Hemoglobin in the blood acts as a chemosensory signal via the mouse vomeronasal system

Abstract

The vomeronasal system plays an essential role in sensing various environmental chemical cues. Here we show that mice exposed to blood and, consequently, hemoglobin results in the activation of vomeronasal sensory neurons expressing a specific vomeronasal G protein-coupled receptor, Vmn2r88, which is mediated by the interaction site, Gly17, on hemoglobin. The hemoglobin signal reaches the medial amygdala (MeA) in both male and female mice. However, it activates the dorsal part of ventromedial hypothalamus (VMHd) only in lactating female mice. As a result, in lactating mothers, hemoglobin enhances digging and rearing behavior. Manipulation of steroidogenic factor 1 (SF1)-expressing neurons in the VMHd is sufficient to induce the hemoglobin-mediated behaviors. Our results suggest that the oxygen-carrier hemoglobin plays a role as a chemosensory signal, eliciting behavioral responses in mice in a state-dependent fashion.

Fig. 1. Mouse VSNs activated by C57BL/6 male mice blood hemoglobin.

a Representative immunohistochemical image (left) and number of c-Fos-positive cells per VNO section from male mice stimulated by blood (1 µl) or control buffer (right). n = 3 for control and blood. Error bars, S.E.M. Arrowheads represent example c-Fos-positive cells. Scale bar, 100 µm. b Representative immunohistochemical images of anti-pS6 and anti-Gαo staining (left) and number of total pS6-positive cells in the Gαo and Gα_i2 zone of the VNO section from blood- or control buffer- stimulated male mice (right). Arrowheads represent example pS6-positive cells (left) and double-positive VSNs (right). n = 3 for control and blood. Error bars, S.E.M. Scale bar, 100 µm. c Representative immunohistochemical image (left) and number of total c-Fos-positive cells in the glomerular layer (Gl) and mitral/tufted cell layer (M/T) of the AOB section from male mice stimulated by blood (1 µl) (right). Arrowheads represent c-Fos positive cells. n = 3 for control and blood. Error bars, S.E.M. Arrowheads represent example c-Fos-positive cells. Scale bar, 100 µm. d The number of c-Fos-expressing cells in the M/T cell layer of the AOB sections from male mice stimulated with the indicated amount of blood. n = 3 for 0, 0.01, 0.03, 0.3, and 3 µl, and n = 4 for 0.1, 1, and 10 µl. Error bars, S.E.M. e Separation of blood by centrifuge. Separated blood components: plasma, cell lysate and residue were used for SDS-PAGE analysis and a c-Fos-inducing assay in C57BL/6 male mice. n = 3. Error bars, S.E.M. f, g Two-step HPLC purification with DEAE (f) and C4 columns (g). Chromatogram and c-Fos-inducing activity of each fraction are shown. The cell lysate fraction was used for DEAE column chromatography and the resultant active fraction was used for C4 column chromatography. The three fraction peaks were defined as heme, α-globin and β-globin by absorption spectrometry and mass spectrometry (Supplementary Fig. 1b-d). h c-Fos-inducing activity of recombinant β-globin. n = 3. Error bars, S.E.M. Scale bar, 100 µm. Arrowheads highlight example c-Fos-positive cells in the VNO (top) and in the M/T cell layer of the AOB (bottom). i Representative EVG recording of hemoglobin-dependent negative change in local field potential of the VNO. j Dose-dependent electrical responses of male VNO to hemoglobin. n = 10. Error bars, S.E.M.

Fig. 2. Gly17 on hemoglobin acts as a crucial interaction site for ligand-receptor binding.

a Types of hemoglobin in BALB/c and C57BL/6 strain blood cell lysate. Two types of hemoglobin (β-globin) in the BALB/c mouse strain were separated by HPLC with a DEAE column. b Number of c-Fos-positive cells in the M/T cell layer of the AOB sections from male mice stimulated with hemoglobin from BALB/c and C57BL/6 male mice blood and blood from various kinds of vertebrates. n = 6 for C57BL/6, Rat, Horse, Human, Guinea pig, Frog, and Zebrafish, n = 8 for BALB/c-major, and n = 9 for BALB/c-minor. Error bars, S.E.M. c Alignment of β-globin amino acid sequences. An amino acid G17 is different in BALB/c minor, horse, frog, and zebrafish β-globin. d Representative immunohistochemical images of total c-Fos-positive cells in the M/T of the AOB sections from G17A-mutant-, H78N-mutant-, and control buffer-stimulated male mice. n = 3 for H78N, n = 9 for control and G17A. Arrowheads represent example c-Fos-positive cells. Scale bar, 100 µm. e Number of c-Fos-inducing cells in the M/T cell layer of the AOB sections from G17A-mutant-, H78N-mutant-, hemoglobin (Hb) from BALB/c-, and control buffer-stimulated male mice. n = 3 for H78N, n = 6 for BALB/c Hb, and n = 9 for control and G17A. Error bars, S.E.M. control vs. BALB/c Hb; p = 0.007, G17A vs. BALB/c Hb; p = 0.046, control vs. H78N; p = 0.058, and G17A vs. H78N; p = 0.060 by the two-sided Steel-Dwass test. f Three dimensional structure of human hemoglobin (1A3N, RCSB Protein Data Bank, https://www.rcsb.org/structure/1a3n). Blue and green cartoons in the model represent α-globin and β-globin, respectively. The position of the 17th glycine is highlighted by a red dot.

