A cluster of neuropeptide S neurons regulates breathing and arousal

Abstract

Neuropeptide S (NPS) is a highly conserved peptide found in all tetrapods that functions in the brain to promote heightened arousal; however, the subpopulations mediating these phenomena remain unknown. We generated mice expressing Cre recombinase from the Nps gene locus (Nps^Cre) and examined populations of NPS+ neurons in the lateral parabrachial area (LPBA), the peri-locus coeruleus (peri-LC) region of the pons, and the dorsomedial thalamus (DMT). We performed brain-wide mapping of input and output regions of NPS+ clusters and characterized expression patterns of the NPS receptor 1 (NPSR1). While the activity of all three NPS+ subpopulations tracked with vigilance state, only NPS+ neurons of the LPBA exhibited both increased activity prior to wakefulness and decreased activity during REM sleep, similar to the behavioral phenotype observed upon NPSR1 activation. Accordingly, we found that activation of the LPBA but not the peri-LC NPS+ neurons increased wake and reduced REM sleep. Furthermore, given the extended role of the LPBA in respiration and the link between behavioral arousal and breathing rate, we demonstrated that the LPBA but not the peri-LC NPS+ neuronal activation increased respiratory rate. Together, our data suggest that NPS+ neurons of the LPBA represent an unexplored subpopulation regulating breathing, and they are sufficient to recapitulate the sleep/wake phenotypes observed with broad NPS system activation.

Keywords: arousal; breathing; fiber photometry; lateral parabrachial; neuropeptide S; neuropeptide S receptor; neuropeptide circuits; respiration; sleep.

Figure 1.. Chemogenetic manipulation of NPS system reveals a role in arousal and REM sleep suppression

