Figures and data
Oxtr reduces the amount of time naive animals take to show partner preference
A. Schematic of the partner preference tests (PPTs) and the day on which they were conducted to study pair bonding behaviors. The prosocial, agonistic and mating behaviors scored are shown below. During the bond formation phase, the ratio of partner-directed vs stranger-directed affiliative behaviors increases in wild types and following the bond formation in the bond maintenance phase, wild types show decreased prosocial behaviors and increased agonistic behaviors towards strangers.
B. A preference for partner is induced in the WTs but not Oxtr1-/- animals following short cohabitation.
C. WT females spend more time huddling with their partners.
D. Oxtr1-/- females spend more time huddling with strangers.
E. WT females show a significant increase in partner huddling time towards the end of a 3-hour assay.
F. Oxtr1-/- males show a significant increase in stranger huddling time in the middle of a 3-hour assay.
G. No difference in partner huddling time between WT and Oxtr1-/- animals following long cohabitation.
H. No difference in stranger huddling time between WT and Oxtr1-/- animals following long cohabitation.
See Supplementary Figure 1 and Supplementary Table 1.
Mean +/- SD, n =>8, *p =< 0.05, FWT = female WT, FOxtr1-/- = female Oxtr1-/-, MWT = male WT, MOxtr1-/- = male Oxtr1-/-.
Oxtr suppresses promiscuous behaviors in a state dependent manner in females
A. Timeline of the behavioral battery carried out on females. Bottom row shows the behaviors scored.
B. Oxtr1-/- female long huddle duration is shorter with partners and longer with strangers after a short cohabitation.
C. Oxtr1-/- female long huddle frequency is greater with strangers, but not partners after a short cohabitation.
D. No difference in median long huddle durations between WT and Oxtr1-/- females post long cohabitation.
E. No difference in long huddle frequencies between WT and Oxtr1-/- females post long cohabitation.
F. Increasing cohabitation time increases the median long huddle duration an Oxtr1-/- female executes with the partner.
G. There is no significant change in the median long huddle duration conducted by females with strangers when cohabitation time is increased.
H. Naive Oxtr1-/- and WT females are equally prosocial towards a naive WT male.
I. There is no difference in the counts of mating behaviors between Oxtr1-/- and WT females.
J. Oxtr1-/- and WT females are equally prosocial towards their partners after a brief separation.
K. Oxtr1-/- females are significantly more prosocial towards a naive WT stranger male compared to WT females.
See Supplementary Figure 2 and Supplementary Table 1.
Mean ± SD, n =>8, *p < 0.05, FWT = female WT, FOxtr1-/- = female Oxtr1-/-, MWT = male WT, MOxtr1-/- = male Oxtr1-/-.
Oxtr function influences different prosocial behaviors across pair bonding in males
A. Timeline of the behavioral battery carried out on males. Bottom row shows the behaviors scored.
B. Oxtr1-/- male long huddle duration is shorter with partners and strangers after a short cohabitation.
C. Oxtr1-/- male long huddle frequency is greater with strangers, but not partners after a short cohabitation.
D. The median long huddle executed by Oxtr1-/- males with the stranger after a long cohabitation, is shorter than their WT siblings.
E. No difference in the long huddle frequencies between WT and Oxtr1-/- males post long cohabitation.
F. There is no change in the frequency of long huddles with partner between Oxtr1-/- and WT males.
G. Increasing cohabitation time decreases the frequency of long huddles with strangers in Oxtr1-/- males.
H. Naive Oxtr1-/- and WT males are equally prosocial towards naive WT females.
I. Oxtr1-/- males execute higher counts of mating behaviors compared to WT.
J. Prosocial behaviors following a brief separation with their partner are not different between Oxtr1-/- and WT males.
K. Oxtr1-/- males are significantly more prosocial towards a naive WT stranger female compared to WT males.
See Supplementary Figure 3 and Supplementary Table 1.
Mean +/- SD, n =>8, *p =< 0.05, FWT = female WT, FOxtr1-/- = female Oxtr1-/-, MWT = male WT, MOxtr1-/- = male Oxtr1-/-
Oxtr sex-specifically influences social interactions between potential mates
A. Schematic of naive social choice with a naive WT animal choosing between an opposite sex WT and opposite sex Oxtr1-/- animal (red triangles).
