Construction of Koi lines.

(A) Example of the remarkable species difference of brain OXTR distribution. OXTR receptor autoradiography in dorsal caudate putamen (CP) and nucleus accumbens shell (NAccSh) of rat, mouse, and prairie voleis (Adapted Froemke and Young(Froemke & Young, 2021)). (B) The pvOxtr BAC clone contains the entire sequence of pvOxtr gene (colored green), the complete intergenic sequence both upstream and downstream of pvOxtr gene, the 12th exon of Rad18 (colored blue) and the 2nd exon of Cav3 (colored purple). The yellow arrows indicate the direction of transcription. (C) Schematic diagram of the strategy to generate Koi mice. The P2A-NLS-Cre-Frt-Amp-Frt cassette was inserted in-frame right before the stop codon of Oxtr. And the ampicillin selection marker was deleted by 706-Flpe. PCR primers for verifying the targeted alleles are shown as red arrows. (D) Two correctly modified vector clones were verified by using PCR with p2 and p8. (E) Koi lines were verified by 5 pairs of primers. From Lane1 to Lane8 are 100bp marker, p3+p7, p2+p6, p2+p7, p4+p8, p2+p8, blank and 1kb marker.

Transgenic CRE mediated Lac-Z reporter gene expression in the brain and mammary gland of Koi lines.

X-gal staining of the brains and mammary glands are from the double positive (Cre+, LacZ+) offspring from Koi-1 (A, F, K, P), Koi-2 (B, G, L, Q), Koi-3 (G, H, M, R), Koi-4 (D, I, N, S) and mOxtr-Ires-Cre knock-in line (E, J, O, T). The mOxtr-Ires-Cre KI line used contains Cre (and an HA tag) inserted in the third exon, right after the start codon. Note that regardless of the insertion site of Cre in the mouse genome DNA or pvOxtr-BAC transgene, Oxtr expression visualized by Lac-Z staining in the mammary gland was comparable between the mOxtr-Ires-Cre KI and Koi lines. This contrasts with the diverse expression patterns of Oxtr expression in the brain. U, V, M, X, Y are the magnified images of the areas within the yellow-border squares from D, N, E, J, O, respectively, to show the distribution of low-density Lac-Z positive cells in these areas. Scale bar=1mm (A-O), 500um (P-T) and 200um (U-Y). (Z) The estimated copy number of transgene of all Koi lines. See Supplementary Figure 2 for images of Koi-5-8 lines. PFC=prefrontal cortex, LS=lateral septum, BLA=basolateral amygdala, VMH=ventromedial nucleus of the hypothalamus, Tha=thalamus.

The expression of OXTR in adult brain of Koi lines.

Autoradiographs illustrate the distribution of OXTR binding in adult Koi-1: mOxtr-/- (A, F, K), Koi-2: mOxtr-/- (B, G, L), Koi-3: mOxtr-/- (C, H, M), Koi-4: mOxtr-/- (D, I, N) and wild type (WT) (E, J, O). All Koi mice used for this analysis were pvOxtr-P2A-Cre: mOxtr-/- mice in which the pvOXTR was expressed while endogenous mOXTR was absent. PFC=prefrontal cortex, LS=lateral septum, BLA=basolateral amgydala, VMH=ventromedial nucleus of the hypothalamus, Tha=thalamus. Scale bar=1mm.

mRNA expression of Oxtr in the adult brain of Koi lines.

Oxtr mRNA (red) was labeled with in situ hybridization against DAPI for nuclei labelling (purple). (A) and (C) show clear expression of Oxtr in the PFA and Ctx in Koi-3: mOxtr-/-. (B) and(D) show clear expression of Oxtr in the NAc and Tha in Koi-4: mOxtr-/-. All Koi mice used for this analysis were pvOxtr-P2A-Cre: mOxtr-/- mice, in which the pvOXTR was expressed while endogenous mOXTR was absent. PFC=prefrontal cortex, NAc= Nucleus accumbens, Ctx=Cortex, Tha=thalamus. Scale bar=400um for A and B, 500um for C and D.

Quantitative PCR analysis of Oxtr mRNA expression in different brain regions of Koi lines.

