While breathing seems to come easy, it is a complex process in which many muscles coordinate to allow air to flow into the lungs. These muscles also control the flow of air we breathe out to allow us to talk, sing, eat, or drink. The brain circuits that control these muscles, can also influence other parts of the brain.

The preBötzinger Complex, which is a key region of brainstem circuits that generate and control breathing, contains neurons that also project widely, connecting to other regions of the brain. This helps to modulate the sense of smell, emotional state, heart rate, and even blood pressure. Understanding how the preBötzinger Complex is organized can untangle how breathing can influence these other processes.

Melo et al. wanted to learn whether they could manipulate the activity of a subgroup of preBötzinger Complex neurons that project into the facial nucleus – a region of the brain that controls the muscles of the face when we breathe – without affecting breathing. If this can be done, it might also be possible to affect blood pressure by manipulating selective preBötzinger neurons, and thus the development of hypertension, without having any impact on breathing.

To test this hypothesis, Melo et al. used rats in which the activation of preBötzinger Complex neurons that project into the facial nucleus was blocked. This decreased the activity of the muscles around the nose with hardly any effect on breathing. Melo et al. also found that the state of consciousness of the rat (anesthetized or conscious) could affect how preBötzinger Complex neurons control these muscles.

Melo et al. also observed that preBötzinger Complex neurons projecting into the facial nucleus had projections into many other regions in the brainstem. This might help to the coordinate respiratory, cardiovascular, orofacial, and potentially other physiological functions.

The findings of Melo et al. set a technical foundation for exploring the influence of specific subgroups of preBötzinger Complex neurons on respiratory modulation of other physiological activities, including blood pressure and heart rate and in conditions, such as hypertension and heart failure. More broadly, most brain regions contain complex and heterogeneous groups of neurons and the strategy validated by Melo et. al. could be applied to unravel other brain-function relationships.

Breathing is a complex behavior requiring the coordination of multiple motor patterns in the face, upper airways, thorax, and abdomen to enable gas exchange between the external and internal environment and maintenance of blood gas homeostasis (Del Negro et al., 2018). Strict coordination of breathing with the activity of facial and upper airway muscles underlies the rhythmic orofacial behaviors, such as sniffing, sucking, licking, whisking, vocalization, and swallowing, which are fundamental for ingestive and exploratory behavior, and social interaction (Deschênes et al., 2016; Deschênes et al., 2015; Huff et al., 2022; Moore et al., 2014; Takatoh et al., 2022). Breathing-related information permeates the brain with respiratory modulation of suprapontine brain structures affecting cognitive function and emotional expression (Ashhad et al., 2022; Lavretsky and Feldman PhD, 2021; Yackle et al., 2017). Similarly, coordination of this motor activity with autonomic nervous system activity produces phasic fluctuations of blood pressure (BP) and heart rate (HR) that are essential for optimizing cardiac efficiency and organ blood flow (Fisher et al., 2022; Menuet et al., 2020; O’Callaghan et al., 2020; Shanks et al., 2022).

The network of neurons coordinating breathing is distributed throughout the brainstem, with a major group responsible for generating inspiration located in the pre-Bötzinger complex (preBötC) within the ventrolateral medulla oblongata (Smith et al., 1991). In rodents, the preBötC consists of a heterogeneous population of ~3000 neurons with distinct, but intermingled, excitatory and inhibitory subgroups (Ashhad and Feldman, 2020; Del Negro et al., 2018). Anatomical studies showed that preBötC sends parallel excitatory and inhibitory projections to target nuclei throughout the brain, including to regions responsible for respiratory motor activity (Yang and Feldman, 2018).

As indicated by their extensive projections, preBötC neurons modulate many circuits in addition to those required for breathing activity. For example, preBötC neurons project monosynaptically to key medullary centers for autonomic regulation, and modulate cardiorespiratory coupling (Dempsey et al., 2017; Menuet et al., 2020; Menuet et al., 2017). Photoinhibition of preBötC neurons produces the expected apnea, but also causes a significant reduction of sympathetic vasomotor activity, reduces respiratory modulation of BP, and increases cardiac parasympathetic nerve activity to decrease HR and respiratory-sinus arrhythmia (Menuet et al., 2020). Neurons in the preBötC also modulate orofacial behaviors. Activation of the developing brain homeobox1 (Dbx1)-expressing preBötC neurons alters the networks responsible for the coordination of upper airways and swallowing, to decrease the probability of swallowing during inspiration (Huff et al., 2022). The preBötC also contributes to motor patterns that are necessary for exploratory behaviors, such as sniffing and whisking. In this case, a group of inhibitory preBötC neurons provide monosynaptic inputs onto the vibrissal premotor neurons in the intermediate reticular nucleus (vIRt) to facilitate synchronous whisking. In addition, preBötC contains facial premotor neurons that modulate nasal dilation and mystacial pad (MP) muscle activity to couple whisking to the breathing rhythm (Deschênes et al., 2016; Takatoh et al., 2022).

Understanding and identifying the preBötC subgroups that coordinate and integrate breathing with different motor and physiological behaviors is of fundamental importance, given that abnormal respiratory modulation of autonomic activity and orofacial behaviors has been associated with the development and progression of diseases (El-Omar et al., 2001; Huff et al., 2022; Menuet et al., 2017; Simms et al., 2009). Although several studies have attempted to understand the physiological function/s of each preBötC neuronal subpopulation, disturbances in breathing, that occur when preBötC neurons are either inhibited or activated, impact our understanding of whether the observed changes are due to the direct effects of preBötC neurons or alterations resulting from changes in the respiratory rhythm and blood gases (Cui et al., 2016; Huff et al., 2022; Menuet et al., 2020; Tan et al., 2008; Yackle et al., 2017). Furthermore, whilst substantial gains have been made in understanding the expression of selective molecular markers in subgroups of preBötC neurons, it is rarely possible to reliably assign particular neurochemical signatures to unique functions (Ashhad and Feldman, 2020; Del Negro et al., 2018; Yackle et al., 2017). For example, while Dbx1⁺somatostatin (SST)⁺ predominantly affect the breathing pattern (Cui et al., 2016; Del Negro et al., 2018; Ashhad and Feldman, 2020), some Dbx1⁺SST⁺ which also express the neurokinin-1 receptor (NK1R) and, in some cases, cadherin-9 (Cdh9+), promote generalized behavioral arousal (Yackle et al., 2017).

To further understand the organization of the preBötC, in this study, we have tested the hypothesis that the preBötC region is composed of segregated subgroups of output neurons, potentially driven by a separate group of rhythmogenic neurons, that modulate specific behaviors, such as orofacial muscle activity. If correct, we predict that selective photoinhibition of preBötC neurons projecting to the facial nucleus (7n) should modulate orofacial behavior with minimal impact on respiratory motor activity. To test this hypothesis, we utilized a combinatorial viral transgenic approach with one virus providing cre-recombinase (Cre)-dependent expression of an optically activated chloride channel, GtACR2, and another retrograde axonal delivery of Cre. We showed that selective transduction and photoinhibition of preBötC facial premotor neurons affects the MP activity while minimally affecting breathing.

We employed a combinatorial Cre-dependent approach to transduce a subgroup of preBötC neurons based on their axonal projections to the facial nucleus. The preBötC is the kernel for breathing (Smith et al., 1991), but also acts as a master oscillator controlling cardiovascular and orofacial activities (Del Negro et al., 2018; Deschênes et al., 2016; Huff et al., 2022; Takatoh et al., 2022), and our study aimed to determine whether these functions involved independent subgroups of preBötC neurons. We validated our approach by co-injections of a Cre virus and the Cre-dependent anion channel virus - non-selective preBötC transduction. In agreement with previous results (Menuet et al., 2020), non-selective photoinhibition of preBötC neurons induced apnea, bradycardia, and biphasic effects on BP. We also observed loss of inspiratory modulation of mystacial pad activity, with increased tonic activity, except in rats anesthetized with ketamine/medetomidine where there was no ongoing mystacial pad activity. By contrast, selective photoinhibition of preBötC→7n neurons had minimal effect on breathing or cardiovascular activity, but altered mystacial pad activity with the effect altered by the type of anesthetic or state. The preBötC→7n neurons showed a restricted anatomical distribution within the preBötC but sent substantial collateral projections to several areas in the brainstem involved with the cardiorespiratory regulation. We conclude that these collaterals are functionally active and responsible for the small effects of photoinhibition of preBötC→7n neurons on BP, HR, and respiratory frequency. Both selective preBötC→7n neuron transduction and non-selective preBötC transduction resulted in transgene expression in both inhibitory and glutamatergic neurons.

The observation that photoinhibition of preBötC→7n neurons significantly affects mystacial pad activity, without affecting breathing, BP, or HR, clearly demonstrates that modulation of preBötC subgroups of neurons based on their axonal projections is a useful strategy. It allows an understanding of the anatomical distribution and neurochemical phenotype of subgroups of preBötC neurons. It also enables assessment of the physiological relevance of these subgroups without interference and potentially the development of secondary physiological changes that occur with profound interruption of breathing. The compartmentalization of the preBötC into segregated subgroups of neurons based on their connections impacts our comprehension of mechanisms that coordinate and integrate breathing with different motor and physiological behaviors. This is of fundamental importance, given that abnormal respiratory modulation of autonomic activity and orofacial behaviors have been associated with the development and progression of diseases (El-Omar et al., 2001; Huff et al., 2022; Menuet et al., 2017; Simms et al., 2009). For example, exaggerated respiratory-sympathetic modulation seems to be an important mechanism for the development and progression of hypertension (Barnett et al., 2020; Menuet et al., 2017; Moraes et al., 2014; Simms et al., 2009; Tatasciore et al., 2007). Reduced inspiratory modulation of HR is associated with the severity of diverse pathological conditions such as hypertension, heart failure, depression, and anxiety (El-Omar et al., 2001; Masi et al., 2007; Thayer et al., 2012). In addition, loss of the synchronization between breathing and orofacial and oropharyngeal behaviors, such as swallowing, has significant clinical implications, as it increases the risk of aspiration pneumonia associated with dysphagia, which can lead to death (Beard et al., 1996; Heemskerk and Roos, 2012; Huff et al., 2022).

