Integration of Excitatory Postsynaptic Potentials in Dendrites of Motoneurons of Rat Spinal Cord Slice Cultures

Abstract

Larkum, Matthew E., Thomas Launey, Alexander Dityatev, and Hans-R. Lüscher. Integration of excitatory postsynaptic potentials in dendrites of motoneurons of rat spinal cord slice cultures. J. Neurophysiol. 80: 924–935, 1998. We examined the attenuation and integration of spontaneous excitatory postsynaptic potentials (sEPSPs) in the dendrites of presumed motoneurons (MNs) of organotypic rat spinal cord cultures. Simultaneous whole cell recordings in current-clamp mode were made from either the soma and a dendrite or from two dendrites. Direct comparison of the two voltage recordings revealed that the membrane potentials at the two recording sites followed each other very closely except for the fast-rising phases of the EPSPs. The dendritic recording represented a low-pass filtered version of the somatic recording and vice versa. A computer-assisted method was developed to fit the sEPSPs with a generalized α-function for measuring their amplitudes and rise times (10–90%). The mean EPSP peak attenuation between the two recording electrodes was determined by a maximum likelihood analysis that extracted populations of similar amplitude ratios from the fitted events at each electrode. For each pair of recordings, the amplitude attenuation ratio for EPSP traveling from dendrite to soma was larger than that traveling from soma to dendrite. The linear relation between mean ln attenuation and distance between recording electrodes was used to map 1/ e attenuations into units of distance (μm). For EPSPs with typical time course traveling from the somatic to the dendritic recording electrode, the mean 1/ e attenuation corresponded to 714 μm; for EPSPs traveling in the opposite direction, the mean 1/ e attenuation corresponded to 263 μm. As predicted from cable analysis, fast EPSPs attenuated more in both the somatofugal and somatopetal direction than did slow EPSPs. For EPSPs with rise times shorter than ∼2.0 ms, the attenuation factor increased steeply. Compartmental computer modeling of the experiments with biocytin-filled and reconstructed MNs that used passive membrane properties revealed amplitude attenuation ratios of the EPSP traveling in both the somatofugal and somatopetal direction that were comparable to those observed in real experiments. The modeling of a barrage of sEPSPs further confirmed that the somato-dendritic compartments of a MN are virtually isopotential except for the fast-rising phase of EPSPs. Large, transient differences in membrane potential are locally confined to the site of EPSP generation. Comparing the modeling results with the experiments suggests that the observed attenuation ratios are adequately explained by passive membrane properties alone.