Intracellular recording of identified neostriatal patch and matrix spiny cells in a slice preparation preserving cortical inputs

The morphology, electrical membrane properties, and corticostriatal excitatory postsynaptic potentials (EPSPs) of two groups of neostriatal projection cells, patch cells, and matrix spiny cells were compared in the rat by the use of an in vitro slice preparation that preserves inputs from medial agranular cortex. Spiny cells were stained intracellularly with biocytin and identified as belonging to the patch (striosomal) compartment or to the matrix by immunohistochemistry for the 28 kD calcium-binding protein calbindin on the same sections. 2. Patch and matrix neurons had very similar axonal and dendritic morphology. Both patch and matrix cells extended their dendrites and local axon collaterals almost exclusively in their respective compartments. Patch cells and most matrix cells had local axon collaterals within or near the parent dendritic domain. However there was a class of matrix cells that extended axon collaterals over a much wider portion of the neostriatum but still restricted to the matrix compartment. 3. Input resistance and membrane time constant were estimated from the membrane response to intracellularly applied current pulses. The average values of matrix cells were and 8.41 ms. The values of patch cells were 31.8 M omega and 8.19 ms and were within the range of those of matrix cells. Both types of cells showed the same kinds of membrane nonlinearities when tested with the use of current pulses. Input resistance and time constant were both strongly affected by a fast anomalous rectification and were thus voltage-dependent, decreasing with membrane polarization. Slow ramplike depolarizing responses were observed in response to depolarizing current steps. 4. Repetitive firing was examined with the use of depolarizing current pulses. In both types of spiny cells, trains of action potentials showed little adaptation of spike frequency and linearly increased with current intensities less than 1 nA. The slopes frequency, calculated from the first and second intervals, were 115.0 and 107.2 Hz/nA, respectively, for matrix cells and 86.0 and 82.8 Hz/nA for patch cells. 5. Stimulation of the medial agranular cortex induced EPSPs in some striatal cells in both compartments. EPSP in matrix cells often showed both short-latency and long-latency components, corresponding to two early components of the response observed in vivo. Some matrix cells, and all patch cells, showed only the longer latency EPSP component. The average latency was 6.3 ms in matrix cells and 9.1 ms in patch cells. The relationship between EPSP amplitude and membrane potential was nonlinear, with EPSP amplitude and duration increasing with decreasing membrane polarization.(ABSTRACT TRUNCATED AT 400 WORDS)