Inhibition of preBötC→7n only transiently affects the mystacial pad activity despite the sustained laser light delivery. These results are similar to our previous study, which showed that non-selective photoinhibition of preBötC neurons transiently affected inspiration and sympathetic nerve activity, while the bradycardia, induced by vagal activation, was retained for the length of light delivery (Menuet et al., 2020). We do not know the mechanisms underlying this phenomenon and further investigation is needed. We can only speculate that intrinsic biophysical properties, such as maintenance of the chloride gradient, may differ between distinct neuronal populations, enabling some neurons to escape prolonged photoinhibition.

Nasofacial and orofacial activity are sensitive to anesthesia and animal state (Deschênes et al., 2016; Deschênes et al., 2015). Naris dilation and whisker motion are silent when animals are deeply anesthetized with ketamine/medetomidine but resume in synchrony with the inspiratory phase of the respiratory cycle, as rats recover from anesthesia (Deschênes et al., 2015). We also observed that mystacial pad activity is minimal and non-rhythmic under surgical anesthesia with ketamine/medetomidine, but resumes, with inspiratory modulation, as the animal recovers consciousness. In contrast, strong inspiratory-related mystacial pad activity occurs when surgical anesthesia is provided by urethane. In both cases, surgical anesthesia was defined by loss of the pedal withdrawal and corneal reflexes. The mechanisms by which deep ketamine and urethane anesthesia differentially affect respiratory-related motor activity are unknown. Under light ketamine anesthesia, GABAergic cells of vIRt displayed phase-locked firing patterns for protraction or retraction phases of whisking, while only retraction unit activity was recorded under light urethane anesthesia, suggesting that only inhibitory vIRt cells, whose activity is correlated with vibrissa retraction, are active in urethane-anesthetized rats (Deschênes et al., 2016).

Likewise, the effect of photoinhibition of preBötC→7n neurons on mystacial pad activity is both state- and anesthetic-dependent. Under urethane anesthesia, selective photoinhibition of preBötC→7n neurons silenced mystacial pad activity, while photoinhibition during different phases of recovery from ketamine/medetomidine anesthesia interrupted the inspiratory-related and increased the overall tonic mystacial pad activity. The reason for this substantial state-dependent effect of the preBötC→7n on mystacial pad activity remains unclear, but it has been reported in other systems. Injection of L-glutamate in the NTS induces a substantial pressor response in awake normotensive rats but a depressor response in the same rats anesthetized with urethane (Machado and Bonagamba, 1992). Photoinhibition of the C1-RVLM neurons in awake rats produces a very small depressor response (Wenker et al., 2017) when compared to that observed in anesthetized animals (Marina et al., 2011). Likewise, electrical stimulation of the central amygdala induces a significant pressor response in conscious rats but a depressor response under anesthesia (Chiou et al., 2009). Changes in the activity of several fast ionotropic transmitter systems, including glycine, GABA, acetylcholine, and glutamate, have been reported in urethane anesthesia (Accorsi-Mendonça et al., 2007; Hara and Harris, 2002). While speculative, we hypothesize that under urethane, the glutamatergic preBötC→7n neurons play a major role in regulating the respiration-related mystacial pad activity, while the inhibitory neurons are more active in the conscious state.

Under urethane anesthesia, and during the early recovery phase following ketamine/medetomidine anesthesia, inspiratory-related mystacial pad activity was clearly evident. In the conscious state, heterogeneous respiratory-related mystacial pad activity was observed. The electrodes we used to record mystacial pad activity were inserted near the rostral tip of the snout, which is known to be the site of origin of the nasolabialis profundus. However, due to the large number of distinct muscles that compose the mystacial pad (Haidarliu et al., 2010), the difficulty of dissecting the rodent snout (Haidarliu et al., 2012), the substantial size of the electrode’s suture pads, and the fact that the nasolabialis profundus electromyogenic activity was not recorded prior to the electrode implantation, it is possible that the mystacial pad activity we recorded reflects the activity of multiple inspiratory- and expiratory-related muscles. Nonetheless, given that nasolabialis profundus is active during basal respiration, while extrinsic muscles are silent, and preferentially contract during sniffing and whisking (de Britto et al., 2020; Deschênes et al., 2015), we conclude that the inspiratory component of the mystacial pad activity predominantly reflects the nasolabialis profundus muscle contraction during urethane anesthesia and the early recovery phase following ketamine/medetomidine. The absence of respiratory-related activity of the mystacial pad in some conscious rats was unexpected, but could be the result of sniffing and/or whisking behavior, as indicated by the high respiratory frequency.

Non-selective inhibition of preBötC induced apnea and immediate interruption of the inspiratory-related mystacial pad activity, which increased in amplitude and became tonic. The inspiratory-related mystacial pad activity recovered when breathing resumed. This coordinated activity between breathing and orofacial/nasofacial activity has been described before (Deschênes et al., 2015). It was demonstrated that both naris dilation and vibrissae retraction are abolished during apnea induced by the application of ammonia to the snout, and their activity synchronously recovers with the resumption of breathing. Nonetheless, the neural mechanisms that underly these responses are still unclear. Non-selective photoinhibition of preBötC neurons may lead to inhibition of preBötC→7n, and at the same time, to inhibition of preBötC inhibitory neurons that project to parvalbumin-expressing inhibitory neurons of the vIRt neurons (vIRt_PV). The vIRt_PV was recently identified as the whisking oscillator (Takatoh et al., 2022), and direct projections from preBötC have been shown to reset the vIRt_PV activity (Deschênes et al., 2016; Golomb et al., 2022; Takatoh et al., 2022). The vIRt_PV neurons also receive local presynaptic inputs from non-PV-glutamatergic neurons, which are not required for whisking, and from non-PV-GABAergic neurons, whose role in whisking is unclear (Takatoh et al., 2022). The post-mortem analysis showed that the non-selective approach transduced some vIRt_PV and non-PV-vIRt neurons, while the selective transduction only sparsely transduced the non-PV-vIRt neurons. Based on this observation, it is likely that the vIRt_PV activity could have been directly affected by the laser delivery, producing desynchronization of facial motoneurons activity and suppression of the rhythmic whisking (Takatoh et al., 2022).

All data generated or analysed during this study are included in the manuscript and supporting file; Source Data files have been provided for Figures 2, 3, 8, Figures 8 -supplement 1 and 3, and Figure 9.

1. 1. Accorsi-Mendonça D
   2. Leão RM
   3. Aguiar JF
   4. Varanda WA
   5. Machado BH

   (2007) Urethane inhibits the GABAergic neurotransmission in the nucleus of the solitary tract of rat brain stem slices

   American Journal of Physiology. Regulatory, Integrative and Comparative Physiology 292:R396–R402.

   https://doi.org/10.1152/ajpregu.00776.2005

   • PubMed
   • Google Scholar
2. 1. Alheid GF
   2. Gray PA
   3. Jiang MC
   4. Feldman JL
   5. McCrimmon DR

   (2002) Parvalbumin in respiratory neurons of the ventrolateral medulla of the adult rat

   Journal of Neurocytology 31:693–717.

   https://doi.org/10.1023/a:1025799830302

   • PubMed
   • Google Scholar
3. 1. Ashhad S
   2. Feldman JL

   (2020) Emergent elements of inspiratory rhythmogenesis: network synchronization and synchrony propagation

   Neuron 106:482–497.

   https://doi.org/10.1016/j.neuron.2020.02.005

   • PubMed
   • Google Scholar
4. 1. Ashhad S
   2. Kam K
   3. Del Negro CA
   4. Feldman JL

   (2022) Breathing rhythm and pattern and their influence on emotion

   Annual Review of Neuroscience 45:223–247.

   https://doi.org/10.1146/annurev-neuro-090121-014424

   • PubMed
   • Google Scholar
5. 1. Barnett WH
   2. Latash EM
   3. Capps RA
   4. Dick TE
   5. Wehrwein EA
   6. Molkov YI

   (2020) Traube-Hering waves are formed by interaction of respiratory sinus arrhythmia and pulse pressure modulation in healthy men

   Journal of Applied Physiology 129:1193–1202.

   https://doi.org/10.1152/japplphysiol.00452.2020

   • PubMed
   • Google Scholar
6. 1. Bassi JK
   2. Connelly AA
   3. Butler AG
   4. Liu Y
   5. Ghanbari A
   6. Farmer DGS
   7. Jenkins MW
   8. Melo MR
   9. McDougall SJ
   10. Allen AM

   (2022) Analysis of the distribution of vagal afferent projections from different peripheral organs to the nucleus of the solitary tract in rats

   The Journal of Comparative Neurology 530:3072–3103.

   https://doi.org/10.1002/cne.25398

   • PubMed
   • Google Scholar
7. 1. Beard CM
   2. Kokmen E
   3. Sigler C
   4. Smith GE
   5. Petterson T
   6. O’Brien PC

   (1996) Cause of death in Alzheimer’s disease

   Annals of Epidemiology 6:195–200.

   https://doi.org/10.1016/1047-2797(95)00068-2

   • Google Scholar
8. 1. Butler AG
   2. O’Callaghan EL
   3. Allen AM
   4. McDougall SJ

   (2021) Use of a physiological reflex to standardize vagal nerve stimulation intensity improves data reproducibility in a memory extinction assay

   Brain Stimulation 14:450–459.

   https://doi.org/10.1016/j.brs.2021.02.012

   • PubMed
   • Google Scholar
9. 1. Chiou R-J
   2. Kuo C-C
   3. Liang K-C
   4. Yen C-T

   (2009) State-dependent amygdala stimulation-induced cardiovascular effects in rats

   The Chinese Journal of Physiology 52:432–440.

   https://doi.org/10.4077/cjp.2009.amh046

   • PubMed
   • Google Scholar
10. 1. Cui Y
    2. Kam K
    3. Sherman D
    4. Janczewski WA
    5. Zheng Y
    6. Feldman JL

    (2016) Defining prebötzinger complex rhythm- and pattern-generating neural microcircuits in vivo

    Neuron 91:602–614.

    https://doi.org/10.1016/j.neuron.2016.07.003

    • PubMed
    • Google Scholar
11. 1. de Britto AA
    2. Magalhães KS
    3. da Silva MP
    4. Paton JFR
    5. Moraes DJA

