Properties and physiological function of Ca2+-dependent K+ currents in uniglomerular olfactory projection neurons

Abstract

Ca(2+)-activated potassium currents [IK(Ca)] are an important link between the intracellular signaling system and the membrane potential, which shapes intrinsic electrophysiological properties. To better understand the ionic mechanisms that mediate intrinsic firing properties of olfactory uniglomerular projection neurons (uPNs), we used whole cell patch-clamp recordings in an intact adult brain preparation of the male cockroach Periplaneta americana to analyze IK(Ca) In the insect brain, uPNs form the principal pathway from the antennal lobe to the protocerebrum, where centers for multimodal sensory processing and learning are located. In uPNs the activation of IK(Ca) was clearly voltage and Ca(2+) dependent. Thus under physiological conditions IK(Ca) is strongly dependent on Ca(2+) influx kinetics and on the membrane potential. The biophysical characterization suggests that IK(Ca) is generated by big-conductance (BK) channels. A small-conductance (SK) channel-generated current could not be detected. IK(Ca) was sensitive to charybdotoxin (CTX) and iberiotoxin (IbTX) but not to apamin. The functional role of IK(Ca) was analyzed in occlusion experiments under current clamp, in which portions of IK(Ca) were blocked by CTX or IbTX. Blockade of IK(Ca) showed that IK(Ca) contributes significantly to intrinsic electrophysiological properties such as the action potential waveform and membrane excitability.

Keywords: Periplaneta americana; antennal lobe; chemosensory; glomerulus; olfaction.

Fig. 1.

Morphology of uniglomerular projection neurons (uPNs) and isolation of Ca²⁺-activated potassium currents [I_K(Ca)]. A: morphology of the recorded uPN revealed by labeling with biocytin-streptavidin via the patch pipette. A1: overview. The neuron innervated a single glomerulus (GL) and sent a single axon along the medial antennal lobe tract (mALT) to the mushroom body's calyces (CA) and the lateral horn (LH). The position of the soma, which was lost during processing, is marked (*). AL, antennal lobe; na, neural anterior; np, neural posterior; l, lateral; m, medial. A2–A4: higher magnification of the framed areas in A1. A2 and A3: boutons in the calyces (A2) and in the LH (A3). A4: neurites in the single innervated glomerulus. B–E: isolation of I_K(Ca). The holding potential was −60 mV. B: currents were activated by 300-ms depolarizing steps from −60 mV to 60 mV in 10-mV increments. TTX, tetrodotoxin; 4-AP, 4-aminopyridine. C: current traces elicited by the same depolarizing steps as in B, during the application of 5 × 10⁻⁴ M Cd²⁺. D: subtraction of the C traces from the B traces yields I_K(Ca). E: voltage dependence for activation of I_K(Ca): current-voltage (I-V) relation and current density-V relation of I_K(Ca). E_M, membrane potential; C_M, cell capacitance.

Fig. 2.

Voltage dependence of I_K(Ca). A: I-V relation. B: current density-V relation. Current density was calculated from the ratio between I_K(Ca) and C_M. C and D: normalized I-V relation [I_K(Ca) as fractions of the maximal I_K(Ca) for each neuron]. A–C are based on the same original recordings with extracellular Ca²⁺ concentration ([Ca²⁺]_o) = 1 mM. The recordings in D were performed in [Ca²⁺]_o = 6 mM. Gray, individual cells; black: mean ± SD.

Fig. 3.

I_K(Ca) is dependent on the amplitude and duration of the “Ca²⁺ loading pulses.” Test pulses depolarized the membrane to +60 mV, where virtually no voltage-activated Ca²⁺ influx occurs, since the membrane approaches the Ca²⁺ equilibrium potential. The test pulses were preceded by “Ca²⁺ loading pulses” of varying amplitude (A1–A4) or duration (B1–B3). A1-A4: the +60 mV test pulses were preceded by 200-ms depolarizing Ca²⁺ loading pulses of varying amplitude (−60 mV to +60 mV; 10-mV increments). A1: currents in response to the +60 mV test pulse preceded by loading pulses to +10 mV (large Ca²⁺ influx) and to +60 mV (virtually no Ca²⁺ influx). A2: same experiment as in A1 with the whole series of loading pulses from −60 mV to +60 mV. A3: framed area of A2 in higher resolution. A4: I-V_loading relation of I_K(Ca). The I-V_loading relation shows I_K(Ca) at the beginning of the +60 mV test pulse plotted as a function of the loading-pulse voltage. B1–B3: the +60 mV test pulses were preceded by Ca²⁺ loading pulses of varying duration (5–400 ms). The loading-pulse amplitude (*) was adjusted for each neuron to values where maximal loading occurred during the first part of the experiments (see A4). B1: currents in response to the +60 mV test pulse that were preceded by 10-mV loading pulses of varying duration. B2: I_K(Ca) at the beginning of the +60 mV test pulses [normalized to the maximal I_K(Ca) of each cell] plotted as a function of the loading-pulse duration. B3: decay time constant τ of I_K(Ca) in dependence of the loading-pulse duration.

