БИОЛОГИЧЕСКИЕ МЕМБРАНЫ, 2007, том 24, № 1, с. 50-61


УДК 577.352


© 2007 r. A. I. Sobolevsky

Vollum Institute, Oregon Health and Science University, L474, 3181 SW Sam Jackson Park Road, Portland, OR 97239-3098, USA; e-mail: sobolevs@ohsu.edu

Received 30.07.2006

Ionotropic glutamate receptors (iGluRs) are modular proteins that contain ion channel permeable to different cations including calcium. The physiological role of iGluRs is mainly defined by the fact that currents conducted by their channels underlie communication between neurons in the majority of synapses in our brain. Knowing the structure and function of iGluR channel will not only give us a clue to how our brain works but also may help us to develop drugs for the treatment of multiple neurological disorders. Here I give a brief historical overview of the progress made in studies of iGluR channel structure and function that started more than two decades ago with studies of ion channel block.


The question of how our brain deciphers the complexity of the surrounding world and expresses itself in art, business, science and technology has fascinated investigators for many generations. The last few decades have provided significant breakthroughs in our understanding of the brain. It turns out that higher brain functions such as cognition and memory are results of complex dynamic processes of communication between brain cells, or neurons, mainly at specific contact points called synapses. The way it happens involves an enormous variety of biochemical and biophysical processes but it would be fair to say that the key players in these communications are neurotransmitter-activated receptors. Many of these receptors are ligand-gated ion channels that, in response to presynaptic release of neurotransmitter, open pores or ion channels in the postsyn-aptic membrane. Ion flow through these channels represents an electrical signal propagating from presyn-aptic to postsynaptic neurons throughout the brain. The majority of excitatory neurotransmission in our brain is mediated by ionotropic glutamate receptors (iGluRs), a family of ligand-gated ion channels that include three major pharmacologically defined classes named after selective synthetic agonists: dicarboxylic amino acids N-methyl-D-aspartate (NMDA), a-amino-3-hydroxy-5-methyl-4-isoxalone propionic acid (AMPA), and kainate. From a simplistic point of view, iGluRs are receptor-channel complexes: binding of extracellular glutamate to the receptor triggers conformational changes that ultimately lead to opening of the associated ion channel (Fig. 1).


Transport of ions across the lipid bilayer of the plasma membrane is an energetically unfavorable process. Ion channels solve this problem by embedding themselves into the membrane via their hydrophobic surfaces and creating hydrophilic pathways (pores) that connect the intracellular and extracellular sides of the membrane and have specific structural features critical for trafficking charged particles [1]. Importantly, ion channel pores can be gated (can close or open) in response to specific stimuli (binding of glutamate in the case of iGluRs). A number of studies have addressed the biophysical properties and gross architecture of iGluR channels and the major features of their gating mechanism.

Geometry and permeability of the iGluR channel pore

Measurements of reversal potential for iGluRs expressed in neurons or heterologous expression systems showed that iGluR channels are permeable to cations but not to anions. Of the different iGluR subtypes, NMDA receptor channels are highly permeable to Ca2+ ions giving them additional physiological importance as triggers in secondary messenger cascades. The dimensions of the iGluR channel pore have been gauged by applying ions and organic molecules of various size and electrophysiologically monitoring currents through the channel. The diameter of the narrowest portion of the channel pore was estimated as ~5.5 A for NMDA receptors [2-4] and 7.0-7.4 A for AMPA and kainate receptors [5]. Studies of ion channel block, including voltage dependence and kinetics, for differently sized blockers gave estimates for dimensions in intracellular and extracellular vestibules [2, 6-14]. An example of a

simplified geometric model of the NMDA receptor channel pore is shown in Fig. 2C. Apart from being gauges of pore dimensions, channel blockers appeared to be valuable tools in studies of iGluR gating.