    (2020) Active expiratory oscillator regulates nasofacial and oral motor activities in rats

    Experimental Physiology 105:379–392.

    https://doi.org/10.1113/EP088046

    • PubMed
    • Google Scholar
12. 1. Del Negro CA
    2. Funk GD
    3. Feldman JL

    (2018) Breathing matters

    Nature Reviews. Neuroscience 19:351–367.

    https://doi.org/10.1038/s41583-018-0003-6

    • PubMed
    • Google Scholar
13. 1. Dempsey B
    2. Le S
    3. Turner A
    4. Bokiniec P
    5. Ramadas R
    6. Bjaalie JG
    7. Menuet C
    8. Neve R
    9. Allen AM
    10. Goodchild AK
    11. McMullan S

    (2017) Mapping and analysis of the connectome of sympathetic premotor neurons in the rostral ventrolateral medulla of the rat using a volumetric brain atlas

    Frontiers in Neural Circuits 11:9.

    https://doi.org/10.3389/fncir.2017.00009

    • PubMed
    • Google Scholar
14. 1. Deschênes M
    2. Haidarliu S
    3. Demers M
    4. Moore J
    5. Kleinfeld D
    6. Ahissar E

    (2015) Muscles involved in naris dilation and nose motion in rat

    Anatomical Record 298:546–553.

    https://doi.org/10.1002/ar.23053

    • PubMed
    • Google Scholar
15. 1. Deschênes M
    2. Takatoh J
    3. Kurnikova A
    4. Moore JD
    5. Demers M
    6. Elbaz M
    7. Furuta T
    8. Wang F
    9. Kleinfeld D

    (2016) Inhibition, not excitation, drives rhythmic whisking

    Neuron 90:374–387.

    https://doi.org/10.1016/j.neuron.2016.03.007

    • PubMed
    • Google Scholar
16. 1. El-Omar M
    2. Kardos A
    3. Casadei B

    (2001) Mechanisms of respiratory sinus arrhythmia in patients with mild heart failure

    American Journal of Physiology. Heart and Circulatory Physiology 280:H125–H131.

    https://doi.org/10.1152/ajpheart.2001.280.1.H125

    • PubMed
    • Google Scholar
17. 1. Fisher JP
    2. Zera T
    3. Paton JFR

    (2022) Respiratory-cardiovascular interactions

    Clinical Neurology 188:279–308.

    https://doi.org/10.1016/B978-0-323-91534-2.00006-0

    • PubMed
    • Google Scholar
18. 1. Ganella DE
    2. Callander GE
    3. Ma S
    4. Bye CR
    5. Gundlach AL
    6. Bathgate RAD

    (2013) Modulation of feeding by chronic rAAV expression of a relaxin-3 peptide agonist in rat hypothalamus

    Gene Therapy 20:703–716.

    https://doi.org/10.1038/gt.2012.83

    • PubMed
    • Google Scholar
19. 1. Golomb D
    2. Moore JD
    3. Fassihi A
    4. Takatoh J
    5. Prevosto V
    6. Wang F
    7. Kleinfeld D

    (2022) Theory of hierarchically organized neuronal oscillator dynamics that mediate rodent rhythmic whisking

    Neuron 110:3833–3851.

    https://doi.org/10.1016/j.neuron.2022.08.020

    • PubMed
    • Google Scholar
20. 1. Grimm D
    2. Kay MA
    3. Kleinschmidt JA

    (2003) Helper virus-free, optically controllable, and two-plasmid-based production of adeno-associated virus vectors of serotypes 1 to 6

    Molecular Therapy 7:839–850.

    https://doi.org/10.1016/s1525-0016(03)00095-9

    • PubMed
    • Google Scholar
21. 1. Haidarliu S
    2. Simony E
    3. Golomb D
    4. Ahissar E

    (2010) Muscle architecture in the mystacial pad of the rat

    Anatomical Record 293:1192–1206.

    https://doi.org/10.1002/ar.21156

    • PubMed
    • Google Scholar
22. 1. Haidarliu S
    2. Golomb D
    3. Kleinfeld D
    4. Ahissar E

    (2012) Dorsorostral snout muscles in the rat subserve coordinated movement for whisking and sniffing

    Anatomical Record 295:1181–1191.

    https://doi.org/10.1002/ar.22501

    • PubMed
    • Google Scholar
23. 1. Hara K
    2. Harris RA

    (2002) The anesthetic mechanism of urethane: the effects on neurotransmitter-gated ion channels

    Anesthesia and Analgesia 94:313–318.

    https://doi.org/10.1097/00000539-200202000-00015

    • PubMed
    • Google Scholar
24. 1. Heemskerk AW
    2. Roos RAC

    (2012) Aspiration pneumonia and death in Huntington’s disease

    PLOS Currents 4:RRN1293.

    https://doi.org/10.1371/currents.RRN1293

    • PubMed
    • Google Scholar
25. 1. Huff A
    2. Karlen-Amarante M
    3. Pitts T
    4. Ramirez JM

    (2022) Optogenetic stimulation of pre-Bötzinger complex reveals novel circuit interactions in swallowing-breathing coordination

    PNAS 119:e2121095119.

    https://doi.org/10.1073/pnas.2121095119

    • PubMed
    • Google Scholar
26. 1. Koizumi H
    2. Koshiya N
    3. Chia JX
    4. Cao F
    5. Nugent J
    6. Zhang R
    7. Smith JC

    (2013) Structural-functional properties of identified excitatory and inhibitory interneurons within pre-Botzinger complex respiratory microcircuits

    The Journal of Neuroscience 33:2994–3009.

    https://doi.org/10.1523/JNEUROSCI.4427-12.2013

    • PubMed
    • Google Scholar
27. 1. Lavretsky H
    2. Feldman PhD JL

    (2021) Precision medicine for breath-focused mind-body therapies for stress and anxiety: are we ready yet?

    Global Advances in Health and Medicine 10:2164956120986129.

    https://doi.org/10.1177/2164956120986129

    • PubMed
    • Google Scholar
28. 1. Li SJ
    2. Vaughan A
    3. Sturgill JF
    4. Kepecs A

    (2018) A viral receptor complementation strategy to overcome CAV-2 tropism for efficient retrograde targeting of neurons

    Neuron 98:905–917.

    https://doi.org/10.1016/j.neuron.2018.05.028

    • PubMed
    • Google Scholar
29. 1. Lim ST
    2. Antonucci DE
    3. Scannevin RH
    4. Trimmer JS

    (2000) A novel targeting signal for proximal clustering of the Kv2.1 K+ channel in hippocampal neurons

    Neuron 25:385–397.

    https://doi.org/10.1016/s0896-6273(00)80902-2

    • PubMed
    • Google Scholar
30. 1. Ma S
    2. Allocca G
    3. Ong-Pålsson EKE
    4. Singleton CE
    5. Hawkes D
    6. McDougall SJ
    7. Williams SJ
    8. Bathgate RAD
    9. Gundlach AL

    (2017) Nucleus incertus promotes cortical desynchronization and behavioral arousal

    Brain Structure & Function 222:515–537.

    https://doi.org/10.1007/s00429-016-1230-0

    • PubMed
    • Google Scholar
31. 1. Machado BH
    2. Bonagamba LGH

    (1992) Microinjection of L-glutamate into the nucleus tractus solitarii increases arterial pressure in conscious rats

    Brain Research 576:131–138.

    https://doi.org/10.1016/0006-8993(92)90618-j

    • PubMed
    • Google Scholar
32. 1. Magalhães KS
    2. da Silva MP
    3. Mecawi AS
    4. Paton JFR
    5. Machado BH
    6. Moraes DJA

    (2021) Intrinsic and synaptic mechanisms controlling the expiratory activity of excitatory lateral parafacial neurones of rats

    The Journal of Physiology 599:4925–4948.

    https://doi.org/10.1113/JP281545

    • PubMed
    • Google Scholar
33. 1. Mahn M
    2. Gibor L
    3. Patil P
    4. Cohen-Kashi Malina K
    5. Oring S
    6. Printz Y
    7. Levy R
    8. Lampl I
    9. Yizhar O

    (2018) High-efficiency optogenetic silencing with soma-targeted anion-conducting channelrhodopsins

    Nature Communications 9:4125.

    https://doi.org/10.1038/s41467-018-06511-8

    • PubMed
    • Google Scholar
34. 1. Marina N
    2. Abdala APL
    3. Korsak A
    4. Simms AE
    5. Allen AM
    6. Paton JFR
    7. Gourine AV

    (2011) Control of sympathetic vasomotor tone by catecholaminergic C1 neurones of the rostral ventrolateral medulla oblongata

    Cardiovascular Research 91:703–710.

    https://doi.org/10.1093/cvr/cvr128

    • PubMed
    • Google Scholar
35. 1. Masi CM
    2. Hawkley LC
    3. Rickett EM
    4. Cacioppo JT

    (2007) Respiratory sinus arrhythmia and diseases of aging: obesity, diabetes mellitus, and hypertension

    Biological Psychology 74:212–223.

    https://doi.org/10.1016/j.biopsycho.2006.07.006

    • PubMed
    • Google Scholar
36. 1. Menuet C
    2. Le S
    3. Dempsey B
    4. Connelly AA
    5. Kamar JL
    6. Jancovski N
    7. Bassi JK
    8. Walters K
    9. Simms AE
    10. Hammond A
    11. Fong AY
    12. Goodchild AK
    13. McMullan S
    14. Allen AM

    (2017) Excessive respiratory modulation of blood pressure triggers hypertension

    Cell Metabolism 25:739–748.

    https://doi.org/10.1016/j.cmet.2017.01.019

    • PubMed
    • Google Scholar
37. 1. Menuet C
    2. Connelly AA
    3. Bassi JK
    4. Melo MR
    5. Le S
    6. Kamar J
    7. Kumar NN
    8. McDougall SJ
    9. McMullan S
    10. Allen AM

    (2020) PreBötzinger complex neurons drive respiratory modulation of blood pressure and heart rate

    eLife 9:e57288.

