Optogenetic Evidence for a Direct Circuit Linking Nociceptive Transmission through the Parabrachial Complex with Pain-Modulating Neurons of the Rostral Ventromedial Medulla (RVM)

Vector injection sites in PB. Injection sites were distributed among sections at +0.24 to −0.36 mm relative to the interaural line. Injections inside the lateral PB (lPB) area were considered on-target. KF, Kölliker-Fuse; mPB, medial PB area; SCP, superior cerebellar peduncle.

Recording sites within the RVM. Recording sites for both ArchT (gray circles) and ChR2 (black circles) experiments were distributed between −1.04 and −2.30 mm (relative to the interaural line). The majority of cells were isolated between −1.04 and −1.80 mm.

Validation of ChR2 and ArchT function on PB neurons. A, Activation of PB neuron by light-induced activation of ChR2 (1- to 2-ms pulses at 2, 10, or 20 Hz). Firing of the neuron reliably followed the light trains. B, Suppression of activity of a PB neuron during light-induced activation of ArchT.

In vitro electrophysiology shows evidence for direct, functional PB inputs to individual RVM neurons. A, Micrograph of slice showing area of PB microinjection. B, Expression of reporter for ChR2 construct in PB in area outlined by square in A. C, Low magnitude (4×) picture of RVM slice containing a biocytin-labeled neuron that was filled during recording. D, High magnification (20×) of biocytin-labeled cell in C. E, Light-evoked stimulation elicited either glutamatergic synaptic currents that were inhibited by KA but not BIC, or GABAergic synaptic currents that were inhibited by BIC but not KA. Note the difference in synaptic decay kinetics. Arrow indicates delivery of light stimulus.

Effect of ChR2-induced activation of PB terminals on ongoing firing of ON-, OFF-, and NEUTRAL-cells in RVM. Terminal activation significantly increased the net firing rate of both ON- and OFF-cells (A, B) but had no net effect on NEUTRAL cells (C). *p < 0.05 and **p < 0.01 compared with baseline, paired t test, n = 9–14 cells per class. Effects of different stimulation parameters are pooled for this analysis.

Effect of ChR2-induced activation of PB terminals on RVM neurons when using different stimulation protocols. As a class, ON-cells (A) and OFF-cells (B) showed significant and comparable responses to all light stimulation trains, whereas NEUTRAL-cells (C) did not respond during any protocol. Reported as geometric mean with 95% confidence intervals, *p < 0.05 and **p < 0.01 compared with the hypothetical value of 1 (no change), Wilcoxon’s signed rank test, n = 9–14 cells per class.

Effects of ArchT-induced inhibition of PB terminals and cell bodies on ON- and OFF-cell nociception-related activity. A, Representative examples show ON- and OFF-cell activity during withdrawal from noxious heat stimulus at baseline compared with during Arch-T inhibition of PB terminals in RVM or at the cell bodies in PB itself. Arrow denotes heat onset. In both cases, nociception-related changes in firing were substantially attenuated. Summary data are provided in B, C. B, ON-cells: effect of ArchT-induced inhibition of PB terminals and cell bodies on the ON-cell burst. C, OFF-cells: the effect of ArchT-induced inhibition of PB terminals and cell bodies on the OFF-cell pause. Reported as geometric mean with 95% confidence intervals, *p < 0.05 and **p < 0.01 compared with baseline, one-way ANOVA with repeated measures and post hoc Dunn’s multiple comparisons test, n = 15 ON-cells, 14 OFF-cells.