A critical residue for isoform difference in tetrodotoxin affinity is a molecular determinant of the external access path for local anesthetics in the cardiac sodium channel

Abstract

Membrane-impermeant quaternary derivatives of lidocaine (QX222 and QX314) block cardiac Na(+) channels when applied from either side of the membrane, but they block neuronal and skeletal muscle channels poorly from the outside. To find the molecular determinants of the cardiac external QX access path, mutations of adult rat skeletal muscle (micro1) and rat heart (rH1) Na(+) channels were studied by two-electrode voltage clamp in Xenopus oocytes. Mutating the micro1 domain I P-loop Y401, which is the critical residue for isoform differences in tetrodotoxin block, to the heart sequence (Y401C) allowed outside QX222 block, but its mutation to brain type (Y401F) showed little block. mu1-Y401C accelerated recovery from block by internal QX222. Block by external QX222 in mu1-Y401C was diminished by chemical modification with methanethiosulfonate ethylammonium (MTSEA) to the outer vestibule or by a double mutant (mu1-Y401C/F1579A), which altered the putative local anesthetic binding site. The reverse mutation in heart rH1-C374Y reduced outside QX314 block and slowed dissociation of internal QX222. Mutation of mu1-C1572 in IVS6 to Thr, the cardiac isoform residue (C1572T), allowed external QX222 block, and accelerated recovery from internal QX222 block, as reported. Blocking efficacy of outside QX222 in mu1-Y401C was more than that in mu1-C1572T, and the double mutant (mu1-Y401C/C1572T) accelerated internal QX recovery more than mu1-Y401C or mu1-C1572T alone. We conclude that the isoform-specific residue (Tyr/Phe/Cys) in the P-loop of domain I plays an important role in drug access as well as in tetrodotoxin binding. Isoform-specific residues in the IP-loop and IVS6 determine outside drug access to an internal binding site.

Figure 1

Sequence alignment of pore-forming regions (P-loops) of four domains of different Na⁺ channel isoforms. Amino acid sequences are from μ1, rat brain IIA, and rH1. Nine residues are assigned to P-loop from each domain and numbers indicate the amino acid position of μ1. Identical amino acids are indicated with dashes.

Figure 2

Mutation of the isoform-specific residue of μ1 IP-loop to heart sequence (μ1-Y401C) allows block by external QX222. (A and B) Typical current traces before (con) and 12 min after external application of 500 μM QX222 recorded from (A) μ1-WT and (B) μ1-Y401C. (C) Effects of externally applied 500 μM QX222 on μ1-WT (○) and μ1-Y401C (●). The bar indicates the period of exposure to 500 μM QX222 in the bath solution. Data represent the means ± SEM from eight oocytes for μ1-WT and nine for μ1-Y401C. Currents were elicited by 35-ms pulses to −10 mV from a holding potential of −100 mV at 20-s intervals (A–C), and peak currents were normalized to peak current in control and plotted as relative Na⁺ currents (relative I_Na) (C).

Figure 3

μ1-Y401C creates an access pathway for external QX222 to reach the internal binding site. (A) Effects of MTSEA modification in the outer vestibule on block by external QX222 in μ1-Y401C. MTSEA modification was done by bath application of 2.5 mM MTSEA for 3–5 min and verified by irreversibility of current reduction after washout for at least 5 min before external application of QX222. (B) Effects of mutation of the putative local anesthetic binding site (μ1-F1579) on block by externally applied QX222 in μ1-Y401C. For this, actions of externally applied QX222 were studied for the double mutant, μ1-Y401C/F1579A. In A and B, the protocol was the same as that in Fig. 2, and peak currents were normalized to that in the control. Data represent the means ± SEM from four oocytes for MTSEA-bound μ1-Y401C (■) and five for μ1-Y401C/F1579A (▴). In each panel, the mean values (●) of block for μ1-Y401C shown in Fig. 2C were represented to compare the block development. The bar indicates the period during exposure to 500 μM QX222 in the bath solution. (C) Effects of μ1-Y401C on recovery from internally applied QX222 block. QX222 was internally applied by microinjecting 50 nl of a 3 mM QX222 solution into oocytes. Twenty to thirty minutes after microinjection, a 1-Hz train of 20 pulses with 35-ms duration was applied to −10 mV from a holding potential of −100 mV to produce use-dependent block by QX222. Then, recovery from the QX222 block was monitored at the indicated intervals after the end of a train for μ1-WT (○) and μ1-Y401C (●). Peak currents during recovery were normalized by the difference between the peak current during the first pulse of a train and the 10-s recovery test pulse, and plotted as recovery %. The 10-s recovery was assumed to allow recovery from inactivation of unblocked channels but be short enough to prevent QX222 dissociation from blocked channels. The smooth lines are single exponential fits, and data represent the means ± SEM from six oocytes for μ1-WT and five to seven for μ1-Y401C.

