Activated OT neurons in the PVN of mandarin voles during pup care (n =4) and infanticide (n = 4).

a. Representative histological images of OT (green) and c-Fos (red) positive cells in PVN: Blue, DAPI: Yellow, merged cells. Enlarged views of the boxed area were shown to the right of each image, white arrows indicated the overlap of OT and c-fos positive cells. Scale bars, 50 μm. b. Number of OT positive cells in pup care and infanticide female voles. c. Number of c-Fos positive cells in pup care and infanticide female voles. d. Percentage of c-Fos-expressing cells in OT cells of PVN from pup care and infanticide female voles. **P < 0.01. Independent sample t-tests. e. Number of OT cells in pup care and infanticide male voles. f. Number of c-Fos cells in pup care and infanticide male voles. g. Percentage of c-Fos-expressing cells in OT cells of PVN from pup care and infanticide male voles. *P < 0.05. Independent sample t-tests. Erroe bars indicate SEM.

Figure 1-souce data 1.

Statistical results of data on the number of OT-positive cells, the number of cfos-positive cells, and the rate of overlap between OT and cfos in the PVN of female and male pup-care and infanticide voles.

Effects of optogenetic activation of PVN OT neurons on pup-directed behaviors

a. Schematic of virus injection and optical fiber implantation. b. Schematic diagram of the behavioral test. c. Representative histologic images of CHR2 (red) expression and OT staining (green), enlarged view of the boxed area is on the right side. Blue, DAPI. Scale bars, 50 μm. d. Statistics on the specificity of CHR2 expression in 3 voles, more than 89% of CHR2 positive neurons overlapped with OT positive neurons. e. Time line of the experiment. f-i, Approach (f, h) and infanticide (g, i) latency of infanticide 7 female and 6 male voles in mCherry (control virus group) and CHR2 groups. **P < 0.01 vs. CHR2 OFF; ***P < 0.001 vs. CHR2 OFF; ****P < 0.0001 vs. CHR2 OFF. Two-way rmANOVA (factors: treatment × stimulus). j-n, Latency to approach (j), latency to retrieve (k), grooming time (l), sniffing time (m) and crouching time (n) of 7 pup-care female voles in female control virus and CHR2 groups. o-s, Latency to approach (o), latency to retrieve (p), grooming time (q), sniffing time (r) and crouching time (s) of 8 pup-care male voles in male control virus and CHR2 groups. *P < 0.05 vs. CHR2 OFF, ***P < 0.001 vs. CHR2 OFF. Two-way rmANOVA (factors: treatment × stimulus). Error bars indicate SEM.

Figure 2-source data 1.

Statistical results of the number of cells expressing only CHR2 and co-expressing CHR2 and OT in the PVN of 3 optogenetic virus-injected voles, the approach latency and infanticide latency of infanticide voles, and the approach latency, retrieval latency, grooming time, sniffing time, and crouching time of pup-care voles.

Effects of optogenetic inhibition of PVN OT neurons on pup-directed behaviors a. Schematic of virus injection and optical fiber implantation. . b. Schematic diagram of the behavior test. c. Representative histological images of OT staining (green) and eNpHR (red) expression, enlarged view of the boxed area is shown on the right side. Blue, DAPI. Scale bars, 50 μm. d. Statistics on the specificity of CHR2 expression in 3 voles, more than 90% of eNpHR expression overlapped with OT. e. Time line of the experiment. f-i, Approach (f, h) and infanticide (g, i) latency in 6 female (f, g) and 6 male (h, i) infanticide voles. *P < 0.05 vs. eNpHR OFF; **P < 0.01 vs. eNpHR OFF; ****P < 0.0001 vs. eNpHR OFF. Two-way rmANOVA (factors: treatment × stimulus). j-n, Approach latency (j), retrieval latency (k), grooming time (l), sniffing time (m) and crouching time (n) in 6 control mCherry and eNpHR group of pup-care female voles. o-s, Approach latency (o), retrieval latency (p), grooming time (q), sniffing time (r) and crouching time (s) in 6 control mCherry and eNpHR group of pup-care male voles. Error bars indicate SEM.

Figure 3-source data 1.

Statistical results of the number of cells expressing only eNpHR and co-expressing eNpHR and OT in the PVN of 3 optogenetic virus-injected voles, the approach latency and infanticide latency of infanticide voles, and the approach latency, retrieval latency, grooming time, sniffing time, and crouching time of pup-care voles.

