Content uploaded by Huadong Xu
Author content
All content in this area was uploaded by Huadong Xu on Jan 14, 2025
Content may be subject to copyright.
RESEARCH ARTICLE SUMMARY
◥
NEUROSCIENCE
Sexually dimorphic dopaminergic circuits
determine sex preference
Anqi Wei†, Anran Zhao†, Chaowen Zheng†, Nan Dong†, Xu Cheng, Xueting Duan, Shuaijie Zhong,
Xiaoying Liu, Jie Jian, Yuhao Qin, Yuxin Yang, Yuhao Gu, Bianbian Wang, Niki Gooya, Jingxiao Huo,
Jingyu Yao, Weiwei Li, Kai Huang, Haiyao Liu, Fenghan Mao, Ruolin Wang, Mingjie Shao, Botao Wang,
Yichi Zhang, Yang Chen, Qian Song, Rong Huang, Qiumin Qu, Chunxiang Zhang*, Xinjiang Kang*,
Huadong Xu*, Changhe Wang*
INTRODUCTION: Innate social behaviors are es-
sential for survival and reproduction. Animals
should make correct social decisions (e.g., when,
where, how, and with whom) to reach a maxi-
mal benefit from social interactions, especially
when confronted with conflicts between in-
nate requirements and external threats. How-
ever, how social decisions are convergently
encoded by the internal-drive condition and
external-environment context remains unclear.
Furthermore, the sex of a social partner is a fun-
damental component affecting social decision-
making. Interactions with individuals of the
opposite sex are vital for the innate requirements
of mating and reproduction, whereas same-sex
social interaction provides social support and
facilitates collaboration for shared goals. How-
ever, the neural mechanisms underlying socio-
sexual preference remain virtually unknown.
RATIONALE: We investigated the sociosexual
preference of male and female mice under both
normal conditions and when exposed to external
threats. By using dual-color fiber photometry Ca
2+
recordings and projection-specific chemogenetic
and optogenetic manipulations of dopamine (DA)
neurons in the ventral tegmental ar ea (VTA) , we
defined the sexually dimorphic DA circuits respon-
siblefortheswitchingofsociosexualpreferences.
RESULTS: Bothmaleandfemalemiceexhib-
ited a preference for social interaction with
females but shifted to male preference when
facing survival threats mediated via different
sensory pathways, such as olfaction [through
testing with the stressor trimethylthiazoline
(TMT)], vision (contextual fear conditioning),
and auditory (cued fear conditioning), indicat-
ing the integrated encoding of social decisions
in response to innate requirements and exter-
nal environmental factors.
Using c-Fos staining and photometric Ca
2+
recordings, we observed a strong correlation
between the excitation of VTA
DA
neurons and
the switching of sexual preference when con-
fronted with survival stress. Chemogenetic ac-
tivation of VTA dopaminergic (VTA
DA
)neurons
facilitated male preference, whereas inhibiting
these neurons blocked TMT’s effects on the
switching of sexual preference in both sexes,
validating the critical role of VTA
DA
neurons
in orchestrating the shift in social preference.
Dual-color fiber photometry Ca
2+
recordings
and projection-specific chemogenetic manipu-
lations demonstrated that sexually dimorphic
alterations in VTA
DA
circuits dictate the switch-
ing of sociosexual preference in response to ex-
ternal survival threats. The competition between
two VTA
DA
pathways, representing the bal-
ance between innate requirements and exter-
nal threats, was used by males to encode their
sexual preferences. VTA
DA
projections to the
nucleus accumbens (NAc) were predominant
under normal conditions to promote female
preference, whereas projection to the medial
preoptic area (mPOA) mediated male prefer-
ence in response to survival threats.
By contrast, firing-pattern alteration of the
VTA
DA
-NAc projection was utilized by females
to determine their sexual preference. Female
interactions were associated with stronger and
faster Ca
2+
transients, indicating the occurrence
of phasic-like action potential (AP) firings of
NAc-projecting VTA
DA
neurons in female mice.
On the other hand, male interactions facili-
tated by environmental threats were correlated
with Ca
2+
signals exhibiting slower kinetics,
reflecting the sustained tonic-like AP firings
of these neurons. Notably, the phasic firing–like
optogenetic excitation of VTA
DA
-NAc terminals
resulted in larger transients of DA release, pro-
moting female preference through the en-
hanced DA-D1R transmission (D1R, type 1 DA
receptor). Conversely, the tonic firing–like ex-
citation of these terminals induced a lower
sustained DA release and thus led to male pre-
ference through the predominant DA-D2R
transmission (D2R, type 2 DA receptor).
