Molecular Neurobiology and Genetics: Review Investigation of ...

Neuron, Vol. 20, 427–444, March, 1998, Copyright 1998 by Cell Press

Molecular Neurobiology and Genetics: ReviewInvestigation of Neural Functionand Dysfunction

of related proteins, to defined genetic manipulation tocreate transgenic mouse “knockouts,” molecular neuro-biology has underpinned many of our recent insightsinto how the brain functions. A consequence of thesedevelopments is the identification and study of genes

Tim Green, Stephen F. Heinemann,†and Jim F. Gusella*†

Molecular Neurobiology LaboratoryThe Salk Institute for Biological StudiesLa Jolla, California 92037*Molecular Neurogenetics Unit that cause neurological disease, revealing the role that

affected genes play in normal nervous system functionMassachusetts General HospitalCharlestown, Massachusetts 02129 and increasing the chances for treatment and cure.

Much work in the field of neurobiology has focusedon the study of the synapse and membrane excitability,since it is generally believed that the regulation of these

During the last 60 years, our understanding of the brain processes underlies much of higher brain function, in-has advanced at a spectacular rate due to the efforts of cluding memory, learning, and cognition. Synaptic trans-researchers trained in physics, mathematics, anatomy, mission is mediated by neurotransmitter molecules re-physiology, biochemistry, medicine, and most recently leased from the presynaptic cell, which activate twogenetics and molecular biology. The differing perspec- classes of receptors. One is the G protein–coupled re-tives provided by these diverse specializations have ceptor class, in which ligand binding leads to activationprovided insights into brain function that might have of G proteins and which usually operates in time framesbeen missed if neurobiology had been a more homoge- of seconds. The activated G proteins act as secondneous discipline. Neuroscience is now an integral part messengers to regulate the metabolism of the cell, asof modern biology, thanks in part to the unifying effects well as by directly affecting channel proteins. Theseof the newest entrants to the field, geneticists and mo- “slow” G protein–coupled receptors play a role in modu-lecular biologists. This review will focus on some of the lating synaptic transmission, cell excitability, and sen-advances which have resulted from the application of sory systems. The other class of receptors is the neuro-molecular–genetic techniques to the study of the brain. transmitter- or ligand-gated ion channels, which on

The human brain is estimated to have about a hundred ligand binding undergo a fast conformational change tobillion nerve cells, two million miles of axons, and a open an integral pore within a time frame of 1–100 ms.million billion synapses, making it the most complex The majority of synaptic transmission in the brain isstructure, natural or artificial, on earth. The advantage mediated by these “fast” receptors.and power of the genetic approach is grounded in the Following theold adage towrite about what you know,fact that the information to build the brain is coded in and because of space limitations, we will restrict de-the genome, a much simpler structure than the brain tailed discussion to these “fast” receptors. The historyitself. Mammals are thought to have about one hundred of the use of molecular techniques in neurobiology is inthousand genes, only a subset of which are expressed many ways encapsulated in the efforts to understandin the brain, and all of which will soon be sequenced on these channels. We present here a personal historicalcompletion of the genome project a few years from now. view of the cloning of two neurotransmitter receptors:In addition, a steady stream of ingenious techniques are the first receptor type cloned, thenicotinic acetylcholinebeing developed to manipulate and analyze the genome receptor; and the cloning of the major excitatory recep-with ever increasing ease and power. tor class in the mammalian brain, the glutamate-gated

Molecular cloning has made it possible to identify the ion channel receptor (GluR). In addition we discuss thegenes and determine the primary structure of many of impact, advantages, and limitations of the molecularthe proteins that mediate synaptic function and mem- approach in the investigation of channel structure andbrane excitability. Site-directed mutagenesis and func- function and the mutations that have been discoveredtional studies in heterologous expression systems have to cause brain dysfunction and disease.begun to elucidate how proteins such as ion channelswork at the molecular level. High resolution structures

Ion Channel Cloning: an Historical Perspectiveof domains from channels and receptors are being ob-Cloning of the Acetylcholine-Gated Iontained for the first time, while structures for whole recep-Channel Receptortor molecules are hopefully not far behind. These struc-The contribution of cDNA cloning to the determinationtures will make it possible, at long last, to understandof primary structure is both obvious and overwhelming.the mechanisms underlying channel gating and ion flowThe first receptor cDNAs cloned were four subunits ofat the atomic level. Most recently, the technology ofthe nicotinic receptor (nAChR), which is activated by thehomologous recombination or “gene targeting” hasneurotransmitter acetylcholine as well as nicotine, onemade it possible to create mutations in specific mouseof the most widely consumed drugs. Cloning was madegenes and in defined brain regions, enabling the dissec-possible by the serendipitous evolution of two systems.tion of the role of individual proteins in brain function.The first was the electric organ of Torpedo electric fish,From the initial use of gene cloning to identify familiesa modified muscle that uses the nicotinic receptor ex-pressed at high levels to generate a large current toparalyze its prey. The secondwas the evolution of snake†To whom correspondence should be addressed.

Neuron428

Figure 1. Timelines of Cloning and Topological Models

(A) The dates when the first members of each main subfamily were cloned are indicated for four of the main neuronal ion channel families.The asterisk denotes the publication of the first 55 amino acids of Torpedo nAChR subunit protein sequences (Raftery et al., 1980).(B) The topological models are placed to indicate the approximate dates that they were first proposed, where space allows. Membranedomains are shown as either rectangles for transmembrane domains or loops for reentrant domains. Regions implicated in the pore regionare colored green, and the lipid bilayer is indicated in gray. In all cases, the extracellular space is to the top. References for topologicalmodels, reading left to right: nAChR-type: first and fourth (Claudio et al., 1983; Devillers-Thiery et al., 1983; Noda et al., 1983), second (Finer-Moore and Stroud, 1984), and third (Ratnam et al., 1986); voltage-gated: first (Noda et al., 1984), 8-crossing (data not shown; Greenblatt etal., 1985; Guy and Seetharamulu, 1986), second (Noda et al., 1986), and third (Guy and Conti, 1990); ATP-gated: first (Brake et al., 1994) andsecond (Rassendren et al., 1997); iGluR (ionotropic glutamate receptor): first (Hollmann et al., 1989), second (Seeburg, 1993), and third (Woand Oswald, 1994; Hollmann et al., 1994; Bennett and Dingledine, 1995).

peptide neurotoxins, which bind with high affinity and group at Stanford. At around the same time, ShosakuNuma and colleagues in Japan and Jean-Pierre Chan-specificity to nicotinic receptors in the electric organ.

In the early 1970s, a number of laboratories built affinity geux and colleagues in France used the newly availableprotein sequence to clone the nicotinic receptor cDNAs.columns with these neurotoxins and purified nicotinic

receptor from the electric organ of fish in milligram quan- Two years after the work was first started, cDNA se-quences for all four constituent Torpedo nicotinic recep-tities. The availability of large quantities of nicotinic re-

ceptor made it possible to produce specific antibodies tor subunits were reported by these laboratories in thespace of six months (Ballivet et al., 1982; Noda et al.,and to determine N-terminal protein sequence of the

nicotinic receptor subunits (Devillers-Thiery et al., 1979; 1982, 1983; Claudio et al., 1983; Devillers-Thiery et al.,1983; Figure 1A). This presaged a flood of nicotinic re-Raftery et al., 1980; reviewed by Karlin and Akabas,

1995). The availability of large qutities of purified recep- ceptor cloning by homology screening, including cDNAscoding for the related mammalian muscle nicotinic ace-tor also enabled the unambiguous determination of sub-

unit stoichiometry, something that still eludes us for tylcholine receptor expressed at the neuromuscularjunction. One unanticipated discovery was the findingmany of the other receptors, and revealed the first in-

stance of what has turned out to be a common theme that during the development of the neuromuscular junc-tion there is a switch in the composition of the structureamongst ion channels: receptors made up of a number

of distinct but homologous subunits (Raftery et al., of the nicotinic receptor due to the substitution of thegamma subunit with the epsilon subunit, conferring new1980).

The work to isolate the cDNAs for the nicotinic recep- biophysical properties to the receptor (Mishina et al.,1986).tor began in the early 1980s. In the laboratories of Steve

Heinemann and Jim Patrick at the Salk institute, Marc Perhaps one of the most important findings comingfrom the cloning of the nicotinic receptor has been theBallivet started a project using immunological methods

and cloning vectors newly developed by Paul Berg’s discovery of a large family of related genes that code

Review: Green, Heinemann, and Gusella429

for neuronally expressed nicotinic receptors. Using low kainate, or NMDA (Hollmann and Heinemann, 1994;stringency homology screening, the laboratories of Sprengel and Seeburg, 1995). The characterization ofBoulter, Patrick, and Heinemann at the Salk Institute recombinant glutamate receptors has revealed a num-cloned the first nicotinic receptor subunit expressed in ber of unexpected properties. The first was that non-the brain, a3 (Boulter et al., 1986). Further screening NMDA receptor subtypes have a high permeability toled to the discovery of a large family of about a dozen calcium ions and that this is regulated by the subunitnicotinic receptor genes, expressed in almost every composition of the receptor (Hollmann et al., 1991). Thisregion of the mammalian brain. In addition, these sub- has important implications, since it is well establishedunits have been demonstrated to assemble in a com- that glutamate receptors are involved in the two best-binatorial manner, increasing the potential molecular studied examples of synaptic plasticity, long-term po-diversity of native nicotinic receptors. One of these sub- tentiation (LTP) and long term depression (LTD), whichunits, a7 (Couturier et al., 1990; Schoepfer et al., 1990), are thought to play a role in the formation of memoryhas proved extremely useful in molecular studies as (Bliss and Collingridge,1993). The elevation of intracellu-a “model” nicotinic receptor, since it forms functional lar calcium is thought to be the signal that is essentialhomomeric channels. for initiating LTP and LTD, evidence that the regulation

The cloning of large numbers of nicotinic receptor of calcium permeability is a crucial process for normalsubunits was unexpected because there was, and still brain function (reviewed by Nicoll et al., 1988; Malenkais, little evidence for synaptic transmission mediated and Nicoll, 1990). The finding that excess calcium canby nicotinic acetylcholine receptors in the mammalian lead to neuronal cell death also suggests that the regula-brain. While synaptic transmission has not been demon- tion of calcium permeability plays an important role instrated, there is strong evidence that functional nicotinic maintaining the health of the brain (Choi, 1994).acetylcholine receptors are present in many areas of Permeability to calcium ions in non-NMDA receptorsthe brain, including the medial habenula, the interpedun- was found to be largely determined by a single aminocular nucleus, the retina, the lateral and medial genicu- acid in the ion channel. Receptor subunits containinglate, and the neocortex. Working at the Salk Institute, the neutral amino acid glutamine have high calcium per-Ana Elgoyhen discovered the latest nicotinic subunit, meability, while subunits containing the basic residuea9, using a genomic screen. This subunit most likely arginine have low calcium permeability (Hume et al.,forms a postsynaptic receptor mediating the cholinergic

1991; Verdoorn et al., 1991). As might have been ex-efferent input to hair cells of the inner ear (Elgoyhen et

pected, the highly calcium-permeable NMDA receptorsal., 1994). Further confirming the functional importance

have the neutral amino acid asparagine at this position.of nicotinic receptors, a nonopioid analgesic was re-

In AMPA receptors, permeability to calcium is deter-cently described that acted through selective modula-

mined by the presence or absence of the GluR2 subunit.tion of neuronal nicotinic receptors (Bannon et al., 1998).

