The EEEE locus is the sole high-affinity Ca(2+) binding structure in the pore of a voltage-gated Ca(2+) channel: block by ca(2+) entering from the intracellular pore entrance

Abstract

Selective permeability in voltage-gated Ca(2+) channels is dependent upon a quartet of pore-localized glutamate residues (EEEE locus). The EEEE locus is widely believed to comprise the sole high-affinity Ca(2+) binding site in the pore, which represents an overturning of earlier models that had postulated two high-affinity Ca(2+) binding sites. The current view is based on site-directed mutagenesis work in which Ca(2+) binding affinity was attenuated by single and double substitutions in the EEEE locus, and eliminated by quadruple alanine (AAAA), glutamine (QQQQ), or aspartate (DDDD) substitutions. However, interpretation of the mutagenesis work can be criticized on the grounds that EEEE locus mutations may have additionally disrupted the integrity of a second, non-EEEE locus high-affinity site, and that such a second site may have remained undetected because the mutated pore was probed only from the extracellular pore entrance. Here, we describe the results of experiments designed to test the strength of these criticisms of the single high-affinity locus model of selective permeability in Ca(2+) channels. First, substituted-cysteine accessibility experiments indicate that pore structure in the vicinity of the EEEE locus is not extensively disrupted as a consequence of the quadruple AAAA mutations, suggesting in turn that the quadruple mutations do not distort pore structure to such an extent that a second high affinity site would likely be destroyed. Second, the postulated second high-affinity site was not detected by probing from the intracellularly oriented pore entrance of AAAA and QQQQ mutants. Using inside-out patches, we found that, whereas micromolar Ca(2+) produced substantial block of outward Li(+) current in wild-type channels, internal Ca(2+) concentrations up to 1 mM did not produce detectable block of outward Li(+) current in the AAAA or QQQQ mutants. These results indicate that the EEEE locus is indeed the sole high-affinity Ca(2+) binding locus in the pore of voltage-gated Ca(2+) channels.

Figure 1

Determination of side-chain orientation of pore-lining amino acids in the AAAA channel. (A) Superimposed two-electrode voltage clamp records illustrating the effects of MTSEA upon whole-cell Ca²⁺ channel currents. Currents (100 mM Li⁺) were elicited by step depolarization (150 ms) from the holding potential of −80 to −20 mV every 5 s. In each recording, after a stable baseline was established, MTSEA·Br (final concentration of MTSEA: 2 mM) was dissolved in the 100 mM Li⁺ solution and applied immediately to the oocyte via bath perfusion. (B) Time course of MTSEA action on peak inward Li⁺ currents.

Figure 2

Summary of MTSEA block of substituted-cysteine mutants. (A) Percent block (±SEM) for the quadruple alanine mutant (AAAA) and cysteine substitutions in the AAAA channel. Block was calculated as the average for n = 4–9 oocytes, except for F1144C/AAAA, for which n = 3. Cysteine-substituted mutants differed from the AAAA channel in that block of the mutants was not reversible with wash, and their block was larger (P < 0.00005). Current was carried by Li⁺, and data were collected as described for Fig. 1. (B) Percent block (±SEM) for the EEEE channel and cysteine substitutions in the EEEE channel. Block was calculated as the average for n = 4–6 oocytes. Block of the cysteine-substituted mutant versions of the EEEE channel was not reversible with wash, unlike the rapidly reversible block of the EEEE channel, and block of the mutants was much larger than for the EEEE channel (P < 0.00002). Current was carried by Ba²⁺ (40 mM Ba²⁺ solution), holding potential was −80 mV, test potential was +20 mV, and 150-ms step depolarizations were applied every 15 s. Open bars indicate reversible block (AAAA and EEEE channels only), filled bars indicate −1 position cysteine substitution mutants, and hatched bars indicate 0-position cysteine substitutions. For reference, EEEE locus glutamates are located at positions 393, 736, 1145, and 1446 in motifs I–IV, respectively.

