Optical and electrophysiological recordings of corticospinal synaptic activity and its developmental change in in vitro rat slice co-cultures.

Electrophysiological recordings and optical imaging with a fast voltage-sensitive dye (di-4-ANNEPS) were used to directly examine the spatiotemporal properties of in vitro corticospinal synapses formed in co-cultures of cerebral cortex and spinal cord slices. Whole cell recordings from spinal cord cells showed both monosynaptic and polysynaptic excitatory postsynaptic currents (EPSCs) in response to stimulation of corticospinal axons. Monosynaptic EPSCs and excitatory postsynaptic potentials (EPSPs) were isolated in artificial cerebrospinal fluid containing high concentrations of divalent cations. Optical imaging and extracellular recordings were done simultaneously. Both EPSPs and optically recorded excitatory postsynaptic potentials (optEPSPs) lasted 300-500 ms and were almost always positive. The major component of these long-lasting potentials was blocked by ifenprodil, a specific antagonist of the NR2B subunit-containing N-methyl-d-aspartate receptor (NMDAR). The spatial distribution of corticospinal optEPSPs paralleled that of the corticospinal field excitatory postsynaptic potentials (fEPSPs), suggesting that positive fEPSP amplitude is a reliable indicator of the distribution of corticospinal synapses. Corticospinal optEPSPs spread into the ventrolateral region by 6-7 days in vitro (DIV), but were restricted to the dorsomedial area by 11-13 DIV, suggesting synapses were eliminated from the ventrolateral side of the spinal cord. After the recordings were complete, corticospinal fibers were often anterogradely labeled with biocytin to assess the relation between presynaptic fiber distribution and the optical signals (optically-recorded presynaptic fiber volley (opt-prevolley) and optEPSP). The distributions of the opt-prevolleys and optEPSPs correlated well with the distribution of presynaptic fibers, suggesting the opt-prevolley reflects corticospinal fiber activity and that the fibers made synapses relatively evenly along their axons. The NR2B-mediated component of the corticospinal synaptic response declined during the interval between 6 and 7 DIV and 11-13 DIV, suggesting that a shift in the NMDAR subtype from NR2B to something else (perhaps NR2A) may be involved in regulating developmental plasticity in the rat spinal cord and the process of corticospinal synapse elimination.