Fig. 3. Vmn2r88 is a specific vomeronasal receptor for hemoglobin.

a Dual-color ISH staining of a VNO section from a hemoglobin-stimulated C57BL/6 mouse labeled with the Egr1 cRNA probe (green) and V2Rf clade-specific cRNA probe (magenta). n = 3. Scale bar, 50 µm. b, c Bar graph representing the average cell number in VNO sections from C57BL/6 mice stimulated with hemoglobin. In Fig. 3b, probes to distinguish each V2R clade were used. In Fig. 3c, probes to narrow down candidate genes in V2Rf clade (named V2Rf1-V2Rf5) and to further distinguish each gene in V2Rf5 clade (Vmn2r88, 89 and 122) were used. Black, yellow, and gray parts represent only V2R-positive (b) or V2Rf-positive (c), V2R and Egr1 double-positive, and only Egr1-positive cells respectively. Hybridization temperatures are shown below the graphs. d Vmn2r88 and pS6 immunostaining of VNO sections from Vmn2r88^+/+or Vmn2r88^−/− male mice exposed to hemoglobin (Hb) or distilled water (control). Open arrowheads show pS6-positive cells; closed arrowheads show cells double-labeled for pS6 and Vmn2r88. In the mutant mice, corresponding pS6 and Vmn2r88 expression completely disappeared. n = 3 for Vmn2r88^+/+-control, and Vmn2r88^-/--Hb, and n = 5 for Vmn2r88^+/+-Hb. Scale bar, 50 µm. e The number of pS6- (green), Vmn2r88- (magenta), and double-positive cells (yellow) per VNO section. In the group of Vmn2r88^+/+-Hb 16 sections from each of 5 animals were quantified and 16 sections from each of 3 animals were counted in the other groups. Double-positive cells completely disappeared in the sections from Hb-stimulated Vmn2r88^-/- mice.

Fig. 4. Hemoglobin enhances c-Fos expressing cells in the dorsal region of the VMH and PAG only in lactating mothers.

a Representative ISH images of the posterior region of the MeA from C57BL/6 male, virgin female, and lactating mothers pre-stimulated with control- or hemoglobin (Hb)-cotton swabs. c-Fos cRNA probe (red) was used in conjunction with nuclear DAPI staining (blue). Abbreviations: MeApd, MeA posterodorsal region; MeApv, MeA posteroventral region; D, dorsal; and V, ventral. n = 3 for male-control and Hb, virgin female-Hb, n = 4 for virgin female-control, n = 6 for mother-control, and n = 7 for mother-Hb. Scale bar, 100 µm. b The number of c-Fos-positive cells in each sub-region of the MeA stimulated with control buffer or Hb. Abbreviations: MeAa, MeA anterior region. For counting MeAa; n = 3 for male-control and Hb, virgin female-Hb, n = 4 for virgin female-control, n = 6 for mother-control, and n = 7 for mother-Hb. 6 (MeAa) and 8 (MeAp) sections from each animal were quantified. Error bars, S.E.M. (MeApd) p = 0.004, and (MeApv) virgin female; p = 0.057, and mother; p = 0.001 by the two-sided Wilcoxon rank-sum test. c The number of c-Fos-positive cells in the BNST and PMCo stimulated with control buffer or Hb. n = 3 for virgin female-Hb, n = 4 for male-control and Hb, virgin female-control, n = 6 for mother-control, and n = 7 for mother-Hb. 7 (BNST) and 5 (PMCo) sections from each animal were quantified. Error bars, S.E.M. d Representative ISH images of the VMH from males, virgin females, and lactating mothers, stimulated with control buffer or hemoglobin. c-Fos cRNA probe (red) was used in conjunction with nuclear DAPI staining (blue). Abbreviations: d, dorsal region; vl, ventrolateral region. n = 3 for virgin female-Hb, n = 4 for male-control and Hb, virgin female-control, n = 6 for mother-control, and n = 7 for mother-Hb. Scale bar, 100 µm. e Quantification of c-Fos-positive neurons in the VMH. The number of sections counted to determine the number of c-Fos-positive neurons in each brain area of each animal was 10. Error bars, S.E.M. n = 3 for virgin female-Hb, n = 4 for male-control and Hb, virgin female-control, n = 6 for mother-control, and n = 7 for mother-Hb. p = 0.034 by the two-sided Wilcoxon rank-sum test. f Number of c-Fos-positive cells in the MPA stimulated with control buffer or Hb. n = 3 for virgin female-Hb, n = 4 for male-control and Hb, virgin female-control, n = 6 for mother-control, and n = 7 for mother-Hb. 4 sections from each animal were quantified. Error bars, S.E.M. g Representative ISH images of the dorsal region of the PAG from lactating mothers stimulated with control buffer or hemoglobin. c-Fos cRNA probe (red) was used in conjunction with nuclear DAPI staining (blue). n = 5 for mother-control and Hb. Scale bar, 100 µm. h Quantification of c-Fos-expressing neurons in the PAGd. n = 3 for virgin female-Hb, n = 4 for male-control and Hb, virgin female-control, n = 5 for mother-control and Hb. 4 sections from each animal were quantified. Error bars, S.E.M. p = 0.007 by the two-sided Wilcoxon rank-sum test.

