Selective Inhibition of Pre‐Bötzinger Neurons Projecting to the Facial Nucleus Decreases Nasofacial Activity Without Affecting Breathing

The pre‐Bötzinger Complex (preBötC) is a major generator of respiratory rhythm and is crucial for respiratory modulation of orofacial behaviours. The nasolabialis profundus (NLP) is a facial muscle located in the rostral part of the rat snout that causes vibrissae protraction. NLP motoneurons receive direct input from the preBötC, but the identity of the responsible preBötC neurons is not fully elucidated. The aim of this study was to identify this preBötC neurons sub‐group based on their axonal projections and physiological function. To achieve this, we developed a cre‐recombinase (Cre) dependent adeno‐associated virus (AAV) that expressed the light‐activated chloride channel (GtACR2) predominantly in the soma (AAV‐DIO‐GtACR2‐muGFP‐Kv2.1). To validate this new AAV, it was injected with AAV‐Cre‐TdTomato into the preBötC of Sprague‐Dawley rats (60‐80 g, n=2). Three weeks later, rats were anesthetised with intravenous urethane, and blood pressure (BP), heart rate (HR), diaphragm (dEMG) and NLP electromyography (EMG) were recorded. Optical fibers were bilaterally implanted into the preBötC, and 470 nm light (50 Hz, 5 ms pulse, 20 mW) delivered for 30 s to photoinhibit neurons. The results are expressed as mean ± SD. We observed effective Cre recombination, resulting in GtACR2‐muGFP expression in the preBötC. During photoinhibition of preBötC, dEMG initially ceased, but this did not last for the entire photoinhibition period (8 ± 2 s); HR decreased (ΔHR: ‐10 ± 1.2% from baseline); and BP decreased during apnea (ΔMAP: ‐10 ± 10 % from baseline), but returned toward baseline levels as breathing resumed. Under baseline conditions, activity of the NLP occurred almost exclusively during inspiration, with a limited pre‐I component activity. All NLP EMG activity ceased during initial photoinhibition and then returned with activity in the dEMG. During the photoinhibition, when breathing had resumed, we observed an enlargement of the duration of the pre‐I component of NLP (0.27 ± 0.02 vs. baseline: 0.06 ± 0.0004 s). To selectively inhibit preBötC neurons that project to the NLP motoneurons, we injected a retrogradely transporting AAV expressing Cre (AAVrg‐mCherry‐Cre) into the lateral edge of the 7n, and AAV‐DIO‐GtACR2‐muGFP‐Kv2.1 into the preBötC (n=2). Three weeks later, we performed the same experimental protocol as described above. Postmortem histology showed strong muGFP expression in a sub‐group of preBötC neurons. Selective photoinhibition of these preBötC neurons that project to the 7n neurons reduced NLP activity (Δ amplitude: ‐42 ± 6% from baseline) and the pre‐I component (0.03 ± 0.02 vs. baseline: 0.08 ± 0.001 s s), without affecting breathing, BP or HR. In conclusion, beyond validating a new optogenetic tool, our results suggest that separate populations of preBötC neurons, with coordinated respiratory rhythmicity, drive dEMG and NLP EMG activity.