B. Male WT choosers prefer WT females.
C. Preference for WT females over Oxtr1-/- females by WT males manifests during the middle of a 6-hour assay.
D. Schematic of comparison of time in contact with the MOAT from WT-Oxtr1-/- condition to contact with the MOAT from WT-WT condition.
E. WT female chooser spends more time in contact with a MOAT from a WT-Oxtr1-/- condition than the WT-WT. The color of the dots represents the genotype of the MOAT.
F. WT female choosers transition between contacts more frequently in the WT-WT condition compared to the WT-Oxtr1-/- condition.
G. WT female chooser spend significantly more time in contact with the MOAT from the WT-Oxtr1-/- condition in the last 40 mins of the assay.
H. WT female choosers prefer their MOAT over longer periods of time in the WT-Oxtr1-/- condition. WT male choosers prefer their MOAT earlier in the WT-Oxtr1-/- condition, but then show preference for the same amount of time as the WT-WT.
See Supplementary Figure 4 and Supplementary Table 1.
Mean +/- SD, n = 8, *p =< 0.05, FWT = female WT, FOxtr1-/- = female Oxtr1-/-, MWT = male WT, MOxtr1-/- = male Oxtr1-/-
Post pair bonding gene expression is driven by female specific changes dependent on Oxtr signaling
A. Schematic of experimental groups used for sequencing highlighting pre- vs post-pairing comparisons between groups. Animals with symbols represent those used for tissue collection.
B. Scatterplot illustrating the relationship between fold change and mean expression, emphasizing the 13 differentially expressed (DE) genes in post-pairing individuals compared to pre-pairing. Gray dots indicate non-significant genes (padj > 0.05); Blue dots indicate downregulated genes (padj < 0.05, log2FC <0) and upregulated genes in red (padj < 0.05, log2FC>0). The right panel shows box plots of normalized gene expression for the top six DE genes by fold-change between pre- and post-pairing conditions.
C. Bar graph emphasizing how each combination of genotype and sex (denoted at bottom as ‘subgroup’) influence the 13 overall pairing-DE genes identified. The top graph depicts the absolute fold change and standard error of the differential expression analysis including all pre- and post-bonding individuals in which the 13 genes of the x axis were identified. The bottom graph depicts the absolute fold change and standard error in analyses when only individuals of that subgroup were considered. (* denotes padj < 0.05 for the pre- vs post-pairing analysis).
D. Box plots of select genes that show sex- and genotype-specific changes in expression with pairing condition. Adora2A and Fkbp5 are DE when considering WT females, Spred3 is DE when considering WT and Oxtr1-/- females, and Sema3b is DE when considering Oxtr1-/- males. Note: Sema3b is the mouse ortholog gene symbol for vole gene “ENSMOCG00000019704” which does not have an assigned gene symbol.
E. Heatmap showing correlations between module eigengenes (MEs) for modulesWT and pairing condition (Pre=1, Post=0). Heatmap color and value reflect Pearson correlation coefficient; adjusted p values shown in parentheses (padj = p * number of modules).
F. Preservation of modulesWT in OXTR-mutant samples. The Zsummary measure combines module density and intramodular connectivity metrics to a composite statistic where Z>2 suggests moderate preservation and Z>10 suggests high preservation119. Of all modulesWT(n=15), MpinkWT had the lowest Zsummary (Zsummary=2) in mutant samples.
G. Box plots showing comparison of MEpinkWT for pre- vs post-pairing condition separated by sex and genotype. (Wilcoxon rank sum test, Bonferroni adjusted for 4 comparisons).
For all box plots center = median, box boundaries = 1st and 3rd quartile, whiskers =
1.5*IQR from boundaries. (n = 16 pre-pairing and 15 post-pairing) See Supplementary Figure 5.
Oxtr regulated genes are common to social behavior relevant processes across species and are similarly regulated across brain regions
A. Schematic of experimental groups for sequencing, highlighting genotype comparisons between groups. Animals with symbols represent those used for tissue collection.