A-I) Sleep/wake time and expression of Nps^Cre∷Rosa26^LSL-hM4Di mice. A) Total wake, NREM, and REM sleep time expressed as a percentage of total time for the 3 hours following 1.0 mg/kg CNO or saline injection at ZT14. n = 9 mice (3 males, 6 females). [Wake: Student’s paired t-test, t(8) = 2.792, p = 0.023; NREM: Student’s paired t-test, t(8) = −2.799, p = 0.023; REM: Student’s paired t-test, t(8) = −1.776, p = 0.114]. B-D) Percentage of time in B) wake, C) NREM, and D) REM sleep expressed in half hour bins following CNO or saline injection at ZT14. n = 9 mice (3 males, 6 females). [Wake: repeated-measures ANOVA, main effect of drug: F(1,16) = 6.600, p = 0.021, drug × time interaction: F(5,80) = 0.526, p = 0.756; NREM: repeated-measures ANOVA, main effect of drug: F(1,16) = 6.889, p = 0.018, drug × time interaction: F(5,80) = 0.527, p = 0.755; REM: repeated-measures ANOVA, main effect of drug: F(1,16) = 1.745, p = 0.205, drug × time interaction: F(5,80) = 0.616, p = 0.688]. E) Individual animals time spent in REM (green), NREM (red), and wake (blue), stacked and expressed as a percentage of total time for the 3 hours following CNO or saline injection at ZT14. The top plot represents saline treated animals, while bottom plot represents the same animals treated with CNO. F) Sleep stage hypnograms for individual animals. Each 4 s epoch is plotted temporally and marked by a green (REM), red (NREM), or blue (wake) bar for the 3 hours following saline (top plot) or CNO (bottom plot) injection at ZT14. White line at 30 min represents the approximate time that CNO should begin inactivating NPS+ cells in the brain. G-I) Representative images of mCitrine expression (green) marking Gi expression in NPS+ cells of the G) peri-LC, H) LPBA, and I) DMT. The signal was amplified with an eGFP antibody. J-Q) Sleep/wake time and expression of Nps^Cre∷Rosa26 ^LSL-hM3Dq mice. J) Total wake, NREM, and REM sleep time expressed as a percentage of total time for the 3 hours following 1.0 mg/kg CNO or saline injection at ZT2. n = 9 mice (6 males, 3 females). [Wake: Student’s paired t-test, t(8) = −0.680, p = 0.516; NREM: Student’s paired t-test, t(8) = −0.974, p = 0.358; REM: Student’s paired t-test, t(8) = 4.162, p = 0.003]. K-M) Percentage of time in K) wake, L) NREM, and M) REM sleep expressed in half hour bins following CNO or saline injection at ZT2. n = 9 mice (6 males, 3 females). [Wake: repeated-measures ANOVA, main effect of drug: F(1,16) = 0.285, p = 0.601, drug × time interaction: F(2.740,43.834) = 1.075, p = 0.366; NREM: repeated-measures ANOVA, main effect of drug: F(1,16) = 0.744, p = 0.401, drug × time interaction: F(2.879,46.062) = 1.320, p = 0.279; REM: repeated-measures ANOVA, main effect of drug: F(1,16) = 8.966, p = 0.009, drug × time interaction: F(2.839,45.425) = 1.793, p = 0.164]. N) Individual animals time spent in REM (green), NREM (red), and wake (blue), stacked and expressed as a percentage of total time for the 3 hours following CNO or saline injection at ZT2. The top plot represents saline treated animals, while bottom plot represents the same animals treated with CNO. O) Sleep stage hypnograms for individual animals. Each 4 s epoch is plotted temporally and marked by a green (REM), red (NREM), or blue (wake) bar for the 3 hours following saline (top plot) or CNO (bottom plot) injection at ZT2. White line at 30 min represents the approximate time that CNO should begin activating NPS+ cells in the brain. P,Q) Representative images of mCitrine expression (green) marking Gq expression in NPS+ cells of the P) peri-LC and Q) DMT/PVT. Signal was amplified with an eGFP antibody. R-Ab) Sleep/wake time and expression of Npsr1^Cre∷Rosa26 ^LSL-hM3Dq mice. R) Total wake, NREM, and REM sleep time expressed as a percentage of total time for the 3 hours following 1.0 mg/kg CNO or saline injection at ZT2. n = 12 mice (5 males, 7 females). [Wake: Student’s paired t-test, t(11) = −2.847, p = 0.016; NREM: Student’s paired t-test, t(11) = 2.671, p = 0.022; REM: Student’s paired t-test, t(11) = 2.738, p = 0.019]. S-U) Percentage of time in S) wake, T) NREM, and U) REM sleep expressed in half hour bins following CNO or saline injection at ZT2. n = 12 mice (5 males, 7 females). [Wake: repeated-measures ANOVA, main effect of drug: F(1,22) = 6.814, p = 0.016, drug × time interaction: F(2,321,51.071) = 0.516, p = 0.627; NREM: repeated-measures ANOVA, main effect of drug: F(1,22) = 4.963, p = 0.036, drug × time interaction: F(2.385,52.476) = 0.453, p = 0.672; REM: repeated-measures ANOVA, main effect of drug: F(1,22) = 4.457, p = 0.046, drug × time interaction: F(2.126,46.767) = 0.834, p = 0.447]. V) Individual animals time spent in REM (green), NREM (red), and wake (blue), stacked and expressed as a percentage of total time for the 3 hours following CNO or saline injection at ZT2. The top plot represents saline treated animals, while bottom plot represents the same animals treated with CNO. W) Sleep stage hypnograms for individual animals. Each 4 s epoch is plotted temporally and marked by a green (REM), red (NREM), or blue (wake) bar for the 3 hours following saline (top plot) or CNO (bottom plot) injection at ZT2. White line at 30 min represents the approximate time that CNO should begin activating NPS+ cells in the brain. X-Ab) Representative images of mCitrine expression (green) marking Gq expression in NPSR1+ cells of the X) basolateral amygdala, Y) endopiriform cortex, Z) laterodorsal thalamus, ventrolateral nucleus, Aa) retrosplenial cortex, and Ab) subiculum. The signal was amplified with an eGFP antibody. Sleep state was staged in 4 s epochs and discerned from EEG/EMG recordings. All graphs are expressed as mean ± s.e.m. * represents p<0.05, ** p<0.01, ***<0.001 See also Figure S1 and Figure S2.

