Fig. 2: Manipulation of the PMv^DAT neuron activity allows bidirectional control of maternal aggression.

a Bilateral Cre-dependent ChR2 transduction and fiber implants in the PMv of lactating dams (left) and confocal image of a sample section counterstained with the pan-neuronal marker NeuN, used to validate ChR2 expression and stereotactic implantation of optic fiber (ft) coordinates (right). b Schematic of the experimental design used to perform ChR2 PMv^DAT neuron photoactivation in the resident lactating dam during resident-intruder (RI) testing. Adult male and female intruders were used to minimise the levels of maternal aggression expressed in baseline conditions^54,62. c Sample behaviour raster plots at baseline (top) and during ChR2 stimulation (bottom) in a lactating dam during a RI test. d Attack duration with and without photostimulation in lactating dams injected with eYFP or ChR2, during the RI test against adult male or adult female intruders (n = 5 mice against male intruders per group, n = 9 mice against female intruders per group, ordinary one-way ANOVA with Tukey’s test for correcting for multiple comparisons, P = 0.1401, P = 0.1534, P = 0.9601 and P < 0.0001 respectively). e Attack latency with and without photostimulation in lactating dams injected with eYFP or ChR2, during the RI test against adult male or adult female intruders (n = 5 mice against male intruders per group, n = 9 mice against female intruders per group, ordinary one-way ANOVA with Tukey’s test for correcting for multiple comparisons, P = 0.7149, P = 0.9468, P = 0.1043 and P = 0.034 respectively). f Bilateral Cre-dependent eNpHR3 transduction and fiber implants in the PMv of lactating dams (left) and confocal image of a sample section counterstained with the pan-neuronal marker NeuN, used to validate eNpHR3 expression and stereotactic implantation of optic fiber (ft) coordinates (right). g Schematic of the experimental design used to perform eNpHR3 PMv^DAT neuron photoinhibition studies in the resident lactating dam during RI testing. h Sample behaviour raster plots during baseline and eNpHR3 mediated photoinhibition in a lactating dam during a RI test. i Quantification of aggression parameters (n = 8 mice per group, two-tailed paired t-test, P < 0.0001, P = 0.0011, and P = 0.0071 respectively). j Quantification of the latency to an attack bout following a photoinhibition episode (n = 8 mice per group, two-tailed paired t-test, P = 0.0131). k Bilateral Cre-dependent taCasp3 transduction in the PMv of lactating dams (left) and sample confocal image sections following injection of eYFP (middle) vs. taCasp3 (right), used to validate the extent of the PMv^DAT cell lesion. l Schematic of the experimental design used to perform RI tests in lactating dams with complete or partial lesion of the PMv^DAT cells. m Sample behaviour raster plots in control vs. PMv^DAT neuron lesioned lactating dams during a RI test. n Quantification of aggression parameters (n = 34 trials per group, 12 mice per group, two-tailed unpaired t-test, P < 0.0001, P < 0.0001, and P < 0.0001 respectively). All but two mice were tested three times in the RI test. Two mice, one from each group, were tested twice in the RI test. o Quantification of PMv^DAT neurons in eYFP and taCasp3 injected mice (n = 24 sections per group, duplicates from 12 mice per group, two-tailed unpaired t-test, P < 0.0001). See also Figs. S1 and S2. All box plots show the median (center line), 25th and 75th percentiles (box bounds), and minima/maxima (whiskers). Exact P values are provided where P > 0.0001; for P ≤ 0.0001, significance is indicated using asterisks (**** for P ≤ 0.0001, *** for P ≤ 0.001, ** for P ≤ 0.01, * for P ≤ 0.05, ns not significant). Source data are provided as Source Data file.