Seeburg’s group made the discovery that in GluR2 theMice with null or “knockout” mutations in two nicotinicresidue at the Q/R site was determined by RNA editingreceptor subunit genes, a7 and b2, have been reported(Sommer et al., 1991). The GluR2 gene codes for a gluta-(Orr-Urteger et al., 1997; Picciotto etal., 1998). Both micemine at the Q/R site, while virtually all GluR2 mRNA isare viable and have altered physiological responses toedited to the codon for arginine, resulting in low calciumnicotine in specific areas of the brain. Most significantly,permeability. Investigation of the mechanism of editingthe mice lacking the b2 subunit have altered avoidancerevealed that the RNA editing process is dependent onlearning and behavioral responses to nicotine (Picciottointronic sequences 39 to the edited site (Higuchi et al.,et al., 1998). We expect that with further such character-1993; Egebjerg et al., 1994). The GluR5 and GluR6 kai-ization of the nicotinic receptors, the data will help tonate receptors have now also been shown to be regu-explain the powerful and diverse consequences of nico-lated by RNA editing at the Q/R site (Kohler et al., 1993).tine consumption, including nicotine addiction.For some glutamate receptor subtypes, the efficiencyCloning of the Glutamate-Gated Ionof editing has been shown to vary during developmentChannel Receptorand between different brain regions (Sommer et al., 1991;In 1988, Michael Hollmann came to the Salk InstitutePaschen and Djuricic, 1994). Recently, a similar form ofand began a project to clone the glutamate receptor byRNA editing has been found in a G protein–coupledlooking for functional expression from pooled cDNAs inreceptor activated by serotonin, the 5-HT2C receptorXenopus oocytes. This “expression cloning” method,(Burns et al., 1997), making RNA editing an increasinglyfirst developed by Shigetada Nakanishi’s group in Tokyoimportant modulator of brain receptor function.(Masu et al., 1987), was necessitated by the absence

In common with virtually every other field of biologicalof high affinity ligands, which precluded methods rely-research, transgenic mice are now being generated withing on the initial isolation of protein that had beentargeted mutations to glutamate receptor genes. Muta-used previously toclone neurotransmitter receptors andtions to the GluR2 gene have provided interesting addi-channels. The successful cloning of the first glutamate-tions to the debate over the role of editing in receptoractivated channel receptor, GluR1 (Hollmann et al.,function. Mice lacking the GluR2 gene are viable, and1989), has been followed by the isolation of 17 relatedexpress calcium-permeable AMPA receptors (Jia et al.,genes in mammals, including subunits for the NMDA1996). In contrast, when the intronic sequence control-receptor, the first of which was cloned by Shigetadaling editing was removed and replaced with a selec-Nakanishi’s group in Kyoto by expression cloning (Mori-tion marker sequence, the mice died as heterozygotesyoshi et al., 1991). Most of the proteins encoded by(Brusa et al., 1995). The explanation for this differencethese genes have been shown to participate in func-

tional glutamate receptors selective for either AMPA, awaits the arrival of transgenic mice with ever more

Neuron430

subtle mutations. The generation of transgenic mice is Ultimately, the question of how many sequences remainto be discovered will be answered by another scion ofalso beginning to illuminate the function of kainate re-

ceptors, which is poorly understood in part due to the molecular biology, the comprehensive genome sequenc-ing project.lack of specific drugs. A newly described knockout of

the kainate-selective GluR6 subunit exhibits reducedlocomotion, altered synaptic physiology in the hippo- Ligand-Gated Channel Receptors:

Structures in Fluxcampus, and reduced susceptibility to kainate-inducedseizures (Mulle et al., 1998). Analysis of Primary and Secondary Structure

Despite early (and continued) expectations that three-Unsurprisingly, given the supposed role of NMDA re-ceptors in a wide range of brain processes, null mutants dimensional structure would routinely be modeled from

primary sequence alone, little has been learned fromfor the NR1 subunit are homozygous lethal (see Ebralidzet al., 1996; McHugh et al., 1996; Tsien et al., 1996). In deduced primary sequences in isolation. Rather, pattern

matching has been used to identify functionally impor-work that perhaps foreshadows the future power andelegance of transgenic techniques, Tonegawa and col- tant domains. Initial topological models were based on

hydropathy plots, used to identify those regions mostleagues produced a mouse lacking the NR1 subunit onlyin the CA1 region of the hippocampus. The mutation is likely to be in the membrane (Kyte and Doolittle, 1982).

Figure 1B shows the development of these topologicalonly induced after birth, avoiding the complication ofdevelopmental artifacts. Analysis of these mice pro- models for four neuronal channel families. Modifications

were made to these models over time to fit emergingvided genetic evidence for the importance of both theNMDA receptor and the CA1 region in LTP and spatial (and sometimes contradictory) data. It is also apparent

that models were heavily influenced, often erroneously,learning (Ebralidz et al., 1996; McHugh et al., 1996; Tsienet al., 1996). This experiment provides the strongest by what was known of the topologies of the other chan-

nel families at the time new subunits were cloned. Theseevidence so far that the most extensively studied formof synaptic plasticity, LTP, mediates some forms of models therefore demonstrate both the great strength

of molecular biology, the generation of hypotheses andlearning. Null mutations have also been made for theother NMDA subunits. Mice lacking NR2B die around provision of the tools to test them, and its great weak-

ness, taking the resulting models rather too seriously.birth, while NR2A and NR2C null mutants display vari-ous, less severe, neuronal deficits (Ebralidz et al., 1996). Which of the current incarnations (Figure 2) will prove

to be correct remains to be seen, but a recurring limita-In work that provides an intriguing insight into the mech-anisms underlying thesephenotypes, C-terminal trunca- tion of these models has been the difficulty in detecting

short membrane domains. An a helix must be 25–30tions of these same subunits also gave the same pheno-types as the null mutants (Sprengel et al., 1998). In a final amino acids to cross the bilayer, but b strands or other

extended conformations require only 7–8 residues tounexpected twist, mice lacking the “orphan” glutamatereceptor subunit d2, which has not yet been demon- cross, and reentrant loops can be any length and are

not necessarily hydrophobic. The general assumptionstrated to form part of a functional glutamate receptor,had deficits in motor coordination, synapse formation, has been that the membrane domains are a helices.

However, in the 9 A structure of the T. marmorataand LTD (Kashiwabuchi et al., 1995). Taken together,these results demonstrate the potential for transgenic nAChR, only the pore-lining domain has the electron

density expected of an a helix (Unwin, 1993b). This sug-techniques to confirm the expected, reveal the unex-pected, and perhaps ultimately help provide a complete gests that the other three hydrophobic domains may

instead cross multiple times as b strands (Figure 2A).genetic dissection of brain function.Ion Channel Cloning: Too Much of a Good Thing? The presence of b structure is supported by spectro-

scopic measurements of nAChR membrane domainsFollowing the cloning of the nicotinic receptors, cDNAsfor other receptors and channels were soon identified, (Gorne-Tschlenokow et al., 1994), as well as data sug-

gesting at least part of M1 has a nonhelical conformationusing either protein sequence, homology screens, orfunctional expression. Beyond the four receptor families (Akabas and Karlin, 1995). To what extent the normally

rectangular representations of transmembrane domainsshown in Figure 1, a large number of cation channels(both voltage-gated and “open channel”) have also been in topological models are in reality zig-zag crossings

(with the intriguing potential for small segments to ex-identified and cloned, many having new and unantici-pated structures (Figure 2; see also Armstrong and Hille, tend beyond the bilayer), will have to await high resolu-

tion structures.1998, for a review of voltage-gated channels [this issueof Neuron]). In addition, for many of the ion channels, Tertiary and Quaternary Structure

The three dimensional (3-D) structure of synaptic ionnumerous subunit subtypes have been identified, pro-viding, at least in theory, the potential for tremendous channels is perhaps the area in which molecular neuro-

biology has had the least impact, but can be expectedfunctional, temporal, and spatial diversity. The possiblecombinations appear far in excess of what is required to make the most significant contribution in the future.

Beyond topological models,what we knowabout synap-to explain the known functional diversity, much of whichis in any case derived from artificial pharmacological tic ion channel structure is essentially limited to the

Torpedo nAChR structure (Figure 3). This structure wasdistinctions. This suggests a subtlety of function thatmay require additional analytical tools to characterize. determined by cryo–electron microscopy (cryo-EM) us-

ing tubular crystals grown directly from membranes iso-The rate of discovery of new receptor genes has inevita-bly slowed, and the recent cloning of theGABAB receptor lated from the electric organ of Torpedo electric fish

(Unwin, 1993b). In this structure, the subunits are ar-marks the identification and sequencing of the lastknown major receptor class (Kaupmann et al., 1997). ranged in a ring around the central ion channel, and


Figure 2. A Topological Menagerie of Mammalian Neuronal Channels

Domains are colored as for Figure 1. Except for the nicotinic receptor, no structure is implied for the transmembrane domains (rectangles).The agonists for the three ligand-gated channels are shown in red.(A) Nicotinic acetylcholine receptor family. Family includes both anion- and cation-selective channels. Ligand binds in the N-terminal domain,and all subunits have a conserved 13 amino acid disulfide loop. M1, M3, and M4 are thought to cross as b strands, and M2 as an a helix(see enlarged region). The channel is formed by the M2 segment from five homologous subunits (see inset).(B) Ionotropic glutamate receptor family. Channels are permeable to Na1, K1, and Ca21. Ligand binds to two regions, one before M1 and theother between M3 and M4. The pore is formed by the reentrant loop M2, and current evidence suggests the channel is a pentamer.(C) P2X ATP-gated receptor. An ATP binding motif is present in the M1–M2 loop. Residues in M2 are involved in the pore (Rassendren et al.,1997), but the stoichiometry is unknown.(D) Voltage-gated ion channels (Armstrong and Hille, 1998). Channels highly selective for either Na1, K1, or Ca21. Voltage changes are detectedby S4. The reentrant loop H5 contributes to the pore. Channels have 4-fold symmetry (inset), from either four internal repeats (Na1 and Ca21)or four subunits (K1).(E) Inward rectifying K1 channels. Channels are tetrameric (Yang et al., 1995), with two putative membrane domains and a reentrant loop.There is some evidence of an alternative topology with four membrane domains and no loop (Schwalbe et al., 1997).(F) Open rectifying K1 channels (Leonoudakis et al., 1998). Channels activate instantaneously, independent of voltage. Membrane domainsand loops identified by hydropathy analysis and homology. Pore suggested to be formed by reentrant loops, although there is no experimentalevidence for this.

each contributes homologous sequences to the pore. see also Armstrong and Hille, 1998). The picture is lessclear for ionotropic glutamate receptors. The size ofThe same basic arrangement is generally assumed to

apply to the other neuronal ion channels, although such the native receptor complex determined by biochemicaltechniques is consistent with either four or five subunitsassumptions should be made with caution: for example,

the aquaporin water channels appear by low resolution (Brose et al., 1993). Electrophysiological analysis ofmixed wild-type/mutant receptor populations supportsEM to have a central pore, but the water is fluxed inde-

pendently within each of the four subunits (Li et al., a pentameric structure (Ferrer-Montiel and Montal, 1996;Premkumar and Auerbach, 1997), although this is far1997).

The muscle nAChR is a pentamer composed of four from universally accepted and is the subject of activeresearch. The stoichiometry and arrangement of thedifferent subunits in the stoichiometry a2bgd or a2bed in

the case of adult muscle. The other members of this ATP-gated and open rectifying channels shown inFigure2 is currently unknown.family (see Figure 1A) are also pentamers, demonstrated

either directly using EM (Nayeem et al., 1994; Boess et Further progress on 3-D structure determination hasbeen limited, primarily due to the difficulties encoun-al., 1995), or indirectly using biochemical or electrophys-

iological techniques (Langosch et al., 1990; Cooper et tered when working with membrane proteins. While het-erologous expression of soluble proteins has fueled theal., 1991). In contrast, the voltage-gated ion channels

have 4-fold symmetry, either from the four internal re- enormous expansion in structure determination by X-raycrystallography and nuclear magnetic resonance (NMR)peats in Na1 and Ca21 channels, or by the association

of four subunits as for K1 channels (Figures 2D and 2E; spectroscopy, membrane protein expression remains

Neuron432

subunits is to use molecular cloning techniques for theexpression of functional subdomains for crystallization.This technique is being used increasingly to study solu-ble proteins that have proved refractile to expression orcrystallization. This approach, however, requires accu-rate identification of functional domains so that indepen-dently folding units can be stably expressed. Not leastbecause of the uncertainties in subunit topology, identi-fication of such domains in ion channels has been slow.

This now appears to be changing, as independentfunctional domains from nACh receptors, glutamate re-ceptors, and voltage-gated channels are now beingidentified and expressed (e.g., Kuusinen et al., 1995;Kim et al., 1997; West et al., 1997). Indeed, the firstdomain structures are starting to appear. The structureof the well-characterized inactivation domain from anN-type KV channel has been determined by NMR (Antzet al., 1997), and the structure of the tetramerizationdomain from Shaker KV has just been determined byX-ray crystallography (Kreusch et al., 1998). This struc-ture reveals an unexpectedly narrow channel at the cen-ter of the tetramer. If this constriction indeed forms acytoplasmic vestibule, the entryway to the pore is notonly quite far from the presumed selectivity filter but isalso only wide enough for dehydrated ions to pass. Oneprediction of the earlier biophysical studies is that atsome point the ions must be stripped of H2O as theypass through the channel (Armstrong and Hille, 1998).Further important functional insights will undoubtedlyemerge as more structures are determined. It is alsohoped that the next few years will yield a more reliablesystem for the recombinant expression of integral mem-brane proteins, an advance that would help reveal thestructures of these proteins in the same way gene clon-ing has revealed their sequences over the last 15 years.

