Inactivation

Inactivation is an im portant regulatory m echanism for tightly controlling the am ount of Ca^^ entering a cell during an action potential. As a result, neurosecretion is m ore

Chapter 1 hUrcxiuction

precisely controlled, even suppressed during periods o f short-term synaptic depression (Forsythe et a i , 1998), and cytosolic Ca^ concentration can be prevented from reaching cytotoxic levels. Inactivation occurs when the channel moves from the open activated state to an inactivated impermeant state, as opposed to a resting closed state. Unlike resting channels that are readily available for opening upon arrival o f an action potential, channels that have progressed to an inactivated state will no longer gate the flow o f ions on the arrival o f new a depolarising impulse, and will have to return to a closed state before it is able to activate once again in response to a voltage change (Figure 1.8). Three different modes o f inactivation have been reported for VDCCs: Fast (Zhang et a i , 1994), slow (Sokolov et al., 2000) and C a^-dependent (Peterson et ai.,

1999). Each type o f inactivation is intrinsically controlled by the structural properties o f the a i subunit and its interactions with the auxiliary subunits. The contribution o f one o f four auxiliary p subunits to this process is covered later in the section describing P subunit function (Section 1.2.2.3 page 72).

Open

F a s t-i S lo w -i

II?

CSosed

Figure 1.8 - channel state transitions during an action potential

channels reside in different conformational populations: closed (resting) states, activated (open) states and several inactivated states. Channel activation occurs rapidly in response to prolonged depolarisations or following trains of stimuli. Progression to inactivated states can be fast or slow and voltage or calcium dependent. The fast and slow voltage dependent inactivated states are closely interrelated. Transitions to inactivation may occur from the open as well as from the resting state. Negative shifts of the membrane potential promote transitions of open and/or inactivated channels to the resting state. Diagram modified from Hering et al. (20(X)).

1.2.2.2.3.1 Voltage dependent

Unlike the voltage-dependent Na^ and K channels that inactivate rapidly due to block o f the pore by a particle or cytoplasmic loop that is part o f the same channel (Hoshi et a i , 1990; Vassilev et al., 1988), inactivation o f VDCCs is much less understood and for a long time was not believed to involve blocking o f the pore by another part o f the channel. During a prolonged depolarisation or a rapid train o f stimuli, channels can proceed rapidly or slowly to inactivation and moreover channels can move to the slow

C hapter 1__________________________________________________________Introduction

inactivating state from the fast inactivating state (Sokolov et al., 2000). H ow ever, they do not necessarily have to pass through the open state, a phenom enon that is represented in plots o f activation and steady state inactivation curves, w hich barely overlap. In fact recent reports suggest inactivation occurs m ore readily in channels in a “partially activated” closed state; a intermediate state where the voltage sensor is in the activated position, but the channel is not open and from which it can progress directly to an inactivated state (Patil et a l, 1998).

The fast and slow com ponents o f inactivation can be m easured by fitting a double exponential to recordings of decaying whole cell currents (See Chapters 5 and 6). H ow ever it is not uncom mon for inactivating currents to only present a single com ponent that can be fitted by a single exponential. The tim e constants calculated from the fitting o f these curves (Tjnactivation) are often referred to as T\ or Tfast, and %2 or Tsiow and are m easured in the order o f milliseconds. The Tinactivation and its voltage dependence is a defining characteristic for each particular channel type.

M utations of the a i subunit, either deliberately introduced by the researcher (H erlitze et a i , 1997; Stotz et a i , 2000; Zhang et a i , 1994) or those associated with disease states related w ith altered VDCC inactivation (e.g. Fam ilial H em iplegic M igraine, FH M - (Kraus et a i , 1998)), coupled with the different inactivation properties of splice variants of the sam e V D CC a i subunit (Bourinet et al., 1999)^ have helped to elucidate som e regions or residues of the a i subunit that control channel inactivation. Transplantation o f IS6 betw een rabbit Cay2.1 and m arine ray Cav2.3 (doe-1) transferred the inactivation properties o f the donor channel to the acceptor (Zhang et al., 1994). H owever, because these channels norm ally dem onstrate m oderate to fast inactivation and are closely phylogenetically related (Ertel et a l , 2000), it was highly probable that in these experim ents other regions were also contributing to inactivation. Indeed^ point m utations in the pore lining regions o f IIS6, IIIS6 and IV S6 (H ering et al., 2000; Kraus

et al., 1998), the S5-S6 linker o f dom ains II, III and IV (Hans et al., 1999; Kraus et al.,

1998), and in the I II linker (Bourinet et al., 1999; H erlitze et a l , 1997) o f the ai2 .1 subunit all influence its rate and voltage-dependence o f inactivation. M ore recent investigations have determ ined that all four transm em brane dom ains are involved controlling the voltage dependence o f activation process, and dom ains II, III, and IV the

