sensitive channel of 35 pS conductance (IK channel) found in the plasma ... dissociation constant (Ka(0)) for TEA block of the IK channel was reduced from 44 ...
Bioscience Reports, Vol. 15, No. 6, 1995
Novel Actions of Ryanodine and Analogues Perturbers of Potassium Channels H . V a i s a n d P. N . R. U s h e r w o o d ~ The effects of ryanodine, 9,21-didehydroryanodine and 9,21-didehydroryanodol on two types of K* channel (a maxi, Ca2--activated, 170pS channel (BK channel) and an inward rectifier, stretchsensitive channel of 35 pS conductance (IK channel) found in the plasma membrane of locust skeletal muscle have been investigated. 10-gM-10-SM ryanodine irreversibly induced a dose-dependent reduction of the reversal potential (Vrev) of the currents of both channels, i.e. from - 6 0 mV in the absence of the alkaloid to - 1 5 mV for t0-5M ryanodine, measured under physiologically normal K § and Na + gradients. In both cases the change in the ionic selectivity was Ca2+-independent. 9,21-didehydroryanodine and 9,2]-didehyroryanodol also reduced Vrev, but only to -35 mV during application of 10-5M of these compounds. Additionally, 9,21-didehydroryanodine reversibly diminished the conductances of the two K § channels. To test the hypothesis that ryanoids increase Napermeability by enlarging the K § channels, the channels were probed with quaternary ammonium ions during ryanoid application. When applied to the cytoplasmic face of inside-out patches exised from locust muscle membrane, TEA blocked the K § channels in a voltage-dependent fashion. The dissociation constant (Ka(0)) for TEA block of the IK channel was reduced from 44 mM to 1 mM by 10.7 M ryanodine, but the voltage-dependence of the block was unaffected. Qualitatively similar data were obtained for the BK channel. Ryanodine had no effect on the Kd for cytoplasmically-applied TMA. However, the voltage-dependence for TMA block was increased for both K- channels, from 0.47 to ~0.8 with 10-6M ryanodine. The effects of ryanodine on TEA and TMA block support the hypothesis that ryanodine enlarges the K § channels so as to facilitate permeation of partially hydrated Na-- ions.
KEY WORDS: Ryanodine; 9,21-didehydroryanodine; 9-21-didehydroryanodol; potassium channels; locust muscle; quaternary ammonium ions.
INTRODUCTION D u r i n g t h e p a s t d e c a d e o r so a w e a l t h o f l i t e r a t u r e o n r y a n o d i n e h a s b e e n p u b l i s h e d , m o s t o f w h i c h has a d d r e s s e d t h e a c t i o n o f r y a n o d i n e o n r e c e p t o r s (so-called ryanodine receptors) associated with endoplasmic reticulum of mamm a l i a n m u s c l e a n d n e u r o n e . I n fact, w h e n r e v i e w i n g r e c e n t w o r k o n r y a n o d i n e one might be forgiven for concluding that studies of the biological properties of this c o m p o u n d a r e r e l a t i v e l y n e w , y e t t h e i r o r i g i n d a t e s b a c k t o t h e 1940's w h e n 1 Department of Life Science. The University of Nottingham, Nottingham NG7 2RD, U.K. z To whom correspondence should be addressed.
