Abstract
The delayed rectifier potassium current plays a critical role in cellular physiology. This current (IK) inDrosophila larvae is believed to be a single current. However, a likely null mutation in the ShabK+ channel gene (Shab3) reducesIK but does not eliminate it. This raises a question as to whether or not the entire IKpasses through channels encoded by one gene. Similarly, an incomplete blockade of IK by high concentrations of quinidine, a selective IK blocker, raises a question as to whether IK consists of two components that are differentially sensitive to quinidine. We have addressed these questions by a combined use of genetics, pharmacology, and physiology. The current component removed by theShab3 mutation differed from the remaining component in activation kinetics, inactivation kinetics, threshold of activation, and voltage dependence. The two components showed strong differences in sensitivity to quinidine. Physiological properties of the current component removed by theShab3 mutation were similar to those of the quinidine-sensitive fraction of IK. Complementary to this, properties of the current component remaining in the Shab3 mutant muscles were similar to those of the quinidine-resistant fraction ofIK. These observations strongly suggest that, in contrast to the current belief, IKconsists of two components in Drosophila, which are genetically, pharmacologically, and physiologically distinct. These components are being called IKS andIKF. IKS is carried via Shab-encoded channels.IKF defines a new voltage-activated K+ current in Drosophila.
Diversity of K+ channels provides a basis for a wide spectrum of physiological properties among excitable and nonexcitable cells. For example, K+ channels with diverse characteristics play a vital role in several phenomena, including repolarization of membrane potential, cardiac and neuronal pacemaker activity, repetitive firing, sensory receptor potentials, secretion, fertilization, and learning (Rudy, 1988; Colatsky, 1990;Cook, 1990; Hille, 1992; Wu and Ganetzky, 1992; Jan and Jan, 1997;Armstrong and Hille, 1998). Our understanding of diversity of K+ channels, as well as of properties and function of a variety of K+ channels, has been advanced greatly by single gene mutations ofDrosophila. Combining mutations that selectively disrupt channels with drugs that block specific channels has helped in resolving various ionic currents and in determining the role of specific currents in excitability of nerve and muscle cell membranes (Salkoff, 1983; Wu et al., 1983; Gho and Mallart, 1986; Elkins and Ganetzky, 1988; Singh and Wu, 1989; Singh and Wu, 1990; Gho and Ganetzky, 1992). Voltage-activated K+current in the larval muscles of Drosophila has been resolved into two components, a fast transient current (IA) and a delayed sustained current (IK) (Salkoff, 1983; Wu et al., 1983;Wu and Haugland, 1985; Singh and Wu, 1989).
Among the two voltage-activated K+currents, IA is disrupted by mutations in the Shaker gene, which codes for the structure of theIA channels (Kaplan and Trout, 1969;Kamb et al., 1987; Papazian et al., 1987; Pongs et al., 1988). This current is blocked by 4-aminopyridine (4-AP) (Wu and Haugland, 1985; Wu and Ganetzky, 1992). Whereas it has been possible to partially block IK with quinidine, a cinchona alkaloid used as an antiarrhythmic agent in humans (Singh and Wu, 1989;Kraliz et al., 1998), a mutational analysis of this current has not been possible because of absence of mutations that disrupt this current. However, a recently identified mutation (Shab1) in the Shab gene selectively reduces IK without affecting other known ionic currents in the larval muscles (Chopra, 1994; M. Chopra, G.-G. Gu, and S. Singh, unpublished observations). Mutations at the Shab locus, including a likely null allele (Shab3), enable us to ask questions about this current that have not been possible before.
Until now, IK in Drosophilahas been believed to be one homogenous current (Wu and Haugland, 1985;Singh and Wu, 1989, 1999; Wu and Ganetzky, 1992; Tsunoda and Salkoff, 1995b). However, the current is not completely removed byShab3, which appears to be a genetically null allele of the gene that codes for the structure of the channels (Hegde et al., 1999). This raises a question as to whetherIK is indeed a single current or whether it consists of two components, only one of which is carried by the Shab-encoded channels. The current is also not blocked completely by high concentrations of quinidine and its analogs, which selectively block IK. The data presented here strongly argue that the delayed sustained current in the larval muscles of Drosophila may consist of two distinct components. Identification of a new current component (IKF) in these experiments raises important questions on the identity of the gene that codes for the channels carrying IKF, in vivo physiological role of IKF in muscle excitability, and pharmacological specificity of the current.
