G protein G(alpha)o is essential for vomeronasal function and aggressive behavior in mice

Abstract

The rodent vomeronasal organ (VNO) mediates the regulation of species-specific and interspecies social behaviors. We have used gene targeting to examine the role of the G protein Gαo, encoded by the gene Gnao1, in vomeronasal function. We used the Cre-loxP system to delete Gαo in those cells that express olfactory marker protein, which includes all vomeronasal sensory neurons of the basal layer of the VNO sensory epithelium. Using electrophysiology and calcium imaging, we show that the conditional null mice exhibit strikingly reduced sensory responses in V2R receptor-expressing vomeronasal sensory neurons to specific molecular cues, including MHC1 antigens, major urinary proteins, and exocrine gland-secreting peptide. Gαo is also vital for vomeronasal sensing of two N-formylated mitochondrially encoded peptides derived from NADH dehydrogenase 1. Furthermore, we show that Gαo is an essential requirement for the display of male-male territorial aggression as well as maternal aggression in mice. Finally, we show that Gαo-dependent maternal aggression can be induced by major urinary proteins. These cellular and behavioral phenotypes identify Gαo as the primary G-protein α-subunit mediating the detection of peptide and protein pheromones by sensory neurons of the VNO.

Fig. 1.

Patterns of protein expression in B6 and cGαo^−/− mice. (A) Loss of Gαo immunoreactivity in VNO and AOB (encircled) of cGao^−/− mice (Right). In B6 control mice, Gαo immunoreactivity (green) is present in basal VSNs and caudal AOB (Left). (Scale bars: VNO, 50 μm; AOB, 100 μm.) A, anterior; P, posterior. (B) Comparison of Gαo1 expression (copies per 1 μg total RNA) in VNO, MOE, and brain of cGαo^−/− vs. B6 mice. LSD: ***P < 0.0001; ns, not significant. (C) Immunohistochemistry using antibodies specific for PDE4A or V2R2 receptors shows that laminar VNO organization persists in the absence of Gαo. (Scale bars: 50 μm.) (D) Analysis of the number of V2R2-positive VSNs indicates a significant reduction of VSNs in the basal VNO layer of cGαo^−/− vs. B6 mice (unpaired t test: P < 0.0001). (E) The number of PDE4A-positive VSNs in cGαo^−/− vs. B6 mice did not change (unpaired t test: P = 0.59).

Fig. 2.

Essential role of Gαo in molecular sensing by V2R-expressing VSNs. (A and B) Local field potentials (EVG) generated by VSNs of B6 and cGαo^−/− mice in response to 500-ms pulses of SYFPEITHI (10⁻¹¹ M), ESP1 (10⁻⁷ M), MUP25 (10⁻⁷ M), 2-heptanone (10⁻⁷ M), and isobutylamine (10⁻⁷ M). (B) EVG peak responses in B6 and cGαo^−/− mice. Number of independent recordings indicated above each bar. Each stimulus was tested in at least four mice. LSD: **P < 0.001, ***P < 0.0001. No difference was observed between B6 and Gnao1^fx/fx control mice (LSD: P = 0.10–0.87). (C and E) Examples of Ca²⁺ transients imaged in three V2R2-positive neurons of B6 (C) or cGαo^−/− mice (E). Each cell was stimulated successively with SYFPEITHI (10⁻¹¹ M), ESP1 (10⁻⁷ M), MUP25 (10⁻⁷ M), and KCl (50 mM). Responses to peptide and protein cues were mutually exclusive. Note that Gαo-deficient, V2R2-positive VSNs did not produce Ca²⁺ responses to the three ligands but showed normal responses to high K⁺. (D and F) Examples of Fura-2 ratio and post hoc immunohistochemistry images showing V2R2-positive neurons from B6 (D; c18) or cGαo^−/− mice (F; c35), respectively. Time points at which Ca²⁺ images were acquired are indicated by arrowheads (gray) in C and E. BF, bright field; DAPI, nuclear stain. (G) Analysis of ligand-evoked Ca²⁺ responses in single VSNs from cGαo^−/− (n = 13) vs. B6 mice (n = 15). Number of cells analyzed is indicated above each bar. (H) Analysis of ligand-evoked Ca²⁺ responses in identified V2R2-positive VSNs from cGαo^−/− (n = 8) vs. B6 mice (n = 8). Number of V2R2-positive neurons analyzed is indicated above each bar. Stimulus concentrations were as in E. Data are expressed as means ± SEM.

Fig. 3.

Essential role of Gαo in vomeronasal recognition of mitochondrial formyl peptides. (A and B) Local field potentials generated in the VNO of B6 (A) and cGαo^−/− mice (B) in response to 500-ms pulses of f-MFFINTLTL, f-MFFINILTL, or f-MLF (each at 10⁻⁷ M). (C) Field potential peak responses in B6 and cGαo^−/− mice. Numbers of independent recordings are indicated above each bar. Each stimulus was tested in at least four mice. Data are expressed as means ± SEM. LSD: **P < 0.001; ns, not significant; P = 0.49. No difference was observed between B6 and Gnao1^fx/fx control mice (LSD: P = 0.08–0.29).

Fig. 4.

Loss of male–male aggression in cGαo^−/− mice. (A–C) Analysis of attack duration (A), number of attacks (B), and latency to first attack (C) in cGαo^−/− males (n = 19) vs. B6 males (n = 22). LSD: ***P < 0.0001. Castrated intruder males each were swabbed with 40 μL LMW or B6-MUPs urine fractions. (D) cGαo^−/− males (n = 19) did not show increased intermale mounting compared with B6 mice (n = 22). Mann–Whitney u test: ns, not significant; P = 0.85. Number of individual measurements is indicated in brackets above each bar. Each value consists of three repeated measurements (performed on different days) per animal.

Fig. 5.

Loss of maternal aggression in lactating cGαo^−/− females. (A–C) Analysis of attack duration (A), number of attacks (B), and latency to first attack (C) in cGαo^−/− (n = 7) vs. B6 females (n = 8). LSD: *P < 0.05; **P < 0.001; ***P < 0.0001. In contrast to cGαo^−/− females, all B6 females responded aggressively to castrated intruder males swabbed with LMW or B6-MUPs. Maternal aggressive behavior in cGαo^−/− females (B and C) was not significantly different from B6 females exposed to intruders swabbed with PBS (LSD: P = 0.26–0.55). Numbers of experiments are indicated above each bar. (D) Pup retrieval abilities did not differ significantly between lactating B6 (n = 4) and cGαo^−/− (n = 6) females. LSD: P = 0.09–0.81. (E) Schematic model outlining the impact of multiple vomeronasal circuits on the promotion of aggressive behaviors in mice.