Top-down acetylcholine signaling via olfactory bulb vasopressin cells contributes to social discrimination in rats

Abstract

Social discrimination in rats requires activation of the intrinsic bulbar vasopressin system, but it is unclear how this system comes into operation, as olfactory nerve stimulation primarily inhibits bulbar vasopressin cells (VPCs). Here we show that stimulation with a conspecific can activate bulbar VPCs, indicating that VPC activation depends on more than olfactory cues during social interaction. A series of in vitro electrophysiology, pharmacology and immunohistochemistry experiments implies that acetylcholine, probably originating from centrifugal projections, can enable olfactory nerve-evoked action potentials in VPCs. Finally, cholinergic activation of the vasopressin system contributes to vasopressin-dependent social discrimination, since recognition of a known rat was blocked by bulbar infusion of the muscarinic acetylcholine receptor antagonist atropine and rescued by additional bulbar application of vasopressin. Thus, our results implicate that top-down cholinergic modulation of bulbar VPC activity is involved in social discrimination in rats.

Fig. 1. Social interaction activates bulbar VPCs.

Representative average z-projections of the olfactory bulb that were immune-stained for eGFP (green, CF488) and pERK (magenta, CF 594). Scale bar, 50 µm valid for all images in the same panel. a VPCs following water, urine, or rat stimulation. Arrows indicate cells that are double-labeled for eGFP and pERK. b pERK⁺ MCs. c Averaged fraction of pERK⁺ VPCs of all VPCs in different stimulation groups (%). One-way ANOVA, LSD for single comparison, *p < 0.05 rat vs. water. d Cumulative probability of averaged fraction pERK⁺ VPCs of all VPCs in different stimulation groups (%). e Averaged number of pERK⁺ MCs per section in different stimulation groups. One-way ANOVA, LSD for single comparison, *p < 0.05 rat vs. water and rat vs. urine. Data are presented as box-plots including first, median, and third quartiles with whiskers representing the range of data points and distribution of single data points. n = 10 rats (water), n = 9 rats (urine), n = 10 rats (rat).

Fig. 2. Cholinergic modulation triggers excitatory responses and action potentials during ON stimulation of VPCs.

a Schematic drawing of the experimental setup. Whole-cell patch-clamp recordings in 300 µm in vitro slices of responses from VPCs to electrical ON stimulation (50 µA, 100 µs, 30 s intervals). b Representative averaged traces of responses following ON stimulation in the ACSF condition (artificial cerebrospinal fluid, gray) and during bath application of 5-HT (serotonin 20 µM, red). c Cumulative analysis of normalized IPSP amplitudes of evoked IPSPs to the ACSF control condition during bath application of 5-HT (n = 8 cells). Data are presented as means including distribution of single data points. Lines between data points represent related measurements. Related-Samples Wilcoxon Signed Rank Test, *p < 0.05. d Representative averaged traces of responses following ON stimulation in the ACSF condition (artificial cerebrospinal fluid, gray) and during bath application of NA (noradrenaline 20 µM, red). e Cumulative analysis of normalized IPSP amplitudes of evoked IPSPs to the ACSF control condition during bath application of NA (n = 9 cells). Data are presented as means including distribution of single data points. Lines between data points represent related measurements. Related-Samples Wilcoxon Signed Rank Test, *p < 0.05. f Representative averaged traces of responses to ON stimulation in the ACSF condition (gray) and a single trace during bath application of ACh showing APs (acetylcholine, red). g Pie-chart represents the proportion of VPCs showing either APs, EPSPs or IPSPs in the ACSF and ACh condition. Dot-plots represent cumulative analysis of amplitudes of PSPs or APs (set as 100 mV) in the ACSF and ACh condition. Related-Samples Wilcoxon Signed Rank Test, *p < 0.05. n = 20 cells. (h + i) Cumulative probability of evoked PSP amplitudes in the ACSF, 5-HT, NA, or ACh condition (n = 20/8/9/20 cells). The amplitudes of APs were set as 100 mV. Kruskal–Wallis test for variation comparison, Bonferroni post-hoc, *p < 0.05 ACh vs. ACSF, 5-HT, NA. ON, olfactory nerve, G glomerulus, GL glomerular layer, EPL external plexiform layer, VPC vasopressin cell.