    https://doi.org/10.7554/eLife.57288

    • PubMed
    • Google Scholar
38. 1. Messerli FH
    2. Hofstetter L
    3. Rimoldi SF
    4. Rexhaj E
    5. Bangalore S

    (2019) Risk factor variability and cardiovascular outcome

    Journal of the American College of Cardiology 73:2596–2603.

    https://doi.org/10.1016/j.jacc.2019.02.063

    • Google Scholar
39. 1. Messier JE
    2. Chen H
    3. Cai ZL
    4. Xue M

    (2018) Targeting light-gated chloride channels to neuronal somatodendritic domain reduces their excitatory effect in the axon

    eLife 7:e38506.

    https://doi.org/10.7554/eLife.38506

    • PubMed
    • Google Scholar
40. 1. Moore JD
    2. Kleinfeld D
    3. Wang F

    (2014) How the brainstem controls orofacial behaviors comprised of rhythmic actions

    Trends in Neurosciences 37:370–380.

    https://doi.org/10.1016/j.tins.2014.05.001

    • PubMed
    • Google Scholar
41. 1. Moraes DJA
    2. Machado BH
    3. Paton JFR

    (2014) Specific respiratory neuron types have increased excitability that drive presympathetic neurones in neurogenic hypertension

    Hypertension 63:1309–1318.

    https://doi.org/10.1161/HYPERTENSIONAHA.113.02283

    • PubMed
    • Google Scholar
42. 1. Ngo HB
    2. Melo MR
    3. Layfield S
    4. Connelly AA
    5. Bassi JK
    6. Xie L
    7. Menuet C
    8. McDougall SJ
    9. Bathgate RAD
    10. Allen AM

    (2020) A chemogenetic tool that enables functional neural circuit analysis

    Cell Reports 32:108139.

    https://doi.org/10.1016/j.celrep.2020.108139

    • PubMed
    • Google Scholar
43. 1. O’Callaghan EL
    2. Lataro RM
    3. Roloff EL
    4. Chauhan AS
    5. Salgado HC
    6. Duncan E
    7. Nogaret A
    8. Paton JFR

    (2020) Enhancing respiratory sinus arrhythmia increases cardiac output in rats with left ventricular dysfunction

    The Journal of Physiology 598:455–471.

    https://doi.org/10.1113/JP277293

    • PubMed
    • Google Scholar
44. 1. Pagliardini S
    2. Janczewski WA
    3. Tan W
    4. Dickson CT
    5. Deisseroth K
    6. Feldman JL

    (2011) Active expiration induced by excitation of ventral medulla in adult anesthetized rats

    The Journal of Neuroscience 31:2895–2905.

    https://doi.org/10.1523/JNEUROSCI.5338-10.2011

    • PubMed
    • Google Scholar
45. 1. Palkovits M

    (1983) The rat brain in stereotaxic coordinates

    Neuropeptides 3:319.

    https://doi.org/10.1016/0143-4179(83)90049-5

    • Google Scholar
46. 1. Scott DJ
    2. Gunn NJ
    3. Yong KJ
    4. Wimmer VC
    5. Veldhuis NA
    6. Challis LM
    7. Haidar M
    8. Petrou S
    9. Bathgate RAD
    10. Griffin MDW

    (2018) A novel ultra-stable, monomeric green fluorescent protein for direct volumetric imaging of whole organs using CLARITY

    Scientific Reports 8:667.

    https://doi.org/10.1038/s41598-017-18045-y

    • PubMed
    • Google Scholar
47. 1. Shanks J
    2. Abukar Y
    3. Lever NA
    4. Pachen M
    5. LeGrice IJ
    6. Crossman DJ
    7. Nogaret A
    8. Paton JFR
    9. Ramchandra R

    (2022) Reverse re-modelling chronic heart failure by reinstating heart rate variability

    Basic Research in Cardiology 117:4.

    https://doi.org/10.1007/s00395-022-00911-0

    • PubMed
    • Google Scholar
48. 1. Simms AE
    2. Paton JFR
    3. Pickering AE
    4. Allen AM

    (2009) Amplified respiratory-sympathetic coupling in the spontaneously hypertensive rat: does it contribute to hypertension?

    The Journal of Physiology 587:597–610.

    https://doi.org/10.1113/jphysiol.2008.165902

    • PubMed
    • Google Scholar
49. 1. Smith JC
    2. Ellenberger HH
    3. Ballanyi K
    4. Richter DW
    5. Feldman JL

    (1991) Pre-Bötzinger complex: a brainstem region that may generate respiratory rhythm in mammals

    Science 254:726–729.

    https://doi.org/10.1126/science.1683005

    • PubMed
    • Google Scholar
50. 1. Takatoh J
    2. Nelson A
    3. Zhou X
    4. Bolton MM
    5. Ehlers MD
    6. Arenkiel BR
    7. Mooney R
    8. Wang F

    (2013) New modules are added to vibrissal premotor circuitry with the emergence of exploratory whisking

    Neuron 77:346–360.

    https://doi.org/10.1016/j.neuron.2012.11.010

    • PubMed
    • Google Scholar
51. 1. Takatoh J
    2. Prevosto V
    3. Thompson PM
    4. Lu J
    5. Chung L
    6. Harrahill A
    7. Li S
    8. Zhao S
    9. He Z
    10. Golomb D
    11. Kleinfeld D
    12. Wang F

    (2022) The whisking oscillator circuit

    Nature 609:560–568.

    https://doi.org/10.1038/s41586-022-05144-8

    • PubMed
    • Google Scholar
52. 1. Tan W
    2. Janczewski WA
    3. Yang P
    4. Shao XM
    5. Callaway EM
    6. Feldman JL

    (2008) Silencing preBötzinger complex somatostatin-expressing neurons induces persistent apnea in awake rat

    Nature Neuroscience 11:538–540.

    https://doi.org/10.1038/nn.2104

    • PubMed
    • Google Scholar
53. 1. Tatasciore A
    2. Renda G
    3. Zimarino M
    4. Soccio M
    5. Bilo G
    6. Parati G
    7. Schillaci G
    8. De Caterina R

    (2007) Awake systolic blood pressure variability correlates with target-organ damage in hypertensive subjects

    Hypertension 50:325–332.

    https://doi.org/10.1161/HYPERTENSIONAHA.107.090084

    • PubMed
    • Google Scholar
54. 1. Thayer JF
    2. Åhs F
    3. Fredrikson M
    4. Sollers JJ III
    5. Wager TD

    (2012) A meta-analysis of heart rate variability and neuroimaging studies: Implications for heart rate variability as A marker of stress and health

    Neuroscience & Biobehavioral Reviews 36:747–756.

    https://doi.org/10.1016/j.neubiorev.2011.11.009

    • Google Scholar
55. 1. Van Duyne GD

    (2015) Cre Recombinase

    Microbiology Spectrum 3:MDNA3-0014.

    https://doi.org/10.1128/microbiolspec.MDNA3-0014-2014

    • PubMed
    • Google Scholar
56. 1. Wenker IC
    2. Abe C
    3. Viar KE
    4. Stornetta DS
    5. Stornetta RL
    6. Guyenet PG

    (2017) Blood pressure regulation by the rostral ventrolateral medulla in conscious rats: effects of hypoxia, hypercapnia, baroreceptor denervation, and anesthesia

    The Journal of Neuroscience 37:4565–4583.

    https://doi.org/10.1523/JNEUROSCI.3922-16.2017

    • PubMed
    • Google Scholar
57. 1. Wu J
    2. Capelli P
    3. Bouvier J
    4. Goulding M
    5. Arber S
    6. Fortin G

    (2017) A V0 core neuronal circuit for inspiration

    Nature Communications 8:544.

    https://doi.org/10.1038/s41467-017-00589-2

    • Google Scholar
58. 1. Yackle K
    2. Schwarz LA
    3. Kam K
    4. Sorokin JM
    5. Huguenard JR
    6. Feldman JL
    7. Luo L
    8. Krasnow MA

    (2017) Breathing control center neurons that promote arousal in mice

    Science 355:1411–1415.

    https://doi.org/10.1126/science.aai7984

    • Google Scholar
59. 1. Yang CF
    2. Feldman JL

    (2018) Efferent projections of excitatory and inhibitory preBötzinger Complex neurons

    The Journal of Comparative Neurology 526:1389–1402.

    https://doi.org/10.1002/cne.24415

    • PubMed
    • Google Scholar
60. 1. Zolotukhin S
    2. Byrne BJ
    3. Mason E
    4. Zolotukhin I
    5. Potter M
    6. Chesnut K
    7. Summerford C
    8. Samulski RJ
    9. Muzyczka N

    (1999) Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield

    Gene Therapy 6:973–985.

    https://doi.org/10.1038/sj.gt.3300938

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Selective transduction and photoinhibition of pre-botzinger neurons that project to the facial nucleus in rats affect the nasofacial activity" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, including Jeffrey C Smith as Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Andrew King as the Senior Editor. The following individual involved in the review of your submission has agreed to reveal their identity: Muriel Thoby-Brisson (Reviewer #2).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

1) As indicated by the reviewers, the authors should use traditional markers as a counterstain of the preBötC region to confirm the anatomical location. How can the authors be sure that they have transduced all preBötC neurons in the region ventral and caudal to the NAc, given that preBötC neurons are not definitively defined? Also please note that the coronal section shown in Figure supplement 1B appears to be caudal to the medullary levels containing the preBötC, and hence does not appear to illustrate distinct CTB retrogradely labeled preBötC subpopulations. Additional data addressing the above should be included.

2) The authors should quantify the expression of mCherry within GFP positive neurons as a control for the specificity of the CN7 projecting neurons by ruling out Cre-independent viral expression, as indicated by Reviewer #3. Please include this data.

3) There are cells non-selectively and selectively transduced by the vector combination in the vLRt (Figure 2—figure supplement 2C and D and Figure 3—figure supplement 1C and D), raising the possibility that some of the perturbations of MP activity might be attributable to photoinhibition of these cells, which is worth comment in the text.

4) The authors should explain why some preBötC neurons express muGFP that were not VGluT2⁺ or VGAT+ (Figure 4B and F).

5) The authors should discuss why the inhibition of MP EMG activity is only transient during sustained photoinhibition at the very high laser power employed.

