Membrane conductances of mouse cone photoreceptors

Abstract

Vertebrate photoreceptor cells respond to light through a closure of CNG channels located in the outer segment. Multiple voltage-sensitive channels in the photoreceptor inner segment serve to transform and transmit the light-induced outer-segment current response. Despite extensive studies in lower vertebrates, we do not know how these channels produce the photoresponse of mammalian photoreceptors. Here we examined these ionic conductances recorded from single mouse cones in unlabeled, dark-adapted retinal slices. First, we show measurements of the voltage dependence of the light response. After block of voltage-gated Ca2+ channels, the light-dependent current was nearly linear within the physiological range of voltages with constant chord conductance and a reversal potential similar to that previously determined in lower vertebrate photoreceptors. At a dark resting membrane potential of -45 mV, cones maintain a standing Ca2+ current (iCa) between 15 and 20 pA. We characterized the time and voltage dependence of iCa and a calcium-activated anion channel. After constitutive closure of the CNG channels by the nonhydrolysable analogue GTP-γ-S, we observed a light-dependent increase in iCa followed by a Ca2+-activated K+ current, both probably the result of feedback from horizontal cells. We also recorded the hyperpolarization-activated cyclic nucleotide-gated (HCN) conductance (ih) and measured its current-voltage relationship and reversal potential. With small hyperpolarizations, ih activated with a time constant of 25 ms; activation was speeded with larger hyperpolarizations. Finally, we characterized two voltage-gated K+-conductances (iK). Depolarizing steps beginning at -10 mV activated a transient, outwardly rectifying iK blocked by 4-AP and insensitive to TEA. A sustained iK isolated through subtraction was blocked by TEA but was insensitive to 4-AP. The sustained iK had a nearly linear voltage dependence throughout the physiological voltage range of the cone. Together these data constitute the first comprehensive study of the channel conductances of mouse photoreceptors.

Figure 1.

Reversal of mouse cone photocurrent. (A–D) Whole-cell patch recordings from Gnat1^−/− mice (A and B) and from Cx36^−/− mice (C and D). Holding potential was initially −50 mV and was changed to test potentials from −90 mV to +70 mV in 20-mV steps. Sample command step shown in A. At 2.5 s after the change in holding potential, the cones were stimulated with a saturating flash bleaching 1.2 × 10⁴ P* (arrows). The left column shows raw current traces. The right column shows photoresponses at each V_m overlaid after baseline subtraction. For C and D, 10 µM ISR and 25 mM TEA had been added to the bath solution. (E) Curves give peak amplitude of responses to the saturating flash as a function of holding potential, as follows: black circles, Gnat1^−/−, K-Asp internal, Ames’ bicarbonate external (n = 3); blue inverted triangles, Cx36^−/−, K-Asp internal, Ames’ bicarbonate external (n = 8); red triangles, Cx36^−/−, K-Asp internal, Ames’ bicarbonate external with addition of 25 mM TEA and 10 µM ISR (n = 8). Data points are means ± SEM. (F) Comparison of mouse cone current-voltage curve (red triangles) to previous results of current-voltage curves from salamander rods (fuschia diamonds) and bass (fish) cones (blue pentagons).

Figure 2.

Voltage-gated Ca²⁺ and Ca²⁺-activated Cl⁻ currents. Whole-cell patch recordings from Cx36^−/− cones. Recordings were made with a Cs⁺ internal (pipette) solution, which contained both Cs⁺ and TEA in sufficient concentrations to block BK channels (see Materials and methods). (A–C) Representative, leak-subtracted examples from single cones. (A) In the absence of light stimulation, the depolarization from a holding potential of −50 mV to −30 mV (see command monitor above plot) produced an inward current followed by an outward component. (B) As in A but with the addition of 200 or 250 µM NFA to the external solution to block the calcium-activated chloride conductance. (C) As in A but with the addition of 10 µM ISR to the external solution to block the voltage-gated Ca²⁺ conductance. (D) Isolated calcium-activated chloride currents were averaged from 11 cones. Currents were isolated by recording responses to voltage steps as in A and B and then subtracting current traces recorded with NFA from those recorded without NFA. Cones were held at −70 mV before stepping for 750 ms to depolarizing V_ms from −50 mV to 0 mV in 10 mV increments and then returning to −70 mV (see command pulse above plot). (E) i_Ca was isolated by recording in NFA as in B with the same voltage protocol as in D. Responses to voltage steps from −50 mV to 0 mV in 10-mV increments were leak subtracted and averaged (n = 16). (F) Peak and steady-state currents (mean ± SEM) at each voltage from D and E were used to derive the current–voltage relationship of mouse cone i_Cl(Ca) shown above (red triangles) and i_Ca shown below (filled blue squares, peak current; open blue squares, steady-state current). Shaded area indicates approximate physiological range of membrane voltages.

