Stimulation of the midbrain periaqueductal gray modulates preinspiratory neurons in the ventrolateral medulla in the rat in vivo

Abstract

The midbrain periaqueductal gray (PAG) is involved in many basic survival behaviors that affect respiration. We hypothesized that the PAG promotes these behaviors by changing the firing of preinspiratory (pre-I) neurons in the pre-Bötzinger complex, a cell group thought to be important in generating respiratory rhythm. We tested this hypothesis by recording single unit activity of pre-Bötzinger pre-I neurons during stimulation in different parts of the PAG. Stimulation in the dorsal PAG increased the firing of pre-I neurons, resulting in tachypnea. Stimulation in the medial part of the lateral PAG converted the pre-I neurons into inspiratory phase-spanning cells, resulting in inspiratory apneusis. Stimulation in the lateral part of the lateral PAG generated an early onset of the pre-I neuronal discharge, which continued throughout the inspiratory phase, while at the same time attenuating diaphragm contraction. Stimulation in the ventral part of the lateral PAG induced tachypnea but inhibited pre-I cell firing, whereas stimulation in the ventrolateral PAG inhibited not only pre-I cells but also the diaphragm, leading to apnea. These findings show that PAG stimulation changes the activity of the pre-Bötzinger pre-I neurons. These changes are in line with the different behaviors generated by the PAG, such as the dorsal PAG generating avoidance behavior, the lateral PAG generating fight and flight, and the ventrolateral PAG generating freezing and immobility.

Keywords: inspiration; periaqueductal gray; pre-Bötzinger; pre-I neuron; respiration; emotional behavior.

Figure 1

A: Extracellular recording of a pre-BötC pre-I neuron during eupnea, combined with diaphragm EMG. B: Power density spectrum of the pre-I neuron during eupnea (computed from n = 10). C: Coronal sections of the medulla (adapted from Paxinos and Watson, 1997) showing histologically identified extracellular recording sites (eight cells marked by rhodamine in three rats) from the pre-BötC region. D: Spatial distribution of 72 recorded pre-I neurons in the ventrolateral medulla from rostral (−12.9 mm caudal to bregma) to caudal (−12.55 mm caudal to bregma). This zone corresponds to the pre-BötC region. The figure represents a composite of 22 experiments. E: Peak frequency discharge spread computed from 29 pre-I neurons.

Figure 2

A: Phasic activation of a pre-I neuron following DLH stimulation of the recorded area in the pre-BötC region. B: Histogram illustrating dose-dependent phasic excitation of pre-I neurons (n = 10). C: Pre-I neuron recording site.

Figure 3

A: Phasic activation of a pre-I neuron following stimulation in the dorsal PAG. B: Histogram illustrating dose-dependent phasic excitation of pre-I neurons (n = 15). A 40-nl DLH dose induced a greater increase in spiking of the pre-I cells than a 20-nl DLH dose. C: PAG stimulation site. D: Pre-I neuron recording site. aq, Midbrain cerebral aqueduct; dm, dorsomedial; dl, dorsolateral; lat, lateral; vlat, ventrolateral, NA, nucleus ambiguus; sp5I, spinal trigeminal nucleus pars interpolaris; sp5, spinal trigeminal tract; NTS, nucleus tractus solitarius; MLF, medial longitudinal fasciculus; IO, inferior olive. Scale bar = 0.5 mm.

Figure 4

A: Phasic activation of a pre-I neuron following stimulation in the dorsomedial PAG in a different rat. B: PAG stimulation site. C: Pre-I neuron recording site. For abbreviations see Figure 3.

Figure 5

A: Stimulation in the medial part of the lateral PAG (bregma −7.0 mm) converted the pre-I cell into an inspiratory phase-spanning neuron. The pre-I cell started firing 15–30 msec before the diaphragm, as in eupnea, but continued firing throughout inspiration and did not show postinspiratory bursting. B: Quantitative changes to pre-I neuronal spike discharge following stimulation in the dorsal part of the lateral PAG (n = 15). C: PAG stimulation site. D: Pre-I neuron recording site. For abbreviations see Figure 3. Scale bar = 0.5 mm.

Figure 6

A: Stimulation in the medial part of the lateral PAG (bregma −7.0 mm) in a rat different from that represented in Figure 5 also converted the pre-I cell into an inspiratory phase-spanning neuron. B: PAG stimulation site. C: Pre-I neuron recording site. For abbreviations see Figure 3.

Figure 7

A: Stimulation in the lateral part of the lateral PAG (bregma −7.6 mm) induced an early onset of preinspiratory neuronal discharge. The pre-I neuron started firing 200–400 msec prior to the onset of the diaphragm and continued to discharge throughout the inspiratory phase. B: Quantitative changes to pre-I neuronal spike discharge following lateral PAG stimulation (n = 15). C: PAG stimulation site. D: Pre-I neuron recording site. For abbreviations see Figure 3. Scale bar = 0.5 mm.

Figure 8

A: Stimulation in the lateral part of the lateral PAG (bregma −7.6 mm) in a rat different from that represented in Figure 7 also induced an early onset of preinspiratory neuronal discharge. B: PAG stimulation site. C: Pre-I neuron recording site. For abbreviations see Figure 3.

Figure 9

A: Stimulation in the ventral part of the lateral PAG (bregma −8.0 mm) induced tachypnea and tonic activation of the pre-I cell over 1 second, followed by inhibition of the same pre-I cell. B: Quantitative changes to pre-I neuronal spike discharge prior to tachypnea and during tachypnea (n = 15). C: PAG stimulation site. D: Pre-I neuron recording site. For abbreviations see Figure 3. Scale bar = 0.5 mm.

Figure 10

A: Stimulation in a rat different from that represented in Figure 9 in the ventral part of the lateral PAG (bregma −8.0 mm) also induced tachypnea and tonic activation of the pre-I cell over 1 second, followed by inhibition of the same pre-I cell. B: PAG stimulation site. C: Pre-I neuron recording site. For abbreviations see Figure 3.

Figure 11

A: Stimulation in the caudal part of the ventrolateral PAG (bregma −8.3 mm) induced apnea and inhibition of the pre-I neuron. Immediately prior to this inhibition, the pre-I cell was tonically activated during two breaths. Following the return of breathing after 17 seconds, the pre-I neuron showed tonic activation during seven breaths before returning to normal eupneic values. B: Quantitative changes to pre-I neuronal spike discharge prior to apnea and during apnea (n = 15). C: PAG stimulation site. D: Pre-I neuron recording site. For abbreviations see Figure 3. Scale bar = 0.5 mm.

Figure 12

A: Stimulation in the caudal part of the ventrolateral PAG (bregma −8.3 mm) in a rat different from that represented in Figure 11 also induced apnea and inhibition of the pre-I neuron. Immediately prior to this inhibition, the pre-I cell was tonically activated during two breaths. Following the return of breathing after 16 seconds, the pre-I neuron showed tonic activation during three breaths before returning to normal eupneic values. B: PAG stimulation site. C: Pre-I neuron recording site. For abbreviations see Figure 3.

Figure 13

Summary diagram of the effects of PAG stimulation on pre-BötC pre-I neurons. Stimulation in the dorsal PAG produces phasic excitation in the context of avoidance behavior, whereas stimulation in the lateral PAG produces tonic excitation in the context of fight and flight. Stimulation in the ventrolateral PAG produces inhibition of pre-I cells in the context of freezing and immobility.