Fig. 3. ACh modulates both ON-evoked inhibition and excitation via muscarinic receptors but does not increase intracellular Ca²⁺ influx.

Whole-cell patch-clamp recordings in 300 µm in vitro slices of responses from VPCs to electrical ON stimulation (50 µA, 100 µs, 30 s intervals). a Representative averaged traces of responses following ON stimulation in the ACSF condition (gray) and during bath application of ACh (acetylcholine, red) showing reduced IPSPs. b Cumulative analysis of normalized evoked IPSP amplitudes to the ACSF condition during bath application of ACh (n = 5 cells). c Representative averaged traces of responses following ON stimulation in the ACSF condition (gray) and during bath application of bicuculline (50 µM, black) and additional application of ACh (100 µM, red). d Pie-chart represents the proportion of cells showing either APs or EPSPs in the ACh condition (n = 6 cells). Bar-charts represent cumulative analysis of normalized evoked EPSP amplitudes to the bicuculline condition during additional application of ACh (n = 5 cells). Data are presented as means including distribution of single data points. Lines between data points represent related measurements. Related-Samples Wilcoxon Signed Rank Test, *p < 0.05 vs. ACSF/Bicuculline. e Representative picture of a VPC filled with OGB-1. Arrows indicate the tuft and the apical dendrite near the soma. Scale bar, 50 µm. f Representative averaged traces of Ca²⁺ influx in the tuft and the soma above the dotted line. Representative averaged traces of responses following ON stimulation below the dotted line (ACSF with 50 Hz somatic stimulation or ON stimulation, black; ACh with ON stimulation, red). g ΔF/F₀ in % from baseline following different stimulation and pharmacology. Data are presented as box-plots including first, median, and third quartiles with whiskers representing the range of data points and distribution of single data points. (2) × (2) mixed model ANOVA (location [within subject] × treatment [within-subject]) with treatment being either 50 Hz vs. ON or ACSF vs. ACh, *p < 0.05 vs. tuft within the condition. n = 8 cells. #p < 0.05 50 Hz in the ACSF (soma) vs. ON in the ACSF (soma). n = 8/8 cells.

Fig. 4. The muscarinic pathway is responsible for reduction of ON-evoked inhibition.

a Representative averaged traces of responses from VPCs following ON stimulation in the ACSF condition (gray) and during bath application of nicotine (100 µM, red). b Pie-chart represents the proportion of cells showing either APs, EPSPs, or IPSPs (n = 8 cells). Bar-charts represent cumulative analysis of normalized evoked IPSP amplitudes to the ASCF condition (n = 8 cells without APs and EPSPs). c Representative averaged traces of responses from VPCs following ON stimulation in the ACSF condition (gray) and during bath application of atropine (10 µM, black) and atropine+ACh (10 µM+100 µM, red). d Pie-chart represents the proportion of cells showing either APs, EPSPs or IPSPs in the atropine+ACh condition (n = 11 cells). Bar-charts represent cumulative analysis of normalized evoked IPSP amplitudes to the ASCF condition (n = 8 cells without APs and EPSPs). e Representative averaged traces of responses from VPCs following ON stimulation in the ACSF condition (gray) and during bath application of muscarine (1 µM, red). f Pie-chart represents the proportion of cells showing either APs, EPSPs or IPSPs in the (n = 8 cells). Bar-charts represent cumulative analysis of normalized evoked IPSP amplitudes to the ASCF condition (n = 6 cells without APs and EPSPs). g Representative averaged traces of responses from VPCs following ON stimulation in the ACSF condition (gray) and during bath application of mecamylamine (20 µM, black) and mecamylamine+ACh (20 µM+100 µM, red). h Pie-chart represents the proportion of cells showing either APs, EPSPs or IPSPs in the mecamylamine+ACh condition (n = 7 cells). Bar-charts represent cumulative analysis of normalized evoked IPSP amplitudes to the ASCF condition (n = 7 cells without APs and EPSPs). Data are presented as means including distribution of single data points. Lines between data points represent related measurements. Related-Samples Wilcoxon Signed Rank Test, *p < 0.05 vs. ACSF in agonist experiments. Friedman test, Dunn for single comparisons, *p < 0.05 vs. Meca (mecamylamine) or vs. ACSF in antagonist experiments. N.S., not significant.

