Fig 2: Ultra-high-performance liquid chromatography analysis of 5-HT contents and its related compounds, 5-HTP and 5-HIAA, in the medulla oblongata of Phox2b mutants, wildtype littermates and OF1 mice. (A), Example of a chromatogram of medulla oblongata obtained from Phox2b mutants (blue; n=8), wildtype littermates (pink; n=8) and OF1 mice (purple; n=10) with indicated peaks of serotonin (5-HT), its precursor 5-hydroxytryptophan (5-HTP) and its metabolite 5-hydroxyindole acetic acid (5-HIAA). (B), Scatter plots showing 5-HT quantity (g of 5-HT per medulla) with superimposed mean ± standard error of the mean. (C, D), Scatter plots showing, respectively, 5-HT/5-HTP and 5-HIAA/5-HT ratio with median [Q1; Q3] superimposed in OF1 (purple filled circles), wildtype littermates (pink filled circles) and Phox2b mutants (blue filled circles). * Indicates a significant difference between OF1, Phox2b mutants and wildtype littermates (**p < 0.01 and ***p < 0.001; one-way analysis of variance or Kruskal-Wallis test followed by Benjamini, Krieger and Yekutieli’s multiple comparison test) ns, non significant.

Fig 3: Comparison of the c-fos expression between acidosis-responder and acidosis-non-responder Phox2b mutant preparations under etonogestrel. Drawings of representative sections from the medulla oblongata at the caudal level (with inferior olives; A, B), intermediate level (between inferior olives and the facial nucleus; C, D) and rostral level (with the facial nucleus; E, F) illustrating the c-FOS distribution in non-acidosis responder (A, C, E) and acidosis-responder (B, D, F) Phox2b mutants under etonogestrel. Photomicrographs of c-FOS/5-HT immunoreactivities in the caudal ROb of acidosis-non responder (G, I) and acidosis-responder (H, J) Phox2b mutants. Photomicrographs in (I, J) correspond to enlargements of the outlined area in (G, H); note the greater number of dual-labeled c-FOS/5-HT neurons (*) in acidosis-responder than in non-acidosis responder Phox2b mutants. Photomicrographs of c-FOS immunoreactivity in the ventral medullary surface, ventromedial to the facial nucleus in the pf-RTN (K, L) showing that the number of c-FOS positive neurons (solid black arrows) was greater in acidosis-responder than in non-acidosis-responder Phox2b mutants. Photomicrographs of immunofluorescence detection for PHOX2B in wildtype littermates (M, O, Q) and Phox2b mutants (N, P, R) in the ventral medullary surface, just below the caudal edge of the facial nucleus, in the cRTN (M, N), ventromedial to the facial nucleus, in the pf-RTN (O, P) and ventrolateral below the facial nucleus, in the pFRG (Q, R). In wildtype littermates, PHOX2B neurons are distributed in all three delimitations with the highest number of neurons in the cRTN as already described (5). In Phox2b mutants, the number of neurons was drastically reduced but some residual cells were still present as previously described (hollow white arrows) (6). Dual c-FOS (S, T, W, X) and PHOX2B (U, V, Y, Z) detections by immunofluorescence under the ventromedial part of the RTN (pf-RTN) in acidosis-non-responder (S, U) and acidosis-responder (T, V) Phox2b mutants and under the pFRG in acidosis-non-responder (W, Y) and acidosis-responder (X, Z) Phox2b mutants; note the presence of c-FOS positive (solid white arrows) but PHOX2B negative cells in acidosis-responder Phox2b mutants. Scale bar = 200 µm (A–F), 100 µm (M–R), 50 µm (G, H), and 20 µm (I, J, K, L, S, T, U, V). 7N, facial nucleus; 10N, dorsal motor nucleus of the vagus; 12N, hypoglossal nucleus; Amb, ambiguus nucleus; AP, area postrema; c/mNTS, commissural and median parts of the nucleus of the tractus solitarius; IO, inferior olives; VLM, ventrolateral medullary reticular nucleus; pFRG, parafacial respiratory group; pyr, pyramidal tract; ROb, raphe obscurus nucleus (caudal part—cROb, from pyramidal decussation to rostral edge of the inferior olive and rostral part—rROb, from rostral edge of the inferior olives to rostral edge of the facial nucleus); RPa, raphe pallidus nucleus (caudal part—cRPa, from pyramidal decussation to rostral edge of the inferior olive and rostral part—rRPa, from rostral edge of the inferior olives to rostral edge of the facial nucleus); RTN, retrotrapezoid nucleus (parafacial RTN—pf-RTN, in ventromedial position under the facial nucleus and caudal RTN—cRTN, immediately to the caudal edge of the facial nucleus); sp5, spinal trigeminal tract; Sp5, spinal trigeminal interpolaris nucleus; vlNTS, ventrolateral part of the nucleus of the tractus solitaries; VLM, ventrolateral medullary reticular nucleus; WT, wildtype. Note that the blue staining along the medial axis in (G–J) photomicrographs corresponds to background observed when c-FOS labeling was revealed with Ni-concentrated DAB solution so that the contrast between c-FOS-positive nuclei and 5-HT-positive soma could be clearly distinguished.

