Vagotomy

Three types of stimulation paradigms were used: (1) For continuous stimulation, inspiratory activity was recorded in 30 s sweeps containing a 10 s continuous light stimulus (2–10 sweeps for each stimulation). The percentage of change in instantaneous frequency and amplitude of each XII burst relative to the averaged values during baseline was calculated, binned in 1 s intervals, and plotted over time. To examine laser-power-dependent effects on breathing, light intensity was varied from 0 to 13 mW. However, non-specific effects were observed in Cre− control mice >7 mW. Therefore, laser power was set at <7 mW for all other in vivo stimulations. (2) For phase-dependent stimulation, integrated XII nerve activity was thresholded in order to trigger a TTL pulse during the rising phase or the falling phase of the integrated XII signal (Clampex Software), thereby isolating the laser stimulus to the inspiratory or expiratory respiratory phase. (3) For randomized brief stimulations, short 100–200 ms light pulses were delivered at 10 s intervals to stimulate the preBötC at random time points during the respiratory cycle (30–150 stimulations/experiment). The respiratory cycle was defined as the onset of inspiration to the onset of the subsequent inspiration. The phase shift elicited by each stimulation was calculated as the duration of the respiratory cycle containing the stimulus, divided by the average of the two preceding respiratory cycles (expected phase). The phase of preBötC stimulation was calculated as the time between the onset of inspiration and the stimulus onset, divided by the expected phase. The average phase shift was then plotted against the stimulus phase in bins containing 1/10 of the expected phase.
Each stimulation paradigm was performed during fast, eupneic breathing with the vagus intact and during slow breathing following vagotomy. XII nerve activity was continuously recorded while the vagus nerves were cut bilaterally. In a subset of vagus-intact mice, pulled glass injection pipettes were used to inject (150 nl) a mixture of strychnine (250 µM) and gabazine (250 µM) diluted in aCSF bilaterally into the preBötC to block synaptic inhibition. Thirty to 60 s of XII nerve activity before and 5–10 min after vagotomy or injection of strychnine and gabazine were used to quantify changes in breathing parameters.
Following electrophysiological experiments, the position of the fiber optic on the ventral surface of the medulla was marked with an injection of fluorescent beads (Fluorospheres). Fresh wet or fixed and cryoprotected (see below) whole brainstems were imaged, and the coordinates of the stimulation site were measured relative to the midline (X direction) and both the most rostral point of the caudal cerebellar artery and the intersection of the vertebral arteries (Y direction). Distances were normalized to the half-width of the brainstem at the level of the caudal cerebellar artery to correct for variations in animal size and differences due to tissue shrinkage from fixation in paraformaldehyde.