MATERIALS AND METHODS
Flies were grown on a standard cornmeal medium at 21°C (Chopra and Singh, 1994). K+ currents were recorded from body-wall muscles 12 and 13 (Gu and Singh, 1997) of wandering third instar larvae by two-microelectrode voltage clamping (Wu and Haugland, 1985). Larvae were dissected from dorsal side, and internal organs were removed. All recordings were completed within 30 min from the start of the dissection (Gielow et al., 1995). Electrodes were made from thin-walled borosilicate glass capillaries with an outside diameter of 1.0 mm (World Precision Instruments, Sarasota, FL). The voltage electrode was filled with 2.5 m KCl and the current electrode with a 3:1 mixture of KCl and potassium citrate (Wu and Haugland, 1985). Resistances of both electrodes were in the range of 10–15 MΩ.
All recordings were made in a Ca2+-free bath solution. This prevents the activation of the two Ca2+ currents and the two Ca2+-activated K+ currents (Wu and Haugland, 1985; Singh et al., 1989). These recording conditions produce only the voltage-activated fast transient (IA) and the delayed sustained (IK) currents. Currents were elicited by 500 msec voltage steps from a holding potential of −80 mV to potentials between -40 and +40 mV, in 10 mV increments. In some experiments, as mentioned in the figure legends, a prepulse of 2 sec duration, to −20 mV, was used to inactivate IA (Wu and Haugland, 1985).
The recording solution contained (in mm): NaCl 77.5, KCl 5, MgCl2 20, NaHCO3 2.5, trehalose 5, sucrose 115, EGTA 0.5, and HEPES 5 (Stewart et al., 1994; Gu and Singh, 1997). In addition, the recording solution in some experiments, as explained in Results, also contained quinidine, 4-AP, and tetraethylammonium (TEA). The pH was adjusted to 7.1 with NaOH.
Voltage stimuli for eliciting the currents were generated with the help of a Macintosh IISi computer through a 12-bit digital-to-analog converter (MacADIOS II/16 board; GW Instruments, Somerville, MA). The resulting current was recorded with the help of an amplifier (TEC 01C/02/03; NPI Electronic GmbH, Haeldenstrasse, Germany) connected to the computer. Data were converted from analog to digital form with the help of a 16-bit analog to digital converter (MacADIOS II/16 board; GW Instruments). Currents were sampled every 500 μs for digital conversion, except during capacitance transients (every 100 μs), which were used for measuring cell capacitance. The digitized data were analyzed off-line with the help of a program written in “C” language.
Control recordings were performed independently for each set of experiments, and digital subtraction was performed between data obtained during the same set of experiments. The current measurements are given as current density (nanoampere per nanofaraday) to avoid differences attributable to fiber size. The values are given as mean ± SE.
RESULTS
The voltage-activated delayed sustained current (IK) recorded from the normal and theShab3 mutant muscles is shown in Figure 1, A and B, respectively. Under the recording conditions used in this experiment (see Materials and Methods), the voltage-clamp traces show a fast transient peak (IA) and the delayed sustained current (IK). Figure1C shows the current–voltage (I–V) relationships forIK recorded from the normal and the mutant muscles. The Shab3 mutation has two small deletions in the N-terminal region of the channel protein upstream of the first transmembrane domain (Hegde et al., 1999). The first deletion of 24 base pairs removes nucleotides 508–531. The second deletion removes nucleotides 656–1011, in turn shifting the reading frame and introducing a stop codon 74 bases downstream of the mutation. The Shab3 protein is thus expected to be truncated before the S1 segment, which starts at amino acid 436. The truncated protein is also expected to lack the N-terminal tetramerization region before the S1 segment (Li et al., 1992; Shen et al., 1993) and is thus not likely to act in a dominant-negative manner.Shab3 is thus most likely a null mutation in the gene. If the entire IKcurrent passes through Shab-encoded channels, theShab3 mutation is expected to completely eliminate IK. However,Shab3 removes the current only partially (Fig. 1), reducing it by ∼65% of the total current. One possibility raised by these data are that the sustained K+ current may consist of two distinct current components, only one of which is carried by channels encoded by the Shab gene.