Figure 3

Absence of divalent cation block or flux in the AAAA mutant Ca²⁺ channel. (A) High-affinity block of single-channel inward Li⁺ currents by external Ca²⁺ was abolished by replacing all four EEEE locus residues with alanine. Inward Li⁺ currents were studied in cell-attached patches on oocytes expressing either WT or AAAA channels. The patch pipet contained 100 mM Li⁺ plus various [Ca²⁺]. Control solution contained 3 nM Ca²⁺. Holding potential was −100 mV, channels were activated by 25–50-ms prepulses to +40 mV, and the illustrated single-channel currents were recorded at −40 mV. (B) Li⁺, but not Ba²⁺, permeates AAAA channels. Example current–voltage relationships and current records from two-electrode voltage-clamp recordings from individual oocytes expressing either WT or AAAA are illustrated. The bath solution contained either 100 mM Li⁺ (•) or 40 mM Ba²⁺ (□). Currents were recorded during step depolarizations from a holding potential of −100 mV (Li⁺) or −80 mV (Ba²⁺) every 15 s. Reversal potentials in Li⁺ were 3.7 ± 1.2 mV (n = 6) for WT and −8.0 ± 1.7 (n = 6) for AAAA. Smooth curves drawn through the data were best-fits calculated according to: I = (1/{1 + exp[(V_0.5 − V_M)/b]}) · ([A ₁ · exp(zFδV_M/RT)] − {A ₂ · exp[−2zF(1 − δ)V _M/RT]}), where I = membrane current, V_M = membrane potential, V_0.5 = half-activation voltage, b = Boltzmann slope factor, A ₁ and A ₂ are amplitude factors related to the concentrations and permeabilities of the internal and external ions, z = ion valence, δ = electrical distance, and F, R, and T are Faraday's constant, ideal gas law constant, and temperature (K).

Figure 4

Whole-cell current densities in HEK293 cells. (A) Examples of current–voltage relationships from individual cells expressing WT (○), AAAA (▴), or GFP (control, □). The bath solution contained 100 mM Li⁺ and the pipet contained 135 mM Cs⁺. Currents were elicited every 5 s by step depolarizations from a holding potential of −100 mV. (B) Average current densities in WT-, AAAA-, and GFP-transfected (control) HEK293 cells. Inward current density was measured at the inward peak of the I-V, and outward current density was measured at +80 mV. Data plotted as mean ± SEM (n = 10–13).

Figure 5

Concentration and voltage dependence of Ca²⁺ block of outward Li⁺ current carried through WT α_1C channels in inside-out patches excised from HEK293 cells. (A) Internal Ca²⁺ blocked outward Li⁺ currents through single WT channels with high affinity. The bath solution contained 300 mM Li⁺ plus various [Ca²⁺], and the patch pipet contained 55 mM Li⁺. Control Li⁺ solution contained 3 nM Ca²⁺. Single-channel currents were recorded during a test depolarization to +20 mV from a holding potential of −100 mV. In some cases, a 25- or 50-ms prepulse to +100 mV was given to facilitate channel activation. Ca²⁺ on (○) and off (▪) rates are plotted as mean ± SEM versus [Ca²⁺] for n = 3–5. The linear regression fit through the on-rate data points has a slope of 2.3 × 10⁸ M⁻¹ s⁻¹. The horizontal line fit to the off-rate data indicates the average off rate was 2,965 s⁻¹. (B) Voltage dependence of internal Ca²⁺ block of outward Li⁺ currents through single WT channels. The bath solution contained 300 mM Li⁺ and 3 μM Ca²⁺, while the pipet solution contained 55 mM Li⁺. Single-channel currents were recorded during a step depolarization from the holding potential of −100 mV. In some experiments, a 25- or 50-ms prepulse to +100 mV was given to facilitate channel activation. Ca²⁺ on (○) and off (▪) rates plotted as mean ± SEM versus test potential (n = 4–5). The horizontal line fit to the on-rate data indicates the average on rate was 846 s⁻¹. The exponential curve fit to the off-rate data indicates that off rate increased e-fold per 19 mV.

Figure 6

Internal Ca²⁺ does not block outward single-channel Li⁺ currents through AAAA channels. Currents were studied in inside-out patches excised from HEK293 cells expressing AAAA channels (WT records shown for comparison). The bath solution contained 300 mM Li⁺ plus various [Ca²⁺], and the patch pipet contained 55 mM Li⁺. [Ca²⁺] was 3 nM in the control solution. Single-channel currents were recorded during a test depolarization to +20 mV, following a 25- or 50-ms prepulse to +100 mV. Holding potential was −100 mV.

Figure 7

Outward Li⁺ currents through QQQQ channels are not blocked by internal Ca²⁺. Currents were studied in inside-out patches excised from HEK293 cells expressing QQQQ channels. The bath contained 300 mM Li⁺ plus either 3 nM Ca²⁺ (control) or 1 mM Ca²⁺, and the patch pipet contained 55 mM Li⁺. Single-channel currents were recorded during a test depolarization to +20 mV, following a 25- or 50-ms prepulse to +100 mV. Holding potential was −100 mV.