6) The relationship between breathing and the change in mystacial activity in Figures 2, 3, 7, and 8 is not directly quantified. One possible measurement to show a loss of synchrony would be the cross-correlation between signals for breathing and mystacial muscle activity, as in Kurnikova 2017.

7) The authors should discuss the issues regarding alternative interpretations of the data raised by Reviewer #3.

8) The authors should address all other questions/issues raised in the major concerns of Reviewers #2 and 3. Additional experiments/data are not required here.

Reviewer #1 (Recommendations for the authors):

Concerns to be addressed:

1) The authors should discuss why the inhibition of MP EMG activity is only transient during sustained photoinhibition at the very high laser power employed.

2) There are cells non-selectively and selectively transduced by the vector combination in the vLRt (Figure 2—figure supplement 2C and D and Figure 3¬ figure supplement 1C and D), raising the possibility that some of the perturbations of MP activity might be attributable to photoinhibition of these cells, which is worth comment.

3) The authors should explain why some preBötC neurons express muGFP that were not VGluT2⁺ or VGAT+ (Figure 4B and F).

Reviewer #2 (Recommendations for the authors):

While the study is very elegant and the illustrations convincing, I have three concerns:

1. It remains unclear to me why the authors consider their neurons of interest (identified based on their axonal projections) as elements of the preBötzinger network. It might just be a problem of semantics: With the term preBötC, are we talking about a region or a network? Are we talking about a function (breathing) or an anatomical location? For instance are the neurons labelled here Dbx1-derived, as is the case for preBötC neurons involved in breathing (Bouvier et al., 2010; Gray et al., 2010)? Are they, for example, somatostatin or NK1R positive? In the present version of the manuscript, it seems that the identification of neurons as being part of the preBötC is only based on their anatomical location. Could the authors provide additional arguments in favor (or not) for these neurons to express any known features of the respiratory preBötC neurons? Or do the authors consider that the neurons of interest are located in the same region as respiratory preBötC neurons but constitute a completely different type of neurons? This should be clarified and addressed in the discussion. More generally, it actually becomes a point that should be addressed in the field, as several labs are identifying small networks in the brainstem that control distinct motor functions all more or less coordinated with breathing behavior. So are we investigating one large network composed of several intermingled small networks or are they really distinct neuronal groups? What is now the definition of the preBötC?

2. The authors provide evidence that the neurons of interest are a mixture of excitatory and inhibitory phenotypes. The photo-inhibition targets indistinctly either type. It is obvious to expect opposite effects when inhibiting inhibitory neurons versus excitatory neurons. So how do the authors explain the overall effects? Which sub-population (excitatory or inhibitory) is mainly involved in this effect? Are inhibitory and excitatory neurons involved in controlling the same orofacial muscles? Are they sending collaterals to the same brainstem nuclei? This point deserves to be developed further in the discussion (within the text line 435, page 19).

And of course, this raises the question of what would happen if the labelled neurons would be photo-activated instead of photo-inhibited. Answering this question would imply performing an additional set of experiments with different viral construction, and I don't think this could reasonably be asked of the authors, but they could at least discuss this point.

3. Sites of projection: it is surprising that the projections of preBötC neurons with the non-specific transduction do not invade more globally the hypoglossal nucleus (Figure 5). In contrast, the labelling appears strongly restricted to a very limited number of hypoglossal motoneurons. Is this just due to the choice of the image or is it really representative? Even more surprising when we compare the same image with the one obtained with selective transduction (Figure 6). Here the labelling invades a large portion of the XII nucleus. It is difficult to understand how fewer neurons could send more projections to the same target. Also, can you explain why the selective transduction does not result in the specific labelling of lateral and dorsal lateral border of the facial nucleus, as expected due to the approach used??

4. Personal curiosity: Do the authors have indications about the pattern of discharge of preBötC→7n neurons?

Reviewer #3 (Recommendations for the authors):

If these can be addressed, the manuscript will be more decisive.

Does inhibition of a similar number of preBötC neurons impact breathing?

While silencing many preBötC versus the subset that projects to the facial nucleus impact breathing to different magnitudes, none-the-less, both manipulations slow/stop breathing (most obvious in Figure 8). The magnitude of this effect changes upon the use of anesthesia and the type and depth. An alternative interpretation from the authors of this experiment is that inhibition of a random small number of preBötC neurons (matching the number that projects to the facial nucleus) would similarly slow breathing. The experiments conducted cannot rule this out and such a finding would re-sculpt the main message of the manuscript.

Does inhibitions of preBötC neurons that do not project to the mystacial pad prevent respiratory modulation of the nares?

The preBötC neurons labeled by retrograde tracing from the facial nucleus are presumed to directly control the breathing modulation of the nares. While this is a simple conclusion, the experiments cannot exclude that the partial slowing of breathing is indirectly changing the movement of the nares. For example, alternative premotor populations, like the nIRT or nRF (Kurnikova 2019), that receive preBötC input are those that also modulate the activity of the muscles in the nares. In such a case, the axonal targeting of CN7 would have a small effect on the modulation of the mystacial pad. This might be addressed by inhibiting the axons of these preBötC neurons within the facial nucleus.

Cross correlation of mystacial activity with breathing.

The authors conclude that the respiratory activity of the mystacial pad is lost by inhibition of the CN7 projecting preBötC neurons. While the activity of the mystacial pad muscle changes (mostly to tonic states), the relationship between breathing and the change in mystacial activity in Figures 2, 3, 7, and 8 is not directly quantified. One possible measurement to show a loss of synchrony would be the cross-correlation between breathing and mystacial muscle activity, as in Kurnikova 2017.

Other points to address

– Figure 1C: It would be helpful to use a molecular marker to denote the preBötC (like sst/Nk1R/oprm1) to demonstrate that the labeled neurons are within the anatomical location of the preBötC. In panel C, many labeled cells appear ventral to the preBötC intermingled with other medullary groups that express TH. A similar analysis of the retrogradely labeled neurons in Figure 4F would demonstrate preBötC localization.

– Figure 2 and 3: Quantification of the expression of mCherry within GFP positive neurons. This is expected to be 100% and serves as a control for claiming specificity for the CN7 projecting neurons by ruling out cre-independent viral expression.

– Figure 2 D/K vs. 3 D/K: Explanation for why mystacial pad activity increases when the entire preBötC is inhibited but the pad activity decreases when the CN7 projecting neurons are silenced. Also, as in Figure 2, the remaining figures show the mystacial pad activity increasing in a similar direction between both experimental cohorts.

Essential revisions:

1) As indicated by the reviewers, the authors should use traditional markers as a counterstain of the preBötC region to confirm the anatomical location. How can the authors be sure that they have transduced all preBötC neurons in the region ventral and caudal to the NAc, given that preBötC neurons are not definitively defined? Also please note that the coronal section shown in Figure supplement 1B appears to be caudal to the medullary levels containing the preBötC, and hence does not appear to illustrate distinct CTB retrogradely labeled preBötC subpopulations. Additional data addressing the above should be included.

We agree with the reviewers that due to technical limitations and the lack of specific neurochemical signatures that define preBötC, we cannot claim that all preBötC neurons have been transduced by the viral vector used in this study. This statement has been modified in the revised version of the manuscript. To better characterize, anatomically and phenotypically, the neuronal population of preBötC transduced by viral injections, we also have included in the manuscript new data (Figures 4 and 5) showing the co-localization between muGFP and reeling, somatostatin, NK1R, which are typical markers used to define preBötC neurons (pages 9).

2) The authors should quantify the expression of mCherry within GFP positive neurons as a control for the specificity of the CN7 projecting neurons by ruling out Cre-independent viral expression, as indicated by Reviewer #3. Please include this data.

As suggested by the reviewer, we quantified the expression of mCherry within muGFP-positive neurons. We found that 48.5% (95% CI: 12.77 to 84.23%) of muGFP neurons also expressed mCherry. We recognize that the percentage of co-localization seems to be lower than expected. However, considering that the Cre-loxP system requires only a single molecule of Cre-recombinase enzyme (Van Duyne, 2015), it is plausible that low levels of Cre, that were not detected by immunohistochemistry, had resulted in the expression of muGFP. We believe that the results shown in Figure 2—figure supplement 2 are more important to rule out the Cre-independent viral expression, showing that the control injections of AAV-DIO-GtACR2-MuGFP into preBötC did not produce any expression of muGFP in the absence of expression of Cre (pages 9-10, lines 212-217).

3) There are cells non-selectively and selectively transduced by the vector combination in the vLRt (Figure 2—figure supplement 2C and D and Figure 3—figure supplement 1C and D), raising the possibility that some of the perturbations of MP activity might be attributable to photoinhibition of these cells, which is worth comment in the text.

We thank the Reviewer for raising this point. Whilst every attempt is made to develop injection protocols that effectively target the cell population of interest, this is next to impossible and other neurons are affected. We have added a short section to the Discussion (page 17-18, lines 399-406) to address this issue and draw the reader’s attention to additional neurons transduced in other nuclei.

4) The authors should explain why some preBötC neurons express muGFP that were not VGluT2⁺ or VGAT+ (Figure 4B and F).

We thank the reviewer for noting this, however we do not have a clear understanding of why a small proportion of the transduced neurons express neither VGluT2 nor VGAT. The simplest explanation is that it is due to technical limitations and the strict criteria employed for quantifying expression. A neuron was only considered positive for VGAT or Vglut2 if it contained at least eight transcript dots for the mRNA, and that these were co-localized with a nucleus labeled with DAPI. During the quantification process, it was common to observe MuGFP-positive neurons that expressed four or fewer dots of transcript for VGAT or VGlut2 mRNA or neurons that were not co-localized with DAPI. These neurons were not included in either of the groups MuGFP + VGlut2 or MuGFP +

VGAT. Alternative hypotheses are the existence of other vesicle transporters that haven’t been identified yet, or that these neurons utilize transmitters other than glutamate, GABA, and glycine. We do not have further insight into this neurochemistry.

5) The authors should discuss why the inhibition of MP EMG activity is only transient during sustained photoinhibition at the very high laser power employed.

