Hyperpolarization-activated cationic currents (Ih) in neurones of the trigeminal mesencephalic nucleus of the rat

Abstract

1. We studied the voltage-dependent current activated by membrane hyperpolarization in sensory proprioceptive trigeminal mesencephalic nucleus (MNV) neurones. 2. Membrane hyperpolarization (from -62 to -132 mV in 10 mV steps) activated slowly activating and non-inactivating inward currents. The hyperpolarization-activated currents could be described by activation curves with a half-maximal activation potential (V ) of -93 mV, slope (k) of 8.4 mV, and maximally activated currents (Imax) of around 1 nA. The reversal potential of the hyperpolarization-activated currents was -57 mV. 3. Extracellular Cs+ blocked hyperpolarization-activated currents rapidly and reversibly in a concentration-dependent manner with an IC50 of 100 microM and Hill slope of 0.8. ZD7288 (1 microM; 4-(N-ethyl-N-phenylamino)-1,2-dimethyl-6-(methylamino) pyridinium chloride), the compound developed as an inhibitor of the cardiac hyperpolarization-activated current (If), also blocked the hyperpolarization-activated currents in MNV neurones. Extracellular Ba2+ (1 mM) did not affect hyperpolarization-activated currents. We tested whether the hyperpolarization-activated currents contribute to the somatic membrane properties of MNV neurones by performing some experiments using current-clamp recording. In such experiments application of Cs+ (1 mM) produced no effect on neuronal resting membrane potentials. 4. During the course of our experiments we noticed that activating ATP-gated non-selective cation channels (P2X receptors) caused an inhibition of Ih associated with a V shift of 10 mV in the hyperpolarizing direction. This P2X receptor-mediated inhibition of Ih was blocked in recordings made with the rapid calcium chelator BAPTA (11 mM) in the pipette solution. 5. We conclude that the current activated by membrane hyperpolarization in MNV neurones is Ih on the basis of its similarity to Ih observed in other neuronal preparations. Activation of Ih can account for the anomalous time-dependent inward rectification that has previously been described in MNV neurones.

Figure 1. Current-clamp recording of time-dependent anomalous rectification

The injection of hyperpolarizing current (top) caused a change in the membrane potential (bottom) which was followed by a sag in the membrane voltage response (indicated by arrowhead in the left trace). This time-dependent anomalous rectification was blocked by the extracellular application of Cs⁺ (1 mm; right trace).

Figure 2. Voltage-clamp recording of hyperpolarization-activated currents

Hyperpolarizing steps from -62 to -132 mV evoked membrane responses in MNV neurones that consisted of three components. There was an initial change in instantaneous holding current, which was followed by a slowly developing inward current (shown as I_h activation in the top trace). On returning the membrane potential back to -62 mV, the tail current was measured as a residual current. The middle trace shows data from the same neurone in the presence of extracellular Cs⁺; note here there is no slow inward current. The bottom trace shows a digital subtraction of the middle trace from the top trace (Control - Cs⁺); the current that is revealed represents the pure hyperpolarization-activated current (see also Banks et al. 1993).

Figure 3. Voltage-dependent activation of I_h

A, upper panel shows the voltage waveform used to activate I_h from a holding potential of -62 mV to defined potentials in 10 mV steps of 500 ms duration. Lower panel shows a family of I_h currents recorded from an MNV neurone. The data from this cell are plotted in B. The activation curve for the steady-state currents did not reach a maximum but that for the tail currents did. The data for tail currents from a number of cells (n = 6) are shown in C. In this graph the data from individual cells have been normalized to the size of the current at -132 mV using steps that are 2 s in duration, and the pooled data were fitted to the Boltzmann function of the form: where V_m is the membrane potential, V_½ is the membrane potential at which I_h is half-activated, I is the tail current amplitude recorded after the voltage step back to -62 mV, I_max is the maximum tail current amplitude recorded after a step from -132 mV, and k is the slope factor that determines the steepness of the fitted curve.