Channel Function: a Molecular PerspectiveBoth ligand- and voltage-gated ion channels share acommon functional paradigm. Detection of neurotrans-mitters or voltage changes is directly coupled, throughthe process of gating, to the opening of the permeationpathway. For several of these channels, significant prog-ress has been made in identifying the domains responsi-ble both for detection or ligand binding and ion perme-ation. In large measure, this progress has been due tothe application of molecular biology, through the gener-ation of chimeric proteins and single-site mutants.Again, we will concentrate here on work relating to the

Figure 3. Three-Dimensional Representation of the Torpedo Nico- ligand-gated channels; the voltage-gated channels aretinic Acetylcholine Receptor

discussed in detail elsewhere in this issue (ArmstrongThe image is based on data at 9 A resolution obtained using cryo–

and Hille, 1998).electron microscopy (Unwin, 1993b). The receptor is viewed fromLigand-Gated Ion Channel Agonist Binding Sitesthe side (top) and from above (bottom). The top two-thirds of theOf the three families of neurotransmitter-gated ion chan-receptor (z70 A) are located synaptically, followed by the mem-

brane-spanning region (colored band) and the smaller cytoplasmic nels (Figures 2A–2C), the agonist binding sites for thedomain. The subunits are located with 5-fold symmetry around the acetylcholine and glutamate receptors are best charac-central pore, visible in the bottom view. Reproduced from Unwin, terized. In nACh and related receptors, the agonist bind-1993a, with permission.

ing site is located in the large N-terminal domain beforethe first transmembrane domain. A combination of pho-toaffinity labeling and site-directed mutagenesis has ledproblematic. Indeed, the membrane proteins for which

structures have been determined can all be purified in to the identification of regions involved in the agonistbinding sites of the ion channel receptors for acetylcho-relatively large quantities from natural sources. One al-

ternative to determining the structure of intact channel line (Changeux et al., 1992), glycine (Schmieden et al.,


1992; Rajendra et al., 1995) and GABA (Buhr et al., 1996; changes, but using current techniques it should soonbe possible to identify the molecular changes underlyingWingrove et al., 1997). In all these receptors, severalthis theoretical framework. Part of the difficulty of study-noncontiguous regions have been identified as beinging gating is that it is part of the same equilibrium withimportant for ligand binding. The acetylcholine bindingligand binding, makingthe twoevents difficult todissect.site in the Torpedo nicotinic receptor appears from theIndividual residues might affect this equilibrium, andEM structure to be contained within the a subunit, half-therefore control gating, by stabilizing or destabilizingway up the extracellular domain (Unwin, 1993b). Thisthe open or closed states or the transition between thecontrasts somewhat with data from photoaffinity label-two. Intuitively, such residues would be expected toing suggesting the binding site might be located in thehave interacting complements, which might be identi-interface between subunits (Dunn et al., 1993) and evi-fied by rescue of nonfunctional mutants. Such an ap-dence that mutations within the non-a subunits affectproach will have to await the generation of putativeagonist binding. However, without a structure it is diffi-gating mutants. To complicate the issue further, thesecult to distinguish between mutations that are directlychannels also have at least one state (desensitized) inwithin the binding site and those that perturb its struc-addition to open and closed. In the continued presenceture indirectly.of agonist, many receptors enter a closed, unresponsiveRegions comprising the agonist binding site of iono-state. As with gating, the underlying mechanism istropic glutamate receptors have also been identified, inpoorly understood, although two studies have now iden-work that serves as a testament to the power of molecu-tified regions in the N-terminal domain mediating onelar techniques. This work has been guided by the obser-type of desensitization in the NMDA receptor (Krupp etvation, made when the first NMDA receptor subunit wasal., 1998; Villarroel et al., 1998). More such studies willcloned, that part of the sequence of glutamate receptorbe needed, as it is likely that the first structures deter-subunits is similar to that of the bacterial glutaminemined will be of dissected subdomains, missing theperiplasmic binding protein (GlnH). This similarity ex-linkages crucial to understanding gating and desensiti-tends across two noncontiguous domains of the recep-zation. A complete description of how channels opentor, one just before the first membrane domain M1 andand close remains one of the biggest challenges facingthe second between M3 and M4 (by convention count-the field.ing the reentrant loop as M2), although it is very weak.The Ion Channel

At the time, these two regions were thought to be onThe pores of the neurotransmitter-gated channels are

opposite sides of the membrane, but changes to theoften visualized as constrictions whose size and charge

topological models later placed them both on the out-properties determine channel selectivity. Despite their

side (Figure 2B).different topographical architectures, the ion pores in

The importance of thesedomains in agonistactivationnicotinic and glutamate channels appear functionally

of GluRs was first demonstrated with chimeras, con- similar. Both receptor classes have wide vestibules atstructed betweenAMPA- and kainate-selective subunits each end of the channel, narrowing to a central selectiv-(Stern-Bach et al., 1994). This work showed that these ity filter of similar size (Bormann et al., 1987; Villarroeldomains (termed S1 and S2) were together responsible et al., 1995). In molecular terms, a single region hasfor agonist selectivity. Based on this work, Kuusinen et generally been labeled as the pore-lining domain, foral. (1995) joined the GluR2 subunit S1 and S2 domains example M2 in nAChRs and the reentrant loops in thevia a short linker sequence and expressed the construct voltage-gated channels and glutamatereceptors (Figurein insect cells. This fusion was both soluble and bound 2). While this is convenient, and these regions have beento [3H]AMPA and other glutamate receptor ligands with elegantly demonstrated to form part of the permeationthe expected affinities, demonstrating the sufficiency of pathway, it is probably an oversimplification. For thethese domains for ligand binding. Furthermore, changes acetylcholine receptor, the architecture of the ion chan-to residues in both S1 and S2 have been found to affect nel has been investigated using a combination of photo-receptor function or ligand binding affinities, consistent affinity labeling, subunit chimeras, and single-site mu-with a role for these domains in agonist activation (e.g., tants. The second transmembrane domain (M2), wasUchino et al., 1992; Kuryatov et al., 1994; Paas et al., first identified as the pore region using a chimera be-1996; Laube et al., 1997; Swanson et al., 1997). This tween the Torpedo and bovine nAChR d subunits. Thework has spawned numerous models of the S1/S2 bind- chimeras had the conductance properties correspond-ingsite, based on structures determinedfor theperiplas- ing to the d subunit M2 region they contained (Imotomic binding proteins. Given the level of sequence simi- et al., 1986). Further experiments using open channellarity, the utility of these models probably lies more in blockers as photoaffinity labels identified residues inter-providing testable hypotheses than for rational drug de- acting with the pore (reviewed by Unwin, 1993a), andsign. In the long term, the ability to express the agonist rings of positive charges at each end of the channelbinding domain as a soluble polypeptide will probably were demonstrated to affect ion flux using stepwisehave the greatest impact if and when it leads to a 3-D mutagenesis (Imoto et al., 1988).structure. The most comprehensive studies are now being doneCoupling of Binding to Channel Opening using the technique of cysteine-scanning mutagenesis.The mechanisms by which agonist binding is translated This involves the application of molecular biological andinto channel opening is the least understood aspect of chemical techniques, by carrying out systematic muta-ion channel function, both for the ligand- and voltage- tion of single residues to cysteines and using two cyste-gated channels. Gating has long been described as a ine-labeling reagents having different permeability prop-

erties. This method provides a powerful technique to“black-box” process involving allosteric conformational

Neuron434

probe the pore region (Akabas et al., 1992; Akabas et techniques, it became possible to determine the chro-mosomal location of a genetic defect. This is accom-al., 1994). In this way, pore-lining segments have nowplished by tracking the inheritance of naturally occurringbeen identified in a wide range of channel families, andsequence variations (polymorphisms) in human DNA asuse of this technique should allow the identification ofthey are transmitted through the generations of familiesmost channel-lining domains. The results from this workwith an inherited disorder (Gusella, 1986). By identifyingare also often used as evidence of the secondary struc-statistically significant coinheritance (“genetic linkage”)ture of the regions studied. This aspect is less convinc-of a DNA marker with the genetic defect, it is possibleing, however, given the necessary assumption that theto infer the presence of the disease gene in the samecysteine is not disturbing the structure and the likelihoodchromosomal vicinity as the polymorphic DNA sequence.that even small distortions in structural elements willThe chromosomal assignment of the disease gene thenaffect the observed labeling pattern.offers the possibility of identifying the culprit gene onBeyond the Ion Channelthe basis of its location in the genome without any priorIt is known that the ligand-gated receptors can be linkedknowledge of the nature of its protein product. Thisto proteins inside the cell. This was first demonstratedstrategy, which has become knownas positional or loca-for the nicotinic receptor expressed in the electric organtion cloning, is applicable to any inherited disorder butand muscle (see Scotlet al., 1993). Recent work hasis particularly important for nervous system disordersdemonstrated that glycine (Kirsch et al., 1991) and gluta-in which the functional complexity of the target tissuemate receptors are also capable of binding to a diversityhampers the direct identification of defective proteinsof cytoplasmic proteins (reviewed by Sheng, 1996; Ken-(Gusella, 1989; Collins, 1990).nedy, 1997). Mary Kennedy and her collaborators foundMapping the Huntington’s Disease Genethat the NMDA receptor is associated with the postsyn-The first major success with this strategy came withaptic density (PSD) (Moon et al., 1994; reviewed bythe discovery of a DNA marker linked to Huntington’sKennedy, 1997), which triggered a search for proteinsdisease (HD), a then-enigmatic autosomal dominant dis-that associate with glutamate receptors. Using the two-order that was best known as the affliction that felledhybrid genetic screen, Kennedy and Seeburg’s groupsfolk singer Woody Guthrie (Martin and Gusella, 1986).collaborated to demonstrate binding of the NMDA NR2HD typically has its onset inmid-life (mean, z40 years)assubunit to the PDZ domain of the PSD-95 protein, ex-subtle, insidious, adventitious movements and gradually

plaining the previous finding that the NMDA receptor isprogresses to full-blown chorea that consumes all parts

found in the PSD (Kornau et al., 1995). Morgan Shengof the body. In addition to the writhing, dance-like move-

and his colleagues demonstrated that the PSD-95 pro-ment disorder, HD also produces psychiatric changes

tein can causeclustering of NMDAreceptors and Shakerand cognitive decline with death ensuing z15years after

potassium channels and in a nice experiment showedonset. The clinical signs are paralleled by a characteris-

that the clusters are disrupted in a Drosophila PSD- tic pattern of progressive neuropathology that devas-95 null mutant (Tejedor et al., 1997). Huganir and his tates the caudate nucleus and causes extensive neu-colleagues discovered a protein they termed GRIP, ronal loss throughout the basal ganglia and cortex. Inwhich binds through PDZ domains to theAMPA receptor 1983, we used two large HD families, one American andGluR2 subunit (Dong et al., 1997). There is increasing one Venezuelan, to map the HD gene to chromosomeevidence that these receptor-associated proteins are 4, accomplishing the first step in our goal to identify theinvolved in a number of functions such as clustering nature of this genetic defect (Gusella et al., 1983, 1984).and localizing the receptor to specific regions in the cell, Mapping Other Neurogenetic Disordersregulating receptor function, and catalyzing posttransla- The success in HD demonstrated the power of the ge-tional modifications of receptors. The discovery that netic linkage strategy and set off a torrent of similarligand-gated channel receptors associate with cyto- studies in a wide variety of inherited disorders, includingplasmic proteins raises the possibility that the binding many nervous system diseases. Within the next decade,of neurotransmitter to the receptor may directly activate linked DNA markers were found for such neurogenetica signal transduction pathway within the cell, in addition disorders as Alzheimer’s disease, Lou Gehrig’s diseaseto gating or opening the receptor channel and catalyzing (amyotrophic lateral sclerosis or ALS), neurofibromatosision flow through the membrane. This possibility is sup- 1 and 2, ataxia telangiectasia, and torsion dystonia (Ta-ported by the recent report that AMPA receptors can ble 1). Ultimately, the ability to attack genetic diseaseactivate G proteins in the absence of an ion flow (Wang with this approach was a major impetus for the initiationet al., 1997b). of the Human Genome Project to sequence the entire

genome and to identify all its genes. Indeed, the earlyAn Alternative Molecular Biological Approach days of the Human Genome Project were aimed at satu-to Nervous System Function rating the human chromosomes with DNA markers byNeurologic Disease Genes—the Indirect Approach identifying and mapping highly polymorphic stretchesWhile cDNA cloning based on protein sequence and of dinucleotides and other short repeats in human DNAfunctional characteristics was crucial for identifying to facilitate the positional cloning strategy. Disease genegenes of known importance for brain function, such as isolation was also a driving force for the developmentligand-gated ion channel receptors, a valuable approach of new physical mapping and large-insert cloning tech-for identifying unknown mutant genes that cause neuro- nologies.logic dysfunction also emerged in the 1980s. By merging Cloning Based on Map Locationthe tenets of Mendelian genetics with the extensive vari- When the HD locus was mapped, the molecular biologi-

cal tools for cloning and assessing several million baseation in human DNAdetectable with molecular biological


Table 1. Representative Autosomal Neurogenetic Defects Identified from Genetic Linkage Mapping

Year Year OmimNeurogenetic Disorder Linked Cloned Numbera Protein Product

Alzheimer’s disease 1992 1995 104311 presenilin 1Alzheimer’s disease 1995 1995 600759 presenilin 2Amyotrophic lateral sclerosis 1991 1993 105400 superoxide dismutase 1Ataxia telangiectasia 1988 1995 208900 novel, similar to phosphatidylinositol

39 kinase and yeast rad3Nonsyndromic deafness 1 1992 1997 124900 Drosophila diaphanous homologDentatorubro-pallidoluysian 1994 1994 125370 atrophin 1, novel, CAG repeat

AtrophyDystonia with diurnal variation 1993 1994 128230 GTP cyclohydrolase 1Dystonia 1989 1997 128100 torsin A, novel ATP binding proteinFriedreich ataxia 1988 1996 229300 frataxin, novelHereditary multi-infarct 1993 1996 125310 Drosophila Notch3 homolog

DementiaHuntington disease 1983 1993 143100 huntingtin, novel, CAG repeatInfantile neuronal ceroid 1990 1995 256730 palmitoyl-protein thioesterase

