). C h a n n e l gating is also subtly modified by these toxins; i n b o t h cardiac (52) and neuronal (47) tissue the tendency, after l o n g ( > 1 sec) depolarizations, o f the channels to dwell i n a state from w h i c h they can o n l y be slowly made activatable (i.e., a "slow inactivated state"), is furthered by T T X and S T X (Figure 7). T h e underlying mechanism for this gating modulation is completely u n k n o w n . +

- 9

D

+

+

+

+

D

+

C e r t a i n occluders also discriminate among N a channels from neuronal and skeletal muscle. B u t i n this case the blocking ligands are small peptides, the jz-conotoxins from the mollusc Conus geographus. This molecule binds tightly to muscle N a channels, effectively reducing N a * current (53; see Figure 6 A ) , and also can displace b o u n d +




STRICHARTZ & CASTLE

B 1 0 8

11

Control O p e n iMMMMMiMM

closed

I

; i

•

:

l h . ;J

3

1 JJM G I I I A ~I

r u — j

Q

S

^

10s 10nM STX

Figure 6. Blockade o f N a channels by /i-conotoxin ( G I I I A ) or by saxitoxin ( S T X ) (46). A . Macroscopic currents i n frog skeletal muscle recorded i n control solution (a) and 10-15 m i n (b) and c) 2 5 - 3 0 m i n (c) after external perfusion w i t h G I I I A (2 / z M ) . B . Single channel currents from skeletal muscle vesicles which are fused with planar l i p i d bilayers and modified by batrachotoxin, ensuring long open times under c o n t r o l conditions. T h e downward deflecting closing events are caused by binding o f G I I I A or S T X , and the duration o f the closed (blocked) state is inversely proportional to the particular toxin's dissociation rate. N o t e the difference i n time scales. (Reproduced with permission from Ref. 46. Copyright 1986 T h e N e w Y o r k Academy o f Sciences). +


Hall and Strichartz; Marine Toxins ACS Symposium Series; American Chemical Society: Washington, DC, 1990. I

100 +

I

t

200 r e c

(msec)

I

300

test pulse

I

400

I

500

Figure 7. Slow inactivation o f N a channels is potentiated by S T X . T h e graph shows the t i m e required for the recovery o f N a channels t o an activatable state after a l o n g (1 sec, +50 m V ) inactivating depolarization. W h e n tested by a brief test pulse, control currents (A) recovered i n a fast ( r = 233 msec) phase. A d d i t i o n o f S T X ( o , 2 n M , which approximately halved the currents w i t h n o inactivating pulse) approximately doubled the fraction o f currents recovering i n the slow phase and also increased the time constant o f slow recovery. T h e fast recovery rate was unaffected. (Reproduced with permission from Ref. 47. Copyright 1986 T h e N e w Y o r k A c a d e m y o f Sciences).

+

I

0

conditioning pulse


1.


STRICHARTZ & CASTLE

15

radiolabelled S T X , showing that it acts at the same site (54, 55). Reversible blockade o f single muscle channels resembles that by S T X (Figure 6 B ) ; although the potency o f /i-conotoxins is lower, the dissociation rate is also smaller. In neuronal and cardiac Na channels, the blockade o f N a currents and the i n h i b i t i o n o f S T X binding are far weaker, thus /i-conotoxins discriminate between channels that b i n d T T X and S T X about equally. +

+

+

T h e results from such occluders show that N a channels differ i n structure between their internal and external surfaces, have binding sites for a variety o f m o n o valent and divalent cations, and are pharmacologically different despite very similar physiological actions i n nerve, muscle, and mammalian cardiac cells. It is noteworthy that two other groups o f small peptides from the same organism's venom affect other i o n channels. T h e w-conotoxins block certain C a channels, and a-conotoxins block acetylcholine-activated i o n channels as found at the neuromuscular j u n c t i o n (see Figure 1; see ref. 56). Conus venoms contain a variety o f toxins that affect N a channel gating (40, 57) and others that modify K channels, as well as a high concentration o f phospholipase (58). A s w i t h other offensive (predatory) venoms, the mixture o f active components has a synergism that exploits the prey's n o r m a l physiology to effect a rapid, convincing paralysis. 2 +


+

+

+

The Na Channel Is a Modulated Receptor In all o f the examples cited above, changes i n the states o f the channel are affected by toxin action. Reciprocally, the binding and actions o f the toxins are changed by voltage-driven gating among different channel conformations. Such a dynamic interact i o n has been termed the "modulated receptor" model, and was originally proposed as an explanation for effects o f local anesthetics o n N a channels (59). F o r the toxin classes reviewed here, stabilizers have selective affinity for resting over inactivated channels, occluders seem to increase the tendency o f channels to enter the "slow inactivated" state, and most activators b i n d more tightly and certainly more rapidly to an activated state o f the N a channel. A n exception are the brevetoxins, which, like B T X o r veratridine, alter channel kinetics to favor the open state, but whose onset o f action is not accelerated by channel activation. +

+

A second aspect o f the modulated receptor is the interaction among toxins o f different classes. W e previously noted that activators (including brevetoxin), were potentiated by stabilizers. In addition, the binding o f brevetoxin is slightly enhanced by occluders (60), and the blocking action o f occluders ( S T X and T T X ) becomes voltagedependent i n activator ( B T X ) - m o d i f i e d channels (67), whereas it normally exhibits no voltage-dependent action (62). Thus, the N a channel has multiple, separate, yet interactive binding sites for several types o f toxins (Figure 8). Taken together with the broad effects o f membrane potential o n individual toxin actions, it seems probable that structural changes ripple through extensive domains o f the channel during n o r m a l physiological gating. +

Toxin-Induced Permeabilities Certain marine toxins induce "new" i o n permeabilities, rather than modifying existing i o n channels. Palytoxin ( P T X ) , a large organic molecule, for example, irreversibly increases the cation permeability o f many cell membranes, to b o t h N a and other metal and organic cations (63-65; see chapter by O h i z u m i i n this volume). T h e mode o f action may involve the N a / K pump, converting it to a passive channel (or carrier-exchanger), for the depolarizing effect o f P T X can be inhibited by preincubation o f the tissue with ouabain (strophanthin), a glycoside that selectively i n h i bits the N a / K p u m p (65). Alternatively, some other protein o r carbohydrate o n the +

+

+

+

+


Hall and Strichartz; Marine Toxins ACS Symposium Series; American Chemical Society: Washington, DC, 1990. +

Figure 8. A schematic for the toxin binding sites o n the voltage-gated N a channel. Toxin-free o p e n and closed conformations are drawn at the left and center. Separate sites are depicted w i t h i n the membrane for activators such as B T X , V T D ( A ) , and brevetoxin ( B ) ; these are c o u p l e d to each other and to the a-peptide toxin site (a), which is kinetically linked to the ^-peptide toxin site (ft see ref. 20). N e a r the outer o p e n i n g o f the pore is a site ( G ) for S T X and T T X w h i c h is affected by b i n d i n g at site A and w h i c h can modify inactivation gating.

Pharmacology of Marine Toxins - American Chemical Society

Recommend Documents