Figure 4.

Distinct interneuron subtypes have different TC-evoked response strengths, consistent with their spatial distributions in L1. A, Intrinsic physiological properties of an LS interneuron in L1. Top left, Absence of spike frequency adaptation during moderate spike rates (Chu et al., 2003). Middle left, A marked delay from current onset to the time of first spike during a just-suprathreshold current (Chu et al., 2003; Zhu and Zhu, 2004; Lee et al., 2010). Inset shows large afterhyperpolarization at high magnification (dashed arrow) (Chu et al., 2003; Lee et al., 2010). Bottom left voltage trace, Extremely fast voltage sag (arrow) and no conventional slower sag in response to a large hyperpolarizing current. We found this to be a consistent feature that distinguished LS from NLS cells. Injected current amplitudes for top, middle, and bottom left: +165, +155, and −725 pA. Bottom right, A slow afterdepolarization (ADP) after single spikes evoked by short (3 ms), large amplitude (+550 pA), current pulses (Chu et al., 2003). This LS cell was in Cg2, 0.98 mm anterior to bregma, 80 μm from the pia. B, Intrinsic physiological properties of an NLS interneuron in L1. Top left, Relatively strong spike frequency adaptation during moderate spike rates. Middle left, Only a short latency from current onset to the time of the first spike during a just-suprathreshold current. Inset shows a small afterhyperpolarization (dashed arrow). Bottom left voltage trace, A conventional slow voltage sag (arrow) in response to a hyperpolarizing current, with slower kinetics than in LS cells. Current amplitudes: +130, +70, and −150 pA. Bottom right, No slow afterdepolarization after single spikes (current, +300 pA). This NLS cell was in Cg1, 0.86 mm anterior to bregma, 120 μm from the pia. C, Top, Schematic for dual whole-cell recording of synaptic responses from an LS and NLS cell to optical stimulation of ChR2-expressing matrix TC axons. Bottom, TC-evoked EPSP amplitudes of the LS cell were more than twice those of the simultaneously recorded NLS cell. The cell pair was in area Cg1, 1.98 mm anterior to bregma (LS cell was 90 μm and NLS cell was 120 μm below the pia). The thalamic injection was centered on the anteromedial thalamic nucleus, 1.06 mm posterior from bregma. Laser flashes were 11.3 mW, aimed directly between the two cells. Steady-state membrane potentials were between −79 and −81 mV for A–C. D, Mean group TC-evoked EPSP amplitudes of LS cells were more than double those of NLS cells (p < 0.03, repeated-measures ANOVA, LS vs NLS groups). Optical stimuli were 11.3 mW aimed at or near the recorded cells within L1. Thalamic virus injections were centered on the ventromedial, anteromedial, reuniens, or rhomboid thalamic nuclei, and all resulted in clear ChR2-expressing projections to outer L1. The ChR2–EYFP fluorescence intensities in L1 were approximately equal for the slices containing the NLS cells and the slices containing the LS cells from the same brains (within 10%, n = 5; data not shown). E, Sublamination of interneuron types within L1. Histograms indicate number of recorded cells of each type as a function of the percentage distance between the pia (0%) and the L2 border (100%). The NLS cells were found in the lower half of L1, whereas the LS and other (unclassified) L1 cells were more evenly distributed throughout the layer. F, There was a significant correlation between matrix TC-evoked PSP amplitudes and interneuron soma location within L1 (p < 0.03, r = 0.35, n = 41). Responses tended to be large for cells located in outer L1 and smaller for cells of inner L1. Optical stimuli, thalamic virus injections, and their projections same as D.