Random mutagenesis reveals a region important for gating of the yeast K+ channel Ykc1
1997; Springer Nature; Volume: 16; Issue: 16 Linguagem: Inglês
10.1093/emboj/16.16.4817
ISSN1460-2075
AutoresStephen H. Loukin, Brian Vaillant, Xin Liang Zhou, Edgar P. Spalding, Ching Kung, Yoshiro Saimi,
Tópico(s)Cardiac electrophysiology and arrhythmias
ResumoArticle15 August 1997free access Random mutagenesis reveals a region important for gating of the yeast K+ channel Ykc1 Stephen H. Loukin Stephen H. Loukin Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Brian Vaillant Brian Vaillant Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Xin-Liang Zhou Xin-Liang Zhou Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Edgar P. Spalding Edgar P. Spalding Department of Botany, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Ching Kung Ching Kung Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Department of Genetics, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Yoshiro Saimi Corresponding Author Yoshiro Saimi Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Stephen H. Loukin Stephen H. Loukin Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Brian Vaillant Brian Vaillant Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Xin-Liang Zhou Xin-Liang Zhou Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Edgar P. Spalding Edgar P. Spalding Department of Botany, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Ching Kung Ching Kung Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Department of Genetics, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Yoshiro Saimi Corresponding Author Yoshiro Saimi Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA Search for more papers by this author Author Information Stephen H. Loukin1, Brian Vaillant1, Xin-Liang Zhou1, Edgar P. Spalding2, Ching Kung1,3 and Yoshiro Saimi 1 1Laboratory of Molecular Biology, University of Wisconsin, Madison, WI, 53706 USA 2Department of Botany, University of Wisconsin, Madison, WI, 53706 USA 3Department of Genetics, University of Wisconsin, Madison, WI, 53706 USA The EMBO Journal (1997)16:4817-4825https://doi.org/10.1093/emboj/16.16.4817 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info YKC1 (TOK1, DUK1, YORK) encodes the outwardly rectifying K+ channel of the yeast plasma membrane. Non-targeted mutations of YKC1 were isolated by their ability to completely block proliferation when expressed in yeast. All such mutations examined occurred near the cytoplasmic ends of the transmembrane segments following either of the duplicated P loops, which we termed the 'post-P loop' (PP) regions. These PP mutations specifically caused marked defects in the 'C1' states, a set of interrelated closed states that Ykc1 enters and exits at rates of tens to hundreds of milliseconds. These results indicate that the Ykc1 PP region plays a role in determining closed state conformations and that non-targeted mutagenesis and microbial selection can be a valuable tool for probing structure–function relationships of ion channels. Introduction Patch–clamp recordings of the Saccharomyces cerevisiae plasma membrane reveal an outwardly rectifying K+ current (Gustin et al., 1986; Bertl et al., 1993), a stretch activated current (Gustin et al., 1988) and a less prominent inward K+ current conducted through a K+ transporter (Bertl et al., 1995). The gene encoding the outwardly rectifying K+ channel, YKC1 (also known as TOK1, DUK1 and YORK), was identified from the yeast genomic sequence (Miosga et al., 1994) by virtue of its conserved P regions (Ketchum et al., 1995; Zhou et al., 1995; Lesage et al., 1996; Reid et al., 1996). Deletion of YKC1 results in complete elimination of the outwardly rectifying plasma membrane K+ current (Zhou et al., 1995; Reid et al., 1996). Ykc1 has two P regions embedded in a predicted topology of M1-M2-M3-M4-M5-P1-M6-M7-P2-M8. The first M1–M6 topology is reminiscent of many voltage- or cyclic nucleotide-gated (V or CNG) channels, including K+, Na+, Ca2+ and non-selective cation channels. The second P region appears to have arisen by a tandem duplication of the P1 region (Reid et al., 1996). Interestingly, there are now several 'two pored' channels found to have an M1-P1-M2-M3-P2-M4 motif (Fink et al., 1996; Goldstein et al., 1996; Czempinski et al., 1997; Lesage et al., 1997). The P regions of Ykc1 are more similar to the 6-transmembrane K+ channels than to the IRK-type 2-transmembrane K+ channels (Zhou et al., 1995). Outside its P regions Ykc1 shares little primary sequence homology with other channels. When expressed in oocytes, YKC1 produces a K+-selective, outwardly rectifying current with properties similar to those of natively