515 0144-8463/95/1200-0515507.50/09 1995PlenumPublishingCorporation
516
Vais and Usherwood
it was discovered that the powdered root and stem wood of the shrub Ryania speciosa Vahl is insecticidal (Pepper and Carruth, 1945). These botanical materials are rich in ryanodine (Rogers et al., 1948) and 9,21-didehydroryanodine (Waterhouse et al., 1987) and these two compounds are mainly responsible for the insecticidal properties of Ryania extracts (Jefferies and Casida, 1994). Soon after the discovery of its insecticidal properties, ryanodine was shown to have profound effects on muscles of vertebrates and invertebrates. These are described in a seminal review on the pharmacology of ryanodine by Jenden and Fairhurst (1969). Because of its high pharmacological potency, it was anticipated that ryanodine action on muscle would be associated with a highly specific mechanism, possibly involving the Ca2+-transport properties of sarcoplasmic reticulum (SR) corresponding to excitation-contraction coupling. Subsequently, a ryanodinebinding protein was purified and cloned from mammalian muscle that affects the release of Ca 2+ from SR. At nano-molar concentrations, ryanodine opens a Ca2+-release channel in the SR; at micro-molar concentrations it closes this channel (Fleischer and Inui, 1989). The discovery of ryanodine-binding proteins in muscle and neurones, understandably perhaps, has deflected attention away from other possible sites of action for ryanodine on excitable cells. Is there any evidence that ryanodine has other sites of action? Usherwood (1962) observed a profound alteration of electrical excitability during ryanodine application to locust skeletal muscle, in addition to its effect on excitation-contraction coupling. Normally, locust skeletal muscle is gradely electrically excitable, but following application of ryanodine all-or-none action potentials occur. It was suggested that this change was due to an alteration in the K + conductance properties of the muscle. Strong support for this hypothesis came much later from studies by Huddie et al. (1988), who showed that ryanodine increases the Na + permeability of two types of K + channel present in the plasma membrane of locust skeletal muscle. This observation has been recently confirmed and extended by Vais et al. (1994; 1995), who have demonstrated that the effect of ryanodine on the K + channels of locust muscle is independent of its action on excitation-contraction coupling. Here we review the evidence for this assertion and describe how studies of the influence of ryanodine and analogues on the block of the K + channels by quaternary ammonium ions (QAIs) suggest that these ryanoids enlarge the K § channels. MATERIALS A N D METHODS
Experiments were undertaken on fibres of the extensor-tibiae muscle of the methatoracic leg of adult locust (Schistocerca gregaria). The ventral surface of this muscle was exposed by dissection o f the overlying flexor tibiae and retractor unguis muscles. During this procedure, and in the experimental studies which followed, the preparation was bathed in a low-K + saline (raM): 10 KC1; 180 NaC1; 2CAC12; 10HEPES; pH = 6.8 (standard saline). Prior to patch-clamping, the muscle preparation was incubated in 2 mg m1-1 collagenase (type 1A, Sigma) in standard saline for 1-2 h at -20~ the enzyme solution being continuously circulated. After enzyme treatment, the preparation was thoroughly washed with standard saline.
Ryanoids and K§ Channel Selectivity
517
Patch pipettes were pulled from borosilicate glass (GC 150-10, Clark Electromedical Instruments, U.K.), fire-polished and coated with Sylgard TM. The pipette solution had the following composition (mM): 180KC1; 10NaC1; 2CAC12; 10HEPES; pH6.8, to which various amounts of ryanoid were added. Patch pipette resistances were 2-6 M~. In most experiments, the alkaloids were present only in the patch pipettes; by this means it was possible to limit contamination of a preparation with ryanoids. When the reversibility of their action was tested, the ryanoids (in low-K + saline) were applied onto, and subsequently removed from, a membrane patch using a rapid microperfusion system (Bates et al., 1990). Solutions containing QAIs had the same composition as the low-K + saline used in the bath, except that TEAC1 or TMACI (Sigma) replaced NaC1 on an equimolar basis. The QAI-containing salines were rapidly microperfused over membrane patches (Bates et al., 1990). Because "inside-our' patches were mainly used throughout this study, only the internal (cytoplasmic) blocking sites of the K + channels were scrutinized. The recording system consisted of a standard List (EPC-7) patch clamp amplifier linked to a Sony 701es PCM and a Sony VCR. Data stored on video tape were processed with a TL-1 DMA interface (Axon Instruments, USA) and analysed on a PC using pCLAMP software (Axon Instr., USA). Prior to digital acquisition, data were filtered at l kHz. The sample frequency was 5 kHz.
Data Analysis For most experiments, amplitude frequency histograms were fitted to channel current data obtained from membrane patches in which only one channel was present. Data on the time course of ryanoid action on the amplitude of the single channel current were analyzed using the cursor system of Axotape software (Axon Instr., USA). Single channel data recorded after ryanoid application were analysed visually to select 400 ms epics containing at least 10 openings. The mean amplitude of channel openings present in each of 10 or more successive epics of 400 ms duration was measured. The epics were separated by 2-3 s. The mean amplitude of the single channel current for each epic was taken as being representative for the centre time value of that epic. By plotting mean current amplitudes versus time post-ryanoid application, an account of the time course of ryanoid action on single channel current amplitude was obtained.