Fig. 5. Cholinergic cells in the horizontal limb of the diagonal band of Broca are activated during social interaction.

a Representative average z-projections of the horizontal limb of the diagonal band of Broca (HDB) that were immune-stained for ChAT (choline-acetyltransferase, green, CF488) and pERK (magenta, CF 594) in the different experimental groups. Arrows indicate cells that are double-labeled for pERK and ChAT. Insets show enlarged pictures of the cell indicated by blank arrows. Scale bar, 50 µm valid for all images in the same panel. b + c Averaged number of pERK⁺ cells or pERK⁺ cholinergic cells per section in different experimental groups. Data are presented as box-plots including first, median, and third quartiles with whiskers representing the range of data points and distribution of single data points. One-way ANOVA, LSD for single comparisons, N.S., not significant, *p < 0.05 rat vs. urine, n = 5 rats (water), n = 5 rats (urine), n = 6 rats (rat).

Fig. 6. VP microinjection into the OB rescues atropine-induced impairment of social discrimination in male rats.

a Schematic time course of the experimental design. b Rat (5 weeks) without markings representing anogenital sniffing toward a stimulus rat (3 weeks) with red markings. c Representative picture of the OB with dye injection via drug injection system, counterstained with Cresyl violet and a schematic picture of the olfactory bulb. The cross in the schematic picture indicates the approximate injection position in the left picture. ONL olfactory nerve layer, GL glomerular layer, EPL external plexiform layer, GCL granule cell layer. d–f Amount of time (s) that rats investigate stimulus rats in the sampling phase and in the discrimination phase (known or novel rat) following microinjection (1 µL per bulb) with only vehicle, atropine (1 µg/ 1 µL) and vehicle or atropine and VP (1 ng/ 1 µL). Data are presented as box-plots including first, median, and third quartiles with whiskers representing the range of data points and distribution of single data points. Paired Samples t test between known and novel, *p < 0.05, n = 13 rats (Veh/Veh), n = 14 rats (Atr/Veh), n = 14 rats (Atr/VP). g Amount of time (s) that rats are engaged in playing with the stimulus rat during the sampling phase. Data are presented as box-plots including first, median, and third quartiles with whiskers representing the range of data points and distribution of single data points. Kruskal–Wallis Test, n = 13 rats (Veh/Veh), n = 14 rats (Atr/Veh, 1 µg), n = 14 rats (Atr/VP, 1 µg/1 ng). h Amount of time (s) within time bins of 1 min that rats investigate the stimulus rat during the sampling phase following different drug applications. Data are presented as means ± SEM including distribution of single data points. (4) × (3) mixed model ANOVA (time bin [within subject] × treatment [between-subject]), n = 13 rats (Veh/Veh), n = 14 rats (Atr/Veh, 1 µg), n = 14 rats (Atr/VP, 1 µg/1 ng). N.S., not significant.

Fig. 7. Graphic summary of OB pathways involved in social discrimination.

Blue, green, and gray arrows represent excitatory and red lines inhibitory pathways indicated by this study and our previous publication. The cells within the dashed circle indicate the excitatory network within the same home glomerulus. Dashed pathways were confirmed but not fully dissected on the synaptic level. Green pathways are related to vasopressin signaling, gray pathways to sensory input or behavioral output and black pathways are based on findings from literature. Question marks indicate potential dendritic and axonal release of VP in the OB. Axonal projections innervate GL, EPL, MCL, and sGCL. EPL external plexiform layer, eTC external tufted cell, GL glomerular layer, HDB horizontal limb of the diagonal band of broca, MC mitral cell, MCL mitral cell layer, mTC middle tufted cell, ON olfactory nerve, sGCL superficial granule cell layer, VPC vasopressin cell.