expressed YKC1. These properties include a high degree of K+ selectivity, strong outward rectification and a characteristic flickery open channel behavior. There is disagreement as to whether the outward rectification of YKC1 is a function of the electrochemical driving force for K+ (ΔμK+) (Ketchum et al., 1995; Lesage et al., 1996) or the absolute membrane voltage (Vm) (Zhou et al., 1995; Reid et al., 1996). We agree with Ketcham et al. (1995) and Lesage et al. (1996) that the rectification of oocyte-expressed YKC1 is primarily determined by ΔμK+ (Loukin et al., in preparation). It was originally reported that the rectification of Ykc1 was due to a blockage by external cations (Ketchum et al., 1995), but other investigators have failed to confirm such a divalent ion dependence of Ykc1 rectification (Zhou et al., 1995; Lesage et al., 1996). Activation of Ykc1 is modeled as a C1↔C2↔O transition (Lesage et al., 1996), where C2 is an instantaneously activating state and C1 is slowly activating states. In this report we describe the novel application of non-targeted mutagenesis and classical microbial selection towards a structure–function study of YKC1. Mutations of functional consequence were isolated from randomly mutagenized alleles of YKC1 by their ability to block growth. These alleles were expected to result from loss of regulatory function, such as gating or ionic filtration, since deletion of YKC1 does not affect proliferation under standard conditions. The advantage of this non-targeted strategy over standard targeted mutagenesis is that it does not require nor make a priori assumptions about either the physiological relevance of channel functions or the location of the mutational target. These mutants have furthered our understanding of C1 and C2. In the light of localizations of the gating function in V and CNG channels (Gordon and Zagotta, 1995), our results appear to indicate that the PP region may be part of a conserved gating structure of V and CNG channels. Results Isolation of YKC1 mutations that block growth As reported previously (Zhou et al., 1995; Reid et al., 1996), deletion of YKC1 eliminates the outward rectifying K+ current of the yeast plasma membrane. Ykc1 channel activity can be restored in such deletion strains by expressing YKC1 from plasmid pGALYKC1, which contains the YKC1 open reading frame (ORF) inserted downstream of the galactose-inducible GAL1 promoter. No Ykc1 current could be detected in the plasma membrane of a YKC1 deletion yeast strain (αku8) alone nor in this strain bearing plasmid pGALYKC1 (αku8[pGALYKC1]) cultured in glucose. Ykc1 current increases to levels several fold higher than its native expression when this strain is cultured in the presence of galactose. To screen for YKC1 mutations which cause deleterious channel activities, four pools of pGALYKC1 plasmids were mutagenized with hydroxylamine (HA) and transformed separately into αku8 yeast cells (Figure 1A) (note that HA, though commonly used, primarily induces CG→TA transitions and this limits the range of mutations available; see Discussion). Such a strategy guarantees that mutant plasmids isolated from separate pools have resulted from independent mutagenic events. Note that the plasmids and not the yeast cells were mutagenized in vitro and the chromosomal copy of YKC1 in αku8 had been deleted. These transformed cells were plated onto synthetic glucose medium (SD) lacking uracil (Sherman, 1991), which selected for the presence of the plasmid while repressing mutagenized YKC1. The resultant colonies were replica plated onto a similar medium with glucose replaced by galactose (SG), which induced expression of mutagenized YKC1 in plasmids. Replica that failed to grow on the SG plate were isolated from the SD plate. These galactose-sensitive strains were further plated onto SG containing uracil to test whether the cells which spontaneously lost their plasmids and hence the mutated YKC1 gene (∼1% per division under non-selective conditions; Ausubel et al., 1993) could grow. Figure 1B (right) shows that the galactose-sensitive phenotype was indeed plasmid encoded. Figure 1.Isolation of YKC1 mutant alleles that block yeast proliferation. (A) Summary of the screen used to isolate YKC1 mutants of functional consequence. See Material and methods for a detailed description. (B) Growth blocking phenotype of YKC1 mutants. Equal titers of cells of yeast strain αku8 transformed with the empty plasmid pGALURA, wild-type YKC1-expressing pGALYKC1 (YKC1+) or mutant YKC1-expressing pGALYKC-X (where X is the corresponding mutant allele number) were serially 10-fold diluted (left to right) and allowed to form colonies under conditions which: repressed YKC1 expression (glucose); induced expression of YKC1 (galactose); induced expression of YKC1 but allowed cells which had spontaneously lost their plasmids to proliferate (galactose + uracil). As can be seen, the mutant YKC1 caused galactose sensitivity, which was a plasmid-dependent phenotype since it occurred only when cells were concomitantly required to maintain their plasmids. Download figure Download PowerPoint From ∼40 000 colonies screened, 14 were isolated containing plasmids that stop growth in galactose (Figure 1B, three mutants shown). Those causing slower growth were not included in this report. The plasmids were recovered from these strains and these YKC1 alleles were named YKC1-xxx where the first x indicates which of the four mutagenized plasmid pools they came from to delineate mutational independence. The YKC1 ORFs from two of these plasmids were also subcloned into identical but unmutagenized vectors from which they also conferred the galactose-sensitive phenotype to transformed yeast, verifying that this phenotype was caused by mutations in the YKC1 ORFs and not those occurring elsewhere in the plasmid. Six of these 14 alleles were chosen at random from the four pools for further analysis. Growth blocking mutations all occurred at the cytoplasmic ends of the post-P loop membrane spanning domains The DNA sequences of these six mutant alleles were determined. Two independent isolates, YKC1-301 and YKC1-401, both contain a T322I substitution, YKC1-201, YKC1-304 and YKC1-402 contain a S330F substitution and YKC1-101 contains a V456I substitution (Figure 2A). YKC1-201 and YKC1-401 contain additional silent I248I and F409F nucleotide mutations. The repeated T322I and S330F alleles arose from independent mutagenic events, since they were isolated from separately mutagenized plasmid pools, indicating that at most only a few other hydroxylamine-mutable sites would elicit such a phenotype. The clustering of the mutations is striking, since they all occur at the cytoplasmic ends of the membrane spanning domains following either P loop, M6 or M8, termed the 'PP' (post-P loop) region (Figure 2A). All six mutations occurred within a 10 amino acid window downstream of the aligned P loop, sequences (Figure 2B). Ykc1 comprises 691 residues. Even if the size of this window is doubled to 20 amino acids (solid bordered box in Figure 2B), the chance of six independent mutations randomly appearing in such a window by chance alone is <1 in 1 000 000. Thus, CT→AG transitions that completely block proliferation on SG appear to exclusively occur in the PP regions. Figure 2.Location of growth blocking mutations of Ykc1. Both strands of six mutant alleles of YKC1 which blocked proliferation completely were sequenced. Both strands of all six alleles were sequenced. Non-silent mutations occurred in only three residues. The multiple instances of T322I and S330F were due to independent mutagenic events since they were isolated separately from the original mutagenesis reactions. (A) These mutations cluster in a common topological region, the intracellular junction of the membrane spanning domains immediately following either P loop, a region we called 'PP' for post-P loop. (B) The P1 M6 and P2 M8 regions are aligned and marked. The PP regions, defined as an area of 20 residues surrounding the growth blocking mutations, are marked with solid edged boxes. Identical residues are shaded. Note that the T322I and the V456I mutations are precisely equidistant from the P loops. Download figure Download PowerPoint PP mutations do not affect the O or C2 state The channel activities of the wild-type and all three mutant alleles, YKC1-101 (V456I), YKC1-301 (T322I) and YKC1-402 (S330F), were analyzed. I–V plots show that the steady-state ensemble currents of both wild-type and the mutants, all of which were well expressed in oocytes, rectify outwardly with no substantial inward currents at any potential (Figure 3). Raising external K+ 5-fold from 20 (open symbols) to 100 mM (closed symbols) causes an ∼40 mV rightward shift in the threshold of activation of outward currents in both the wild-type and the mutants, similar to the +40 mV shift this substitution would have on EK. Thus the steady-state current–voltage relationships, being largely determined by ΔμK+, not Vm, are not substantially affected by the PP mutations. Figure 3.Wild-type and PP mutant channels maintain steady-state rectification. Whole cell currents were recorded from oocytes injected with RNA transcribed from wild-type and each of the three different mutant alleles of YKC1. Near steady-state current was measured 900 ms (dotted line) after stepping from the holding potential of −80 mV to test potentials of between −100 and +80 mV at 10 mV intervals in 20 (open symbols) and 100 mM (closed symbols) bath K+. The near steady-state rectification is not significantly affected by PP mutations. For both wild-type and the mutants there was no significant inward current and the current activation threshold was near the predicted EK. Currents were recorded from oocytes 3–5 days after injection with ∼50 ng synthetic mRNA. Bath solutions contained 5 mM HEPES, pH 7.5, 1 mM MgCl2, 1 mM CaCl2 and either 100 or 20 mM KCl and 80 mM NaCl. Traces were neither leak subtracted nor compensated for capacitance. All calibration bars 10 μA×100 ms. Download figure Download PowerPoint Direct examination of the rapid C2→O activation (see Introduction) demonstrates that C2 remains largely intact in the PP mutants. Using recording techniques designed to detect sub-millisecond transitions (see Materials and methods), the time dependence of the 'instantaneous' activation from C2 can in fact be seen (Figure 4A, top; note that the entire trace is only 1 ms here). This C2→O activation occurs notably more slowly than the time it takes for Vm to equilibrate to the command potential (Figure 4A, bottom), ruling out that this time dependence is an artifact of the time dependence of charging the oocyte. When the capacitive component of the outward current is subtracted, this activation from C2 can be seen to have a rate upon depolarization to +80 mV of ∼100 μs (Figure 4B). Activation from C2 of Ykc1-101 (Figure 4C) appears similar to that of the wild-type. Thus, this PP mutation does not significantly affect the rapid transition from C2 to O. Figure 4.PP mutation does not affect activation from C2. Note the rapid time scale displayed, with all the traces in this figure showing only 1 ms or 1/1000 the time course illustrated in the whole cell traces of the previous figures. All calibration bars are 15 μA×0.4 ms. (A) (Upper traces) Whole cell currents of oocytes expressing wild-type Ykc1 upon voltage steps from −20 mV to +80, +70, +60, −100, −110 and −120 mV in 100 mM external K+. The capacitive currents can be seen upon all voltage steps. Rapid activation from the C2 state can clearly be resolved after these capacitive surges upon depolarization (star). (Lower traces) Vm was simultaneously monitored with the current shown in (A) to verify that Vm in fact reached the command potential before significant C2→O activation occurred. The vertical dashed line marks the same time point in both the upper and lower traces to help show that the command potential was reached long before the majority of C2→O activation had occurred. (B) Traces showing C2→O activation with the capacitive currents subtracted by addition of the −120, −110 and −100 mV traces to the +80, +70 and +60 traces respectively in (A). Wild-type Ykc1 was activated from C2 at a rate of ∼100 μs. (C) Capacitance-subtracted C2→O activation of Ykc1-101-expressing oocytes shows that this activation is not affected by this PP mutation. All currents were recorded from oocytes 3 days after injection with ∼100 ng YKC1 RNA to induce high current expression. Electrical recording techniques are described in Materials and methods. Download figure Download PowerPoint The PP mutations also had no substantial effect on the outward conducting properties of O. Substitution of cytoplasmic K+ with either Na+ or NH4+ abolished outward currents in inside-out patches of both YKC1+- and YKC1-101-expressing oocytes (Figure 5), indicating that Ykc1-101 maintained the wild-type selectivity for K+. Similar results were obtained with the other two PP mutant types (data not shown). The apparent unitary slope conductance calculated from single channel excised patches was 22.0 ± 1.9 pS for Ykc1-101, similar to the 20.3 ± 2.5 pS for Ykc1+ (mean ± SD, n = 3). Note that these conductances are probably underestimates of the true open channel conductance, due to the flickery open channel behavior of Ykc1 (Bertl et al., 1993). Figure 5.PP mutant channels maintain ionic selectivity. Ionic selectivities were determined by perfusing bath solutions containing the stated cations over the cytoplasmic face of inside-out patches from oocytes expressing either multiple wild-type or Ykc1-101 channels. Outward currents were observed in both cases only when K+, but not Na+ or NH4+, was present on the cytoplasmic side. Currents were recorded at a Vm of +50 mV. The higher current level of the Ykc1-101 trace results from higher expression levels. Bath (cytoplasmic) solutions contained 140 mM chloride salt of the stated cation and 5 mM EGTA. Pipette (external) solutions contained 14 mM KCl, 126 mM NaCl. Both solutions also contained 1 mM