|
Figure 1. Structure of the anion permeation pathway of CLC proteins. (A) Top left panel. Dimeric structure of bClC-K (PDB entry: 5TQQ) viewed from the membrane plane (extracellular side above and cytoplasmic side below). The two subunits are shown in blue and green. Important residues and the ones analyzed in this work are represented as colored sticks in the blue subunit: S121 in green, V166 in red, Y425 in cyan, F519 in pink, Y520 in orange, N257 in dark red. The color code is the same in all of the panels. The PDB entry for bClC-K did not include bound anions although an electron density was observed at Scen but not Sint. The position of Scen is conserved in bClC-K and EcClC-1.23 Here, we superimposed the structure of bClC-K with the positions of the anions bound at Scen and Sint as found in the structure of EcClC-1 (PDB entry: 1OTS) represented as magenta spheres with the purpose of illustrating the residues involved in anion coordination at Scen in b-ClC-K. The other panels show an expanded representation of the anion permeation pathway of bClC-K, EcCLC-1, and CmCLC (PDB entry: 3ORG). In CmClC, Scen is occupied by the side chain of the “gating” glutamate, whereas anions are bound at Sext and Sint. In all of the panels some transmembrane helices were removed for clarity. (B) Sequence alignment of the residues investigated in this work including their position in the secondary structure of bClC-K. They are S121, Y425, F519, and Y520 and are indicated in bold. The sequence alignment is on the basis of the one of Feng et al.22 indicating a gap in α-helix R next to F519.
|
|
Figure 2. Representative patch-clamp recordings of wild-type ClC-Ka and the mutants S121G, S121T, and S121P. The boxes show representative current traces recorded in extracellular solutions containing chloride (top traces) or nitrate (bottom traces), and I-V curves obtained from average steady-state currents containing chloride (black circles), nitrate (open circles), bromide (black up triangles), and thiocyanate (black down triangles) for wild-type ClC-Ka (n=10), S121G (n=6), S121P (n=6), and S121T (n=6). Currents were normalized to those measured in chloride at +80 mV. Lines in the graphs are polynomial fits to the points. The voltage protocol consisted of 5 milliseconds at the holding potential followed by voltage steps from 2120 to +80 mV with 20-mV increments (100 milliseconds duration) followed by a +40-mV pulse of 20 milliseconds and return to the holding potential for 100 milliseconds. The holding potential was the resting membrane potential. Errors are indicated as SEM. For most of the data points the error is smaller than the symbol. For the analysis we only used cells in which the full sequence of anions could be tested.
|
|
Figure 3. Examples of nonstationary noise analysis of wild-type ClC-Ka, S121G, and S121T. The voltage protocol consisted of 100–200 identical pulses composed of 20 milliseconds at the holding potential followed by a step to +80 mV and a step to 260 mV (each of 200 milliseconds) and return to the holding potential for 200 milliseconds. The holding potential was the resting membrane potential. Data analysis was performed at +80 mV. Mean current (upper trace) and variance (lower trace) are shown on the left. On the right, the variance (symbols) is plotted versus the mean current and fitted with a parabola (line) as described in the Methods section to obtain a single channel conductance of 19.662.2 pS (n=14) for wild type, 8.561.1 pS for S121G (n=5), and 3.460.5 pS for S121T (n=10). WT, wild type.
|
|
Figure 4. Representative patch-clamp recordings of Y425A. (A) Representative current traces recorded from the same cell in extracellular solutions with the anions indicated. The voltage protocol is the same as in Figure 2. (B) I-V curves were obtained from average steady-state currents in extracellular solutions containing chloride (black circles, n=5), nitrate (open circles, n=5), bromide (black up triangles, n=4), and thiocyanate (black down triangle, n=5). Values were normalized to those measured in chloride at +80 mV. Lines are polynomial fits to the points. Errors are indicated as SEM. For most of the data points the error is smaller than the symbol. (C) Mean current (upper trace) and the variance (lower trace) are shown on the left. The voltage protocol is the same as in Figure 3. On the right the variance (symbols) is plotted versus the mean current and fitted with a parabola (line) as described in the Methods section to obtain a single channel conductance of 17.464.8 pS (n=5). The values are not significantly different from wild type (P.0.1).