To examine whether properties of the current component eliminated byShab3 differed from those of the remaining component, the two components were compared for their activation and inactivation kinetics (Fig.2). To enable this comparison, the fast transient current (IA), which masks the rise of IK, needs to be removed.IA was inactivated by using a 2 sec prepulse to −20 mV. Under these conditions, onlyIK is observed (Wu and Haugland, 1985). IK was recorded from wild-type [Canton-S (CS)] and Shab3muscles. The component removed by theShab3 mutation (Fig. 2C) was obtained by digitally subtracting the mutant current (Fig.2B) from the wild-type current (Fig.2A). Figure 2D compares the kinetics of the two current components. In this figure, theShab-independent component was scaled vertically to bring its maximal value to the level of the maximal value of theShab-affected component. Traces are shown for voltage steps to +20 and +40 mV. Activation kinetics of theShab-independent component were faster than those of theShab-affected component. In addition, theShab-independent component showed slight inactivation after reaching the maximum value, whereas the Shab-affected component did not show inactivation until the end of the pulse. Kinetic differences between the Shab-affected and theShab-independent components lend support to the possibility of IK consisting of two distinct components. For the following discussion, the current component eliminated by Shab3 (presumably representing channels encoded by the Shab gene) is designated as IKS (for “slow” activation). The current component unaffected by theShab3 mutation (presumably representing channels not encoded by the Shab gene) is designated as IKF (for “fast” activation).
Figure 2E shows that voltage-dependence of theIKF component is different from that of the IKS component. Activation threshold of the two components is also different, withIKF activating at approximately −30 mV and IKS at approximately −10 mV. This further strengthens arguments for the distinct nature of the two current components.
The possibility that IK may consist of two distinct components is also raised by another independent set of experiments. Quinidine and its analogs, which selectively blockIK in larval muscles (Singh and Wu, 1989, 1990) do not block IKcompletely, even at high concentrations (Kraliz and Singh, 1997; Kraliz et al., 1998). One possibility raised by these data is thatIK may consist of two components, only one of which is sensitive to blockade by quinidine. To determine whether there was any correlation between the two likely current components resolved by quinidine and the two likely components resolved by the Shab3 mutation, we examined the blockade of IKF andIKS by quinidine (Fig.3). As in the experiments mentioned above, IKF was recorded fromShab3, andIKS was obtained by digitally subtracting IKF from the total current obtained in the wild-type (CS). Figure 3, A andB, respectively, show recordings from the CS and theShab3 muscles in the presence of 100 μm quinidine. Currents recorded from CS andShab3 in the absence of quinidine were similar to those shown in Figure 2, A and B, respectively, and are not shown here. Figure 3A represents residual IK (i.e.,IKF +IKS) not affected by quinidine. Figure3B represents residual IKFnot affected by quinidine. Data shown in Figure 3B were digitally subtracted from data shown in Figure 3A to obtain residual IKS not affected by quinidine (Fig. 3C). Figure 3D shows I–V plots for IKF andIKS in the presence of 100 μm quinidine, with I–V plots forIKF andIKS in quinidine-free solution shown for comparison as dotted and dashed lines, respectively. The bar graph in Figure 3E shows percentages of IKF andIKS obtained during a pulse to +40 mV, remaining in the presence of quinidine. Quinidine reducedIKF by ∼35% andIKS by ∼89%. Thus, the two current components showed a strong difference in their blockade by quinidine. The data indicate that the Shab3mutation and quinidine affect the same component of the total current, whereas the component left unaffected by theShab3 mutation is the one that is less sensitive to quinidine.