LipofuscinosisJuvenile neuronal ceroid 1989 1995 204200 novel

LipofuscinosisMyotonic dystrophy 1981b 1992 160900 myotonin kinaseNeurofibromatosis 1 1987 1990 162200 neurofibromin, GAP-related domainNeurofibromatosis 2 1987 1993 101000 merlin, related to cytoskeletal ERMsNiemann-Pick C 1993 1997 257220 novel, cholesterol traffickingProgressive myoclonic epilepsy 1991 1996 254800 cystatin BSpinal muscular atrophy 1990 1995 253300 novelSpinocerebellar ataxia 1 1976b 1993 164400 ataxin 1, novel, CAG repeatSpinocerebellar ataxia 2 1993 1996 183090 ataxin 2, novel, CAG repeatSpinocerebellar ataxia 3 1993 1994 109150 ataxin 3, novel, CAG repeatSpinocerebellar ataxia 7 1995 1997 164500 novel, CAG repeatStargardt’s disease 1992 1997 248200 novel, ABC transporterWilson disease 1985b 1993 277900 Cu21 transporting ATPase

a Full clinical descriptions and references can be obtained from Online Mendelian Inheritance in Man on the World-Wide Web at http://www.ncbi.nlm.nih.gov/Omim.b Genetic linkage was achieved first with expressed protein markers rather than DNA markers.

pairs of DNA surrounding a linked genetic marker to find shared by the greatest proportion of HD patients, as thesite of the defect. In 1993, a decade after the HD locusan unknown disease gene had not been developed. As

the technology advanced, the search for the HD gene was mapped, the disease gene was identified as a novel4p16.3 gene with a very unexpected mutation (Hunting-within the terminal 4p16.3 cytogenetic band applied

such novel technologies as pulsed-field gel electropho- ton’s Disease Collaborative Research Group, 1993).The HD Defectresis, chromosome jumping libraries, CpG island librar-

ies, NotI-end clone libraries, cosmid contig construc- More often than not, when a gene is identified by posi-tional cloning the nature of the defect provides newtion, yeast artificial chromosomes, phage P1 clones,

site-directed chemical cleavage of chromosomal DNA, insight, identifying a previously unsuspected or un-known protein. A primary example of this came with theradiation hybrid mapping, cDNA selection, exon trap-

ping, and many others (Gusella and MacDonald, 1993). 1993 discovery that one form of Lou Gehrig’s disease(amyotrophic lateral sclerosis) is caused by mutationsOther neurogenetic disorders succumbed to this ap-

proach and a number of interesting and often surprising altering superoxide dismutase 1, an enzyme long stud-ied but never implicated in the pathogenesis of motordisease genes were found (Table 1). However, HD

proved difficult because of uncertainty in pinpointing its neurons (Rosen et al., 1993). In HD, both the targetprotein and thenature of the mutationcame as a surpriseprecise location near the tip of the chromosome 4 short

arm. Ultimately, a 2 Mb candidate region was defined (Gusella et al., 1993; Huntington’s Disease CollaborativeResearch Group, 1993; Gusella and MacDonald, 1995;and entirely cloned (Bates et al., 1992; Baxendale et al.,

1993), but identification of the HD gene within this large Sharp and Ross, 1996). HD is due to an expanded, unsta-ble CAG trinucleotide repeat in a gene encoding a largesegment depended on an additional genetic strategy

that has subsequently proved increasingly valuable in novel protein of z350 kDadubbed huntingtin. On normalchromosomes, the CAG repeat is polymorphic, dis-tracking down disease genes. We used highly polymor-

phic DNA markers to identify a common pattern of alleles playing 10–35 triplets, and is inherited in a Mendelianfashion, whereas HD chromosomes possess a repeat(or haplotype) that implicated about one-third of seem-

ingly unrelated HD patients from around the world as of .35 triplets that has a mutation rate approaching 1as it changes in length in most meioses, with a biashaving a common ancestor (MacDonald et al., 1992).

This haplotype analysis ultimately targeted a segment toward size increases. The CAG repeat is within thehuntingtin coding sequence, 17 codons from the aminoof only z150 kb, the region whose marker alleles were

Neuron436

terminus, where it designates a run of consecutive gluta-mine residues immediately upstream from a proline-richregion. The remaining 97% of the protein is not similarto any others that have been reported, except for anumber of HEAT motifs, domains of unknown functionnamed for proteins in which they were first identifiedby computational analysis (Huntingtin-Elongation factor3-A subunit of protein phosphatase 2A-TOR1) (Andradeand Bork, 1995). Huntingtin is expressed widely in boththe nervous system and in other tissues, but its normalphysiological role remains unknown. Mouse knockoutand knock-in experiments have shown that total lackof huntingtin causes embryonic lethality at gastrulation(Duyao et al., 1995; Nasir et al., 1995; Zeitlin et al., 1995),whereas reduced levels of huntingtin support furtherdevelopment but result inabnormal neurogenesis (Whiteet al., 1997). The expanded polyglutamine segment ofhuntingtin, when introduced into the homologous mouseprotein, does not impair these developmental functions(White et al., 1997). Thus, the human HD mutation ap-pears to act by a “gain-of-function” mechanism.

Interestingly, HD was the second of a growing list ofneurodegenerative disorders found to be caused by alengthened polyglutamine segment in unrelated, butwidely expressed, proteins. The first disorder of thistype was spinal bulbar muscular atrophy, caused by aCAG repeat expansion in the androgen receptor gene(Brooks and Fischbeck, 1995). Subsequently, expandedCAG repeats have been found indentatorubropallidoluy-sian atrophy and spinocerebellar ataxias 1, 2, 3, 6, and7 (Koshy and Zoghbi, 1997; Gusella et al., 1997). In eachdisease, a lengthened run of glutamines alters a protein,producing a distinctive pattern of neuronal cell loss, withan age of onset of clinical symptoms that varies inverselywith the number of CAG repeats (Figure 4). Thus, fromthe exercise of positional cloning, a common mecha-nism for neurodegenerative disease has emerged thatappears to have two components: (1) a neuronal toxicitydetermined by the number of consecutive glutaminesand (2) a specificity of neuronal loss determined by theprotein context in which the polyglutamine is presentedto the cell.

Figure 4. Increasing CAG Length Decreases Age at Onset in Poly-The polyglutamine segment in mutant huntingtin altersglutamine Disordersthe physical properties of the protein, disproportionatelyAge at onset of neurologic symptoms is plotted against CAG repeatdelaying its migration on SDS-PAGE gels and dramati-length for spinocerebellar ataxia 2 (SCA2), spinocerebellar ataxia 1

cally increasing its reactivity with certain monoclonal (SCA1), Huntington’s disease (HD), dentatorubropallidoluysian atro-antibodies (Persichetti et al., 1995; Trottier et al., 1995a, phy (DRPLA), and spinocerebellar ataxia 3 (also Machado-Joseph1995b). A truncated amino terminus of the abnormal disease [SCA3]) based on a compilation of published data (Gusella

et al., 1997).protein will aggregate in vitro and in vivo in transgenicmice (Davies et al., 1997; Scherzinger et al., 1997). Thealtered physical properties of mutant huntingtin are also

protein causes a specific disruption of neuronal functionreflected in the HD postmortem brain with the recentthat triggers a cell death program of which aggregateobservation of dystrophic neurites and intranuclearformation is a by-product. The latter possibility is moreand cytoplasmic huntingtin immunoreactive inclusionsattractive in explaining the specificity of neuronal loss(DiFiglia et al., 1997). Similar results have been observeddespite widespread expression of the culprit proteinsfor the mutant versions of the other neurodegenerativeand raises the questionas towhether this specifity lies inpolyglutamine disorders (Davies et al., 1998; Ross, 1997).the normal role of huntingtin and the other polyglutamineHowever, it remains uncertain whether these aggregatesdisease proteins in the function of the adult brain.are a cause or a consequence of pathogenesis. The nextMore Complex Disorders: Alzheimer’s Diseasefew years will see the delineation of whether the alteredand the Psychiatric Disordersphysical properties of these proteins lead directly to theAs the genetic linkage approach has become more pow-development of aggregates that are toxic to theneurons,

or alternatively, whether the physical change in each erful with a dense map of highly polymorphic markers,


increasing attention is being paid to disorders that may The ligand-gated ion channels offer similar potentialfor displaying genetic variation that either causes orbe caused by more than one disease gene or by themodifies human disease. Table 2 lists the genetic lociinteraction of multiple genes. The strategy has met withencoding ligand-gated ion channel receptor subunits ingreat success in Alzheimer’s disease (AD), where threehuman that have been either identified or inferred bydifferent disease genes and one risk factor locus havework in other species. Their cytogenetic map locationsbeen identified (Goate, 1997; Hutton and Hardy, 1997;indicate that these receptor genes are spread through-Kim and Tanzi, 1997). AD involves deposition in the brainout the human chromosomes, with some notable clus-of extracellular amyloid, prompted by a 42 amino acidters of genes, such as the GABAA receptor genes onpeptide cleaved from a much larger protein, and devel-chromosomes 4p, 5q, and 15q, that probably reflectopment of intraneuronal neurofibrillary tangles. Muta-radiations from a common ancestor by duplication.tions in the amyloid protein precursor gene as well asStartling Mutations in the Glycine Receptoralterations in two homologous genes encoding preseni-The first disorder associated with a mutation in a neuro-lins 1 and 2 can increase the production of amyloido-transmitter receptor was found, not surprisingly, by po-genic peptide and cause early-onset AD. In addition,sitional cloning.Hyperekplexia (Startle disease; Kok dis-the E4 allele of the apolipoprotein E gene is a significantease) is characterized by muscle rigidity of nervousrisk factor for the development of late onset AD. Addi-system origin, particularly in neonates, and by an exag-tional genetic factors in AD are actively being sought.gerated startle response to unexpected auditory or tac-A similar approach to psychiatric disorders, most no-tile stimuli. The underlying genetic defect was firsttably bipolar affective disorder and schizophrenia, hasmapped to chromosome 5q by genetic linkage usingmet with much greater difficulty, perhaps because ofDNA markers (Ryan et al., 1992a). As the symptoms ofmore problematic phenotypic characterization and greaterhyperekplexia are markedly reduced by clonazepam,heterogeneity of genetic contributions. Various reportsthis implicated theGABAA receptor gene clusterat 5q33–of genetic linkage and association in each of these dis-5q35. However, more detailed genetic linkage studieseases have not yet produced a clearcut pathogenic lo-with physically mapped DNA markers excluded thesecus. However, as a result of global cDNA sequencinggenes, leaving the GLRA1 glycine receptor gene as theanalyses, there are now a substantial number of discreteprime candidate (Ryan et al., 1992b). Initially, Arg271genes available for analysis as candidates. This re-was identified as the target of missense alterations in

source suggests that a hybrid strategy that utilizes in-four families with autosomal dominant hyperekplexia

formation from genetic transmission to assess individ-(Shiang et al., 1993). Substituting uncharged amino

ual candidate loci (e.g., the transmission-disequilibriumacids (Leu or Gln) for the charged residue appears to

test) can effectively complement genome-wide linkagereduce the agonist sensitivity of the receptor (Rajendra

searches to place the hope of major progress in under-et al., 1994). Mutations affecting other codons have sub-

standing behavioral disorders on the horizon. A major sequently been identified in other families (Shiang et al.,class of candidates for such studies is the collection of 1995; Elmslie et al., 1996; Milani et al., 1996; Seri et al.,genes encoding ligand-gated ion channel receptors. 1997).

The homology of human chromosome 5q and mouseChr 11 led to the discovery that recessive mutations

Ligand-Gated Ion Channel Receptor Genesin the mouse homolog of GLRA1 underlie two mouse

in Disease neurologic mutants, spasmodic and oscillator. spas-Genomic Distribution of Ion Channel modic displays a phenotype similar to hyperekplexiaReceptor Genes and is also caused by a missense mutation (Ryan et al.,The crucial role of ion channels in the operation of the 1994). oscillator has a more severe phenotype, includingbrain and neuromusculature, the high degree of se- progressive tremor and muscle spasms leading to deathquence conservation of ion channel proteins, and the by 3 weeksof age. It is causedby a frameshift-producingcapacity to specifically alter ion channel function all deletion predictedto eliminate the final cytoplasmic loopsupport the view that naturally occurring mutations in and transmembrane domain (Buckwalter et al., 1994).ion channel genes could underlie a variety of human Interestingly, recessive forms of hyperekplexia havedisorders. This was first demonstrated for the muscle now been found in human, including both missense andvoltage-gated Na1 channel a subunit, which coinherits null mutations (Rees et al., 1994; Brune et al., 1996). Thein family studies with hyperkalemic periodic paralysis, a lack of lethality of the latter indicates that the loss ofdisorder involving episodic muscle weakness associated functional glycine receptors can be compensated to awith increased extracellular potassium (Fontaine et al., greater degree in human than in mouse. No mutations1990). Subsequently, it was recognized that different in other glycine receptor genes have yet been found inmissense alterations in this gene can produce a variety human. However, the discovery of a mutation in theof other symptoms, including cold-induced muscle stiff- homolog of GLRB in the spastic mouse suggests thatness (paramyotonia congenita) and fluctuating myoto- these genes should be scanned for alterations in pa-nia. The individual clinical signs in a given patient pre- tients with hyperekplexia-like symptoms without GLRA1sumably reflect differential effects of the specific mutations (Kingsmore et al., 1994; Koch et al., 1996).mutation on the kinetics of Na1 channel function. This Defective Brain and Musclegeneral finding has subsequently been demonstrated Acetylcholine Receptorsfor a number of other muscle and neuronal voltage- Another brain neurotransmitter-associated defect hasgated ion channels in genetic disorders of both human been found in the CHRNA4 nicotinic acetylcholine re-

ceptor gene on chromosome 20. A null mutation at thisand mouse (Bulman, 1997).