Fig 4: Effect of an increase in 5-HT signaling by fluoxetine in Phox2b mutant mice. Scatter plots with a superimposed box and whisker plot (median [Q1; Q3]) showing percentage of change of fR under fluoxetine in Phox2b mutants (A) and wildtype littermates (B). In Phox2b mutants, 3.125 (n=14), 6.25 (n=12) and 12.5 µM (n=10) of fluoxetine. In wildtype littermates, 3.125 (n=15), 6.25 (n=16) and 12.5 µM (n=15) of fluoxetine. * p < 0.05 indicates a significant change in fR relative to pre-fluoxetine values (paired t-test or Wilcoxon matched pairs signed rank test depending on the normality of data distribution). (C–E), Individual phrenic activity traces (fourth cervical ventral nerve root, C4) and integrated C4 activity from medullary-spinal cord from Phox2b mutant mice in metabolic acidosis under drug-free (C, n=23), etonogestrel (D, n=39) and etonogestrel/fluoxetine (E, n=14) exposure. (F), Scatter plot with surperimposed median [Q1; Q3] showing the respiratory-like rhythm (respiratory frequency, fR) during the last five minutes of metabolic acidosis in percentage of pre-metabolic acidosis. In the presence of fluoxetine, the restoration of the respiratory response to metabolic acidosis induced by etonogestrel in Phox2b mutant mice was abolished. *Indicates a significant change in fR relative to pre-metabolic acidosis values (*p < 0.05; paired t-test). §Indicates a significant difference between the 3 conditions (§p < 0.05; Kruskal-Wallis test followed by Benjamini, Krieger and Yekutieli’s multiple comparison test). ?C4, integrated activity of C4 ventral nerve root; C4, electrical activity of the C4 ventral nerve root. Yellow squares and pink circles represent respectively male and female Phox2b mutant mice acidosis-responders (+10% above pre-metabolic values) and green squares and grey circles represent respectively male and female Phox2b mutant mice acidosis-non-responders under, drug-free, etonogestrel and etonogestrel/fluoxetine, respectively.

Fig 5: Effects of etonogestrel on central H+ chemosensitivity in Phox2b mutant mice. (A–H), Individual phrenic activity traces (fourth cervical ventral nerve root, C4) and integrated C4 activity from medullary-spinal cord from wildtype littermates (A–D) and Phox2b mutant mice (E–H) preparations, under drug-free (A, B, E, F) or etonogestrel (C, D, G, H; 5.10-2 µM) exposure, in normal-pH (A, C, E, G; pre-metabolic acidosis) or metabolic acidosis conditions (B, D, F, H). (I, J), Scatter plots with a superimposed box and whisker (median [Q1; Q3] and minimum and maximum values) showing the respiratory-like rhythm (respiratory frequency, fR) during the last five minutes of metabolic acidosis in percentage of pre-metabolic acidosis values under drug-free and etonogestrel exposure in wildtype littermates (I; n=31 and n=38 respectively) and Phox2b mutant mice (J; n=23 and n=39). Etonogestrel induced not only a restoration of the response to metabolic acidosis considering all preparations, but also an increase in the proportion of acidosis-responder (preparations displaying an increase in fR under metabolic acidosis compared to pre-metabolic acidosis values by at least 10%, p<0.05; Fisher’s exact test; J, yellow squares for males and pink circles for females). *Indicates a significant difference in fR relative to pre-metabolic acidosis values (*p < 0.05, ***p < 0.001, ****p < 0.0001; paired t-test or Wilcoxon test). §Indicates a significant difference in Phox2b mutant mice between preparation exposed to metabolic acidosis with and without etonogestrel (§§p < 0.01; Mann-Whitney test). ?C4, integrated activity of C4 ventral nerve root; C4, electrical activity of C4 ventral nerve root, WT, wildtype.