According to the above interpretation, properties of the current carried by the Shab-encoded channels (IKS) are expected to be similar to those of the quinidine-sensitive fraction of the current, and properties of the Shab-independent current (IKF) are expected to be similar to those of the quinidine-resistant fraction of the current, with only a minor deviation attributable to some blockade ofIKF by quinidine. Figure4 compares the properties ofIKF andIKS with those of the quinidine-resistant and the quinidine-sensitive fractions, respectively. As described for Figure 2 above,IKF was measured inShab3 muscles, andIKS was obtained by subtractingIKF from the CS current. Similarly, quinidine-resistant fraction was obtained by measuring current from CS in the presence of 100 μm quinidine, and the quinidine-sensitive fraction was obtained by subtracting quinidine-resistant fraction from the total CS current.IKF,IKS, and the quinidine-resistant fraction of total IK were as shown in Figures 2, B and C, and 3A, respectively, and are not shown in Figure 4. The quinidine-sensitive fraction of total IK is shown in Figure 4A. Figure 4, B and C, compares the activation and inactivation kinetics of the two current components as resolved by the two methods. To avoid clutter, current traces are shown for voltage pulses to +20 and +40 mV only. Traces for quinidine-resistant fraction were scaled up to traces forIKF for comparison of kinetics. Similarly, traces for IKS were scaled up to those for the quinidine-sensitive fraction for comparison. Activation and inactivation kinetics ofIKF were similar to those of the quinidine-resistant fraction (Fig. 4B). Similarly, kinetics of IKS were similar to those of the quinidine-sensitive fraction (Fig. 4C). Figure4D shows that voltage dependence ofIKS andIKF (as seen in Fig.2E and shown here for comparison as dashedand dotted lines, respectively) were similar to those of quinidine-sensitive and quinidine-resistant fractions, respectively. Partial blockage of IKF by quinidine is reflected in the quinidine-resistant fraction being slightly less than IKF and the quinidine-sensitive fraction being slightly more than IKS(Fig. 4D).
Shab3 and quinidine affect the current by very different mechanisms.Shab3 is expected to eliminate the channel protein itself, whereas quinidine blocks the channels present in the membrane. The above data showing similar effects of quinidine and the Shab3 mutation on the amplitude, activation kinetics, inactivation kinetics, threshold of activation, and voltage dependence of two current components suggest that the delayed sustained current consists of two distinct components, one carried by Shab-encoded channels (which are blocked by quinidine) and the other carried by a different set of channels (which are relatively less sensitive to blockade by quinidine).
To examine the sensitivity of IKF andIKS to other K+ channel blockers, we tested the two currents for blockade by 4-AP and TEA. Figures5 and 6show the effect of these drugs on the two currents. 4-AP (5 mm) blocked both IKF andIKS to a similar extent, to ∼56 and ∼60% of the control current, respectively. Thus,IKF andIKS are much less sensitive to blockade by 4-AP than IA, which is almost completely blocked by 50 μm 4-AP (Wu and Ganetzky, 1988). This is consistent with the inability of 1 mm 4-AP to affect Drosophila Shabchannels expressed in Xenopus oocytes (Covarrubias et al., 1991). The effect of 10 mm TEA is shown in Figure6 with IKF andIKS being ∼75 and 84% of the control current, respectively. This compares with ∼100 mm TEA nearly eliminatingIA andICF in the larval muscles (Wu and Ganetzky, 1988). We are currently testing other K+ channel blockers to identify drugs and toxins that can selectively block IKF. Pharmacological agents and mutations that eliminateIKF selectively, and in general a pharmacological profile of the two currents, will be very helpful in analyzing the properties of the two currents, in determining their individual roles in membrane excitability, and in studying the mechanisms underlying their regulation.
With the resolution of IK into two components, the total voltage-activated K+current in the larval muscles of Drosophila, as shown in Figure 1A, can be now resolved into three distinct components (Fig. 7) in several ways. These include the Shaker and the Shab mutations, which eliminate IA andIKS, respectively, and 4-AP and quinidine, which block IA andIKS, respectively. Differences in physiological properties between the three currents further help in resolving these currents. In combination with a similar resolution of the two Ca2+-activated K+ currents (ICF andICS) (Gho and Mallart, 1986; Singh and Wu, 1989, 1990) and two Ca2+ currents (the 1,4-dihydropyridine-sensitive and the amiloride-sensitive current) (Gielow et al., 1995), the larval muscles of Drosophilaprovide an excellent preparation in which all known specific current components can now be resolved and studied individually.
DISCUSSION
Experiments described in this report provide a strong argument for the existence of two distinct current components (IKF andIKS) in the slow sustained voltage-activated K+ current (IK) in the larval muscles ofDrosophila. Voltage-activated K+ current in the larval muscles ofDrosophila has been previously resolved into two distinct currents. With the data presented here, we can now resolve the total voltage-activated K+ current into three components. Resolution of IK intoIKS andIKF will be particularly helpful in analyzing the properties of these two currents, deciphering the functional role of each current in muscle excitability, and studying mechanisms underlying their function and regulation.
