Molecular Studies of Channels L.Y. Jan, D.M. Papazian, T.L. Schwarz, B.L Tempel, L.C. Timpe, and Y.N. Jan Howard Hughes Medical Institute and the Departments of Physiology and Biochemistry, University of California, San Francisco, CA 94143 We are interested in molecules important for neuronal functions, such as voltage-dependent potassium channels. Several different subtypes of such channels have been revealed in biophysical studies. Often, they are expressed in different proportions in different nervous tissues, providing those neurons with particular features of excitability. Do these different channels represent products of genes that are structurally and evolutionarily related? How is their expression regulated in different neurons? How might their structure and/or expression be altered with experience? To approach these questions, one has to study the molecules biochemically. So far only a few, such as the acetylcholine receptor and sodium channels, have been characterized in detail because there are high-affinity ligands and abundant sources for them. To attempt biochemical studies of other important neuronal elements, we have approached the molecular analysis of a potassium channel through genetic means. The Shaker Locus: Gene(s) That Probably Code for Different types of voltage-sensitive K+ channels determine to a large extent the firing pattern and shape of action potentials in individual neurons (Hille 1984). Since the amount of transmitter released from a neuron can be drastically altered by slight alterations of the shape of the action potential (Klein and Kandel 1980; Llinas et al. 1982), K+ channels may play an important role in the control of synaptic efficacy. In fact, some K+ channels are known to be influenced by transmitter molecules and intracellular messengers (Adams et al. 1982; Grega et al. 1987; Dunlap et al. 1987; Piomelli et al. 1987) and have been implicated in learning (Byrne 1987). Thus far, however, no K+ channels have been purified for biochemical studies, because they are present in very low abundance in most tissues characterized. Previous electrophysiological studies by Jan, Jan, and Dennis (1977) suggested that the Shaker mutations affect the function of K+ channels. Further genetic and biophysical studies (Salkoff 1983; Tanouye et al. 1981; Timpe and Jan 1987; Wu and Haugland 1985) provided strong evidence suggesting that the Shaker locus contains structural gene(s) for a rapidly inactivating K* channel, the A channel (Connor and Stevens 1971). Recent molecular studies of the Shaker locus by Diane Papazian, Tom Schwarz, Bruce Tempel, and Leslie Timpe (Papazian et al. 1987; Tempel et al. 1987; Schwarz et al. 1988; Timpe et al. 1988) are described below. Multiple Potassium Channel Components Are Produced by Alternative Splicing at the Shaker Locus About 200 kb of DNA were isolated from the Shaker region. Within this expanse of DNA, we identified molecular lesions of five severe Shaker mutants, spanning over 60 kb of DNA (Papazian et al. 1987). Although the abundance of Shaker gene product(s) turned out to be rather low, we isolated and analyzed cDNAs from the Shaker locus. The seven cDNA clones that we sequenced represent five distinct transcripts and predict four different proteins. All share a central core. For the carboxy-terminal third, two versions were found: that in ShA and ShC and that in ShB and ShD. At the amino-terminal end, three variants diverge at a single splice site: an eight-residue terminus in ShC, a 60-amino acid sequence in ShA and ShB, and a 48-amino acid sequence in ShD. No strong similarity of one amino-terminal variant to another was observed. The carboxy-terminal variants, however, are very similar, probably because the two mutually exclusive sets of 3' exons arose by duplication (Schwarz et al. 1988). The hypothesis that the Shaker locus encodes a structural component of a potassium channel was first advanced on the basis of genetic and physiological studies (Salkoff 1983; Tanouye et al. 1981; Timpe and Jan 1987; Wu and Haugland 1985). From the sequence of these Shaker cDNAs, two arguments were used in support of this hypothesis (Tempel et al. 1987). First, each Shaker product contains six regions whose length and hydrophobicity indicates that they may be membrane spanning. Second, it contains a region similar to the vertebrate sodium channel (Noda et al. 1986), an arginine-rich region called S4-like, that has been hypothesized to be a transmembrane helix responsible for channel opening in voltage-dependent channels (Noda et al. 1986; Greenblatt et al. 1985; Guy and Seetharamulu 1986; Catterall 1986). The function of Shaker products has been tested by transcribing mRNA in vitro from any one of four Shaker cDNAs encoding different Shaker proteins and injecting it into Xenopus oocytes. Expression of A currents in these oocytes demonstrates that a single species of Shaker product is sufficient to form functional A channels (Timpe et al. 1988). The relative abundance of the various Shaker transcripts differed in the head and body of the fly, implying that the Shaker proteins do not exist in a fixed stoichiometry in all tissues (Schwarz et al. 1988). Thus, some of the Shaker products are probably parts of channel subtypes with differential tissue distribution. Expression in Xenopus oocytes further demonstrates that the A currents generated by different Shaker products show different kinetic properties (Timpe et al. 1988). Because the Shaker products are much smaller and resemble one of the four internally homologous domains of the sodium channel, we suspect that the A channels formed in Xenopus oocytes are homomultimers of Shaker products. It is important to determine the subunit structure of the A channel in the fly, as it is possible that those A channels contain more than one Shaker product. The number of possible heteromultimeric complexes made from the known Shaker products is quite large. These potential channels could differ in their tissue distribution, developmental expression, or physiological properties. Cloning of a Probable Potassium Channel Gene From Mouse Brain Sequences encoding the S4-like segment and the hydrophobic portion of a cDNA from the Shaker locus, ShAl, were used to screen a mouse-brain cDNA library at reduced stringency. Four independent cDNA clones were sequenced. They appear to come from a single gene, which