Figure 3.

Feedback-mediated currents elicited by light after dialysis of GTP-γ-S. (A) Examples from three different cones from Cx36^−/− mice show the effect of introducing 2 mM GTP-γ-S from the whole-cell patch pipette into the cone cell body. Currents were recorded at a holding potential of V_m = −50 mV. GTP-γ-S produced a slow decrease in resting current and nearly eliminated responses to a repeated saturating flash stimuli (1.2 × 10⁴ P*). (B) After cones treated as in A had reached a steady-state resting current with little or no response at V_m = −50, responses were recorded to the same light stimulus at holding potentials from −70 to −10 mV in steps of 20 mV. Records show baseline-subtracted averages of 13 cones. A strong biphasic response was detected when cones were held at −30 mV (red). (C) Responses as in B at a holding potential of −30 mV, with (red) and without (control, black) 10 µM ISR (n = 2). (D) Responses as in B and C but first with 200 µM NFA in the bath (black, n = 4) followed by NFA together with 100 nM IBX (red, n = 4). (E) Responses from D were subtracted [NFA – (NFA + IBX)] to give IBX-sensitive BK current (n = 4). For B–E, flashes were given at t = 0.

Figure 4.

i_h (HCN channels). Whole-cell patch recordings from Cx36^−/− cones with 25 mM TEA and 10 µM ISR included in the bath solution. The pipette solution was K-Asp. (A) Currents from a sample cone held at −50 mV and stepped for 400 ms to potentials of −25 to −135 mV in increments of 10 mV. Responses to depolarizing steps were omitted. Colored lines are fits to single exponential functions to give activation time constants, τ_act. See text. (B) Values of τ_act (mean ± SEM, n = 10) from fits like those in A are plotted against holding potential. (C) The reversal potential of i_h was measured by first hyperpolarizing the cone from the resting holding potential of −50 mV to −120 mV for 400 ms to activate the conductance, and then depolarizing from −120 mV for another 400 ms to potentials ranging from −100 mV to +10 mV in increments of 10 mV. See command trace above plot. Cones were returned to −50 mV after the second command step. Dashed lines indicate 5 ms and 250 ms after the second voltage step was applied. (D) The peak values of the currents were plotted as a function of V_m to derive the current–voltage relationship of mouse cone i_h (black, n = 10). The peak current values were averaged from last 50 ms of each voltage step, shown as the thin horizontal bar below the current traces in A. To derive the reversal potential of i_h, we measured the difference in the current just after the beginning of the depolarizing step (vertical dashed red line to left) and at steady-state 250 ms after the beginning of the step (vertical dashed red line to right). This procedure was adopted to compensate for the small leakage current, evident in the nonzero values of current recorded at 250 ms after the second step at positive potentials, where i_h would be expected to be zero. The values of the current differences for voltages near the reversal potential are plotted as the red symbols and line. Data are means ± SEM.

Figure 5.

Voltage-gated K⁺ currents. Whole-cell patch recordings from Cx36^−/− cones. (A) Voltage-gated transient current (i_k,trans) recorded in the presence of 25 mM TEA to block a sustained current, together with 10 µM ISR to block Ca²⁺ currents and Ca²⁺-activated K⁺ and Cl⁻ currents. Recordings are average responses (after leak subtraction, n = 5) to voltage steps from a holding potential of V_m = −70 mV to voltages from −50 mV to +20 mV in steps of 10 mV. (B) Sustained, TEA-sensitive component of K⁺ current (i_k,sus) isolated by recording responses to voltage steps in the presence and absence of TEA. Currents were recorded in external Ames’ medium containing 10 µM ISR, 7.5 mM CsCl, and 2 mM 4-AP, with and without 25 mM TEA. Cones were stimulated with steps from a holding potential of V_m = −50 mV to voltages from −10 mV to +90 mV in steps of 20 mV. Records with TEA were subtracted from those without TEA. Average current traces are shown from eight cones. (C) Means ± SEM of peak currents from A and B are plotted as a function of step voltage. Transient currents from A are plotted in black (n = 5), and sustained currents in red (n = 8).