Fig. 4.

Voltage and Ca²⁺ dependence of I_K(Ca). To measure the voltage and Ca²⁺ dependence of I_K(Ca) independently from the instantaneous Ca²⁺ influx, I_K(Ca) was recorded under voltage and cytosolic Ca²⁺ concentration ([Ca²⁺]_i) clamp. The holding potential was −60 mV, and voltage pulses were applied between −60 mV and +90 mV in 10-mV increments. [Ca²⁺]_i was clamped at 56 μM (n = 5), 143 μM (n = 5), 540 μM (n = 6), or 1,800 μM (n = 6) with an EDTA or EGTA-Ca²⁺ buffering system. Voltage-activated Ca²⁺ currents were blocked by 5 × 10⁻⁴ M CdCl₂. A: under clamped [Ca²⁺]_i I_K(Ca) did not inactivate during a sustained voltage pulse. Increasing Ca²⁺ concentrations increased the I_K(Ca) amplitude, and at all Ca²⁺ concentrations I_K(Ca) increased with increasing depolarization. B: example traces (different neurons) demonstrating the effect of increasing [Ca²⁺]_i on I_K(Ca). C: conductance-voltage (G-V) relations for different [Ca²⁺]_i. Error bars are omitted for better visualization. D: voltages for half-maximal activation (V_0.5,act) for different [Ca²⁺]_i.

Fig. 5.

Charybdotoxin (CTX) and iberiotoxin (IbTX) reduce I_K(Ca). A: example traces (different neurons) demonstrating the effect of increasing concentrations of CTX (left) and IbTX (right) on I_K(Ca). B: concentration-response relation of CTX and IbTX. Curves are fits to a sigmoidal relation (Eq. 1). CTX had an IC₅₀ of 2.7 nM (1.8–3.8 nM), and IbTX had an IC₅₀ of 157 pM (53.7–460 pM).

Fig. 6.

uPNs do not generate small-conductance current (I_SK). A: in dopaminergic (DA) substantia nigra neurons of mice a short depolarizing voltage pulse (Ca²⁺ loading pulse) evoked Ca²⁺ influx that activated an apamin-sensitive I_SK. B: uPNs did not generate I_SK. The 100-ms loading pulse to 0 mV was preceded by a 1-s hyperpolarizing step to −80 mV. The holding potential was −60 mV. The recordings are only shown for the framed region of the voltage protocol. p/n protocols were not applied. The apamin concentration was 100 nM in A and 1 μM in B.

Fig. 7.

Comparison of direct and indirect “block” of I_K(Ca) by CTX and Cd²⁺, respectively. Holding potential in all recordings was −60 mV. Currents were activated by 300-ms depolarizing steps from −60 mV to +60 mV in 10-mV increments. A: whole cell currents of a uPN during TTX and 4-aminopyridine (4-AP) application. B and C: recordings of the same neuron as in A, during the additional application of CTX (B) and during the application of CTX and 5 × 10⁻⁴ M Cd²⁺ (C). D: subtraction of the B traces from the A traces. E: subtraction of the C traces from the B traces. F: I-V relation for the current traces shown in A–E.

Fig. 8.

Example current-clamp recording to demonstrate the CTX effect on action potential waveform.

Fig. 9.

Summary and quantification of CTX (A), IbTX (B), and apamin (Apa; C) effects on electrophysiological properties [rate of repolarization (1), width of half-maximal amplitude (2), amplitude of afterhyperpolarization (AHP) (3), latency to first action potential during a depolarizing voltage step (4), action potential threshold (5), and number of action potentials during 500-ms depolarizing current pulses (6)]. Schematics demonstrate how the parameters were measured; for details see methods. n values are given in the bars of each graph. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. CTL, control; W, wash.