OT release in the mPFC upon pup-directed behaviors a. Recording instrument settings. b. Illustrations of viral expression and optical fibre location. c. Representative histological image of OT1.0 sensor (green) and optical fibre locations. Blue, DAPI. Scale bars, 500 μm. d, e, Area under the curve (AUC) per second for pup-care female (d) and male (e) voles approaching pups for the first, second and third time (n = 8). *P < 0.05 vs. first. **P < 0.01 vs. first. One-way rmANOVA. f. Representative ΔF/F traces in pup-care female (f, top) and male (f, bottom) voles during interaction with pups. g-k, Post-event histograms (PETHs) of z-score of OT1.0 sensor for the following pup-directed behaviors: approach (g), investigate (h), retrieve (i), grooming (j) and crouch (k). l,m, The mean AUC of z-scores for pup-care female (l) and male (m) voles across various pup-directed behaviors (n = 8). Female: **P < 0.01 vs approach. P < 0.01 vs retrieve. Male: *P < 0.05 vs approach. P < 0.05 vs retrieve. One-way rmANOVA. n. Representative ΔF/F traces in infanticide female (n, top) and male (n, bottom) voles during interaction with pups. o,p, PETHs of z-score of OT1.0 sensor for approach and infanticide in infanticide voles. q,r, The mean AUC of z-score of OT1.0 sensor for pre pup, approach and infanticide in infanticide female (q) and male (r) voles (n = 8). **P < 0.01 vs infanticide. One-way rmANOVA. Eerro bars indicate SEM.

Figure 4-source data 1.

Area under the curve per second for pre-pup, approach, and infanticide behaviors in infanticide voles and area under the curve per second for first, second, and third approaches to pups, as well as pre-pup, approach, investigation, retrieval, grooming, and crouching behaviors in pup-care voles.

Determination of PVN to mPFC oxytocin projection. a. Schematic diagram of mPFC virus injection and OT staining. b. Histological pictures of OT (red) and AAV11 (green) co-staining in male and female. Yellow, merged. Blue, DAPI. Scale bars, 50 μm.

Effects of optogenetic activation of the PVN OT neuron projection terminals on pup-directed behaviors a. Schematic of virus injection and optical fiber implantation. b. Schematic diagram of the behavior test. c. Illustration of optical fiber implantation in the target brain region. d. Representative histological pictures of the fiber position and projection terminals. Blue, DAPI. Scale bars, 50 μm. e. Time line of the experiment. f-i, Changes in approach latency (f, h) and infanticide latency (g, i) of female (f, g) and male (h, i) infanticide voles in CHR2 and control mCherry group before and after delivery of light (n = 6). **P < 0.01 vs. CHR2 OFF; ***P < 0.001 vs. CHR2 OFF. Two-way ANOVA (factors: treatment × stimulus). j-n, Approach latency (j), retrieval latency (k), grooming time (l), sniffing time (m) and crouching time (n) in control mCherry and CHR2 group of pup-care female voles (n = 6). o-s, Approach latency (o), retrieval latency (p), grooming time (q), sniffing time (r) and crouching time (s) in control mCherry and CHR2 group of pup-care male voles (n = 6). **P < 0.01 vs. CHR2 OFF. Two-way rmANOVA (factors: treatment × stimulus). Error bars indicate SEM.

Figure 6-source data 1.

Statistical results of the approach latency and infanticide latency in infanticide voles, and the approach latency, retrieval latency, grooming time, sniffing time, and crouching time of pup-care voles.

Optogenetic inhibition of the PVN OT neuron projection terminals promoted the onset of infanticide a. Illustration of virus injection and optical fiber implantation. b. Schematic of the behavior test. c. Diagram of optical fiber implantation in the target brain region. d. Representative histological pictures of the fiber location and projection terminals. Blue, DAPI. Scale bars, 75 μm. e. Time line of the experiment. f-i, Changes in approach (f, h) and infanticide (g, i) latency of female and male infanticide voles in eNpHR (n = 6) and control mCherry groups (n = 6) before and after light delivery. ***P < 0.001 vs. eNpHR OFF. ****P < 0.0001 vs. eNpHR OFF. Two-way rmANOVA (factors: treatment × stimulus). j-n, Approach latency (j), retrieval latency (k), grooming time (l), sniffing time (m) and crouching time (n) in control mCherry (n = 6) and eNpHR (n = 6) group of pup-care female voles. o-s, Approach latency (o), retrieval latency (p), grooming time (q), sniffing time (r) and crouching time (s) in control mCherry (n = 6) and eNpHR group (n = 6) of pup-care male voles. Error bars indicate SEM.

Figure 7-source data 1.

Statistical results of the approach latency and infanticide latency in infanticide voles, and the approach latency, retrieval latency, grooming time, sniffing time, and crouching time of pup-care voles.

