Hydrogen-peroxide-induced toxicity of rat striatal neurones involves activation of a non-selective cation channel

Abstract

Striatal neurones are particularly vulnerable to hypoxia/ischaemia-induced damage, and free radicals are thought to be prime mediators of this neuronal destruction. It has been shown that hydrogen peroxide (H2O2), through the production of free radicals, induces rat insulinoma cell death by activation of a non-selective cation channel, which leads to irreversible cell depolarization and unregulated Ca2+ entry into the cell. In the study presented here, we demonstrate that a subpopulation of striatal neurones (medium spiny neurones) is depolarized by H2O2 through the production of free radicals. Cell-attached recordings from rat cultured striatal neurones demonstrate that exposure to H2O2 opens a large-conductance channel that is characterized by extremely long open times (seconds). Inside-out recordings show that cytoplasmically applied beta-nicotinamide adenine dinucleotide activates a channel with little voltage dependence, a linear current-voltage relationship and a single-channel conductance of between 70 and 90 pS. This channel is permeable to Na+, K+ and Ca2+ ions. Fura-2 imaging from cultured striatal neurones reveals that H2O2 exposure induces a biphasic intracellular Ca2+ increase in a subpopulation of neurones, the second, later phase resulting in Ca2+ overload. This later component of the Ca2+ response is dependent on the presence of extracellular Ca2+ and is independent of synaptic activity or voltage-gated Ca2+ channel opening. Consequently, this channel may be an important contributor to free radical-induced selective striatal neurone destruction. These results are remarkably similar to those observed for insulinoma cells and suggest that this family of non-selective cation channels has a widespread distribution in mammalian tissues.

Figure 1. The effect of hydrogen peroxide (H₂O₂) on the resting electrical properties of striatal neurones

A(i), recording illustrating that the resting membrane properties of visually identified medium-sized spiny neurones, recorded from striatal slices, are stable. A(ii), current-clamp recording illustrating that application of 10 mm H₂O₂ (vertical arrow) caused decreased input resistance and slow depolarization followed by sudden and irreversible depolarization. A(iii), pre-application of the free-radical scavenger dimethylthiourea (DMTU, 1 mm) prevented the H₂O₂-induced depolarization. B, induction of single-channel currents by H₂O₂. Representative continuous cell-attached recording (no applied pipette potential) illustrating the spontaneous, sudden appearance of an inward current, approximately 30 min following exposure of the neurone to 45 mm H₂O₂+ 5 mm sodium azide. The current appears abruptly and expansion of the initial part of the current trace illustrates the rapid recruitment of six channels, giving rise to a sustained inward current. Note the large amplitude of these channel events and the slow kinetics, exemplified by the long, stable open durations.

Figure 2. Single-channel characteristics of the NS_NAD channel

Recordings from inside-out patches at different membrane potentials (given at the side of the traces) in symmetrical 140 mm NaCl (A), a bath solution of 140 mm NaCl and electrode solution containing 140 mm KCl (B) and a bath solution of 140 mm NaCl with an electrode solution containing 110 mm CaCl₂ (C). On the right of each set of traces are the corresponding single-channel current (I)-voltage (V) relationships. The straight lines show the best fit, with slope conductances of 74.9 pS (A), 89.2 pS (B) and 38.4 pS (C), and no obvious rectification. The currents were recorded with 0.2 mm free internal Ca²⁺ and 0.5 mmβ-nicotinamide adenine dinucleotide (β-NAD⁺). C indicates the closed state.

Figure 3. Dependence of the NS_NAD channel on Ca²⁺ and β-NAD⁺

Removal of β-NAD⁺ in the continued presence of Ca²⁺ (A), or removal of Ca²⁺ in the continued presence of β-NAD⁺ (B), resulted in the abrupt cessation of channel activity, an action that was reversible on re-application of β-NAD⁺ (A) or Ca²⁺ (B). Single-channel currents were recorded from inside-out patches at −40 mV in symmetrical 140 mm NaCl, in the presence of 0.2 mm Ca²⁺ and 0.5 mmβ-NAD⁺. C indicates the closed state.

Figure 4. H₂O₂ causes a biphasic rise in [Ca²⁺]_i

A, representative imaging experiment, from three neurones (also B and C), illustrating the biphasic [Ca²⁺]_i response caused by 10 mm H₂O₂. The initial rise occurred within 3 min, and a second, larger rise (> 1 μm) after 20–45 min. B, a separate experiment where addition of 10 mm H₂O₂ in the absence of extracellular Ca²⁺ caused increased resting [Ca²⁺]_i, which was sustained until re-addition of extracellular Ca²⁺, which resulted in an immediate [Ca²⁺]_i increase (> 1 μm). Note that the time of exposure to H₂O₂ in the absence of extracellular Ca²⁺ was longer than the time taken to reach a near-maximal response in A. C, identical experiment to B, except that prior to re-application of extracellular Ca²⁺, a cocktail of blockers (1 μm TTX, 10 μm NBQX, 50 μm D-APV and 100 μm Cd²⁺), was applied; this did not prevent Ca²⁺ entry on addition of extracellular Ca²⁺.