B. Scatterplot illustrating the relationship between fold change and mean expression, emphasizing differentially expressed (DE) genes in Oxtr1-/- individuals compared to WT. Significantly downregulated genes are shown in blue (log2FC<-0.25, padj<0.05) and upregulated genes in red (log2FC>0.25, padj < 0.05). Box plots of normalized gene expression for the top six DE genes by fold change between genotypes.
C. Gene Ontology enrichment analysis for all genotype-DE genes (n=1014) highlights processes related to extracellular channel binding activity, axonal processes, behavior related gene expression, and cation transport. The leaves represent significant GO categories (BH adjusted p value < 0.05) which have been hierarchically clustered by their semantic similarity. Branches are colored by cluster and high-frequency words are displayed to the right. Leaf size and color correspond to the number and median fold change of DE genes in each category.
D. Enrichment analysis shows that genes which are DE in Oxtr1-/- samples are over-represented in some neuropsychiatric disease-associated gene sets. The first DE gene set evaluated (‘Joint’) represents genes which are DE in Oxtr1-/- samples when considering both sexes in the differential expression analysis (padj < 0.01). The second gene set are DE genes when only considering females, and the third when only considering males. Gene sets are further separated by direction of effect across the x-axis to highlight the enrichment in upregulated DE genes. Colors correspond to the Benjamini-Hochberg corrected Chi-square p values; asterisks denote padj < 0.05. Y-axis abbreviations are as follows: SCZ=schizophrenia, NDD=neurodevelopmental disorder, EPI=epilepsy, and ASD=autism spectrum disorder.
E. Box plots of Calcr and dlk1 normalized gene counts by genotype. Log2FC:-0.962 and padj: 1.75e-06 for Calcr and log2FC: -0.276 and padj: 0.00334 for dlk1 from comparison considering all WT and Oxtr1-/- samples.
F. Brain regions selected for in situ analysis. (NAc=nucleus accumbens, PVN=paraventricular nucleus of the hypothalamus, LS=lateral septum)
G. ISH images and quantification for Calcr expression in NAc, PVN, and LS from WT and Oxtr1-/- animals.
H. ISH images and quantification for Dlk-1 expression in NAc, PVN, and LS from WT and Oxtr1-/- animals.
Box plots in G. and H. show comparison of the area-normalized cell number by genotype. (n=3 per condition, Wilcoxon sign rank, *=p<0.05, ***=p<0.001)
See Supplementary Figure 6 and Supplementary Table 1.
Loss of Oxtr results in the decrease in density of Oxt and Avp expressing neurons in the PVN
A. UMAP visualization of mouse hypothalamus single-cell RNAseq dataset66 with proposed PVN clusters labeled. The authors’ original numeric cluster labels and embedding are shown. Each cluster is colored by the Jaccard similarity coefficient between the mouse marker gene sets and the vole Oxtr1-/- DE genes (ONP set). This is also represented in a heat plot to the right. Some of the highest similarity is seen in Cluster 43, which represents Oxt+ cells, and Cluster 26, which represents Avp+ cells in the PVN.
B. Representative images of IHC staining against OT and AVP neuropeptides from FWT and FOxtr1-/- PVN showing the loss of structural integrity and OT and AVP positive cells from the knock-out. (White dotted line denotes the PVN area, scale bar represents 100µm)
C. Mean of positive cells per unit area across all sections is lower for Oxtr1-/-s vs WT for both OT (left) and AVP (right). Each dot represents 1 animal.
D. The number of sections with a high density of OT positive cells is greater in both female (left) and male (right) WT animals.
E. Similarly, WT animals of both sexes show an increase in the number of sections containing AVP positive neurons in high densities.
F. The average OT cell density from matched PVN sections arranged from anterior to posterior shows that the decrease in cell density in Oxtr1-/- animals is biased towards the anterior. Each dot represents the average cell density, the lines represent the sixth order non-linear regression model for cell density by sex and by genotype (n=3) and the ribbons represent the standard deviation.
G. Oxtr1-/- animals show a significant decrease in AVP cell density in the anterior PVN in both sexes.
See Supplementary Figure 7 and Supplementary Table 1.
Mean +/- SD, n = 3, *p =< 0.05, FWT = female WT, FOxtr1-/- = female Oxtr1-/-, MWT = male WT, MOxtr1-/- = male Oxtr1-/-.