We thank the reviewer for their helpful comments and overall support for this manuscript. We have observed that photoinhibition of preBötC transiently inhibits breathing, blood pressure (Menuet et al., 2020), and now MP activity. In contrast, the bradycardic effect of photoinhibition, induced by vagal activation, is retained for the length of laser delivery. In unpublished data of extracellular recordings of preBötC neurons, a similar phenomenon is observed, with continuous inhibition of some, but only transient inhibition of other neurons. We are attempting to understand this mechanism. One hypothesis is that the maintenance of the chloride gradient is different between different neuronal populations. Clearly, we are interested in whether this difference provides insight into the different neuronal groups and plan to explore this further. We have added a section to the Discussion to address this issue (page 15, lines 330-337).

6) The relationship between breathing and the change in mystacial activity in Figures 2, 3, 7, and 8 is not directly quantified. One possible measurement to show a loss of synchrony would be the cross-correlation between signals for breathing and mystacial muscle activity, as in Kurnikova 2017.

In consideration of the reviewer’s helpful suggestion, we have quantified the relationship between breathing and mystacial pad activity during the selective photoinhibition of 7n projecting preBötC neurons. We evaluated the synchronization between MP activity and diaphragm activity by analyzing the event-triggered average (ETA) of MP activity (triggered by the rapid rise of dEMG that occurs at the beginning of inspiration). The comparison was made over 5-8 respiratory cycles during baseline and during the initial period of photoinhibition. We haven’t applied this analysis for the nonselective transduction of preBötC neurons, as photoinhibition of preBötC neurons induced apnea, thus the ETA analysis could not be done. The description of these results (pages 8-9, 11-13), as well as their corresponding new figures (Figure 3E, Figure 8 —figure supplement 3), are included in the revised version of the manuscript.

Selective inhibition of preBötC neurons projecting to 7n in urethane anesthetized rats

In urethane-anesthetized rats, we found that photoinhibition of preBötC→7neurons resulted in the loss of respiratory modulation of MP activity (3.9 mV [95% CI: 1 to 6.8 mV] vs. baseline 26.6 mV [95% CI: 9.4 to 43.8 mV], p=0.02), whilst producing a minimal reduction of dEMG (23.8 mV [95% CI: 9.9 to 37.5 mV] vs. 29.1 mV [95% CI: 11.2 to 47 mV], p=0.04) (Figure 3E) (pages 8-9, lines 187-193).

Selective inhibition of preBötC neurons projecting to 7n in conscious rats

We also quantified the respiratory-related activity of MP in rats under three different conditions: ketamine/medetomidine anesthesia, during the very early stage of regaining consciousness, a few min after the anesthesia was reversed with atipamezole, and 1-2 h after the atipamezole injection, when the animal had recovered from anesthesia and was completely awake and freely behaving. The results show that both activity and respiratory modulation of MP are dependent on the rat’s state/behaviour.

– Surgical ketamine/medetomidine anesthesia

Under ketamine/medetomidine anesthesia, the MP activity is minimal and nonrespiratory modulated (Figure 8—figure supplement 3A). Photoinhibition of preBötC→7n did not produce any changes in MP activity (page 11, lines 254-257).

– Early recovery phase.

During the very early stage of regaining consciousness, the inspiratory modulation of the mystacial pad activity resumed in all animals. Despite this, two different responses were observed on MP respiratory-related activity during photoinhibition of preBötC→7n. We observed that in one sub-cohort (n=3), photoinhibition flattened the inspiratory modulation of MP (1 mV [95% CI: 0.09 to 1.9 mV] vs. baseline: 6.8 mV [95% CI: 6.4 to 7.8 mV], p=0.01) even though it did not affect the dEMG amplitude (40.3 mV [95% CI: 25 to 58 mV] vs. baseline: 32.3 mV [95% CI: 20.2 to 52.8 mV], p=0.25). Interestingly, an increase in respiratory-related MP activity was observed in the second sub-cohort (n=2)two rats tested in this experiment (rat 1: 9.6 mV vs. baseline: 3.3 mV; rat 2: 17.6 mV vs. baseline: 8 mV). Again, dEMG was not affected by photoinhibition (rat 1: 51 mV vs. baseline: 56 mV; rat 2: 43.9 mV vs. baseline: 41.7 mV) (page 12, lines 271-276).

– Conscious phase.

Variable responses on the respiratory modulation of MP to photoinhibition of preBötC→7n were observed in conscious rats. We found out that, in 3 out of 5 rats, the ETA analysis did not show respiratory modulation of MP activity at either baseline or during photoinhibition. Given the high respiratory frequency, we speculate photoinhibition was performed during sniffing/whisking behaviour. Intriguingly, we also found that the same two rats that presented inspiratory-related MP activity during the early recovery phase from ket/med anesthesia, had expiratory-related activity when they were fully conscious. Moreover, while the selective photoinhibition of preBötC→7n increased the magnitude of the expiratory activity of MP in one of these rats, it decreased in the other one. These distinct responses are shown in Figure 8 – supplement 2C. It is important to note that despite these different effects in the respiratory modulation of MP, the overall activity of MP, measured as AUC, increased in all rats tested in this experimental protocol (Figure 8T, Figure 9T) (pages 12-13, lines 278-286).

Our results show that selective photoinhibition of preBötC→7n significantly changes the MP activity without affecting the inspiratory rhythm generation regardless of the rat’s state. We don’t know the reasons why the responses were heterogeneous among the rats. It is possible that the MP activity recorded in animals regaining consciousness or when they were fully conscious reflected the activity of multiple inspiratory and expiratory-related muscles, given the difficulty of dissecting the rodent snout (Haidarliu et al., 2012), the large number of distinct muscles that compose the mystacial pad (Haidarliu et al., 2010) and the substantial size of the electrode's suture pads. Another possibility is that the responses induced by photoinhibition could be related to the animal’s behaviour, such as resting, sniffing, and whisking. Unfortunately, in our experiment, photoinhibition was indiscriminately applied during any behaviour (pages 16-17, lines 370-386).

7) The authors should discuss the issues regarding alternative interpretations of the data raised by Reviewer #3.

We have addressed the issues regarding alternative interpretations of the data raised by Reviewer #3 in the responses to their questions 1 and 2, and have included this in the revised version of the manuscript (page 19, lines 428-443).

8) The authors should address all other questions/issues raised in the major concerns of Reviewers #2 and 3. Additional experiments/data are not required here.

We have addressed all the questions/issues raised by Reviewers #2 and 3. Additional discussion have been added in the revised version of the manuscript (page 19, lines 428-443).

Reviewer #1 (Recommendations for the authors):

Concerns to be addressed:

1) The authors should discuss why the inhibition of MP EMG activity is only transient during sustained photoinhibition at the very high laser power employed.

We thank the reviewer for their helpful comments and overall support for this manuscript. We have observed that photoinhibition of preBötC transiently inhibits breathing, blood pressure (Menuet et al., 2020), and now MP activity. In contrast, the bradycardic effect of photoinhibition, induced by vagal activation, is retained for the length of laser delivery. In unpublished data of extracellular recordings of preBötC neurons, a similar phenomenon is observed, with continuous inhibition of some, but only transient inhibition of other neurons. We are attempting to understand this mechanism. One hypothesis is that the maintenance of the chloride gradient is different between different neuronal populations. Clearly, we are interested in whether this difference provides insight into the different neuronal groups and plan to explore this further. We have added a section to the Discussion to address this issue (page 15, lines 330-337).

2) There are cells non-selectively and selectively transduced by the vector combination in the vLRt (Figure 2—figure supplement 2C and D and Figure 3¬ figure supplement 1C and D), raising the possibility that some of the perturbations of MP activity might be attributable to photoinhibition of these cells, which is worth comment.

We thank the Reviewer for raising this point. Whilst every attempt is made to develop injection protocols that effectively target the cell population of interest, this is next to impossible, and other neurons are affected. We have added a short section to the Discussion (pages 17-18, lines 399-406) to address this issue and draw the reader’s attention to additional neurons transduced in other nuclei.

3) The authors should explain why some preBötC neurons express muGFP that were not VGluT2⁺ or VGAT+ (Figure 4B and F).

We thank the reviewer for noting this, however, we do not have a clear understanding of why a small proportion of the transduced neurons express neither VGluT2 nor VGAT. The simplest explanation is that it is due to technical limitations and the strict criteria employed for quantifying expression. A neuron was only considered positive for VGAT or Vglut2 if it contained at least eight transcript dots for the mRNA, and that these were co-localized with a nucleus labeled with DAPI. During the quantification process, it was common to observe MuGFP-positive neurons that expressed four or fewer dots of transcript for VGAT or VGlut2 mRNA or neurons that were not co-localized with DAPI. These neurons were not included in either of the groups MuGFP + VGlut2 or MuGFP + VGAT. Alternative hypotheses are the existence of other vesicle transporters that haven’t been identified yet, or that these neurons utilize transmitters other than glutamate, GABA, and glycine. We do not have further insight into this neurochemistry.

Reviewer #2 (Recommendations for the authors):

While the study is very elegant and the illustrations convincing, I have three concerns:

1. It remains unclear to me why the authors consider their neurons of interest (identified based on their axonal projections) as elements of the preBötzinger network. It might just be a problem of semantics: With the term preBötC, are we talking about a region or a network? Are we talking about a function (breathing) or an anatomical location? For instance are the neurons labelled here Dbx1-derived, as is the case for preBötC neurons involved in breathing (Bouvier et al., 2010; Gray et al., 2010)? Are they, for example, somatostatin or NK1R positive? In the present version of the manuscript, it seems that the identification of neurons as being part of the preBötC is only based on their anatomical location. Could the authors provide additional arguments in favor (or not) for these neurons to express any known features of the respiratory preBötC neurons? Or do the authors consider that the neurons of interest are located in the same region as respiratory preBötC neurons but constitute a completely different type of neurons? This should be clarified and addressed in the discussion. More generally, it actually becomes a point that should be addressed in the field, as several labs are identifying small networks in the brainstem that control distinct motor functions all more or less coordinated with breathing behavior. So are we investigating one large network composed of several intermingled small networks or are they really distinct neuronal groups? What is now the definition of the preBötC?