Chapter 1__________________________________________________________ Introduction

rate, but dom ains II and III playing the principal role in both processes (Spaetgens and Zam poni, 1999). M oreover, concurrent substitution o f the dom ain I-II linker w ith IIS6

and I l l s6 segm ents o f C av l.2 into Cay2.3 rem oved inactivation from the acceptor subunit (Stotz et al., 2000). The findings reported in this latter article lead the authors to suggest that the fast com ponent o f Ca^^ channel inactivation may after all be m ediated by block o f the pore by an inactivation particle form ed in the I-II linker, sim ilar to that proposed for the domain III-IV linker o f voltage-dependent sodium channels (Stotz et a l , 2000). This hypothesis is supported by the earlier findings of Cens et al. (Cens et al., 1999), who accelerated the rate in voltage-dependent inactivation o f a Cay2.1 channel by concom itantly over-expressing peptides corresponding to its I-II linker. A lthough this effect is possibly a function o f interfering with p subunit interaction, point m utation or alternative splicing o f the I-II linker of Cay 1.2 or Cay2.1 respectively, both support a role of the cytosolic loop in fast voltage- dependent inactivation (Bourinet et a l , 1999; H erlitze et a l , 1997). The m olecular nature o f a putative hinged particle, and w hether it is involved in inactivation o f all VDCC types rem ains to be determined.

The m olecular determ inants of slow voltage-dependent inactivation are m uch m ore o f an enigm a. A ssociation of the p2a subunit (Section 1.2.2.3) appears to increase the probability that channels will inactivate slowly (Sokolov et al., 2000), and binding o f use-dependent V D C C antagonists (e.g. PAAs) induce a slow rate of recovery from inactivation (H ering et al., 1997; Sokolov et a l, 1999). Indeed, it w ould appear that there is a synergism betw een the residues or regions involved in use-dependent block by organic m olecules and those involved in inactivation. Furtherm ore, under certain conditions, channels enter into the slowly inactivating conform ation from the open state m ore w illingly than they do from the fast inactivating conform ation (Sokolov et al.,

2 0 0 0), indicating that slow inactivation is not m erely a progression from fast

inactivation. Further research is required to fully understand this com ponent o f the inactivation process,* how everyt is possible that it may occur via sim ilar m echanism as the pore distortion process proposed for the C-type inactivation on voltage-dependent

Chapter I Introduction

•'

i

Figure 1.9 - Possible model of fast inactivation of high voltage-activated calcium channels.

T h e I-II linker possibly forms a hin ged bali/lid (a)

that physically occlu de s the inner vestibule o f the channel pore during fast v o ltag e-d ep en d en t inactivation by do ck in g with the cyto solic resid ues at the base o f dom ains II and III S6 se g m e n ts (h).

D iagram adap ted from Stotz et a i , (2000).

1.2.2.2.3.2 Ca^^-dependent

In addition to voltage dependent inactivation, the L-type VDCCs display C a “^- dependent inactivation. Originally identified in the C a“^ currents from axotomised Aplysia neurones (Chad et al., 1984; Chad and Eckert, 1984), this process has been most extensively characterised in the L-type VDCC currents conducted by C a v l.2 in the heart but is also reported to occur in some splice variants of the Cav2.1 (Bourinet et a i,

1999; Lee et al., 1999a). During long depolarising stimuli, Ca'^-dependent inactivation underlies the rapid component of Cay 1.2/C av2.1 inactivation, with increases in [Ca'^]i resulting in currents decaying to baseline in milliseconds (Lee et a i, 1985; Nilius and Benndorf, 1986). C a “^-dependent inactivation is therefore the primary determinant of current duration in depolarised cardiac cells and provides a critical autoregulatory means for controlling [Ca""^], under physiological and pathological conditions. The molecular mechanism underlying this process is independent of the participation of regulatory enzymes or processes requiring the hydrolysis of ATP (Haack and Rosenberg, 1994) and is caused by the direct interaction o f Ca~^ and calmodulin with the cytoplasmic C-terminal domain of Cay 1.2 or Cay2.1 (Lee et a!., 2000; Lee et a i,

1999a; Qin et al., 1999; Soldatov et al., 1997; Zuhlke et al., 1998). C a “‘^-dependent inactivation is a cooperative process involving the binding o f Ca"”^ to calmodulin and the association o f the C a “Vcalmodulin complex with a putative IQ calmodulin-binding m otif (Lee et a i, 2000; Lee et al., 1999a; Soldatov et a l, 1997; Zuhlke and Reuter,

1998). Splice variants not possessing this critical m otif lack Ca"^-depedendent inactivation. Additionally, both mutations that preclude calmodulin binding to Cay a i subunits and calmodulin mutants that cannot bind C a“^ also prevent C a'^-dependent inactivation (Lee et a l, 1999a; Peterson et a l, 1999; Qin et a l, 1999; Zuhlke et a l,

1999). These studies indicate that calmodulin is the Ca""^ sensor for channel regulation

Chapter 1_____________________________________________________

Introduction