Passive properties of neostriatal neurons during potassium conductance blockade

Abstract 

Voltage recordings from neostriatal projection neurons were obtained using in vitro intracellular techniques before and during K⁺-conductance blockade. Neurons were stained with the biocytin technique. Somatic surface area (A _S) was determined by both whole-cell recordings in isolated somata and by measuring stained somata recorded in slices. Dendritic measurements were done in reconstructed neurons. Average determinations of dendritic (A _D) and neuronal (A _N) surface areas coincided with previously reported anatomical data. Thus: A _S≈ 6.5 × 10^–6 cm²; A _D≈ 1.9 × 10^–4 cm²; A _N≈A _D + A _S≈ 2 × 10^–4 cm²; A _D/A _S≈ 30. Measurements were done before and after superfusion with K⁺-conductance blockers (K⁺-blockers). Cells whose neuronal morphology was not obviously distorted by K⁺-blockade were chosen for the present study. Electrotonic transients were matched to a somatic shunt equivalent cylinder model adjusted with the generalized correction factor (F _dga) that constrains the parameters for neuronal anatomy. Neuronal input resistance (R _N; mean ± SEM) increased when it was corrected for somatic shunt, from 49 ± 2 MΩ (n = 80) to 179 ± 7 MΩ (n = 32). A difference was also obtained between the slowest time constant, τ₀ = 16 ± 0.9 ms (n = 49), and the dendritic membrane time constant, τ_mD = 33 ± 1.6 ms (n = 36). When these electrophysiological measurements were used to calculate A _N, the value obtained was similar to the anatomical measurements. Combining anatomical and electrophysiological data, somatic and dendritic input resistances were determined: R _D = 182 ± 7 MΩ; R _S (with shunt) = 74 ± 4 MΩ (n = 32). The generalized correction factor, F _dga = 0.91 ± 0.007 (n = 10), implied a short effective electrotonic length for dendrites: L _D = 0.46 ± 0.014 (n = 32). Saturating concentrations of the K⁺-blockers tetraethylammonium, Cs⁺, and Ba²⁺ increased R _N and induced charging curves well fitted by single exponential functions in 56% of neostriatal neurons. Ba²⁺ greatly decreased the somatic shunt (n = 5): (R _N = 216 ± 21 MΩ, τ₀ = 46 ± 2 ms, R _D = 239 ± 25 MΩ, and R _S = 3.2 ± 0.5 GΩ), rendering values similar to those obtained with whole-cell recordings (e.g., R _N≈ 198 MΩ, R_S≈ 2.62 GΩ) (n = 52). Cs⁺ (n = 5) had less effect on the somatic shunt (R _N = 115 ± 19 MΩ, τ₀ = 49 ± 13 ms, R _S = 161 ± 8 MΩ), although dendritic conductance was equally blocked (R _D = 261 ± 16 MΩ). The Cs⁺-sensitive conductance exhibited inward rectifying properties not displayed by the Ba²⁺-sensitive conductance, suggesting that Cs⁺ preferentially acted upon inward rectifier conductances. In contrast, Ba²⁺ significantly acted upon linear conductances making up the somatic shunt. This suggests a differential action of different K⁺-blockers on the somato-dendritic membrane, implying a differential distribution of membrane conductances. Another action of K⁺-blockers, in about 40% of the cells, was to induce dye and probably electrical coupling between neighboring neurons.