Interaction of blockers with iGluR channel gating machinery

Slow trapping blockers. Minimal interaction with gating. The first studies characterizing interactions of channel blockers with iGluR gating involved slow NMDA receptor channel blockers, such as MK-801, phencyclidine and ketamine [15, 16]. These blockers had relatively high binding affinity and koff of the order of 10-3 to 10-1 s-1 [17]. Because of the slow dissociation rate, recovery from block for these compounds was "use-dependent" that turned out to be indicative of so-called trapping. Trapping implies that the blocker molecule can enter the open pore of the channel and reside inside the pore upon channel closure. In other words, the closed channel can stay "pregnant" (B. Khodorov) with the blocker inside before a subsequent channel opening will give this blocker a chance to leave. Hence, trapping blockers do not prevent channel closure and do not interfere with the gating machinery. Opposite examples of blockers that do interfere with gating are fast NMDA receptor channel blockers, such as 9-aminoacridine (9-AA), IEM-1754, and tetrapentylammonium (TPentA).

Hooks and overshoots. Fast blockers that interfere with gating. Among the most fascinating observations made in studies of the NMDA receptor channel block in the beginning of 1990's was the recording of transient currents that appeared at the end of the blocker application. What was amazing about these currents was that regardless of whether they were recorded after the co-application of the agonist and blocker ("hooks", Fig. 2A) or in the continuous presence of the agonist ("overshoots", Fig. 2B), the blocker was able to temporarily increase the current amplitude to a level that exceeded the control current level. Drawing an analogy between NMDA receptor channels and voltage-gated ion channels, it was intuitively clear from the very beginning that hooks and overshoots somehow reflect interference between the blocker and the gating machinery of the channel. Large amplitude hooks and overshoots in recovery current kinetics have been demonstrated for a number of NMDA receptor channel blockers including tetrabutylammonium (TBA), TPentA, IEM-1754, 9-AA, and tacrine [6, 18-23]. Detailed analysis of blocking kinetics confirmed that hooks and overshoots are indicative of interference between the blocker and gating machinery: these blockers may prevent agonist dissociation from the receptor (9-AA), desensitization (TBA), and/or channel closure (TPentA, IEM-1857). The latter (TPentA and IEM-1857) represent a class of so-called sequential or "foot-in-the-door" blockers (see [17, 24, 25] for review). Additionally, all blockers that demonstrated transient current increase events had fast dissociation kinetics. This property is, in fact, a prerequisite for hooks and overshoots to appear.




1 1 1

1 1 y 1 \ 1 1 1 1



Fig. 1. Naive view on iGluR. Glutamate receptor ion channel can be gated (can be opened or closed) when agonist (glutamate, triangle) binds to/dissociates from the receptor. Note a hypothetical connector between the ligand-binding pocket and the gate (black bar) undergoes a conformation change upon agonist binding.

At one time, only fast NMDA receptor channel blockers seemed to interfere with channel gating, while slow blockers were all trapping blockers and thus did not suppress gating functions. Later, however, it became clear that relatively slow blockers are capable of affecting gating (e.g., [26]) and that some fast blockers can easily be trapped in the channel (tetraethylammonium and Mg2+; [6, 27]). Use-dependence of recovery from block cannot be observed for the latter blockers. Therefore, new approaches and criteria were developed to test the interaction between fast blockers and gating [6].

Trapping of Mg2+ in NMDA receptor channels. Since all blockers that interfered with NMDA receptor gating had fast dissociation kinetics, it was originally accepted that the rate of channel block somehow correlated with an ability to interfere with gating. For example, kinetics of Mg2+, an extremely fast NMDA receptor channel blocker that causes steep voltage-dependence and therefore underlies the unique contribution of NMDA receptors to synaptic physiology [28, 29], was originally analyzed using a model of sequential block [30]. However, despite its fast dissociation kinetics, Mg2+ does not display overshoots or large hook currents [27]. Application of multiple fast blocker criteria [6] to Mg2+ clearly showed that at hyperpolarized potentials, Mg2+ acts as a trapping blocker [27]. This conclusion was later confirmed by single channel analysis [31]. More recent studies, however, found that at depolarized membrane potentials, a slow component adds to the overall fast kinetics of Mg2+ [32-34]. This component does not fit into a fully symmetric model of trapping block and assumes that at depolarized membrane potentials Mg2+ starts to interfere with NMDA receptor gating. This interference is apparently subunit dependent [35] and can be explained by Mg2+ acceler-


0.2 nA

0.5 min



0.2 nA

0.5 min

Fig. 2. Classical biophysical view on iGluR channels. A, B - Hooks and overshoots, similar to those observed by Koshelev and Khodorov [18, 19], appear after the bl

Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.

Показать целиком