|
|
Figure 5. Nitrate selectivity for wild-type ClC-0 and the mutant A417Y. The boxes
display representative TEVC recordings in extracellular solution containing chloride (top left) or nitrate (bottom left) and average I-V curves (right) for wild-type ClC0 (n=11) and A417Y (n=7). I-V curves were obtained from average steady-state currents in extracellular solutions containing chloride (black circles) and nitrate (open circles) with values normalized to those measured in chloride at +80 mV. Lines in the graphs are polynomial fits to the points. The voltage protocol consisted of 20 milliseconds at the holding potential followed by a 50-millisecond prepulse at +60 mV and by 100-millisecond voltage steps from 2160 to +80 mV with 20 mV and a tail pulse to 2100 mV for 100 milliseconds. This was followed by a return to the holding potential for 200 milliseconds. The holding potential was the membrane resting potential. Data were obtained for at least three batches of oocytes. Errors are indicated as SEM.
|
|
Figure 6. Anion selectivity for wild-type ClC-Ka and the mutants Y520A, F519A, and Y425A. The boxes display representative TEVC recordings in extracellular solution containing chloride (top left) or nitrate (bottom left) and I-V curves (right) for wild-type CLC-Ka (n=15), Y520A (n=7), F519A (n=10), and Y425A (n=10). The voltage protocol consisted of 50 milliseconds at the holding potential followed by a 100-millisecond prepulse at +60 mV and by 200-millisecond steps to various test values (from 2140 to +80 mV with 20-mV increments) followed by a return to the holding potential (200 milliseconds). The holding potential was the membrane resting potential. I-V curves were obtained from average steady-state currents in extracellular solutions containing chloride, nitrate, bromide, and thiocyanate (with the symbols indicated) with values normalized to those measured in chloride at +80 mV. Lines in the graphs are polynomial fits to the points. Data were obtained for at least three batches of oocytes (n.7). Errors are indicated as SEM. For most of the data points the error is smaller than the symbol. For the analysis we only used cells in which the full sequence of anions could be tested.
|
|
Supplementary figure 1. Representative currents recorded from the same cell in patch-clamp experiments on wild-type CLC-Ka, and the mutants S121T, S121P and S121G in extracellular solutions containing the indicated anion. Currents were normalized to those measured in chloride at + 80 mV. Lines in the graphs are polynomial fits to the points. The voltage protocol consisted of 5 ms at the holding potential followed by voltage steps from -120 mV to +80 mV with 20 mV increments (100 ms duration) followed by a +40 mV pulse of 20 ms and return to
the holding potential for 100 ms. The holding potential was the resting membrane potential. The holding potential is the membrane resting potential.
|
|
Supplementary figure 2. Nitrate selectivity for wild-type ClC-0 and the mutants S123A and S123G. The boxes display representative TEVC recordings in extracellular solution containing chloride (top left) and nitrate (bottom left) and average I-V curves (right) for wild-type ClC-0 (n=8), S123A (n=10) and S123G (n=10). The voltage protocol consisted of 50 ms at the holding potential followed by a 50 ms prepulse at +60 mV and by 100 ms voltage steps from -160 mV to +80 mV with 20 mV and a tail pulse to -100 mV for 100 ms and return to the holding potential for 50 ms. The holding potential was the membrane resting potential. I-V curves were obtained from average steady-state currents in extracellular solutions containing chloride (black circles) and nitrate (open circles) with values normalized to those measured in chloride at + 80 mV. Values for PNO3-/PCl- were 0.58 ± 0.02 for wild-type, 0.72 ± 0.04 for S123A and 0.87 ± 0.01 for S123G, respectively. Data were obtained for at least 3 batches of oocytes. Errors are indicated as SEM. For most of the data points the error is smaller than the symbol.
|
|
Supplementary Figure 3. Anion selectivity for the ClC-Ka mutant N257A. A) Representative current traces recorded by TEVC from the same oocyte in extracellular solutions with the anions indicated (left) and average of I-V curves (right). The voltage protocol consisted of 50 ms at the holding potential followed by a 100 ms prepulse at +60 mV and by 200 ms steps to various test values (from -140 to + 80 mV with 20 mV increments) followed by a return to the holding potential (100 ms). The holding potential was the membrane resting potential. B) I-V curves were obtained from average steady-state currents in extracellular solutions containing chloride (black circles) (n=15), nitrate (open circles) (n=15), bromide (black triangles) (n=8), iodide (black diamonds) (n=8). Values were normalized to those measured in chloride at + 80 mV. Data were obtained for at least 3 batches of oocytes. Errors are indicated as SEM. For most of the data points the error is smaller than the symbol. For the analysis we only used cells
in which the full sequence of anions could be tested.
|
|
Supplementary Figure 4. Representative currents recorded from the same oocytes in TEVC experiments on wild-type CLC-Ka, and the mutants Y520A, F519A, Y425A in extracellular solutions containing the indicated anion. The voltage protocol is the same as in Supplementary Figure 3 A.
|