The dependence of Ag+ block of a potassium channel, murine kir2.1, on a cysteine residue in the selectivity filter

Abstract

Externally applied Ag+ (100-200 nM) irreversibly blocked the strong inwardly rectifying K+ channel, Kir2.1. Mutation to serine of a cysteine residue at position 149 in the pore-forming H5 region of Kir2.1 abolished Ag+ blockage. To determine how many of the binding sites must be occupied by Ag+ before the channel is blocked, we measured the rate of channel block and found that our results were best fitted assuming that only one Ag+ ion need bind to eliminate channel current. We tested our hypothesis further by constructing covalently linked dimers and tetramers of Kir2.1 in which cysteine had been replaced by serine in one (dimer) or three (tetramer) of the linked subunits. When expressed, these constructs yielded functional channels with either two (dimer) or one (tetramer) cysteines per channel at position 149. Blockage in the tetramer was complete after sufficient exposure to 200 nM Ag+, a result that is also consistent with only one Ag+ being required to bind to Cys149 to block fully. The rate of development of blockage was 16 times slower than in wild-type channels; the rate was 4 times slower in channels formed from dimers.

Figure 1. Structure of concatameric channels

A, tetramers are formed by linking four coding regions, leaving the native BsmBI and Afl II sites in the subunit at the 3′ end of the coding region. The positions of the EcoRI and Xho I sites used to manipulate the construct are also shown as are the positions of the start (s) and stop (▪) codons. The start codon is preceded by a Kozak initiation sequence. Subunits are linked with 10 CAG triplets, each encoding a Gln residue. B, the structure of the dimer, with its restriction sites and start and stop codons.

Figure 2. Silver blocks wild-type Kir2.1 currents

A, membrane currents recorded from a single CHO cell expressing the gene for Kir2.1 in response to voltage steps from a holding potential of -17 mV to test potentials ranging from +63 to -107 mV, in 10 mV increments and in the absence (left) and presence (right) of 100 nm externally applied Ag⁺. Extracellular [K⁺] was 70 mm, intracellular [K⁺] was 140 mm. Exposure time to Ag⁺ was 90 s. In 10 similar experiments, the current was reduced to 0.32 ± 0.07 of control levels after 90 s exposure to 100 nm Ag⁺. B, current- voltage relation of the cell shown in A in the absence (•) and presence (○) of 100 nm externally applied Ag⁺.

Figure 3. Removal of cysteine at position 149 in the H5 region renders channels insensitive to Ag⁺

A, amino acid sequence of the H5 region of Kir2.1 from residue 138 to 152. Underlined is the Gly-Tyr-Gly (GYG) motif believed to form part of the selectivity filter for K⁺. The cysteine residue at position 149 is shown in bold. B, membrane currents recorded from a single CHO cell expressing the gene for a Kir2.1 channel in which the cysteine residue at position 149 has been mutated to serine (C149S). The recording conditions were identical to those in Fig. 2. Exposure time to 100 nm Ag⁺ was 90 s. In 6 similar experiments, externally applied Ag⁺ had no effect upon the whole-cell C149S currents. C, current-voltage relation for the cell shown in B in the absence (•) and presence (○) of Ag⁺.

Figure 4. Time course of Ag⁺ block of wild-type Kir2.1

A, fractional current for wild-type Kir2.1 plotted against time for 100 nm (•) and 200 nm (○) Ag⁺. The data are fitted with a single exponential (continuous line). Time constant (200 nm) = 30 s; (100 nm) = 69 s, mean values given in Table 2. B, mean fractional current for 3 cells exposed to 100 nm external Ag⁺ plotted against time. Data are fitted assuming only 1 Ag⁺ has to bind to the channel to block; other lines give the results where 2, 3 or 4 Ag⁺ ions are required for blockage (eqns (2) and (3)). C, same data as B, with fractional current (ordinate) plotted on a log scale.

Figure 5. Ag⁺ blockage of monomers, dimers and tetramers

A, whole-cell currents from CHO cells expressing either wild-type Kir2.1 monomers (left), wild-type-C149S dimers (centre) or wild-type-C149S tetramers (right). Subunit formation is indicated in cartoons above the current records, wild-type subunits being indicated by shaded circles, C149S subunits by open circles. The membrane currents shown were recorded in response to voltage steps from a holding potential of -17 mV to test potentials of +32 and -97 mV in the presence and absence of 200 nm external Ag⁺. The times and arrows to the right of the current records indicate the length of exposure to external Ag⁺. B, fractional current for wild-type monomer (•), wild-type-C149S dimer (□) and wild-type-C149S tetramer (▴) plotted against time. The results are fitted with single exponentials (eqn (3), where k‘changes with the number of Cys residues present). Note that given sufficient time, currents through channels formed from the tetrameric construct (which has only 1 cysteine residue at position 149) are completely inhibited by external Ag⁺. In the absence of extracellular Ag⁺, whole-cell currents reduced by less than 5 % over a 10 min period. C, same results as B, plotted with fractional current (ordinate) on a log scale.