Neuron438

Table 2. Ligand-Gated Ion Channel Receptor Genes

Receptor Human Map Mouse MapHuman Gene Symbol Subunits Locationa Locationa Human Disorder Mouse Disorderb

NicotinicCholinergicReceptor

CHRNA1 a1 2q24–q32 2 Congenital myasthenic syndromeCHRNA2 a2 8p21–q21 14CHRNA3 a3 15q24 9CHRNA4 a4 20q13.2–q13.3 2 Nocturnal frontal lobe epilepsyCHRNA5 a5 15q24 9CHRNA6 a6 UN UNCHRNA7 a7 15q14 7CHRNA8 a8 UN UNCHRNA9 a9 UN UNCHRNB1 b1 17p12–p11 11 Congenital myasthenic syndromeCHRNB2 b2 1p21 3CHRNB3 b3 8p11.2 UNCHRNB4 b4 15q24 9CHRNG g 2q21.1–q21.3 1CHRND d 2q33–q34 1CHRNE e 17p13–p12 11 Congenital myasthenic syndrome

SerotoninReceptor

HTR3 5-HT-3 11q23.1–q23.2 UN

GABAA

Receptor

GABRA1 a1 5q33–q35 11GABRA2 a2 4p13–p12 5GABRA3 a3 Xq28 XGABRA4 a4 4p14–q12 7GABRA5 a5 15q11.2–q12 7GABRA6 a6 5q33–q35 11GABRB1 b1 4p13–p12 5GABRB2 b2 5q33–q35 UNGABRB3 b3 15q11.2–q12 7 cleft palate 1GABRG1 g1 4p14–q12 UNGABRG2 g2 5q33–q35 11GABRG3 g3 15q11.2–q12 7GABRD d 1p UNGABRE e Xq28 UNGABRR1 r1 6q14–q21 4GABRR2 r2 6q14–q12 4GABRP p UN UN

GlycineReceptor

GLRA1 a1 5q32 11 Hyperekplexia spasmodic, oscillatorGLRA2 a2 Xp22.1–p21.3 XGLRA3 a3 UN 8GLRA4 a4 UN XGLRB b 4q32 3 spastic

IonotropicGlutamateReceptor

GRIA1 AMPA 1 5q33 11(GluR1)

GRIA2 AMPA 2 4q32–q33 3(GluR2)

GRIA3 AMPA 3 Xq25–q26 X(GluR3)

GRIA4 AMPA 4 11q22–q23 9(GluR4)

GRID1 d1 UN UNGRID2 d2 UN 6 lurcher

continued


Table 2. continued

Receptor Human Map Mouse MapHuman Gene Symbol Subunits Locationa Locationa Human Disorder Mouse Disorderb

GRIK1 kainate 1 21q22.1–q22.2 16(GluR5)

GRIK2 kainate 2 6q16.3–q21 10 HD modifier(GluR6)

GRIK3 kainate 3 1p34–p33 4(GluR7)

GRIK4 kainate 4 11q23 9GRIK5 kainate 5 19q13.2 7GRIN1 NMDA 1 9q34.3 2GRIN2A NMDA 2A 16p13 16GRIN2B NMDA 2B 12p12 6GRIN2C NMDA 2C 17q25 11GRIN2D NMDA 2D UN 7

P2X ATPGatedReceptor

P2RX1 1 17p UNP2RX2 2 UN UNP2RX3 3 11q12 2P2RX4 4 12q24 UNP2RX5 5 UN UNP2RX6 6 UN UNP2RX7 7 12q24 UN

Gene locus, mapping and phenotype details were compiled from combining the information available in the public databases on the World-Wide Web: Online Mendelian Inheritance in Man (OMIM: www.ncbi.nlm.nih.gov/Omim), the Genome Database (GDB; gdbwww.gdb.org/gdb),Mouse Genome Informatics (MGI; www.informatics.jax.org/index.html), and GenBank (www.ncbi.nlm.nih.gov).a UN 5 Unknown.b Excludes targeted (knockout) mutations.

locus was originally implicated in benign familial neona- opening in the presence of acetylcholine, and acetylcho-line-independent channel opening. These changes com-tal convulsions (Beck et al., 1994), but the subsequent

discovery of missense mutations in a nearby voltage- promise normal transmission at theneuromuscular junc-tion and lead to endplate myopathy, presumably bygated K1 channel gene has raised doubts about this

finding (Biervert et al., 1998; Singh et al., 1998). However, cationic overload. Interestingly, a recessive myasthenicsyndrome involving endplate acetylcholine receptor de-alteration of CHRNA4 does cause a separate disorder,

autosomal dominant nocturnal frontal-lobe epilepsy, a ficiency is caused by CHRNE mutations associated withdecreased subunit production, decreased affinity forpartial epilepsy typified by brief, violent, sleep-associ-

ated seizures (Steinlein et al., 1995, 1997). A missense acetylcholine, diminished acetylcholine-induced chan-nel opening, and/or suppression of desensitization (En-mutation and a codon insertion, both affecting M2 of

the a4 subunit, each can cause this phenotype, although gel et al., 1996b; Ohno et al., 1997). In this case, overalldysfunction of the neuromuscular junction is mitigatedit is not clear whether this effect is due to reduced

function or hyperactivity of the receptor (Forman et al., by continued expression of the fetal CHNRG gene.Disease Candidates in Waiting1996).

Another route to identifying the genetic basis of inher- No inherited human disorder has been connected witha definite mutation in a GABAA receptor gene, althoughited disorders is direct sequencing of reasonable candi-

date genes without prior linkage mapping of the genetic inactivation of the GABAB3 homolog in mouse producesa developmental deficit leading to cleft palate (Culiat etdefect. Among the ligand-gated ion channels, this has

been most successful in the congenital myasthenic syn- al., 1995). Similarly, no pathogenicmutation has yet beendemonstrated in a human glutamate receptor gene, al-dromes, in which the endplate acetylcholine receptor is

an obvious candidate. In the autosomal dominant slow though mouse knockoutmutations of these genes causesubtle to severe neurologic symptoms. Interestingly, po-channel myasthenic syndrome, a variety of missense

mutations have been found in three different subunit sitional cloning revealed recently that the lurcher mouse,which displays ataxia due to loss of cerebellar Purkinjegenes, CHNRA1, CHNRB1, and CHRNE (Gomez and

Gammack, 1995; Ohno et al., 1995; Sine et al., 1995; cells during postnatal development, is caused by a mis-sense mutation in the GRID2 homolog (Zuo et al., 1997).Engel et al., 1996a; Gomez et al., 1996; Croxen et al.,

1997; Milone et al., 1997; Wang et al., 1997a). These The heterozygous substitution of Thr in place of a highlyconserved Ala residue in transmembrane domain 3 ofmutations provide a natural structure–function experi-

ment, in which different alterations have variously been the delta2 glutamate receptor leads to a large constitu-tive inward current, providing a geneticmodel for excito-implicated in producing increased affinity for acetylcho-

line, decreased rate of acetylcholine dissociation, en- toxic cell death in the nervous system. Mutations inglutamate receptor genes may ultimately be found tohanced steady-state desensitization, prolonged channel

Neuron440

Akabas, M.H., Stauffer, D.A., Xu, M., and Karlin, A. (1992). Acetylcho-cause neurodegenerative disease in human. However,line receptor channel structure probed in cysteine-substitution mu-normal polymorphic variations in these genes might alsotants. Science 258, 307–310.act as modifiers of phenotype in any disorder whereAkabas, M.H., Kaufmann, C., Archdeacon, P., and Karlin, A. (1994).excitotoxicity plays a role. For example, it has beenIdentification of acetylcholine receptor channel-lining residues in

suggested that GluR6 variation may have a mild effect the entire M2 segment of the a subunit. Neuron 13, 919–927.on age of onset in Huntington’s disease, although the Andrade, M.A., andBork, P. (1995). HEAT repeats in theHuntington’sprimary determinant of disease severity is the length of disease protein. Nature Genet. 11, 115–116.the HD gene’s expanded CAG repeat (Rubinsztein et Antz, C., Geyer, M., Fakler, B., Schott, M.K., Guy, H.R., Frank, R.,al., 1997). Ruppersberg, J.P., and Kalbitzer, H.R. (1997). NMR structure of inac-

tivation gates from mammalian voltage-dependent potassium chan-The capacity of the ligand-gated ion channels to mod-nel. Nature 385, 272–275.ify synaptic function suggests that they may also playArmstrong, C.M., and Hille, B. (1998). Voltage-gated ion channelsan important genetic role in behavioral disorders, suchand electrical excitability. Neuron 20, this issue, 371–380.as schizophrenia, depression, and anxiety, but to dateBallivet, M., Patrick, J., Lee, J., and Heinemann, S. (1982). Moleculartests of genetic association have not provided conclu-cloning of cDNA coding for the g subunit of Torpedo acetylcholine

sive evidence of their involvement. As these are all com- receptor. Proc. Natl. Acad. Sci. USA 79, 4466–4470.plex and probably heterogeneous disorders, any single Bannon, A.W., Decker, M.W., Holladay, M.W., Curzon, P., Donnelly-receptor defect may operate in only a small proportion Roberts, D., Puttfarcken, P.S., Bitner, R.S., Daiz, A., Dickenson, A.H.,of cases, and only in interaction with other genetic and/ Porsolt, R.D., et al. (1998). Broad-spectrum, non-opioid analgesic

activity by selective modulation of neuronal nicotinic acetylcholineor environmental factors. Consequently, normal poly-receptors. Science 279, 77–81.morphic variation in the ion channel receptors may leadBates, G.P., Valdes, J., Hummerich, H., Baxendale, S., Le Paslier,to combinations of alleles that have behavioral conse-D.L., Monaco, A.P., Tagle, D., MacDonald, M.E., Altherr, M., Ross,quences as a result of quite subtle effects on ion channelM., et al. (1992). Characterization of a yeast artificial chromosome

function. Discovery of such geneticeffects maybe immi- contig spanning the Huntington’s disease gene candidate region.nent, as improvements in technology have made it feasi- Nature Genet. 1, 180–187.ble to sequence individual genes in large populations Baxendale, S., MacDonald, M.E., Mott, R., Francis, F., Lin, C., Kirby,of phenotyped individuals. In any event, it is likely that S.F., James, M.,Zehetner, G., Hummerich, H., Valdes, J., et al. (1993).

A cosmid contig and high resolution restriction map of the 2 mega-geneticists have barely scratched the surface of the fullbase region containing the Huntington’s disease gene. Naturerange of human phenotypes that are associated withGenet. 4, 181–186.sequence differences in the ligand-gated ion channelBeck, C., Moulard, B., Steinlein, O., Guipponi, M., Vallee, L., Mont-genes.pied, P., Baldy-Moulnier, M., and Malafosse, A. (1994). A nonsensemutation in the alpha4 subunit of the nicotinic acetylcholine receptor

The Future (CHRNA4) cosegregates with 20q-linked benign neonatal familialIf the past is any guide to the future, then the continued convulsions (EBNI). Neurobiol. Dis. 1, 95–99.application of molecular neurobiology will answer many Bennett, J.A., and Dingledine, R. (1995). Topology profile for a gluta-

mate receptor: three transmembrane domains and a channel-liningof the questions we now have, as well as revealing newre-entrant membrane loop. Neuron 14, 373–384.and unexpected aspects of channel function. It is to beBiervert, C., Schroeder, B.C., Kubisch, C., Berkovic, S.F., Propping,hoped that this research will lead to the elucidation ofP., Jentsch, T.J., and Steinlein, O.K. (1998). A potassium channelthe molecular processes underlying neuronal disease,mutation in neonatal human epilepsy. Science 279, 403–406.

and in the near future we hope to see the fruit of this inBliss, T.V., and Collingridge, G.L. (1993). A synaptic model of mem-

new therapies for the now devastating neurodegenera- ory: long-term potentiation in the hippocampus. Nature 361, 31–39.tive and mental disorders. There is an additional hope, Boess, F.G., Beroukhim, R., and Martin, I.L. (1995). Ultrastructureor some might say a fantasy: molecular genetics gives of the 5-hydroxytryptamine3 receptor. J. Neurochem. 64, 1401–1405.us the power to use engineering design to alter brain Bormann, J., Hamill, O.P., and Sakmann, B. (1987). Mechanism offunction and perhaps the “mind” itself. By applying ge- anion permeation through channels gated by glycine and g-amino-

butyric acid in mouse cultured spinal neurons. J. Physiol. 385,netics with the development of new tools to study be-243–286.havior in the mouse, we may be able to discover impor-Boulter, J., Evans, K., Goldman, D., Martin, G.,Treco, D., Heinemann,tant principles about brain function that will lead to aS., and Patrick, J. (1986). Isolation of a cDNA clone coding for abetter understanding of our own brain and what makespossible neural nicotinic acetylcholine receptor a-subunit. Nature

us human. 319, 368–374.