Figure 4

Effects of substitution of μ1-Y401 with other amino acids on external QX222 block. Actions of externally applied 500 μM QX222 were studied for μ1-Y401F (▪), μ1-Y401A (▴), and μ1-Y401D (▾). For comparison, the mean values of block for μ1-WT (○) and μ1-Y401C (•) shown in Fig. 2C were represented. The bar indicates the period of exposure to 500 μM QX222 in the bath solution. The protocol was the same as that in Fig. 2, and peak currents were normalized to that in control. Data represent the means ± SEM from four oocytes for μ1-Y401F, seven for μ1-Y401A, and five for μ1-Y401D.

Figure 5

Reverse mutation of rH1 to μ1 sequence (rH1-C374Y) inhibits block by external QX and slows down dissociation of internally applied QX. (A) Effects of externally applied QX314 on rH1-WT (○) and rH1-C374Y (●). The bar indicates the period during exposure to 500 μM QX314 in the bath solution. Currents were elicited by 35-ms pulses to −20 mV from a holding potential of −110 mV at 5-s intervals, and peak currents were normalized to peak current in control. Data represent the means ± SEM from seven oocytes for rH1-WT and five for rH1-C374Y. (B) Effects of rH1-C374Y on recovery from internally applied QX222 block. QX222 was internally applied by microinjecting 50 nl of a 3 mM QX222 solution into oocytes. Twenty to thirty min after microinjection, a 1-Hz train of 20 pulses with 35-ms duration was applied to −20 mV from a holding potential of −110 mV to produce use-dependent block by QX222. Then, recovery from QX222 block was monitored at the indicated intervals after the end of a train for rH1-WT (○) and rH1-C374Y (●). Peak currents during recovery were normalized by the difference between the peak current during the 1st and 20th pulses of a train, and plotted as recovery %. The smooth lines are single exponential fits, and data represent the means ± SEM from four to eight oocytes for rH1-WT and three to four for rH1-C374Y.

Figure 6

Isoform-specific residues of IP-loop and IVS6 determine external QX access. (A) Effects of IVS6 mutant and double mutant of IP-loop and IVS6 on block by externally applied QX222. Actions of externally applied 500 μM QX222 were studied for μ1-C1572T (▪) and μ1-Y401C/C1572T (▴). The mean values of block for μ1-WT (○) and μ1-Y401C (•) are from Fig. 2C. The bar indicates the period of exposure to 500 μM QX222 in the bath solution. The protocol was the same as that in Fig. 2, and peak currents were normalized to that in control. Data represent the means ± SEM from six oocytes for μ1-C1572T and five for μ1-Y401C/C1572T. (B) Effects of μ1-C1572T and μ1-Y401C/C1572T on recovery from internally applied QX222 block. Recovery from QX222 block was monitored at the indicated intervals after use-dependent block by 1-Hz stimulation for μ1-C1572T (■) and μ1-Y401C/C1572T (▴). Recovery data for μ1-WT (○) and μ1-Y401C (●) are from Fig. 3C. The smooth lines are single exponential fits, and data represent the means ± SEM from four to five oocytes for μ1-C1572T and three for μ1-Y401C/C1572T. For others, see Fig. 3C.

Figure 7

(A) A model for isoform differences of QX permeation in the Na⁺ channels. In the heart channel, QX can access to the internal binding site from the outside, and escape to the outside from the cytoplasmic side (16, 21). In skeletal muscle and brain channels, access and escape of QX are limited because of their aromatic residues on the IP-loop (present study), and Cys and Val on IVS6 (16, the present study). (B) Hypothetical pore structure of the Na⁺ channel composed of four P-loops and IVS6 with respect to access and binding of local anesthetics. Residues shown by open circles (D400, Y401, E755, A1529, and C1572; numbering of μ1) represent the amino acids affecting QX access from the outside (refs. and , and the present study) and closed circles (K1237, F1579, and Y1586) affecting local anesthetic binding (15, 17, 27). Ile in square (I1575) appears to affect both access and binding of local anesthetics (15). The DEKA locus (Asp-Glu-Lys-Ala) forms the selectivity ring, shown here as an oval (23).