We thank the reviewer for their insight and constructive comments. This is an interesting comment and the reviewer’s concerns regarding the definition of the preBötC resonate with us. Our interpretation of the large body of work from other laboratories is that the preBötC is a complex, as opposed to a nucleus, that occupies an anatomical location. Within that location are neurons that generate inspiratory rhythm. These neurons have a particular neurochemistry (Dbx-1; VGluT2; SST; NK1R) but none of these markers are exclusive identifiers. Neurons with these neurochemistries project from the preBötC to other brain regions and are presumably involved in functions other than inspiratory rhythm generation.

We have transduced neurons in the preBötC on the basis of anatomical coordinates, relative to obex, and the strong inspiratory activity recorded at the site of injection with the injection pipette. In our initial study (Menuet et al., 2020), we demonstrated that these injections transduced neurons caudal to the parvalbumin-expressing neurons of the Bötzinger nucleus, ventral to the subcompact formation of the nucleus ambiguus and predominantly dorsal to, but intermingled with, the catecholaminergic neurons of the ventrolateral medulla. We also observed that the caudal extent of the transduction had some overlap with the very rostral extent of the parvalbumin-expressing neurons of the rVRG. The distribution of transduced neurons in this study is near identical.

We did not provide evidence in this study of the neurochemical phenotype of the transduced neurons, other than primary ionotropic neurotransmitter, as we did not consider any particular marker definitive. However, in response to the reviewer’s request, we have now included evidence that different groups of transduced neurons express reelin, SST, or NK1R, thus marking them a neurons traditionally ascribed to the preBötC (new Figure 5EG). We also show that some non-selectively transduced preBötC neurons also express NK1R, SST, reelin, or both (new figure 4E-G).

We do not consider that our data is sufficient to adequately address the question of the exact nature of the neurons we have studied, relative to rhythm-generating neurons. As indicated by Reviewer 3, there are several potential explanations for our data. Our working hypothesis is that the facial nucleus projecting preBötC neurons are transmitting respiratory rhythm, rather than being responsible for generating it. The small effect of photoinhibiting these neurons on inspiratory activity is due to collaterals. However, it is possible, as suggested by Reviewer 3, that the neurons projecting to the facial nucleus are a sub-group of the rhythm-generating neurons with a collateral to the facial nucleus – collateral being a potentially confusing term. Without being over-speculative, we have attempted to briefly address this issue in the Discussion (page 19, lines 428-443), but know that a different experimental dataset is required to directly answer the Reviewer’s interesting question.

2. The authors provide evidence that the neurons of interest are a mixture of excitatory and inhibitory phenotypes. The photo-inhibition targets indistinctly either type. It is obvious to expect opposite effects when inhibiting inhibitory neurons versus excitatory neurons. So how do the authors explain the overall effects? Which sub-population (excitatory or inhibitory) is mainly involved in this effect? Are inhibitory and excitatory neurons involved in controlling the same orofacial muscles? Are they sending collaterals to the same brainstem nuclei? This point deserves to be developed further in the discussion (within the text line 435, page 19).

And of course, this raises the question of what would happen if the labelled neurons would be photo-activated instead of photo-inhibited. Answering this question would imply performing an additional set of experiments with different viral construction, and I don't think this could reasonably be asked of the authors, but they could at least discuss this point.

The reviewer raises another very interesting question which we can only partially address with our data. Previous studies, e.g. Yang and Feldman, J Comp Neurol 526: 13891402 (2018) (Yang and Feldman, 2018) have clearly demonstrated that the preBötC sends parallel inhibitory and excitatory projections widely throughout the neuraxis. Our data confirm this in relation to projections to the facial nucleus. An interesting study would be to examine the effect of photoinhibiting one sub-group, either excitatory or inhibitory, however, that is not within our genetic targeting tool-kit at this stage. Our working hypothesis is that the influence of the preBötC on orofacial motor activity will be statedependent, with a predominance of excitatory or inhibitory influence under different conditions. Whilst potentially interesting, we are not sure that photoactivation would provide a particularly informative result as the coordinated, coincident activation of all neurons of a particular phenotype is unlikely to be physiological. For this reason, we favour photoinhibition and predict that definitive answers will arise from more selective transduction and inhibition under different conditions. Without being over-speculative, we have attempted to briefly address this issue in the Discussion (pages 20-21, lines 467-477).

3. Sites of projection: it is surprising that the projections of preBötC neurons with the non-specific transduction do not invade more globally the hypoglossal nucleus (Figure 5). In contrast, the labelling appears strongly restricted to a very limited number of hypoglossal motoneurons. Is this just due to the choice of the image or is it really representative? Even more surprising when we compare the same image with the one obtained with selective transduction (Figure 6). Here the labelling invades a large portion of the XII nucleus. It is difficult to understand how fewer neurons could send more projections to the same target. Also, can you explain why the selective transduction does not result in the specific labelling of lateral and dorsal lateral border of the facial nucleus, as expected due to the approach used??

We thank the reviewer for noting this. We made a technical error in adjusting the image for publication and in the process of reducing the background fluorescence, reduced detectability of some axonal projections. We have re-adjusted the same image to correctly depict the level of axon terminal input. In both cases, the axonal projections from preBötC invades the whole hypoglossal nucleus.

4. Personal curiosity: Do the authors have indications about the pattern of discharge of preBötC→7n neurons?

This is an interesting question, but we do not have further data in relation to that at this stage.

Reviewer #3 (Recommendations for the authors):

If these can be addressed, the manuscript will be more decisive.

Does inhibition of a similar number of preBötC neurons impact breathing?

While silencing many preBötC versus the subset that projects to the facial nucleus impact breathing to different magnitudes, none-the-less, both manipulations slow/stop breathing (most obvious in Figure 8). The magnitude of this effect changes upon the use of anesthesia and the type and depth. An alternative interpretation from the authors of this experiment is that inhibition of a random small number of preBötC neurons (matching the number that projects to the facial nucleus) would similarly slow breathing. The experiments conducted cannot rule this out and such a finding would re-sculpt the main message of the manuscript.

We accept the reviewer’s comment and potential alternative interpretation of the data. The hypothesis would be that sub-groups of rhythm-generating/transmitting preBötC neurons have collaterals to another specific region. Selective transduction, and photoinhibition, of all of these sub-groups would result in total inhibition of inspiratory rhythm generation. We have added a short section to the Discussion to address this issue (page 19, lines 428-443).

Does inhibitions of preBötC neurons that do not project to the mystacial pad prevent respiratory modulation of the nares?

The preBötC neurons labeled by retrograde tracing from the facial nucleus are presumed to directly control the breathing modulation of the nares. While this is a simple conclusion, the experiments cannot exclude that the partial slowing of breathing is indirectly changing the movement of the nares. For example, alternative premotor populations, like the nIRT or nRF (Kurnikova 2019), that receive preBötC input are those that also modulate the activity of the muscles in the nares. In such a case, the axonal targeting of CN7 would have a small effect on the modulation of the mystacial pad. This might be addressed by inhibiting the axons of these preBötC neurons within the facial nucleus.

Again we thank the reviewer for their insight and helpful alternative explanations for our data. We contemplated placing the optical fibres in the facial nucleus to see what effect this might have. Our data, including unpublished extracellular recordings of preBötC neurons, demonstrate inhibition when light is directed to the somata. However, through the observations of others, mostly unpublished, we are cautious about using GtACR2 to ‘inhibit’ axonal terminals. Possibly due to differences in chloride gradients in different neuronal compartments, others have reported increased synaptic activity in response to light-induced activation of GtACR2 (Messier et al., 2018). The reviewer’s suggestion might be better tested using the more recently described light-activated K⁺ channels, or the Gi coupled eOPN3. We have added a short section to the Discussion to address the issue raised by the Reviewer’s comment (page 19, lines 428-443).

Cross correlation of mystacial activity with breathing.

The authors conclude that the respiratory activity of the mystacial pad is lost by inhibition of the CN7 projecting preBötC neurons. While the activity of the mystacial pad muscle changes (mostly to tonic states), the relationship between breathing and the change in mystacial activity in Figures 2, 3, 7, and 8 is not directly quantified. One possible measurement to show a loss of synchrony would be the cross-correlation between breathing and mystacial muscle activity, as in Kurnikova 2017.

In consideration of the reviewer’s helpful suggestion, we have quantified the relationship between breathing and mystacial pad activity during the selective photoinhibition of 7n projecting preBötC neurons. We evaluated the synchronization between MP activity and diaphragm activity by analyzing the event-triggered average (ETA) of MP activity (triggered by the rapid rise of dEMG that occurs at the beginning of inspiration). The comparison was made over 5-8 respiratory cycles during baseline and during the initial period of photoinhibition. We haven’t applied this analysis for the nonselective transduction of preBötC neurons, as photoinhibition of preBötC neurons induced apnea, thus the ETA analysis could not be done. The description of these results (pages 8-9, 11-13), as well as their corresponding new figures (Figure 3E, Figure 8 —figure supplement 3), are included in the revised version of the manuscript.

Selective inhibition of preBötC neurons projecting to 7n in urethane anesthetized rats

In urethane-anesthetized rats, we found that photoinhibition of preBötC→7neurons resulted in the loss of respiratory modulation of MP activity (3.9 mV [95% CI: 1 to 6.8 mV] vs. baseline 26.6 mV [95% CI: 9.4 to 43.8 mV], p=0.02), whilst producing a minimal reduction of dEMG (23.8 mV [95% CI: 9.9 to 37.5 mV] vs. 29.1 mV [95% CI: 11.2 to 47 mV], p=0.04) (Figure 3E) (pages 8-9, lines 187-193).

Selective inhibition of preBötC neurons projecting to 7n in conscious rats

We also quantified the respiratory-related activity of MP in rats under three different conditions: ketamine/medetomidine anesthesia, during the very early stage of regaining consciousness, a few min after the anesthesia was reversed with atipamezole, and 1-2 h after the atipamezole injection, when the animal had recovered from anesthesia and was completely awake and freely behaving. The results show that both activity and respiratory modulation of MP are dependent on the rat’s state/behaviour.

– Surgical ketamine/medetomidine anesthesia

Under ketamine/medetomidine anesthesia, the MP activity is minimal and nonrespiratory modulated (Figure 8—figure supplement 3A). Photoinhibition of preBötC→7n did not produce any changes in MP activity (page 11, lines 254-257).