Brake, A.J., Wagenbach, M.J., and Julius, D. (1994). New structuralAcknowledgments motif for ligand-gated ion channels defined by an ionotropic ATP

receptor. Nature 371, 519–523.We are grateful to Nigel Unwin for permission to use Figure 3 and

Brooks, B.P., and Fischbeck, K.H. (1995). Spinal and bulbar muscu-to Senyon Choe and Andreas Kreusch for an early look at theirlar atrophy: a trinucleotide-repeat expansion neurodegenerative dis-tetramerization domain structure. We also thank Marc Ballivet andease. Trends Neurosci. 18, 459–461.

members of the Heinemann laboratory, including Christine Schul-Brose, N., Gasic, G.P., Vetter, D.E., Sullivan, J.M., and Heinemann,teis, Geoff Swanson, Anis Contractor, and Rolf Sternglanz, for help-S.F. (1993). Protein chemical characterization and immunologicalful comments. The work was funded by grants to S. F. H. from thelocalization of the NMDA receptor subunit NMDA R1. J. Biol. Chem.National Institutes of Health and the Mcknight Foundation.268, 22663–22671.

Brune, W., Weber, R.G., Saul, B., von Knebel Doeberitz, M., Grond-ReferencesGinsbach, C., Kellerman, K., Meinck, H.M., and Becker, C.M. (1996).A GLRA1 null mutation in recessive hyperekplexia challenges theAkabas, M.H., and Karlin, A. (1995). Identification of acetylcholinefunctional role of glycine receptors. Am. J. Hum. Genet. 58, 989–997.receptor channel-lining residues in the M1 segment of the a-subunit.

Biochemistry 34, 12496–12500. Brusa, R., Zimmermann, F., Koh, D.-S., Feldmeyer, D., Gass, P.,


Sakmann, B., Seeburg, P.H., and Sprengel, R. (1995). Early-onset Dunn, S.M.J., Conti-Tronconi, B.M., and Raftery, M.A. (1993). A high-affinity site for acetylcholine occurs close to the a–g subunit inter-epilepsy and postnatal lethality associated with an editing-deficientface of the Torpedo nicotinic acetylcholine receptor. BiochemistryGluR-B allele in mice. Science 270, 1677–1680.32, 8616–8621.Buckwalter, M.S., Cook, S.A., Davisson, M.T., White, W.F., andDuyao, M.P., Auerbach, A.B., Ryan, A., Persichetti, F., Barnes, G.T.,Camper, S.A. (1994). A frameshift mutation in the mouse alpha1McNeil, S.M., Ge, P., Vonsattel, J.P., Gusella, J.F., Joyner, A.L.,glycine receptor gene (Glra1) results in progressive neurologicalet al. (1995). Inactivation of the mouse Huntington’s disease genesymptoms and juvenile death. Hum. Mol. Genet. 3, 2025–2030.homolog Hdh. Science 269, 407–410.Buhr, A., Baur, R., Malherbe, P., and Sigel, E. (1996). Point mutationsEbralidz, A.K., Rossi, D.J., Tonegawa, S., and Slater, N.T. (1996).of the a1b2g2 GABAA receptor affecting modulation of the channelModification of NMDA receptor channels and synaptic transmissionby ligands of the benzodiazepine binding-site. Mol. Pharmacol. 49,by targeteddisruption of the NR2C gene. J. Neurosci. 16,5014–5025.1080–1084.Egebjerg, J., Kukekov, V., and Heinemann, S.F. (1994). Intron se-Bulman, D.E. (1997). Phenotypic variation and newcomers in ionquence directs RNA editing of the glutamate receptor subunit GluR2channel disorders. Hum. Mol. Genet. 6, 1679–1685.coding sequence. Proc. Natl. Acad. Sci. USA 91, 10270–10274.Burns, C.M., Chu, H., Rueter, S.M., Hutchinson, L.K., Canton, H.,Elgoyhen, A.B., Johnson, D.S., Boulter, J., Vetter, D.E., and Heine-Sanders-Bush, E., and Emeson, R.B. (1997). Regulation of serotoninmann, S. (1994). a9: an acetylcholine receptor with novel pharmaco-2C receptor G protein coupling by RNA editing. Nature 387, 303–308.logical properties expressed in rat cochlear hair cells. Cell 79,Changeux, J.-P., Galzi, J.L., Devillers-Thiery, A., and Bertrand, D.705–715.(1992). The functional architecture of the acetylcholine nicotinic re-Elmslie, F.V., Hutchings, S.M., Spencer, V., Curtis, A., Covanis, T.,ceptor explored by affinity labelling and site-directed mutagenesis.Gardiner, R.M., and Rees, M. (1996). Analysis of GLRA1 in hereditaryQuart. Rev. Biophys. 25, 395–432.and sporadic hyperekplexia: a novel mutation in a family cosegre-Choi, D.W. (1994). Calcium and excitotoxic neuronal injury. Ann.gating for hyperekplexia and spastic paraparesis. J. Med. Genet.N. Y. Acad. Sci. 747, 162–171.33, 435–436.

Claudio, T., Ballivet, M., Patrick, J., and Heinemann, S. (1983). Nucle-Engel, A.G., Ohno, K., Bouzat, C., Sine, S.M., and Griggs, R.C.otide and deduced amino acid sequences of Torpedo californica(1996a). End-plate acetylcholine receptor deficiency due to non-acetylcholine receptor g subunit. Proc. Natl. Acad. Sci. USA 80,sense mutations in the epsilon subunit. Ann. Neurol. 40, 810–817.1111–1115.Engel, A.G., Ohno, K., Milone, M., Wang, H.L., Nakano, S., Bouzat,Collins, F.S. (1990). Identifying human disease genes by positionalC., Pruitt, J.N., Hutchinson, D.O., Brengman, J.M., Bren, N., et al.cloning. Harvey Lect. 86, 149–164.(1996b). New mutations in acetylcholine receptor subunit genes

Cooper, E., Couturier, S., and Ballivet, M. (1991). Pentameric struc- reveal heterogeneity in the slow-channel congenitalmyasthenic syn-ture and subunit stoichiometry of a neuronal nicotinic acetylcholine drome. Hum. Mol. Genet. 5, 1217–1227.receptor. Nature 350, 235–238.

Ferrer-Montiel, A.V., and Montal, M. (1996). Pentameric subunit stoi-Couturier, S., Bertrand, D., Matter, J.-M., Hernandez, M.-C., Ber- chiometry of a neuronal glutamate receptor. Proc. Natl. Acad. Sci.trand, S., Millar, N., Valera, S., Barkas, T., and Ballivet, M. (1990). A USA 93, 2741–2744.neuronal nicotinic acetylcholine receptor subunit (a7) is develop-

Finer-Moore, J., and Stroud, R.M. (1984). Amphipathic analysis andmentally regulated and forms a homo-oligomeric channel blockedpossible formation of the ion channel in an acetylcholine receptor.by a-BTX. Neuron 5, 847–856.Proc. Natl. Acad. Sci. USA 81, 155–159.

Croxen, R., Newland, C., Beeson, D., Oosterhuis, H., Chauplannaz,Fontaine, B., Khurana, T.S., Hoffman, E.P., Bruns, G.A., Haines, J.L.,G., Vincent, A., and Newsom-Davis, J. (1997). Mutations in differentTrofatter, J.A., Hanson, M.P., Rich, J., McFarlane, H., Yasek, D.M.,functional domains of the human muscle acetylcholine receptoret al. (1990). Hyperkalemic periodic paralysis and the adult musclealpha subunit in patients with the slow-channel congenital myas-sodium channel alpha-subunit gene. Science 250, 1000–1002.thenic syndrome. Hum. Mol. Genet. 6, 767–774.Forman, S.A., Yellen, G., and Thiele, E.A. (1996). Alternative mecha-Culiat, C.T., Stubbs, L.J., Woychik, R.P., Russell, L.B., Johnson,nism for pathogenesis of an inherited epilepsy by a nicotinic AChRD.K., and Rinchik, E.M. (1995). Deficiency of the beta3 subunit ofmutation. Nature Genet. 13, 396–397.the type A gamma-aminobutyric acid receptor causes cleft palateGoate, A.M. (1997). Molecular genetics of Alzheimer’s disease. Geri-in mice. Nature Genet. 11, 344–346.atrics 52 (suppl.), S9–S12.Davies, S.W., Turmaine, M., Cozens, B.A., DiFiglia, M., Sharp, A.H.,Gomez, C.M., and Gammack, J.T. (1995). A leucine-to-phenylalanineRoss, C.A., Scherzinger, E., Wanker, E.E., Mangiarini, L., and Bates,substitution in the acetylcholine receptor ion channel in a familyG.P. (1997). Formation of neuronal intranuclear inclusions underlieswith the slow-channel syndrome. Neurology 45, 982–985.the neurological dysfunction in mice transgenic for the HD mutation.

Cell 90, 537–548. Gomez, C.M., Maselli, R., Gammack, J., Lasalde, J., Tamamizu, S.,Cornblath, D.R., Lehar, M., McNamee, M., and Kuncl, R.W. (1996).Davies, S.W., Beardsall, K., Turmaine, M., DiFiglia, M., Aronin, N.,A beta-subunit mutation in the acetylcholine receptor channel gateand Bates, G.P. (1998). Are neuronal intranuclear inclusions thecauses severe slow-channel syndrome. Ann. Neurol. 39, 712–723.common neuropathology of triplet-repeat disorders with polyglu-Gorne-Tschlenokow, U., Strecker, A., Kaduk, C., Naumann, D., andtamine-repeat expansions? Lancet 351, 131–133.Hucho, F. (1994). The transmembrane domains of the nicotinic ace-Devillers-Thiery, A., Changeux, J.-P., Paroutaud, P., and Strosberg,tylcholine receptor contain a-helical and b structures. EMBO J. 13,A.D. (1979). The amino-terminal sequence of the 40,000 molecular338–341.weight subunit of the acetylcholine receptor protein from TorpedoGreenblatt, R.E., Blatt, Y., and Montal, M. (1985). The structure ofmarmorata. FEBS Lett. 104, 99–105.the voltage-sensitive sodium channel. FEBS Lett. 193, 125–134.Devillers-Thiery, A., Giraudat, J., Bentaboulet, M., and Changeux,Gusella, J.F. (1986). DNA polymorphism and human disease. Annu.J.-P. (1983). Complete mRNA coding sequence of the acetylcholineRev. Biochem. 55, 831–54.binding a-subunit of Torpedo marmorata acetylcholine receptor: a

model for the transmembrane organization of the polypeptide chain. Gusella, J.F. (1989). Location cloning strategy for characterizingProc. Natl. Acad. Sci. USA 80, 2067–2071. genetic defects in Huntington’s disease and Alzheimer’s disease.

FASEB J. 3, 2036–2041.DiFiglia, M.,Sapp, E., Chase, K.O., Davies,S.W., Bates, G.P., Vonsat-tel, J.P., and Aronin, N. (1997). Aggregation of huntingtin in neuronal Gusella, J.F., and MacDonald, M.E. (1993). Hunting for Huntington’sintranuclear inclusions and dystrophic neurites in brain. Science disease. Mol. Genet. Med. 3, 139–158.277, 1990–1993. Gusella, J.F., Wexler, N.S., Conneally, P.M., Naylor, S.L., Anderson,Dong, H., O’Brien, R.J., Fung, E.T., Lanahan, A.A., Worley, P.F., M.A., Tanzi, R.E.,Watkins, P.C., Ottina, K., Wallace, M.R., Sakaguchi,and Huganir, R.L. (1997). GRIP: a synaptic PDZ domain–containing A.Y., et al. (1983). A polymorphic DNA marker genetically linked to

Huntington’s disease. Nature 306, 234–238.protein that interacts with AMPA receptors. Nature 386, 279–284.

Neuron442

Gusella, J.F., Tanzi, R.E., Anderson, M.A., Hobbs, W., Gibbons, K., Kim, M., Yoo, O.J., and Choe, S. (1997). Molecular assembly of theextracellular domain of P2X2, an ATP-gated ion channel. Biochem.Raschtchian, R., Gilliam, T.C., Wallace, M.R., Wexler, N.S., and Con-Biophys. Res. Comm. 240, 618–622.neally, P.M. (1984). DNA markers for nervous system diseases. Sci-

ence 225, 1320–1326. Kingsmore, S.F., Giros, B., Suh, D., Bieniarz, M., Caron, M.G., andSeldin, M.F. (1994). Glycine receptor beta-subunit gene mutation inGusella, J.F., MacDonald, M.E., Ambrose, C.M., and Duyao, M.P.spastic mouse associated with LINE-1 element insertion. Nature(1993). Molecular genetics of Huntington’s disease. Arch. Neurol.Genet. 7, 136–141.50, 1157–1163.

Kirsch, J., Langosch, D., Prior, P., Littauer, U.Z., Schmitt, B. andGusella, J.F., and MacDonald, M.E. (1995). Huntington’s disease:Betz, H. (1991). The 93-kDa glycine receptor-associated proteinCAG genetics expands neurobiology. Curr. Opin. Neurobiol. 5,binds to tubulin. J. Biol. Chem. 266, 22242–22245.656–662.Koch, M., Kling, C., and Becker, C.M. (1996). Increased startle re-Gusella, J.F., Persichetti, F., and MacDonald, M.E. (1997). The ge-sponses in mice carrying mutations of glycine receptor subunitnetic defect causing Huntington’s disease: repeated in other con-genes. Neuroreport 7, 806–808.texts? Mol. Med. 3, 238–246.Kohler, M., Burnashev, N., Sakmann, B., and Seeburg, P.H. (1993).Guy, H.R., and Seetharamulu, P. (1986). Molecular model of theDeterminants of Ca21 permeability in both TM1 and TM2 of highaction potential sodium channel. Proc. Natl. Acad. Sci. USA 83,affinity kainate receptor channels: diversity by RNA editing. Neuron508–512.10, 491–500.Guy, H.R., and Conti, F. (1990). Pursuing the structure and functionKornau, H.C., Schenker, L.T., Kennedy, M.B., and Seeburg, P.H.of voltage-gated channels. Trends Neurosci. 13, 201–206.(1995). Domain interaction between NMDA receptor subunits andHiguchi, M., Single, F.N., Kohler, M., Sommer, B., Sprengel, R., andthe postsynaptic density protein PSD-95. Science 269, 1737–1740.Seeburg, P.H. (1993). RNA editing of AMPA receptor subunit GluR-Koshy, B.T., and Zoghbi, H.Y. (1997). The CAG/polyglutamine tractB: a base-paired intron–exon structure determines position and effi-diseases: gene products and molecular pathogenesis. Brain Pathol.ciency. Cell 75, 1361–1370.7, 927–942.