Figure 2.. Nps^Cre is expressed in distinct neuronal clusters with broad afferent inputs

A-D) tdTomato reporter expression representing NPS+ cells (red). NPS cells are highly clustered in the A) lateral parabrachial area (LPBA), B) medial to the locus coeruleus (peri-LC), C) dorsomedial thalamus (DMT), D) paraventricular thalamus (PVT). In (B), tyrosine hydroxylase staining (green) marks the LC. n = 2 mice (1 male, 1 female). E-S) Rabies tracing and quantification of monosynaptic inputs to NPS+ cells in the LPBA (E-I), peri-LC (J-N), and habenula/PVT/DMT (O-S). E) Representative image of NPS+ starter cells in the LPBA (yellow), expressing both Flex-TVA / Rabies G helper virus (red) and EnvA G-deleted Rabies virus (green). Green cells represent upstream inputs to NPS+ starter cells. F-H) Representative image of cell populations in the F) inferior colliculus, G) bed nucleus of stria terminalis (BNST), and H) central amygdala that project to NPS+ cells in the LPBA. I) Quantification of monosynaptic inputs to LPBA NPS+ cells, expressed as a percentage of total. n = 2 mice (1 male, 1 female). Schematic of injection protocol and LPBA target (I inset). J) Representative image of NPS+ starter cells in the peri-LC (yellow), expressing both Flex-TVA / Rabies G helper virus (red) and EnvA G-deleted Rabies virus (green). K-M) Representative image of cell populations in the K) median raphe, L) interpeduncular nuclei, and M) zona incerta that project to NPS+ cells in the peri-LC. N) Quantification of monosynaptic inputs to peri-LC NPS+ cells, expressed as a percentage of total. n = 2 mice (1 male, 1 female). Schematic of injection protocol and peri-LC target (N inset). O) Representative image of NPS+ starter cells in the habenula (yellow), expressing both Flex-TVA / Rabies G helper virus (red) and EnvA G-deleted Rabies virus (green). P-R) Representative image of cell populations in the P) PVT and Q-R) DMT that project to NPS+ cells in the habenula. S) Quantification of monosynaptic inputs to habenula NPS+ cells, expressed as a percentage of total. n = 4 mice (3 male, 1 female). Schematic of injection protocol and DMT target (S inset). All brain regions with >1% of total monosynaptic inputs to target NPS+ region(s) are plotted. See also Figure S1, Figure S3, and Table S1.

Figure 3.. NPS+ outputs and summary of NPS+ tracing studies

A) LPBA injection site of Cre-dependent Ypet-2a-mGFP (green) / synaptophysin-mRuby (red) in Nps^Cre mouse shown at 63x magnification. B,C) Peri-LC injection site of Cre-dependent Ypet-2a-mGFP (green) / synaptophysin-mRuby (red) in Nps^Cre mouse. Shown at 63x magnification (B) and 20x magnification (C). D) Nucleus incertus / pontine central gray injection site of Cre-dependent Ypet-2a-mGFP (green) / synaptophysin-mRuby (red) in Nps^Cre mouse shown at 20x magnification. E) Schematic summarizing the most prominent outputs of (red) and inputs to (blue) NPS-expressing cells of the LPBA. N = 2 mice (1 male, 1 female). Left panel is a sagittal section, 1.3 mm lateral from midline, with the LPBA in-plane. Right panel is a coronal section of the LPBA injection area (−5.02 mm from Bregma, indicated by black bar in left panel, rotated 90° and displayed coronally). F) Schematic summarizing the most prominent outputs of (red) and inputs to (blue) NPS-expressing cells of the peri-LC / NI. N = 2 mice (1 male, 1 female). Left panel is a sagittal section, 0.6 mm lateral from midline, with the peri-LC in-plane. The right panel is a coronal section of the peri-LC injection area (−5.52 mm from Bregma, indicated by black bar in left panel, rotated 90° and displayed coronally). Note: pontine central gray (PCG) and laterodorsal tegmental nucleus (LDTg) input, and pontine reticular (PnR) output regions are omitted from the sagittal image (left panel) for viewability. These regions are included in the coronal section (right panel). White scale bars (lower right in A-D) represent 20 μm in (A) and (B), 50m in (C), and 25um in (D). E and F) Blue brain regions represent inputs to NPS-expressing cells. Red brain regions represent efferent targets of NPS-expressing cells. Solid lines outlining brain regions represent regions that are in-plane. Dotted lines outlining brain regions represent regions that are out-of-plane. For both schematics, all inputs comprising >3% of total inputs are included, and all outputs with relatively more synaptophysin-mRuby puncta (Table S1, columns 7–8) are included. Abbreviations: BNST = bed nucleus of stria terminalis, CeM = central amygdala, DMTg = dorsomedial tegmental area, DPGi = dorsal paragigantocellular nucleus, IC = inferior colliculus, IOK = inferior olive cap of Kooy, IP = interpeduncular nuclei, LDTg = laterodorsal tegmental nucleus, LH = lateral hypothalamus, LPBA = lateral parabrachial area, LPO = lateral preoptic area, MnR = median raphe, NTS = solitary tract nucleus, PAG = periaqueductal gray, peri-LC = peri-locus coeruleus, PCG = pontine central gray, PnR = pontine reticular nucleus, PSTH = posterior subthalamic nucleus, PVT = paraventricular thalamus, RRF = retrorubral field, VCA = ventral cochlear area, Ve = vestibular nucleus, VLL = ventral lateral lemniscus, VTA = ventral tegmental area. See also Figure S3 and Table S1.