CaCl2, 1 mM MgCl2 and 5 mM HEPES, pH 7.2. Dashed lines represent zero current levels. Leakage was not subtracted. Download figure Download PowerPoint PP mutations decrease residence in and alter activation from the C1 states Fitting the delayed activation from the C1 states (see Introduction) of the wild-type required three rate constants of 14.9 ± 2.9, 71.8 ± 9.0 and 275 ± 13 ms when depolarized from −80 to +80 mV at 23°C in 100 mM K+ (n = 6 separate oocytes) (Figure 6A, individual rate components illustrated by dotted traces at the bottom of the YKC1+ plot). Thus, C1 is actually a set of closed states. Entry into the C1 states is favored by negative voltage (Figure 6B). Comparing the distribution of channels in the C1 states in 100 mM K+ (Figure 6B, ●) with that in 20 mM K+ (Figure 6D, ○), it can be seen that high external K+ also favors residence in C1. The 31 mV negative shift of the Boltzmann distribution between C1 and C2 caused by decreasing external K+ from 100 to 20 mM is close to the −38 mV shift such a lowering of K+ would be predicted to have on EK, suggesting that distribution between C1 and C2 is determined, at least in part, by ΔμK+ (a further examination of the wild-type C1–C2 distribution will be presented in Loukin et al., in preparation). Figure 6.PP mutations severely affect residence in and activation from the C1 closed states. (A) Current activations upon stepping to +80 mV from holding potentials between −100 and +20 mV at 10 mV intervals were recorded from whole oocytes expressing wild-type (●), Ykc1-101 (▪), Ykc1-301 (▴) and Ykc1-402 (▾) in 100 mM bath K+. Three exponential components were required to adequately fit the current activation from −80 to +80 mV of the Ykc1+ current, whereas only two such components were required for Ykc1-101 current activation (dotted lines show extrapolated contributions from individual activations to scale). Thus delayed activation of Ykc1 occurs from three distinct C1 states; the single V456I mutation of Ykc1-101 alters all three. (B) The fraction of steady-state current which was activated with delayed kinetics is plotted as a function of holding potential. Lines are Boltzmann fits of the raw data (symbols). V1/2 (curve midpoint), k (slope factor) and A (maximum amplitude) of the Boltzmann fits were −19.8 mV, 21.5% and 102.3% for wild-type, −79.7 mV, 18.8% and 54.7% for YKC1-101, −95.3 mV, 20.1% and 36.3% for YKC1-301 and −72.7 mV, 20.1% and 52.8% for YKC1-402. Thus, a negative extrapolation of these fits indicates that while wild-type channels reside exclusively in C1 at sufficiently negative voltages, the maximal extent to which Ykc1-101 and Ykc1-402 dwelt in C1 was 50 mV negative shift in the holding potential at which Ykc1 starts to significantly dwell in C1 (Figure 6 B and D). Whereas wild-type Ykc1 exclusively resides in C1 at −100 mV in 100 mM K+ (Figure 6B, ●), extrapolation of the Boltzmann fits of C1 residence versus holding potential showed that <55% of Ykc1-101 and Ykc-402 and <32% of Ykc1-301 channels maximally reside in the delayed C1 states even at highly negative voltages (Figure 6B). In 20 mM K+ <32% of Ykc1-101 and Ykc-402 and <10% of Ykc1-301 channels maximally dwell in C1. Thus at mildly negative potentials, where a significant fraction of wild-type channels are in C1 (e.g. 0 to −50 mV in 100 mM K+), almost none of the mutant channels are. At highly negative potentials, the likely yeast resting potential, wild-type Ykc1 channels will exclusively reside in the delayed C1 states, whereas at most 55% of the mutant channels will, with the remainder residing in rapidly activating C2. The remnant of the mutant Ykc1-101 delayed current activation can be fitted to two exponentials of 33.8 ± 11.4 and 111 ± 20 ms when depolarized from −80 to +80 mV in 100 mM K+ at 23°C (n = 5 separate oocytes, illustrated as dotted lines in Figure 6A, YKC1-101). Thus, all three wild-type delayed C1 states activations were altered by the single V456I mutation, indicating the structural relatedness of the individual C1 states. That the rapid C2→O transition was unaffected by this same mutation (Figure 4) in contrast demonstrates the physical distinction between C2 and C1. The mutants lack of delayed activation from the C1 states is clearly seen at the single channel level as well (Figure 7). Latency analysis on single channel patches demonstrated a lack residence in the delayed C1 states in the mutant. When patches containing single wild-type Ykc1 channels were depolarized from −80 mV, there was a characteristic lag of tens to hundreds of milliseconds before the channel opened (Figure 7, *). This same degree of latency was not seen when the patch was stepped from +40 mV (Figure 7, top right), demonstrating that this latency was due to a transition from the C1 states. When stepped from −80 mV, Ykc1-101 channels did not show such a latency (Figure 7, bottom). Thus, activation from the delayed C1 states in the wild-type, as well as its absence in the PP mutants, can be observed at the single channel level. Figure 7.Single channel analysis of emergence from the C1 states. Latency analysis was conducted on Ykc1+ (upper traces) and Ykc1-101 (lower traces). Excised inside-out single channel patches were held at either −80 (left) or +40 mV (right) for 10 s before stepping to the test potential of +50 mV. The latency to the first opening event at +50 mV was plotted on logarithmic event histograms below three representative traces shown for each condition. These histograms are each a compilation of at least 50 such measurements from patches from at least three separate oocytes. Bath (cytoplasmic) solutions contained 140 mM KCl and 5 mM EGTA. Pipette (external) solutions contained 14 mM KCl, 126 mM NaCl. Both solutions also contained 1 mM CaCl2, 1 mM MgCl2 and 5 mM HEPES, pH 7.2. Calibration bar 10 pA×250 ms. Download figure Download PowerPoint Both wild-type and Ykc1-101 single channels rapidly flicker, which may be due to rapid C2↔O transitions at positive potentials. Visual inspection discerned little difference in this flicker between wild-type and mutant channels. It is also notable that the mutant channels closed for up to 200 ms at positive potentials, possibly reflecting an inactivating state of Ykc1 unrelated to the C1 states. Attempts were made to quantify residence in these closed states, but were thwarted by channel run-down over the course of the extended depolarization of excised patches. PP mutations slow entry into the C1 states In addition to activation, entry into the delayed C1 states was also examined. Twenty percent of the outward current resulted from activation from the delayed C1 states when YKC1+-expressing oocytes were stepped from 0 to +80 mV (100 mM bath K+) (Figure 8A, uppermost trace of YKC1+ traces; Figure 8B, ●, t = 0). Inward currents were not observed during a brief intervening step from 0 to −100 mV, yet still only 20% of the outward current was activated with delayed kinetics upon depolarization if the dwell at −100 mV was brief. Thus Ykc1 first enters C2, not the delayed C1 states, initially upon polarization to −100 mV. If the length of the −100 mV conditioning pulse was increased, an increasing fraction of the current was activated from the C1 states (Figure 8A, YKC1+, lower traces). When the fraction of the eventual steady-state current which was activated from the C1 states was plotted against the dwell time at the −100mV conditioning pulse, it can be seen that all wild-type Ykc1 channels eventually migrated from the C2 to the C1 states at a combined rate of 280 and 2700 ms at −100 mV (Figure 8B, ●). Figure 8.PP mutation decreases C2→C1 transition rate. (A) Oocytes expressing either YKC1+ (●) or YKC1-101 (▪) RNA were repeatedly stepped to +80 mV in 100 mM bath K+ from a holding potential of 0 mV with increasing intervening conditioning pulses at −100 mV. Traces show currents, from top, activating from conditioning pulses lasting 0, 10, 20, 50, 100, 200 and 500 ms and 1, 2, 5, 10 and 20 s. (B) The fraction of current activating with delayed kinetics, calculated as in Figure 6B, is plotted as a function of dwell time at the −100 mV conditioning pulse. Wild-type channels migrated from the C2 to the slowly activating C1 states at an initial rate at least eight times faster than Ykc1-101 channels (inset to B). Recording conditions and calculations were as described in Figure 6. Calibration bars 10 μA×200 ms. Download figure Download PowerPoint A similar voltage protocol was applied to oocytes expressing Ykc1-101 channels (Figure 8A, YKC1-101). Unlike the wild-type, only a minority of the mutant channels dwelt in the delayed C1 states at steady-state at −100 mV. Because of this, the initial rate of C2→C1 transition was determined, so that any reverse C1→C2 transition of the mutant could be ignored. The inset of Figure 8B shows that the C2→C1 transition of the mutant channel was initially eight times slower than that of the wild-type. Thus, the Ykc1-101 PP mutation dramatically slows the transition from C2 to the delayed C1 states upon hyperpolarization. Discussion The findings presented in this report have several-fold significance. We have demonstrated that classical microbial selection can be fruitfully applied to structure–function studies of ion channels. Using microbial selection we have isolated Ykc1 mutants of functional consequence via their ability
Referência(s)