Hollmann, M., and Heinemann, S. (1994). Cloned glutamate recep-Kreusch, A., Pfaffinger, P.J., Stevens, C.F., and Choe, S. (1998).tors. Ann. Rev. Neurosci. 17, 31–108.Crystal structure of the tetramerization domain of Shaker potassium

Hollmann, M., Hartley, M., and Heinemann, S. (1991). Ca21 perme- channel. Nature, in press.ability of KA-AMPA–gated glutamate receptor channels depends on

Krupp, J.J., Vissel, B., Heinemann, S.F., and Westbrook, G.L. (1998).subunit composition. Science 252, 851–853.N-terminal domains in the NR2 subunit control desensitization of

Hollmann, M., Maron, C., and Heinemann, S. (1994). N-glycosylation NMDA receptors. Neuron 20, 317–327.site tagging suggests a three transmembrane domain topology for

Kuryatov, A., Laube, B., Betz, H., and Kuhse, J. (1994). Mutationalthe glutamate receptor GluR1. Neuron 13, 1331–1343.analysis of the glycine-binding site of the NMDA receptor: structural

Hollmann, M., O’Shea-Greenfield, A., Rogers, S.W., and Heinemann, similarity with bacterial amino acid–binding proteins. Neuron 12,S. (1989). Cloning by functional expression of a member of the 1291–1300.glutamate receptor family. Nature 342, 643–648.

Kuusinen, A., Arvola, M., and Keinanen, K. (1995). Molecular dissec-Hume, R.I., Dingledine, R., and Heinemann, S. (1991). Identification tion of the agonist binding site of an AMPA receptor. EMBO J. 14,of a site in glutamate receptor subunits that controls calcium perme- 6327–6332.ability. Science 253, 1028–1031. Kyte, J., and Doolittle, R.F. (1982). A simple method for displayingHuntington’s Disease Collaborative Research Group (1993). A novel the hydropathic character of a protein. J. Mol. Biol. 157, 105–132.gene containing a trinucleotide repeat that is expanded and unstable Langosch, D., Becker, C.M., and Betz, H. (1990). The inhibitory gly-on Huntington’s disease chromosomes. Cell 72, 971–983. cine receptor: a ligand-gated chloride channel of the central nervousHutton, M., and Hardy, J. (1997). The presenilins and Alzheimer’s system. Eur. J. Biochem. 194, 1–8.disease. Hum. Mol. Genet. 6, 1639–1646. Laube, B., Hirai, H., Sturgess, M., Betz, H., and Kuhse, J. (1997).Imoto, K., Methfessel, C., Sakmann, B., Mishina, M., Mori, Y., Konno, Molecular determinants of agonist discrimination by NMDA receptorT., Fukuda, K., Kurasaki, M., Bujo, H., Fujita, Y., and Numa, S. (1986). subunits: analysis of the glutamate binding site on the NR2b subunit.Location of a d-subunit region determining ion transport through Neuron 18, 493–503.the acetylcholine receptor channel. Nature 324, 670–674. Leonoudakis, D., Gray, A.T., Winegar, B.D., Kindler, C.H., Harada,Imoto, K., Busch, C., Sakmann, B., Mishina, M., Konno, T., Nakai, M., Taylor, D.M., Chavez, R.A., Forsayeth, J.R., and Yost, C.S. (1998).J., Bujo, H., Mori, Y., Fukuda, K., and Numa, S. (1988). Rings of An open rectifier potassium channel with two pore domains in tan-negatively charged amino-acids determine the acetylcholine-recep- dem cloned from rat cerebellum. J. Neurosci. 18, 868–877.tor channel conductance. Nature 335, 645–648. Li, H., Lee, S., and Jap, B.K. (1997). Molecular design of aquaporin-1

water channel as revealed by electron crystallography. NatureJia, Z., Agopyan, N., Miu, P., Xiong, Z., Henderson, J., Gerlai, R.,Struct. Biol. 4, 263–265.Taverna, F.A., Velumian, A., MacDonald, J., Carlen, P., et al. (1996).

Enhanced LTP in mice deficient in the AMPAreceptor GluR2. Neuron MacDonald, M.E., Novelletto,A., Lin, C., Tagle, D., Barnes, G., Bates,17, 945–956. G., Taylor, S., Allitto, B., Altherr, M., Myers, R., et al. (1992). The

Huntington’s disease candidate region exhibits many different hap-Karlin, A., and Akabas, M.H. (1995). Toward a structural basis forlotypes. Nature Genet. 1, 99–103.the function of nicotinic acetylcholine receptors and their cousins.

Neuron 15, 1231–1244. Malenka, R.C., and Nicoll, R.A. (1990). Intracellular signals and LTP.Semin. Neurosci. 2, 335–343.Kashiwabuchi, N., Ikeda, K., Araki, K., Hirano, T., Shibuki, K., Taka-

yama, C., Inoue, Y., Kutsuwada, T., Yagi, T., Kang, Y., et al. (1995). Martin, J.B., and Gusella, J.F. (1986). Huntington’s disease. Patho-Impairment of motor coordination, Purkinje cell synapse formation, genesis and management. N. Engl. J. Med. 315, 1267–1276.and cerebellar long-term depression in GluRd2 mutant mice. Cell Masu, Y., Nakayama, K., Tamaki, H., Harada, Y., Kuno, M., and81, 245–252. Nakanishi, S. (1987). cDNA cloning of bovine substance-K receptorKaupmann, K., Huggel, K., Heid, J., Flor, P.J., Bischoff, S., Mickel, through oocyte expression system. Nature 329, 836–838.S.J., McMaster, G., Angst, C., Bittiger, H., Froestl, W. and Bettler, McHugh, T.J., Blum, K.I., Tsien, J.Z., Tonegawa, S., and Wilson,B. (1997). Expression cloning of GABAB receptors uncovers similarity M.A. (1996). Impaired hippocampal representation of space in CA1-to metabotropic glutamate receptors. Nature 386, 239–246. specific NMDAR1 knockout mice. Cell 87, 1339–1349.Kennedy, M.B. (1997). The postsynaptic density at glutamatergic Milani, N., Dalpra, L., del Prete, A., Zanini, R., and Larizza, L. (1996).synapses. Trends Neurosci. 20, 264–268. A novel mutation (Gln266→His) in the alpha1 subunit of the inhibitoryKim, T.W., and Tanzi, R.E. (1997). Presenilins and Alzheimer’s dis- glycine receptor gene (GLRA1) in hereditary hyperekplexia. Am. J.

Hum. Genet. 58, 420–422.ease. Curr. Opin. Neurobiol. 7, 683–688.


Milone, M., Wang, H.-L., Ohno, K., Fukudome, T., Pruitt, J.N., Bren, Persichetti, F., Ambrose, C.M., Ge, P., McNeil, S.M., Srinidhi, J.,Anderson, M.A., Jenkins, B., Barnes, G.T., Duyao, M.P., Kanaley, L.,N., Sine, S.M., and Engel, A.G. (1997). Slow-channel myasthenicet al. (1995). Normal and expanded Huntington’s disease gene al-syndrome caused by enhanced activation, desensitization, and ago-leles produce distinguishable proteins due to translation across thenist binding affinity attributable to mutation in the M2 domain of theCAG repeat. Mol. Med. 1, 374–383.acetylcholine receptor alpha subunit. J. Neurosci. 17, 5651–5665.

Picciotto, M.R., Zoli, M., Rimondini, R., Lena, C., Marubio, L.M., Pich,Mishina, M., Takai, T., Imoto, K., Noda, M., Takahashi, T., Numa,E.M., Fuxe, K., and Changeux, J.-P. (1998). Acetylcholine receptorsS., Methfessel, C., and Sakmann, B. (1986). Molecular distinctioncontaining the b2 subunit are involved in the reinforcing propertiesbetween fetal and adult forms of muscle acetylcholine receptor.of nicotine. Nature 391, 173–177.Nature 321, 406–411.Premkumar, L.S., and Auerbach, A. (1997). Stoichiometry of recom-Moon, I.S., Apperson, M.L., and Kennedy, M.B. (1994). The majorbinant N-methyl-D-aspartate receptor channels inferred from single-tyrosine-phosphorylated protein in the postsynaptic density fractionchannel current patterns. J. Gen. Physiol. 110, 485–502.is N-methyl-D-aspartate receptor subunit 2B. Proc. Natl. Acad. Sci.

USA 91, 3954–3958. Raftery, M.A., Hunkapiller, M.W., Strader, C.D., and Hood, L.E.(1980). Acetylcholine receptor: complex of homologous subunits.Moriyoshi, K., Masu, M., Ishii, T., Shigemoto, R., Mizuno, N., andScience 208, 1454–1456.Nakanishi, S. (1991). Molecular cloning and characterization of the

rat NMDA receptor. Nature 354, 31–37. Rajendra, S., Lynch, J.W., Pierce, K.D., French, C.R., Barry, P.H.,and Schofield, P.R. (1994). Startle disease mutations reduce theMulle, C., Sailer, A., Perez-Otano, I., Dickinson-Anson, H., Castillo,agonist sensitivity of the human inhibitory glycine receptor. J. Biol.P.E., Bureau, I., Maron, C., Gage, F.H., Mann, J.R., Bettler, B., andChem. 269, 18739–18742.Heinemann, S.F. (1998). Altered synaptic physiology and reducedRajendra, S., Vandenberg, R.J., Pierce, K.D., Cunningham, A.M.,susceptibility to kainate-induced seizures in GluR6-deficient mice.French, P.W., Barry, P.H., and Schofield, P.R. (1995). The uniqueNature, in press.extracellular disulphide loop of the glycine receptor is a principalNasir, J.,Floresco, S.B., O’Kusky, J.R., Diewert, V.M., Richman, J.M.,ligand-binding element. EMBO J. 14, 2987–2998.Zeisler, J., Borowski, A., Marth, J.D., Phillips, A.G., and Hayden,Rassendren, F., Buell, G., Newbolt, A., North, R.A., and Surprenant,M.R. (1995). Targeted disruption of the Huntington’s disease geneA. (1997). Identification of amino acid residues contributing to theresults in embryonic lethality and behavioral and morphologicalpore of a P2X receptor. EMBO J. 16, 3446–3454.changes in heterozygotes. Cell 81, 811–823.Ratnam, M., Nguyen, D.L., Rivier, J., Sargent, P.B., and Lindstrom,Nayeem, N., Green, T.P., Martin, I.L., and Barnard, E.A. (1994). Qua-J. (1986). Transmembrane topography of nicotinic acetylcholine re-ternary structure of the native GABAA receptor determined by elec-ceptor: immunochemical tests contradict theoretical predictionstron microscopic image analysis. J. Neurochem. 62, 815–818.based on hydropathy profiles. Biochemistry 25, 2633–2643.