(A) A primer set recognizing both mOxtr and pvOxtr was used to analyze Oxtr mRNA in olfactory bulb, prefrontal cortex, ventral striatum and thalamus from WT, Koi-1: mOxtr-/-, Koi-2: mOxtr-/-, Koi-3: mOxtr-/-, Koi-4: mOxtr-/- and mOxtr-/- mutant mice, n=4 for each genotype. Note that the level of Oxtr mRNA in mOxtr-/- mutant mice was nondetectable (ND). The autoradiograph of OXTR binding in olfactory bulb from each line were shown alongside the legend. There was a significant interaction between position and genotype F (12,64) =26.33, p<0.0001; Post hoc Bonferroni test revealed that Oxtr expression level in olfactory bulb of Koi-1: mOxtr-/- was significantly higher than that of other genotype groups (p<0.0001 for all), and that of Koi-4: mOxtr-/- was significantly higher than that of Koi-3: mOxtr-/- p<0.05. And the striatum of Koi-4: mOxtr-/- expressed significantly higher Oxtr when compared with WT and Koi-3: mOxtr-/- (p<0.05 for both). The Oxtr expression level in thalamus of Koi-4: mOxtr-/- was significantly higher than other genotype groups (p<0.0001 when compared with WT and Koi-3: mOxtr-/-, p<0.001 when compared with Koi-2: mOxtr-/-, p<0.05 when compared with Koi-1: mOxtr-/-) (B) A primer set specifically recognizing pvOxtr. was used to analyze Oxtr mRNA in WT, Koi-1: mOxtr-/-, Koi-2: mOxtr-/-, Koi-3: mOxtr-/- and Koi-4: mOxtr-/- mic, n=4 for each genotype. Note that the level of Oxtr mRNA in WT mice was barely detectable. There was a significant interaction between position and genotype F (9,48) =89.25, p<0.0001; Post Hoc Bonferroni test revealed that Oxtr expression level in the olfactory bulb of Koi-1: mOxtr-/- was significantly higher than that of other genotype groups (p<0.0001 for all), and that of Koi-4: mOxtr-/- was significantly higher than that of Koi-2: mOxtr-/- and Koi-3: mOxtr-/- (p<0.0001 for both). The Oxtr expression level in prefrontal cortex of Koi-3: mOxtr-/- was significantly higher than that of other groups (p<0.01 when comparing with Koi-1: mOxtr-/- and Koi-4: mOxtr-/-; p<0.05 when comparing with Koi-2: mOxtr-/-). And the striatum of Koi-4: mOxtr-/- expressed significantly higher Oxtr when comparing with Koi-2: mOxtr-/- and Koi-3: mOxtr-/- (p<0.01 for both). The Oxtr expression level in thalamus of Koi-4: mOxtr-/- was significantly higher than that of other genotype groups (p<0.0001 for all). All Koi mice used for this analysis were pvOxtr-P2A-Cre: mOxtr-/- mice, in which the pvOXTR was expressed while endogenous mOXTR was absent. The significance of interaction between genotype and brain region: # # # # p<0.0001. The significance of single main effect of genotype:*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Social behaviors of Koi lines.

(A) Illustration of the partner preference test (PPT) paradigm. The experimental mouse was placed in a chamber with an automatic monitoring system, where it could freely explore two pen-boxes localized in two diagonal corners. Habituation phase: two pen-boxes are empty. Recognition phase: a stimulating partner mouse is placed into one pen-box and a stimulating stranger mouse is placed in the other one. Preference phase: The positions of two stimulating mice are switched. (B) Representative traces of one experimental mouse during the three phases. (C-E) The stay time in the surrounding area of each pen box during three phases;(C) Habituation phase;(D) Recognition phase; (E) Preference phase. Samples sizes for the PPT are wild type(WT) (n=12), Koi-1: mOxtr-/- (n=14), Koi-2: mOxtr-/- (n=13) Koi-3: mOxtr-/- (n=12) Koi-4: mOxtr-/- (n=14). All the experimental mice are female. (F) The illustration of the procedure of the maternal behavior test. (G) Pup retrieval latency and (H) crouching time were recorded and analyzed. Sample sizes for the maternal behavior tests are WT (n=15), Koi-3: mOxtr-/- (n=18), Koi-4: mOxtr-/- (n=18). Significance of interaction between independent factors: # p<0.05; # # p<0.01. Significance of main effect or single main effect of genotype:*p<0.05; **p<0.01; ***p<0.001; **** p<0.0001. All Koi mice used for this analysis were pvOxtr-P2A-Cre: mOxtr-/- mice in which the pvOXTR was expressed while endogenous mOXTR was absent.

© 2026, BioRender Inc. All Rights Reserved. Illustrations created with BioRender.com. BioRender citation URL not attainable. Further reproduction of this figure would need permission from the copyright holder.

Heatmap of the chromatin contacts surrounding Oxtr and Oxt across mouse tissues/cells from megabase-size visualization.

(A) Heatmap of the chromatin contacts surrounding Oxtr across mouse tissues/cells: Oxtr is located at a TAD boundary region. The increase of interactions characterizes the TADs structure and the interaction of two neighboring TADs can be more clearly observed in brain-related samples. Blue dashed lines were prepared to help observing the TAD structures, and black circles were prepared to help observing the long-distance interaction obtained from Hi-C interaction matrices surrounding Oxtr. (B) Heatmap of the chromatin contacts surrounding Oxt across mouse tissues/cells. No TAD structure was observed surrounding Oxt locus.

Expression pattern of the Pkcd-Cre BAC transgene.

All five independent Pkcd-Cre BAC transgenic mouse lines (A–E) exhibited expression patterns that closely resembled the endogenous Pkcd expression, as shown by in situ hybridization (F). The Pkcd-Cre BAC construct was generated using the same strategy as that employed for the Koi lines, incorporating the entire intergenic regions both upstream and downstream of the Pkcd gene.

Transgenic CRE mediated Lac-Z reporter gene expression in the brain of Koi lines.

X-gal staining of the brains from the double positive (Cre+, LacZ+) offspring of Koi-5 (A-C), Koi-6 (D-F), Koi-7 (G-I), Koi-8 (J-L). Scale bar=1mm

Negative control of whole-mount mammary gland staining.

X-gal staining of a mammary gland from (Cre-, LacZ+) offspring from heterozygous Koi-4 was shown. The lymph node showed some blue signal, indicating that the lymph node could be stained nonspecifically. All the epithelium ducts did not show any blue signal, verifying the specificity of duct labeling of X-gal staining. Scale=500um

The datasets IDs and sample information for mouse 3D chromatin verification

The database source of each mouse sample, analysis and visualization tool information, and the region analyzed was listed.