QAI-induced Block Both TEA and TMA are specific, fast blockers of K § channels (Stanfield, 1983; Hille, 1992), causing an apparent decrease of the channel open-state current in a dose- and voltage-dependent manner. Their binding and unbinding reactions are very fast. As a result, the limited frequency bandwidth of any patch clamp amplifier precludes full resolution of blocking and unblocking events. QAIs block has been modelled by applying equation (1), which was originally derived to describe the voltage-dependent H-- block of Na § currents in a voltage-clamped axon (Woodhull, 1973): I = I0/[1 + ([QAI]/Kd(O))exp(-zSFVm/RT)]
(1)
518
Vais and Usherwood
where I and I0 represent channel current amplitudes obtained in the presence and in the absence, respectively, of a given concentration of QAI ([QAI]). I and I0 are measured at a certain membrane potential (Vm). Ka(0), the dissociation constant (of the first-order blocking reaction) at zero Vm, indicates the channel affinity for the blocker. The effective valence (z6) is a measure of the voltage-dependence of the block, which in turn provides an electrical indication of the location of the blocker binding site in the transmembrane electric field. F, R, and T have their usual thermodynamic meanings. Equation (1) has previously been used in single channel studies to quantify channel fast block (Villarroel et al., 1988; Shen et al., 1994). Most data are accompanied by standard deviations (SDs), the correspondent number of experiments being indicated in brackets. However, those following a curve fit procedure are presented with standard errors of those fits (SEs). RESULTS Extensive patch clamp studies of the surface membrane of adult locust metathoracic extensor tibiae muscle have shown the presence of two types of K + channel (Gorcynska et aL, 1995). One channel has a maximum conductance of 170pS, fast open-close kinetics and a linear I/V relationship (BK channel). It is activated by Ca 2+, but at very low concentrations of this cation (between 10 -1~ M and 10 -9 M). The other channel has a maximum conductance of 35pS and slow open-close kinetics (IK channel). It is not activated by Ca 2+, but it is stretch-sensitive. This channel is an inward rectifier in cell-attached recordings; its conductance with standard locust saline in the pipette is - 3 times greater for hyperpolarisations than for depolarizations. However, in excised patches it was shown that the rectification results from channel block by cytoplasmic Mg 2+. For both channels, open probability (P0) increased during depolarisations and decreased during hyperpolarisations, resulting in outward rectification in terms of net current flow (Ip0). The Action of Ryanodine and Analogues Gorczynska et al. (1989) noted that ryanodine is equally active when applied to either side of a membrane patch excised from locust muscle. At concentrations as low as 10 -9 M, ryanodine shifted the reversal potentials (Vrev) of the two K + channels from their control values (Vrev(0)) of 60 mV towards zero membrane potential. With 10-6M ryanodine, Vrev for both channels was - 2 0 mV (Fig. la). According to the Goldman-Hodgkin-Katz equation for the membrane potential (Hille, 1992), this is equivalent to an increase in the permeability ratio PNa/PK from 4% to --50%. The permeability of these K + channels for C1- is normally negligible (Gorczynska et al., 1995) and was unaffected by ryanodine (up to 10-5M). The single channel conductances of the two K + channels were unaffected by the alkaloid (up to 10-SM) (Fig. la). Although we have not studied the effects of the alkaloid on the gating kinetics of these channels, visual inspection of many recordings indicate that these do not change,
Ryanoids and K § Channel Selectivity
519
3
/~Y("
/
[rranoai~] (MI
~2 ,/" 1
[1
i
60
40
20
O
-20
r -40
i
Vrn(mY)
b
control
2.2x1[DTMryanodine
E Fig. 1. Ryanodine-induced shift of the reversal potential (VTev) of the IK current, a) control data ( - - O - - ) for which Vrev(0)= 60 + 3 mV; 10 .6 M ryanodine ( - - O - - ) , for which Vrev(0 ) = 20 + 2 mV. Data points are means + SD (n = 6) and are fitted by linear regression. Inset: dose-response relationship for ryanodineinduced alteration of Vrev; data points are means +SD (n = 4) and are fitted with a 4-parameter logistic equation giving an ICso of 28 + 15 nM (SE). Quantitatively similar data (not shown) were obtained for the BK channel, b) Single channel data for IK channel (I) and BK channel (II) recorded at Vm = 0 m V from inside-out patches in control conditions (left), and exposed to 220nM ryanodine (right). Records filtered at 1 KHz cut-off. The alkaloid was added to the high-K + saline filling the pipette, the bath solution being low-K + saline.
at least qualitatively (Fig. lb). The dose-response relationship for the action of ryanodine on Vrev (Fig. l a - - i n s e t ) indicates an ICs0 coefficient of 28 4- 15 nM (SE). The effect of ryanodine on the K + channels was not Ca2+-dependent, i.e. 1 0 - 6 M ryanodine shifted Vrev to 2 0 + 2 m V (n = 5 patches) even when the cytoplasmic face of a patch was continuously perfused with low-K + saline buffered with 9mM E G T A in which the free Ca 2+ concentration [Ca 2+] was