Figure 4.. Npsr1^Cre is expressed in areas critical for sensory processing, behavior, and arousal

A-H) Representative images of tdTomato and DAPI staining in Npsr1^Cre mice. Dense tdTomato expression was observed in the A) anterior olfactory nucleus, B) orbitofrontal cortex, C) nucleus accumbens shell, D) dorsal bed nucleus of stria terminalis, E) paraventricular nucleus of the hypothalamus, F) basolateral amygdala and endopiriform nucleus, G) subiculum, and H) parabrachial nucleus. I) Quantification of tdTomato expression depicted as a percentage of fluorescence per neuroanatomical region. N = 3 mice (1 male, 2 females). All brain regions with >1% of area expressing strong fluorescence are plotted. J) Representative image of defined neuroanatomical regions utilized for HALO software processing and quantification of NPSR1 fluorescence. White scale bars (lower right) represent 100 μm in all panels. Abbreviations: 3V=3^rd ventricle, AO/AOL = anterior olfactory nucleus, AOB = accessory olfactory bulb, BLA = basolateral amygdala, dBNST = dorsal bed nucleus of stria terminalis, dEN = dorsal endopiriform nucleus, FrA = frontal association cortex, GrO = olfactory bulb, LO = lateral orbital cortex, NacS = nucleus accumbens shell, lOFC = lateral orbitofrontal cortex, vOFC = ventral orbitofrontal cortex, PBN = parabrachial nucleus, PVT = paraventricular nucleus of the hypothalamus. Scp = superior cerebellar peduncle See also Figure S1 and Table S1.

Figure 5.. NPS-expressing cells increase activity preceding transitions to wake and suppress activity during REM sleep