Nicoll, R.A., Kanes, J.A., and Malenka, R.C. (1988). The current ex-Rees, M.I., Andrew, M., Jawad, S., and Owen, M.J. (1994). Evidencecitement in long-term potentiation. Neuron 1, 97–103.for recessive as well as dominant forms of startle disease (hyperek-

Noda, M., Takahashi, H., Tanabe, T., Toyosato, M., Furutani, Y., plexia) caused by mutations in the alpha1 subunit of the inhibitoryHirose, T., Asai, M., Inayama, S., Miyata, T., and Numa, S. (1982). glycine receptor. Hum. Mol. Genet. 3, 2175–2179.Primary structure of a-subunit precursor of Torpedo californica ace-

Rosen, D.R., Siddique, T., Patterson, D., Figlewicz, D.A., Sapp, P.,tylcholine receptor deduced from cDNA sequence. Nature 299,Hentati, A., Donaldson, D., Goto, J., et al. (1993). Mutations in Cu/793–797.Zn superoxide dismutase gene are associated with familial amyotro-

Noda, M., Takahashi, H., Tanabe, T., Toyosato, M., Kikyotani, S., phic lateral sclerosis. Nature 362, 59–62.Furutani, Y., Hirose, T., Takashima, H., Inayama, S., Miyata, T., and

Ross, C.A. (1997). Intranuclear neuronal inclusions: a commonNuma, S. (1983). Structural homology of Torpedo californica acetyl- pathogenic mechanism for glutamine-repeat neurodegenerativecholine receptor subunits. Nature 302, 528–532. diseases? Neuron 19, 1147–1150.Noda, M., Shimizu, S., Tanabe, T., Takai, T., Kayano, T., Ikeda, T., Rubinsztein, D.C., Leggo, J., Chiano, M., Dodge, A., Norbury, G.,Takahashi, H., Nakayama, H., Kanaoka, Y., Minamino, N., et al. Rosser, E., and Craufurd, D. (1997). Genotypes at the GluR6 kainate(1984). Primary structure of Electrophorus electricus sodium chan- receptor locus are associated with variation in the age of onset ofnel deduced from cDNA sequence. Nature 312, 121–127. Huntington disease. Proc. Natl. Acad. Sci. USA 94, 3872–3876.Noda, M., Ikeda, T., Kayano, T., Suzuki, H., Takeshima, H., Kurasaki, Ryan, S.G., Dixon, M.J., Nigro, M.A., Kelts, K.A., Markand, O.N.,M., Takahashi, H., and Numa, S. (1986). Existence of distinct sodium Terry, J.C., Shiang, R., Wasmuth, J.J., and O’Connell, P. (1992a).channel messenger RNAs in rat brain. Nature 320, 188–192. Genetic and radiation hybrid mapping of the hyperekplexia regionOhno, K., Hutchinson, D.O., Milone, M., Brengman, J.M., Bouzat, C., on chromosome 5q. Am. J. Hum. Genet. 51, 1334–1343.Sine, S.M., and Engel, A.G. (1995). Congenital myasthenic syndrome Ryan, S.G., Sherman, S.L., Terry, J.C., Sparkes, R.S., Torres, M.C.,caused by prolonged acetylcholine receptor channel openings due and Mackey, R.W. (1992b). Startle disease, or hyperekplexia: re-to a mutation in the M2 domain of the e subunit. Proc. Natl. Acad. sponse to clonazepam and assignment of the gene (STHE) to chro-Sci. USA 92, 758–762. mosome 5q by linkage analysis. Ann. Neurol. 31, 663–668.Ohno, K., Quiram, P.A., Milone, M., Wang, H.L., Harper, M.C., Pruitt, Ryan, S.G., Buckwalter, M.S., Lynch, J.W., Handford, C.A., Segura,J.N., Brengman, J.M., Pao, L., Fischbeck, K.H., Crawford, T.O., et L., Shiang, R., Wasmuth, J.J., Camper, S.A., Schofield, P., andal. (1997). Congenital myasthenic syndromes due to heteroallelic O’Connell, P. (1994). A missense mutation in the gene encoding thenonsense/missense mutations in the acetylcholine receptor epsilon alpha1 subunit of the inhibitory glycine receptor in the spasmodicsubunit gene: identification and functional characterization of six mouse. Nature Genet. 7, 131–135.new mutations. Hum. Mol. Genet. 6, 753–766. Scherzinger, E., Lurz, R., Turmaine, M., Mangiarini, L., Hollenbach,Orr-Urteger, A., Goldner, F.M., Saeki, M., Lorenzo, I., Goldberg, L., B., Hasenbank, R., Bates, G.P., Davies, S.W., Lehrach, H., andDe Biasi, M., Dani, J.A., Patrick, J.W., and Beaudet, A.L. (1997). Mice Wanker, E.E. (1997). Huntingtin-encoded polyglutamine expansionsdeficient in the a7 neuronal nicotinic acetylcholine receptor lack form amyloid-like protein aggregates in vitro and in vivo. Cell 90,a-bungarotoxin binding sites and hippocampal fast nicotinic cur- 549–558.rents. J. Neurosci. 17, 9165–9171. Schmieden, V., Kuhse, J.,and Betz, H. (1992). Agonist pharmacologyPaas, Y., Eisenstein, M., Medevielle, F., Teichberg, V.I., and Dev- of neonatal and adult glycine receptor a-subunits—identification ofillers-Thiery, A. (1996). Identification of the amino acid subsets ac- amino-acid residues involved in taurine activation. EMBO J. 11,counting for the ligand binding specificity of a glutamate receptor. 2025–2032.Neuron 17, 979–990. Schoepfer, R., Conroy, W.G., Whiting, P., Gore, M., and Linstrom,Paschen, W., and Djuricic, B. (1994). Extent of RNA editing of gluta- J. (1990). Brain a-bungarotoxin binding protein cDNAs and MAbsmate receptor subunit GluR5 in different brain regions of the rat. reveal subtypes of this branch of the ligand-gated ion channel gene

superfamily. Neuron 5, 35–48.Cell. Mol. Neurobiol. 14, 259–270.

Neuron444

Schwalbe, R.A., Bianchi, L., and Brown, A.M. (1997). Mapping the Tsien, J.Z., Huerta, P.T., and Tonegawa, S. (1996). The essential rolekidney potassium channel ROMK1. J. Biol. Chem. 272, 25217– of hippocampal CA1 NMDA receptor–dependent synaptic plasticity25223. in spatial memory. Cell 87, 1327–1338.Scotland, P.B., Colledge, M., Melnikova, I., Dai, Z., and Froehner, Uchino, S., Sakimura, K., Nagahari, K., and Mishina, M. (1992). Muta-S.C. (1993). Clustering of the acetylcholine receptor by the 43-kd tions in a putative agonist binding region of the AMPA-selectiveprotein: involvement of the zinc finger domain. J. Cell Biol. 123, glutamate receptor channel. FEBS Lett. 308, 253–257.719–728. Unwin, N. (1993a). Neurotransmitter action: opening of ligand-gatedSeeburg, P.H. (1993). The molecular biology of mammalian gluta- ion channels. Cell 72/Neuron 10 (suppl.), 31–41.mate receptor channels. Trends Neurosci. 16, 359–365. Unwin, N. (1993b). Nicotinic acetylcholine receptor at 9A resolution.Seri, M., Bolino, A., Galietta, L.J., Lerone, M., Silengo, M., and Ro- J. Mol. Biol. 229, 1101–1124.meo, G. (1997). Startle disease in an Italian family by mutation Verdoorn, T.A., Brunashev, N., Monyer, H., Seeburg, P.H., and Sak-(K276E): the alpha-subunit of the inhibiting glycine receptor. Hum. mann, B. (1991). Structural determinants of ion flow through recom-Mut. 9, 185–187.

binant glutamate receptor channels. Science 252, 1715–1718.Sharp, A.H., and Ross, C.A. (1996). Neurobiology of Huntington’s

Villarroel, A., Burnashev, N., and Sakmann, B. (1995). Dimensionsdisease. Neurobiol. Dis. 3, 3–15.of the narrow portion of a recombinant NMDA receptor channel.

Sheng, M. (1996). PDZs and receptor/channel clustering: rounding Biophys. J. 68, 866–875.up the latest suspects. Neuron 17, 575–578.

Villarroel, A., Regaldo, M.P., and Lerma, J. (1998). Glycine-indepen-Shiang, R., Ryan, S.G., Zhu, Y.Z., Hahn, A.F., O’Connel, P., and dent NMDA receptor desensitization: localization of structural deter-Wasmuth, J.J. (1993). Mutations in the alpha1 subunit of the inhibi- minants. Neuron 20, 329–339.tory glycine receptor cause the dominant neurologic disorder, hy-

Wang, H.L., Auerbach, A., Bren, N., Ohno, K., Engel, A.G., and Sine,perekplexia. Nature Genet. 5, 351–358.S.M. (1997a). Mutation in the M1 domain of the acetylcholine recep-

Shiang, R., Ryan, S.G., Zhu, Y.Z., Fielder, T.J., Allen, R.J., Fryer, A., tor alpha subunit decreases the rate of agonist dissociation. J. Gen.Yamashita, S., O’Connell, P., and Wasmuth, J.J. (1995). Mutational

Physiol. 109, 757–766.analysis of familial and sporadic hyperekplexia. Ann. Neurol. 38,

Wang, Y., Small, D.L., Stanimirovic, D.B., Morley, P., and Durkin,85–91.J.P. (1997b). AMPA receptor–mediated regulation of Gi-protein inSine, S.M., Ohno, K., Bouzat, C., Auerbach, A., Milone, M., Pruitt,cortical neurons. Nature 389, 502–504.J.N., and Engel, A.G. (1995). Mutation of the acetylcholine receptorWest, A.P., Bjorkman, P.J., Dougherty, D.A., and Lester, H.A. (1997).alpha subunit causes a slow-channel myasthenic syndrome by en-Expression and circular dichroism studies of the extracellular do-hancing agonist binding affinity. Neuron 15, 229–239.main of the a subunit of the nicotinic acetylcholine receptor. J. Biol.Singh, N.A., Charlier, C., Stauffer, D., DuPont, B.R., Leach, R.J.,Chem. 272, 25468–25473.Melis, R., Ronen, G.M., Bjerre, I., Quattlebaum, T., Murphy, J.V., etWhite, J.K., Auerbach, W., Duyao, M.P., Vonsattel, J.P., Gusella,al. (1998). A novel potassium channel gene, KCNQ2, is mutated inJ.F., Joyner, A.L., and MacDonald, M.E. (1997). Huntingtin is requiredan inherited epilepsy of newborns. Nature Genet. 18, 25–29.for neurogenesis and is not impaired by the Huntington’s diseaseSommer, B., Kohler, M., Sprengel, R., and Seeburg, P.H. (1991).CAG expansion. Nature Genet. 17, 404–410.RNA editing in the brain controls a determinant of ion flow in gluta-

mate-gated channels. Cell 67, 11–19. Wingrove, P.B., Thompson, S.A., Wafford, K.A., and Whiting, P.J.(1997). Key amino acids in theg subunit of the g-aminobutyric acid(A)Sprengel, R., and Seeburg, P.H. (1995). Ionotropic glutamate recep-receptor that determine ligand binding and modulationat the benzo-tors. In Ligand- and Voltage-Gated Ion Channels, A. R. North, ed.diazepine site. Mol. Pharmacol. 52, 874–881.(Boca Raton, FL: CRC Press), pp. 213–263.Wo, Z.G., and Oswald, R.E. (1994). Transmembrane topology of twoSprengel, R., Suchanek, B., Aminco, C., Brusa, R., Burnashev, N.,kainate receptor subunits revealed by N-glycosylation. Proc. Natl.Rozov, A., Hvalby, Ø., Jensen, V., Paulsen, O., Andersen, P., et al.Acad. Sci. USA 91, 7154–7158.(1998). Importance of the intracellular domain of NR2 subunits for

NMDA receptor function in vivo. Cell 92, 279–289. Yang, J., Jan, Y.N., andJan, L.Y. (1995). Determination of the subunitstoichiometry of an inwardly rectifying potassium channel. NeuronSteinlein, O.K., Mulley, J.C., Propping, P., Wallace, R.H., Phillips,

H.A., Sutherland, G.R., Scheffer, I.E., and Berkovic, S.F. (1995). A 15, 1441–1447.missense mutation in the neuronal nicotinic acetylcholine receptor Zeitlin, S., Liu, J.P., Chapman, D.L., Papaioannou, V.E., and Efstrati-a4 subunit is associated with autosomal dominant nocturnal frontal adis, A. (1995). Increased apoptosis and early embryonic lethalitylobe epilepsy. Nature Genet. 11, 201–203. in mice nullizygous for the Huntington’s disease gene homologue.Steinlein, O.K., Magnusson, A., Stoodt, J., Bertrand, S., Weiland, S., Nature Genet. 11, 155–163.Berkovic, S.F., Nakken, K.O., Propping, P., and Bertrand, D. (1997). Zuo, J., De Jager, P.L., Takahashi, K.A., Jiang, W., Linden, D.J., andAn insertion mutation of the CHRNA4 gene in a family with autosomal Heintz, N. (1997). Neurodegeneration in Lurcher mice caused bydominant nocturnal frontal lobe epilepsy. Hum. Mol. Genet. 6, mutation in delta2 glutamate receptor gene. Nature 388, 769–773.943–947.

Stern-Bach, Y., Bettler, B., Hartley, M., Sheppard, P.O., O’Hara,P.J., and Heinemann, S.F. (1994). Agonist selectivity of glutamatereceptors is specified by two domains structurally related to bacte-rial amino acid–binding proteins. Neuron 13, 1345–1357.

Swanson, G.T., Gereau, R.W., IV, Green, T., and Heinemann, S.F.(1997). Identification of amino acid residues that control functionalbehavior in GluR5 and GluR6 kainate receptors. Neuron 19, 913–926.

Tejedor, F.J., Bokhari, A., Rogero, O., Gorczyca, M., Zhang, J., Kim,E., Sheng, M., and Budnik, V. (1997). Essential role for dlg in synapticclustering of Shaker K1 channels in vivo. J. Neurosci. 17, 152–159.

Trottier, Y., Devys, D., Imbert, G., Saudou, F., An, I., Lutz, Y., Weber,C., Agid, Y., Hirsch, E.C., and Mandel, J.L. (1995a). Cellular localiza-tion of the Huntington’s disease protein and discrimination of thenormal and mutated form. Nature Genet. 10, 104–110.

Trottier, Y., Lutz, Y., Stevanin, G., Imbert, G., Devys, D., Cancel, G.,Saudou, F., Weber, C., David, G., Tora, L., et al. (1995b). Polygluta-mine expansion as a pathological epitope in Huntington’s diseaseand four dominant cerebellar ataxias. Nature 378, 403–406.

Molecular Neurobiology and Genetics: Review Investigation of ...

Documents

Transcript of Molecular Neurobiology and Genetics: Review Investigation of ...