Channels carrying IKS are encoded by the Shab gene. IKS shares properties with the current generated by expressing Shabchannels in Xenopus oocytes. These properties include relative resistance to blockade by 4-AP and a relatively slow activation (Covarrubias et al., 1991; Tsunoda and Salkoff, 1995a). However, in contrast to a slight inactivation of the delayed rectifier current recorded from Xenopus oocytes expressingShab channels (Salkoff et al., 1992),IKS shows no observable inactivation of IKS during the 500 msec pulse.
The gene that encodes the channels carrying the new current (IKF) identified in this study remains to be identified. Among the channels that give rise to voltage-activated K+ currents in in vitro expression systems, the Shaker and theShal channels give rise to fast transient currents, whereas the Shab and the Shaw channels give rise to slow sustained currents (Iverson et al., 1988; Timpe et al., 1988; Salkoff et al., 1992). One of the two sustained currents (IKS) resolved in this study is carried by the Shab-encoded channels. These data raise the possibility that the second channel may be encoded by theShaw gene (Tsunoda and Salkoff, 1995a). However, physiological and pharmacological properties ofIKF seen in our recordings differ from those of Drosophila and mammalian Shaw current observed in Xenopus oocyte system (Salkoff et al., 1992;Kirsch and Drewe, 1993; Kanemasa et al., 1995). For example, Kv3.1 channels (a mammalian representative of Shaw) are ∼150 times more sensitive to 4-AP than Kv2.1 (Shab) channels (Kirsch and Drewe, 1993). This contrasts to almost similar blockade of the IKF channels to that of theShab-encoded IKS channels (Fig. 5). Similarly, in contrast toIKF (Fig. 6E), Kv3.1 channels are blocked by TEA with a half-blocking dose of ∼220 μm (Iverson et al., 1988; Timpe et al., 1988;Salkoff et al., 1992). One possibility is that the observed differences arise from an in vitro expression of the channels. On the other hand, IKF may be carried via channels encoded by other genes, such as seizure(sei) or ether-a-go-go (eag) (Warmke et al., 1991; Titus et al., 1997; Wang et al., 1997). Mutations at theeag locus have been shown to reduceIK (Zhong and Wu, 1991), although the exact mechanism for this effect is not yet clear. It is not clear at this stage whether IKF is aShaw current with novel properties, whether it is carried via sei- or eag-encoded channels, or whether it represents an as yet unidentified gene for a voltage-activated K+ channel in Drosophila. It will be very instructive to examine the nature ofIKF and identify the gene that codes for the channels carrying this current. Availability of a pharmacological agent that selectively blocks these channels can provide a valuable tool for this purpose. Similarly, single gene mutations that affect the current can also help greatly. Such mutations and pharmacological agents will also be very valuable for a molecular analysis of the IKF channels and their function in Drosophila.
The delayed rectifier current is a ubiquitous current present in a large variety of cells in most species. In human cardiac cells, it consists of two components that show differential sensitivity to various antiarrhythmic agents (Sanguinetti and Jurkiewicz, 1990; Singh, 1998). It will be interesting to analyze correlations, if any, between the two components of IK inDrosophila and the two components of the delayed rectifier current in human cardiac cells. There is already some indication of a pharmacological overlap between Drosophila and human cardiac K+ currents. Quinidine, which affects the delayed rectifier K+ channels in human heart (Roden, 1996), affects heartbeat in Drosophila (Gu and Singh, 1995), as well as blocks IKS inDrosophila larval muscles (Kraliz et al., 1998). Any correspondence between components of Drosophila and mammalian delayed rectifier currents will be very useful in undertaking a genetic analysis of cardiac excitability, particularly with the help of mutations that affect IKF andIKS in Drosophila.
Footnotes
This work was supported by National Science Foundation Grant MCB-9604457 and National Institutes of Health Grant GM-50779. A.S. would like to thank Karen Snyder of Williamsville North High School for her support during the conduct of these experiments.
Correspondence should be addressed to Dr. Satpal Singh, Department of Biochemical Pharmacology, 308 Hochstetter Hall, State University of New York at Buffalo, Buffalo, NY, 14260-1200.
Dr. A. Singh’s present address: College of Arts and Sciences, Cornell University, Ithaca, NY 14853.