A-G) Photometry with polysomnography in the LPBA. A) Schematic of fiber placement above the LPBA and GcaMP7s virus (green) in the LPBA. B) Representative image of LPBA injected with Cre-dependent GCaMP7s. C) Representative 10-min trace from implanted LPBA animal. Top panel: EEG spectrogram from 0 – 20 Hz, second panel from top: raw EEG, third panel from top: raw EMG, bottom panel: GCaMP signal expressed as a Z-score. Blue = wake, red = NREM sleep, green = REM sleep. D-G) LPBA GCaMP signal expressed as a Z-score 30 s before, and 60 s following sleep state transitions. n=6 mice (3 males, 3 females). Top panels: All transitions of a particular type (e.g., NREM → Wake) within animal were averaged to one data point per time bin. These averages were then averaged across animals and plotted (red line represents mean, gray shading represents s.e.m.). Bottom panels are heatmaps of all individual transitions from all animals. D) NREM to wake transitions (Linear Mixed Effect Model, main effect of time bin, β = −0.094, S.E. = 0.030, 95% CI [−0.156, −0.033], t(34) = −3.136, p = 0.004). E) NREM to REM transitions (Linear Mixed Effect Model, main effect of time bin, β = −0.188, S.E. = 0.0275, 95% CI [−0.243, −0.132], t(34) = −6.819, p = 0.00000008). F) Wake to NREM transitions (Linear Mixed Effect Model, main effect of time bin, β = 0.044, S.E. = 0.014, 95% CI [0.014, 0.074], t(34) = 3.024, p = 0.005). G) REM to wake transitions (Linear Mixed Effect Model, main effect of time bin, β = 0.103, S.E. = 0.031, 95% CI [0.041, 0.166], t(34) = 3.379, p = 0.002). H-N) Photometry with polysomnography in the peri-LC. H) Schematic of fiber placement above the peri-LC and GcaMP7s virus (green) in the peri-LC. I) Representative image of peri-LC injected with Cre-dependent GCaMP7s. J) Representative 10-min trace from implanted peri-LC animal. Top panel: EEG spectrogram from 0 – 20 Hz, second panel from top: raw EEG, third panel from top: raw EMG, bottom panel: GCaMP signal expressed as a Z-score. Blue = wake, red = NREM sleep, green = REM sleep. K-N) Peri-LC GCaMP signal expressed as a Z-score 30 s before, and 60 s following sleep state transitions. n=6 mice (3 males, 3 females).Top panels: All transitions of a particular type within animal were averaged to one data point per time bin. These averages were then averaged across animals and plotted (red line represents mean, gray shading represents s.e.m.). Bottom panels are heatmaps of all individual transitions from all animals. K) NREM to wake transitions (Linear Mixed Effect Model, main effect of time bin, β = −0.023, S.E. = 0.025, 95% CI [−0.075, 0.023], t(34) = −0.931, p = 0.358). L) NREM to REM transitions (Linear Mixed Effect Model, main effect of time bin, β = −0.016, S.E. = 0.039, 95% CI [−0.095, 0.064], t(28) = −0.400, p = 0.693). M) Wake to NREM transitions (Linear Mixed Effect Model, main effect of time bin, β = 0.002, S.E. = 0.024, 95% CI [−0.047, 0.051], t(34) = 0.074, p = 0.941). N) REM to wake transitions (Linear Mixed Effect Model, main effect of time bin, β = 0.088, S.E. = 0.056, 95% CI [−0.026, 0.202], t(28) = 1.582, p = 0.125). O-U) Photometry with polysomnography in the DMT O) Schematic of fiber placement above the DMT and GcaMP7s virus (green) in the DMT. P) Representative image of DMT injected with Cre-dependent GCaMP7s. Q) Representative 10-min trace from implanted DMT animal. Top panel: EEG spectrogram from 0 – 20 Hz, second panel from top: raw EEG, third panel from top: raw EMG, bottom panel: GCaMP signal expressed as a Z-score. Blue = wake, red = NREM sleep, green = REM sleep. R-U) DMT GCaMP signal expressed as a Z-score 30 s before, and 60 s following sleep state transitions. n=6 mice (5 males, 1 female). Top panels: All transitions of a particular type within animal were averaged to one data point per time bin. These averages were then averaged across animals and plotted (red line represents mean, gray shading represents s.e.m.). Bottom panels are heatmaps of all individual transitions from all animals. R) NREM to wake transitions (Linear Mixed Effect Model, main effect of time bin, β = −0.203, S.E. = 0.046, 95% CI [−0.296, −0.109], t(34) = −4.411, p = 0.0001). S) NREM to REM transitions (Linear Mixed Effect Model, main effect of time bin, β = −0.229, S.E. = 0.047, 95% CI [−0.324, −0.133], t(34) = −4.884, p = 0.00002). T) Wake to NREM transitions (Linear Mixed Effect Model, main effect of time bin, β = 0.030, S.E. = 0.014, 95% CI [0.001, 0.059], t(34) = 2.075, p = 0.046). U) REM to wake transitions (Linear Mixed Effect Model, main effect of time bin, β = 0.086, S.E. = 0.034, 95% CI [0.017, 0.155], t(34) = 2.532, p = 0.016). All graphs are expressed as mean ± s.e.m. * represents p<0.05. See also Figure S4.