Figure 4. Pharmacological characterization of I_h

A, extracellular Ba²⁺ (1 mm) had little effect on I_h, whereas in the same cell Cs⁺ (1 mm) caused complete block of I_h. B, the time course of action of Cs⁺ was rapid and was readily reversible. C, Cs⁺ blocked I_h in a concentration-dependent manner. In these experiments the data from a number of cells (n = 4) were normalized to the amplitude of I_h before Cs⁺ was added, and the resulting concentration-effect curve was fitted using iterative methods to the four-parameter logistic equation. D, Cs⁺ blocked I_h in a use-independent manner (n = 5). I_h was activated 5 times every 5 s (^) from -62 to -132 mV, and then the slice was left to incubate in Cs⁺ (1 mm) for 5 min whilst the cell was clamped at -62 mV and I_h not activated. Subsequently, I_h was activated again, 5 times once every 5 s; note that I_h is maximally blocked at the first step (•). The slice was then placed in bathing medium that contained no Cs⁺ for 10 min whilst the cell was clamped at -62 mV. Subsequently, I_h was activated 5 times once every 5 s (□). E, ZD7288 (1 μm) caused ∼70% inhibition of I_h activated by steps from -62 to -132 mV. F, the blocking effect of ZD7288 was slow to develop and did not reverse with up to 17 min of washout. The residual current remaining in ZD7288 (1 μm) could be blocked by Cs⁺ (1 mm; data not shown).

Figure 5. Activation of P2X receptor channels inhibited I_h

A, over a time course of up to 5 min there was no run-down of I_h (n = 4). B, holding current (•) and I_h tail current amplitude (^) are plotted on the same time scale (left). Application of ATPγS (30 μm) caused an inward current of about -600 pA, and there was a concomitant decrease in the amplitude of I_h. However, I_h had fully recovered by 180 s, whereas the P2X current still persisted. The desensitization of the P2X current from 100 to 180 s is reflected in the concomitant decrease in the inhibition of the I_h. The right-hand traces show superimposed tail currents from the cell shown in the graph at the time points indicated as i, ii and iii. C, original traces from a different MNV neurone to that shown in A, illustrating I_h currents at the start of the experiment (left-hand trace), in the presence of ATPγS (10 μm; middle trace) and then after washout of ATPγS (right-hand trace). In this cell, ATPγS (10 μm) evoked an inward current of -350 pA. The dotted line above each trace represents the zero holding current level. D, left-hand bar graph, there was no difference in the inward current evoked by ATPγS (30 μm) in neurones dialysed with (□) or without 11 mm BAPTA (▪). However, the inhibition of I_h (right-hand bar graph) caused by activating P2X receptors was significantly reduced by dialysis with BAPTA (□; *P < 0.05) compared with no BAPTA (▪). The data for the left- and right-hand bar graphs are from the same population of neurones (mean ±s.e.m., n = 6). E, activation curves determined in the absence (^) and presence of ATPγS (10 μm; •). The left-hand graph shows that there was a marked shift in the V_½ for I_h activation in the presence of ATPγS to activate P2X receptors. In four neurones the application of 10 μm ATPγS shifted the V_½ by -10 ± 2 mV (P < 0.05; n = 4), whereas k and I_max changed by only 1.3 ± 0.6 mV and 8 ± 4%, respectively. In these cells ATPγS (10 μm) evoked an inward current of -567 ± 35 pA. The right-hand graph shows that in cells dialysed with 11 mm BAPTA there was no significant shift in the V_½ (-3 ± 2 mV; P > 0.05; n = 6); k and I_max changed by only 0.4 ± 0.9 mV and 1 ± 3%, respectively. In these six neurones the inward current evoked by ATPγS (10 μm) was -490 ± 107 pA. In some graphs the size of the error bar is smaller than the symbol, and unidirectional error bars are shown for clarity.