we call MBKI (MBK stands for mouse-brain K+ channel). The predicted size and amino acid sequence of the MBKI protein are very similar to those of the A channel components encoded by Shaker in Drosophila. Although multiple Shaker proteins are predicted, the sequences showing amino acid similarity with. MBKI are either in the common region or in regions that are highly conserved in all the Shaker proteins. Overall, 65 percent of the amino acids are identical between the MBKI protein and any of the predicted Shaker proteins (Tempel et al. 1988). In contrast, the sodium channel from rat brain (Noda et al. 1986) and the dihydropyridine (DHP) receptor from rabbit muscle (Tanabe et al. 1987) are much larger and show much less amino acid similarity with MBK1 or the Shaker products. This represents the first case in which the characterization of an ion channel in Drosophila has allowed a similar channel to be isolated from mammals. The comparison of amino acids conserved in MBKI and Shaker may be useful for designing and constructing chimeric or in vitro mutagenized channels that could help to define structures important for channel function. Possible Applications to Research on the Etiology or Treatment of Neuropsychiatric Disorders Potassium channels represent a diverse group of ion channels whose expression appears to be under stringent control among neurons as well as a number of nonneural cell types. They control to a large extent the excitability and synaptic efficacy of neurons. Abnormality in the function or regulation of potassium channels probably leads to behavioral or mental disorders, as suggested by the limited number of mutant and pharmacological studies. Before molecular and biochemical information on potassium channels becomes available, such abnormalities are difficult to identify among patients with neuropsychiatric disorders. Cloning of potassium channel genes and mapping these genes on the human genome will allow one to examine possible correlation with loci responsible for hereditary disorders. DNA probes for the potassium channel transcripts and antibodies that recognize the channel can be used to look for possible disruption of normal expression pattern. For any channel that is implicated in a disorder, knowledge of the amino acid sequence and structure of that channel will be of great use in the design of pharmacological agents for treatment. To give an example, the recent characterization of the ATP-sensitive potassium channel in pancreatic beta-islet cells has not only provided a theoretical framework for analyzing the action of glucose concentration increase on insulin release, but also revealed that this potassium channel is likely to be altered in its activity or expression in mutant diabetes and obese mice and that the sulphonylurea type of antidiabetic drugs act on these potassium channels (Petersen and Findlay 1987). Similar advances in understanding and treatment of neuropsychiatric disorders may be facilitated by molecular studies of potassium channels. Future Areas of Research We have just witnessed the beginning of the molecular studies of potassium channels. Thus far, one potassium channel gene has been cloned in Drosophila. The extremely high degree of amino acid conservation between this fly gene and a mouse gene (Tempel et al. 1988) suggests that molecular studies carried out on any experimentally accessible organisms can be readily applied to similar studies in humans. Two major basic questions remain to be approached. 1. The isolation and characterization of other potassium channel genes: This may be approached in a number of ways. For example, if significant structural similarities are found between different potassium channels, one may be able to use one channel gene as a molecular probe to isolate other channel genes. 2. The structure-function analyses of these potassium channels: The determination of potassium channel structure, in conjunction with site-directed mutagenesis experiments, will be useful in understanding how an ion channel works. Such knowledge would be useful in clinical and pharmacological applications as well. REFERENCES Adams, P.R.; Brown, D.A.; and Constanti, A. M-currents and other potassium currents in bullfrog sympathetic neurones. Journal of Physiology (Cambridge) 330:537-572, 1982. Byrne, J.H. Cellular analysis of associative learning. Physiological Reviews 67(2):329439, 1987. Catterall, W.A. Molecular properties of voltage-sensitive sodium channels. Annual Review of Biochemistry 55:953-985, 1986. Connor, J.A., and Stevens, C.F. Prediction of repetitive firing behaviour from voltage clamp data on an isolated neurone soma. Journal of Physiology (Cambridge) 213(1):31-53, 1971. Dunlap, K.; Holz, G.G.; and Rane, S.G. G proteins as regulators of ion channel function. Trends in Neurosciences 10(6):241-244, 1987. Greenblatt, R.E.; Blatt, Y.; and Montal, M. The structure of the voltage-sensitive sodium channel. FEBS Letters (Amsterdam) 193(2):125-134, 1985. Grega, D.S.; Werz, M.A.; and MacDonald, R.L. Forskolin and phorbol esters reduce the same potassium conductance of mouse neurons in culture. Science 235(4786):345-348, 1987. Guy, H.R., and Seetharamulu, P. Molecular model of the action potential sodium channel. Proceedings of the National Academy of Sciences of the United States of America 83(2):508-512, 1986. Hille, B. Ionic Channels of Excitable Membranes. Sunderland, MA: Sinauer, 1984. Jan, Y.N.; Jan, L.Y.; and Dennis, M.J. Two mutations of synaptic transmission in Drosophila. Proceedings of the Royal Society of London. Series B: Biological Sciences (London) 198(1130):87-108, 1977. Klein, M., and Kandel, E.R. Mechanism of calcium current modulation underlying presynaptic facilitation and behavioral sensitization in Aplysia. Proceedings of the National Academy of Sciences of the United States of America 77(11):6912-6916, 1980. Llinas, R.; Sugimori, M.; and Simon, S.M. Transmission by presynaptic spike-like depolarization in the squid giant synapse. Proceedings of the National Academy of Sciences of the United States of America 79(7):2415-2419, 1982. Noda, M.; Ikeda, T.; Kayano, T.; Suzuki, H.; Takeshima, H.; Kurasaki, M.; Takahashi, H.; and Numa, S. Existence of distinct sodium channel messenger RNAs in rat brain. Nature 320(6058):188-192, 1986. Papazian, D.M.; Schwarz, T.L.; Tempel, B.L; Jan, Y.N.; and Jan, L.Y. Cloning of |