– Early recovery phase.

During the very early stage of regaining consciousness, the inspiratory modulation of the mystacial pad activity resumed in all animals. Despite this, two different responses were observed on MP respiratory-related activity during photoinhibition of preBötC→7n. We observed that in one sub-cohort (n=3), photoinhibition flattened the inspiratory modulation of MP (1 mV [95% CI: 0.09 to 1.9 mV] vs. baseline: 6.8 mV [95% CI: 6.4 to 7.8 mV], p=0.01) even though it did not affect the dEMG amplitude (40.3 mV [95% CI: 25 to 58 mV] vs. baseline: 32.3 mV [95% CI: 20.2 to 52.8 mV], p=0.25). Interestingly, an increase in respiratory-related MP activity was observed in the second sub-cohort (n=2)two rats tested in this experiment (rat 1: 9.6 mV vs. baseline: 3.3 mV; rat 2: 17.6 mV vs. baseline: 8 mV). Again, dEMG was not affected by photoinhibition (rat 1: 51 mV vs. baseline: 56 mV; rat 2: 43.9 mV vs. baseline: 41.7 mV) (page 12, lines 271-276).

– Conscious phase.

Variable responses on the respiratory modulation of MP to photoinhibition of preBötC→7n were observed in conscious rats. We found out that, in 3 out of 5 rats, the ETA analysis did not show respiratory modulation of MP activity at either baseline or during photoinhibition. Given the high respiratory frequency, we speculate photoinhibition was performed during sniffing/whisking behaviour. Intriguingly, we also found that the same two rats that presented inspiratory-related MP activity during the early recovery phase from ket/med anesthesia, had expiratory-related activity when they were fully conscious. Moreover, while the selective photoinhibition of preBötC→7n increased the magnitude of the expiratory activity of MP in one of these rats, it decreased in the other one. These distinct responses are shown in Figure 8 – supplement 2C. It is important to note that despite these different effects in the respiratory modulation of MP, the overall activity of MP, measured as AUC, increased in all rats tested in this experimental protocol (Figure 8T, Figure 9T) (pages 12-13, lines 278-286).

Our results show that selective photoinhibition of preBötC→7n significantly changes the MP activity without affecting the inspiratory rhythm generation regardless of the rat’s state. We don’t know the reasons why the responses were heterogeneous among the rats. It is possible that the MP activity recorded in animals regaining consciousness or when they were fully conscious reflected the activity of multiple inspiratory and expiratory-related muscles, given the difficulty of dissecting the rodent snout (Haidarliu et al., 2012), the large number of distinct muscles that compose the mystacial pad (Haidarliu et al., 2010) and the substantial size of the electrode's suture pads. Another possibility is that the responses induced by photoinhibition could be related to the animal’s behaviour, such as resting, sniffing, and whisking. Unfortunately, in our experiment, photoinhibition was indiscriminately applied during any behaviour (pages 16-17, lines 370-386).

Other points to address

– Figure 1C: It would be helpful to use a molecular marker to denote the preBötC (like sst/Nk1R/oprm1) to demonstrate that the labeled neurons are within the anatomical location of the preBötC. In panel C, many labeled cells appear ventral to the preBötC intermingled with other medullary groups that express TH. A similar analysis of the retrogradely labeled neurons in Figure 4F would demonstrate preBötC localization.

In consideration of the reviewer's suggestion, we have now characterized, anatomically and phenotypically, the neuronal population of preBötC transduced by viral injections. These results are shown in the new Figure 5 of the revised version of the manuscript. Briefly, the new data show the expression of muGFP in different groups of neurons expressing reelin, somatostatin, or NK1R – typical markers used to define preBötC neurons (Figure 5E-G).

– Figure 2 and 3: Quantification of the expression of mCherry within GFP positive neurons. This is expected to be 100% and serves as a control for claiming specificity for the CN7 projecting neurons by ruling out cre-independent viral expression.

As suggested by the reviewer, we quantified the expression of mCherry within muGFP-positive neurons. We found that 48.5% (95% CI: 12.77 to 84.23%) of muGFP neurons also expressed mCherry. We recognize that the percentage of co-localization seems to be lower than expected. However, considering that the Cre-loxP system requires only a single molecule of Cre-recombinase enzyme (Van Duyne, 2015), it is plausible that low levels of Cre, that were not detected by immunohistochemistry, had resulted in the expression of muGFP. We believe that the results shown in Figure 2—figure supplement 2 are more important to rule out the Cre-independent viral expression, showing that the control injections of AAV-DIO-GtACR2-MuGFP into preBötC did not produce any expression of muGFP in the absence of expression of Cre (pages 9-10, lines 212-217).

– Figure 2 D/K vs. 3 D/K: Explanation for why mystacial pad activity increases when the entire preBötC is inhibited but the pad activity decreases when the CN7 projecting neurons are silenced. Also, as in Figure 2, the remaining figures show the mystacial pad activity increasing in a similar direction between both experimental cohorts.

We thank the reviewer for raising this important question. We don’t know why silencing preBotC→7n decreases mystacial pad activity in urethane-anesthetized rats, while the opposite response is seen when the entire preBötC is inhibited. One possible explanation might related to the population of neurons transduced – whilst a similar proportion of inhibitory and excitatory, as well as SST+, reelin+, and NK1R+ neurons were transduced by both approaches, the selective approach resulted in a much smaller population of transduced neurons with a more restricted anatomical location. Another explanation might be statedependent activation of different neuronal populations. Interestingly, mystacial pad activity increases with the photoinhibition of preBötC→7n neurons in conscious rats. Unfortunately, the experiments necessary to unravel the mechanisms that might underlie state-dependent effects are beyond this study. However, based on previous studies that reported opposite effects with interventions made in anesthetised versus conscious animals (Machado and Bonagamba, 1992; Wenker et al., 2017; Marina et al., 2011; Chiou et al., 2000), a possible explanation for our results is that under urethane, the glutamatergic preBötC→7n neurons play a major role in regulating the respiration-related mystacial pad activity, whilst the inhibitory neurons are more active in the conscious state.

References

Chitravanshi VC, Sapru HN. 1999. Phrenic nerve responses to chemical stimulation of the subregions of ventral medullary respiratory neuronal group in the rat. Brain Res 821:443– 460. doi:10.1016/S0006-8993(99)01139-7

Haidarliu S, Golomb D, Kleinfeld D, Ahissar E. 2012. Dorsorostral Snout Muscles in the Rat Subserve Coordinated Movement for Whisking and Sniffing. Anat Rec (Hoboken) 295:1181–1191. doi:10.1002/ar.22501

Haidarliu S, Simony E, Golomb D, Ahissar E. 2010. Muscle Architecture in the Mystacial Pad of the Rat. Anat Rec (Hoboken) 293:1192–1206. doi:10.1002/ar.21156

Machado BH, Bonagamba LGH. 1992. Microinjection of l-glutamate into the nucleus tractus solitarii increases arterial pressure in conscious rats. Brain Res 576:131–138. doi:10.1016/0006-8993(92)90618-J

Menuet C, Connelly AA, Bassi JK, Melo MR, Le S, Kamar J, Kumar NN, McDougall SJ, McMullan S, Allen AM. 2020. Prebötzinger complex neurons drive respiratory modulation of blood pressure and heart rate. eLife 9:e57288. doi:10.7554/eLife.57288

Messier JE, Chen H, Cai Z-L, Xue M. 2018. Targeting light-gated chloride channels to neuronal somatodendritic domain reduces their excitatory effect in the axon. eLife 7. doi:10.7554/eLife.38506

Van Duyne GD. 2015. Cre Recombinase. Microbiol Spectr 3.MDNA3-0014-2014. doi:10.1128/microbiolspec.MDNA3-0014-2014

Yang CF, Feldman JL. 2018. Efferent projections of excitatory and inhibitory preBötzinger Complex neurons. J Comp Neurol 526:1389–1402. doi:10.1002/cne.24415

Author details

1. Mariana R Melo

   Department of Anatomy & Physiology, University of Melbourne, Melbourne, Australia

   Contribution

   Conceptualization, Resources, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   mariana.del@unimelb.edu.au

   Competing interests

   No competing interests declared

2. Alexander D Wykes

   1. Florey Institute of Neuroscience and Mental Health, Melbourne, Australia
   2. Florey Department of Neuroscience and Mental Health, University of Melbourne, Melbourne, Australia

   Contribution

   Conceptualization, Resources, Validation, Investigation, Methodology, Writing – original draft

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-8264-7585

3. Angela A Connelly

   Department of Anatomy & Physiology, University of Melbourne, Melbourne, Australia

   Contribution

   Conceptualization, Software, Formal analysis, Validation, Visualization, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4547-4352

4. Jaspreet K Bassi

   Department of Anatomy & Physiology, University of Melbourne, Melbourne, Australia

   Contribution

   Conceptualization, Software, Formal analysis, Validation, Investigation, Visualization, Methodology

   Competing interests

   No competing interests declared

5. Shane D Cheung

   Biological Optical Microscopy Platform (BOMP) - University of Melbourne, Melbourne, Australia

   Contribution

   Resources, Software

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-8733-4756

6. Stuart J McDougall

   Florey Institute of Neuroscience and Mental Health, Melbourne, Australia

   Contribution

   Conceptualization, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8778-675X

7. Clément Menuet

   Institut de Neurobiologie de la Méditerrané, INMED UMR1249, INSERM, Aix-Marseille Université, Marseille, France

   Contribution

   Conceptualization, Data curation, Validation, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-7419-6427

8. Ross AD Bathgate

   1. Florey Institute of Neuroscience and Mental Health, Melbourne, Australia
   2. Department of Biochemistry and Molecular Biology, University of Melbourne, Melbourne, Australia

   Contribution

   Resources, Funding acquisition, Validation, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

9. Andrew M Allen

   1. Department of Anatomy & Physiology, University of Melbourne, Melbourne, Australia
   2. Florey Institute of Neuroscience and Mental Health, Melbourne, Australia

   Contribution

   Conceptualization, Resources, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   a.allen@unimelb.edu.au

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2183-5360

• 627

  Page views

• 99

  Downloads

• 0

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.