Fig. 5. Hemoglobin enhances digging behavior in lactating mothers.

a Schematic illustration of the timeline of the cotton exposure assay. In the trials of lactating mothers, their pups were removed one hour before the cotton exposure. b–d Quantification of the digging time duration (sec) of wild type C57BL/6 lactating mothers (b), Vmn2r88-mutant lactating mothers (c), males, and virgin females (d) with pre-exposure to cotton balls transfused with hemoglobin (Hb), and fresh blood. Wild type C57BL/6 mothers; n = 15 for control, n = 6 for ESP1, n = 11 for Hb, and n = 8 for fresh blood, Vmn2r88-mutant lactating mothers; n = 5–9, virgin females; n = 9, males; n = 8. Error bars, S.E.M. (b) control vs. fresh blood; p = 0.028, control vs. Hb; p = 0.049, ESP1 vs. fresh blood; p = 0.048, and ESP1 vs. Hb; p = 0.044, and (c) +/+ (control) vs. +/+ (Hb); p = 0.017 and + /+ (Hb) vs. -/- (Hb); p = 0.041 by the two-sided Steel-Dwass test in panel (b) and (c). e, f Quantification of the total freezing duration (e) and digging time duration (f) of lactating mothers stimulated with the indicated concentration (from 100-fold to 50000-fold dilution with mineral oil) of a 2MT transfused cotton swab. n = 3 for 1000-fold and 50000-fold, n = 5 for 100-fold, n = 8 for 5000-fold and 10000-fold. Error bars, S.E.M. (e) 100-fold vs. 10000-fold; p = 0.007, 100-fold vs. 5000-fold; p = 0.007, 1000-fold vs. 10000-fold; p = 0.015, and 1000-fold vs. 5000-fold; p = 0.015, and (f) p = 0.043 by the two-sided Steel-Dwass test.

Fig. 6. Hemoglobin enhances rearing, a type of exploratory behavior, in lactating mothers.

a Schematic illustration of the timeline of the open field (OF) assay with cotton pre-exposure. Cotton exposure was performed in their home cage (with their pups). b Quantification of total distance, total center time, moving speed, and rearing time duration of lactating mothers, pre-stimulated with control buffer-, ESP1-, 2MT- (low 2MT: 10000-fold dilution, high 2MT: 10-fold dilution) or hemoglobin (Hb)-cotton swabs in the open field assay for 10 min. n = 8 for control, n = 7 for Hb, and n = 5 for ESP1, low 2MT, and high 2MT. Error bars, S.E.M. p = 0.048 by two-sided Wilcoxon rank-sum test with Dunnett correction. c Quantification of rearing time duration of Vmn2r88-mutant lactating mothers, pre-stimulated with control buffer- or Hb-cotton swabs. n = 6 for + /+ (Hb) and -/- (Hb), n = 10 for + /+ (control). Error bars, S.E.M. + / + (control) vs. +/+ (Hb); p = 0.098 and + /+ (Hb) vs. -/- (Hb); p = 0.011 by the two-sided Steel-Dwass test.

Fig. 7. SF1-positive cells in the VMHd are important for hemoglobin-dependent digging enhancement in lactating female mice.

a Schematic illustration of the animal setup and timeline for pharmacogenetic inhibition of SF1-expressing neurons in the VMHd. AAV-DIO-hM4Di-mCherry is injected into SF1-positive cells in the VMHd. Image adapted from Allen Mouse Brain Atlas. b A representative coronal section showing DREADD-Gi expression (mCherry-positive cells shown in red) in the VMHd. n = 8. Scale bar, 500 µm. c Quantification of the total digging duration (sec) of hemoglobin (Hb)-stimulated SF1-Cre lactating mothers with i.p. injection of saline or CNO. n = 4 for CNO group. n = 4 for saline group. Error bars, S.E.M. p = 0.030 by the two-sided Wilcoxon signed-rank test. d Schematic image and illustration of the setup and timeline for optogenetic activation of SF1-expressing neurons in the VMHd of lactating females. AAV-DIO-ChR2 or AAV-DIO-GFP are injected into SF1-positive cells in the VMHd and optic fibers are implanted above the target region. Image adapted from Allen Mouse Brain Atlas. e A representative coronal section showing ChR2 expression (eYFP-positive cells shown in green) in the VMHd. n = 5. Scale bar, 500 µm. f, g Quantification of digging behavior, with or without weak light stimulation (f 0 mW, g 0.01 mW). Error bars, S.E.M. p = 0.043 by the two-sided Wilcoxon signed-rank test, n = 5 for GFP and ChR2 group.