Pup-directed behaviors before and after intraperitoneal delivery of OT a. Diagram of intraperitoneal OT delivery. b,c, Approach (b) and infanticide latency (c) of infanticide female voles (n = 8). *P < 0.05. Paired-samples t-test. **P < 0.01. Wilcoxon signed ranks test. d,e, Approach (d) and infanticide latency (e) of infanticide male voles (n = 7). *P < 0.05. Paired-samples t-test. f-j, Approach latency (f, n = 7), sniffing time (g, n = 7), latency to retrieve (h, n = 4), grooming time (i, n = 7) and crouching time (j, n = 7) before and after delivery of OT in pup-care female voles. k-o, Approach latency (k), sniffing time (l), latency to retrieve (m), grooming time (n) and crouching time (o) before and after delivery of OT in pup-care male voles (n = 5). *P < 0.05. Paired-samples t-test. p,q Changes in infanticide rates in female (p) and male (q) voles after administration of saline and OT. *P < 0.05. Independent samples t-tests. Error bars indicate SEM.

Figure 8-source data 1.

Statistical results of the approach latency and infanticide latency in infanticide voles, and the approach latency, retrieval latency, grooming time, sniffing time, and crouching time of pup-care voles. Statistical results on the number of infanticide or not infanticide in voles injected with saline and OT.

Light-induced c-Fos expression overlapping with mCherry, CHR2 or eNpHR

a, 473 nm light-induced c-Fos (green) expression overlapping with mCherry (red). Blue, DAPI. Scale bars, 50 μm. b. 473 nm light-induced c-Fos (green) expression overlapping with CHR2 (red). Blue, DAPI. Scale bars, 50 μm. c, 473 nm light induced more c-fos expression in CHR2, which indicated the effectiveness of CHR2 (n = 3). ****P < 0.0001. Independent-samples t-test. d, 589 nm light-induced c-Fos (green) expression overlapping with mCherry (red). Blue, DAPI. Scale bars, 50 μm. e, 589 nm light-induced c-Fos (green) expression overlapping with eNpHR (red). Blue, DAPI. Scale bars, 50 μm. f, 589 nm light induced less c-fos expression in eNpHR, which indicated the effectiveness of eNpHR (n = 3). **P < 0.01. Independent-samples t-test. Error bars indicate SEM. For c,f, n = 3 per group, data are mean ± s.e.m. Statistical analysis was performed using independent-samples T test (c,f). **P < 0.01, ****P < 0.0001. Details of the statistical analyses are as follows: c, t (4) = -23.139, P = 0.000021, f, t (4) = 10.136, P = 0.001.

Supplementary Data Fig. 1-source data 1

Statistical results of light-induced c-Fos expression overlapping with mCherry, CHR2 or eNpHR.

Recordings of OTmut sensor in the mPFC on OT release

a. Recording instrument settings. b. Illustrations of viral expression and optical fibre location. c. Representative histological image of OTmut sensor (green) and optical fibre locations. Blue, DAPI. Scale bars, 500 μm. d-h,j-n, PETHs of z-score of OTmut sensor for the following pup-directed behaviors: approach (d,j), investigate (e,k), retrieve (f,l), grooming (g,m) and crouch (h,n). i,o, AUC per second of z-scores for pup-care female (n = 6, i) and male (n = 6, o) voles across various pup-directed behaviors. p,q,s,t, PETHs of z-score of OTmut sensor for approach and infanticide in infanticide voles. r,u, AUC per second of z-score of OTmut sensor for pre pup, approach and infanticide in infanticide female (n = 6, r) and male (n = 6, u) voles. Error bars indicate SEM. For i,o,r,u, n = 6 per group, data are mean ± s.e.m. Statistical analysis was performed using one-way rmANOVA (i,o,r and u). Details of the statistical analyses are as follows: i, F (5, 25) = 0.680, P = 0.643; o, F (2, 25) = 0.074, P = 0.996; r, F (2, 10) = 0.002, P = 0.998; u, F (2, 10) = 0.665, P = 0.536.

Supplementary Data Fig. 2-source data 1

Area under the curve per second for pre-pup, approach, and infanticide behaviors in infanticide voles and area under the curve per second for pre-pup, approach, investigation, retrieval, grooming, and crouching behaviors in pup-care voles.

Recordings of OT1.0 sensor for investigating object

a,b,d,e, Post-event histograms (PETHs) of z-score of OT1.0 sensor for approaching and investigating object in female (a,b) and male (d,e) voles. c,f, AUC per second of z-scores for pre object, approaching and investigating behaviors in female (n = 4, c) and male (n = 4, f) voles. Error bars indicate SEM. For c,f, n = 4 per group, data are mean ± s.e.m. Statistical analysis was performed using one-way rmANOVA. Details of the statistical analyses are as follows: c, F (1.016, 3.047) = 0.346, P = 0.601; f, F (2,6) = 0.183, P = 0.838.

Supplementary Data Fig. 3-source data 1

Area under the curve per second for pre-objecte, approach, and investigating in female and male voles.