Figure 6.. Lateral parabrachial area-restricted NPS+ neuronal activation augments wake, suppresses REM sleep, and enhances respiratory rate

A) Schematic of injection protocol and Cre-dependent hM3Dq targeted to the LPBA. B-H) Sleep/wake time and expression of LPBA-restricted hM3Dq Nps^Cre mice. (11 males, 1 female). B) Total wake, NREM, and REM sleep time expressed as a percentage of total time for the 3 hours following CNO or saline injection at ZT2. [Wake: Student’s paired t-test, t(11) = −1.97, p = 0.075; NREM: Student’s paired t-test, t(11) = 0.637, p = 0.537; REM: Student’s paired t-test, t(11) = 6.676, p = 0.000035)]. C) Individual animals time spent in REM (green), NREM (red), and wake (blue), stacked and expressed as a percentage of total time for the 3 hours following CNO or saline injection at ZT2. The top plot represents saline treated animals, while bottom plot represents the same animals treated with CNO. D) Sleep stage hypnograms for individual animals. Each 4 s epoch is plotted temporally and marked by a green (REM), red (NREM), or blue (wake) bar for the 3 hours following saline (top plot) or CNO (bottom plot) injection at ZT2. White line at 30 min represents the approximate time that CNO should begin activating NPS+ cells in the brain. E) Representative images of NPS+ cells in the LPBA expressing either AAV-hM3Dq (top) or AAV-mCherry control (bottom) at 10x (left) and 20x (right) magnification. F-H) Percentage of time in F) wake, G) NREM, and H) REM sleep expressed in half hour bins following CNO or saline injection at ZT2. [Wake: repeated-measures ANOVA, main effect of drug: F(1,22) = 2.605, p = 0.121, drug × time interaction: F(5,110) = 1.570, p = 0.174; NREM: repeated-measures ANOVA, main effect of drug: F(1,22) = 0.314, p = 0.581, drug × time interaction: F(5,110) = 1.404, p = 0.228; REM: repeated-measures ANOVA, main effect of drug: F(1,22) = 17.813, p = 0.0004, drug × time interaction: F(2,848,62.655) = 2.747, p = 0.053]. # indicates statistically significant (p = 0.037) if analyzed only over the 1-hour post-CNO onset (ZT2.5 - ZT3.5). I-N) Plethysmography outputs of various breathing parameters. n = 21 mice (AAV-mCherry control: 7 males, 2 females; AAV-hM3Dq: 11 males, 1 female). Mice expressing hM3Dq exclusively in NPS+ cells of the LPBA exhibited I) significantly more breaths per minute, J) significantly shorter inspiratory and K) expiratory times, L) significantly higher peak inspiratory flow, and M) significantly higher peak expiratory flow following CNO injection, in comparison to mCherry-injected control animals. There were no significant differences observed between groups in tidal volume. Data is expressed as a ratio (CNO or saline post-injection period normalized to same-day 2-hour baseline recording). [I: repeated-measures ANOVA, main effect of drug: F(1,19) = 4.729, p = 0.042, drug × virus interaction: F(1,19) = 2.340, p = 0.143; J: repeated-measures ANOVA, main effect of drug: F(1,19) = 11.038, p = 0.004, drug × virus interaction, F(1,19) = 27.090, p = 0.00005; K: repeated-measures ANOVA, main effect of drug: F(1,19) = 0.179, p = 0.677, drug × virus interaction: F(1,19) = 10.189, p = 0.005; L: repeated-measures ANOVA, main effect of drug: F(1,19) = 9.428, p = 0.006, drug × virus interaction: F(1,19) = 25.301, p = 0.00007; M: repeated-measures ANOVA, main effect of drug: F(1,19) = 0.354, p = 0.559, drug × virus interaction: F(1,19) = 9.430, p = 0.006; N: repeated-measures ANOVA, main effect of drug: F(1,19) = 1.553, p = 0.228, drug × virus interaction: F(1,19) = 1.199, p = 0.287]. All graphs are expressed as mean ± s.e.m. * represents p<0.05, ** p<0.01, ***<0.001. See also Figure S5 and Figure S6.