CONCLUSION: Both male and female mice ex-
hibitfemalepreferencebutshifttomaleprefer-
ence when confronted with survival threats. The
sexually dimorphic alterations in VTA
DA
circuits,
including neuronal activity, DA transmission,
and circuit integration, play a key role in en-
coding the switch of sociosexual preference in
both sexes. Our study thus introduces a neural
mechanism for understanding how social de-
cisions can be convergently determined by the
balance between innate requirements and
external survival threats. ▪
RESEARCH
The list of author affiliations is available in the full article online.
*Corresponding author. Email: changhewang@xjtu.edu.cn (C.W.);
hdxu@pku.edu.cn (H.X.); kxj335@163.com (X.K.);
zhangchx999@163.com (C.Z.)
†These authors contributed equally to this work.
Cite this article as A. Wei et al., Science 387, eadq7001
(2025). DOI: 10.1126/science.adq7001
READ THE FULL ARTICLE AT
https://doi.org/10.1126/science.adq7001
NAc
NAc
Male mice
Female mice
Male preference
Female preference
Male preference
Female preference
mPOA VTA
Sexually dimorphic dopamine circuits determine sociosexual preference. Male individuals use the competition
between VTA
DA
-NAc–projecting pathways (promoting female preference) and VTA
DA
-mPOA–projecting pathways
(promoting male preference upon survival stress) to determine their social decisions. Conversely, firing-pattern
alteration of the VTA
DA
-NAc projection is utilized by females to determine their sociosexual preferences,
mediating female preference through the phasic firing–dominated DA-D1R transmission and male preference
through the tonic firing–facilitated DA-D2R transmission when confronted with survival stress.
Wei et al., Science 387, 155 (2025) 10 January 2025 1of1
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
RESEARCH ARTICLE
◥
NEUROSCIENCE
Sexually dimorphic dopaminergic circuits
determine sex preference
Anqi Wei
1
†, Anran Zhao
1
†, Chaowen Zheng
1
†, Nan Dong
1
†, Xu Cheng
2,3
, Xueting Duan
1
,
Shuaijie Zhong
1
, Xiaoying Liu
1
, Jie Jian
2
, Yuhao Qin
1
, Yuxin Yang
1,4
, Yuhao Gu
1
, Bianbian Wang
1
,
Niki Gooya
5
, Jingxiao Huo
1
, Jingyu Yao
1
, Weiwei Li
1,6
, Kai Huang
1
, Haiyao Liu
1
, Fenghan Mao
1
,
Ruolin Wang
1
, Mingjie Shao
1
, Botao Wang
1
, Yichi Zhang
1
, Yang Chen
1
, Qian Song
1
, Rong Huang
1
,
Qiumin Qu
1
, Chunxiang Zhang
2
*, Xinjiang Kang
2,3,4
*, Huadong Xu
1
*, Changhe Wang
1,2,3,7
*
Sociosexual preference is critical for reproduction and survival. However, neural mechanisms encoding
social decisions on sex preference remain unclear. In this study, we show that both male and female
mice exhibit female preference but shift to male preference when facing survival threats; their preference is
mediated by the dimorphic changes in the excitability of ventral tegmental area dopaminergic (VTA
DA
)
neurons. In males, VTA
DA
projections to the nucleus accumbens (NAc) mediate female preference, and those
to the medial preoptic area mediate male preference. In females, firing-pattern (phasic-like versus tonic-like)
alteration of the VTA
DA
-NAc projection determines sociosexual preferences. These findings define VTA
DA
neurons as a key node for social decision-making and reveal the sexually dimorphic DA circuit mechanisms
underlying sociosexual preference.
Innate social behaviors, such as mating, con-
solation, and collaboration, are fundamen-
tal for reproduction, health, and survival,
and provide adaptive benefits by offering
greater capabilities for resource acquisi-
tion and defense against threats (1–3). Social
decisions are dynamically adjusted to reach
a maximal benefit from social engagement,
particularly when confronted with conflicts be-
tween innate requirement and external threat
(4–11). Multiple brain regions, including the
ventral tegmental area (VTA) (12–15), medial
prefrontal cortex (mPFC) (16–18), amygdala
(3,16,17,19,20), nucleus accumbens (NAc)
(13,21,22), dorsal raphe nucleus (23), insular
cortex (24), and medial preoptic area (mPOA)
(19,25,26),have been reported tobe functionally
involved in the modulation of social decision-
making; however, how social decision-making
is convergently encoded by the internal-drive
condition and external-environment context
remains largely unknown (27).
The selection of social encounters is one of
the most fundamental components of social
decision-making and involves consideration of
sex, age, hierarchy, health, and other factors
(7,28). Social interaction with individuals of
the opposite sex, especially in the context of
mating and procreation, is crucial for the in-
nate requirements of reproduction and genetic
continuity across generations (29,30). Concur-
rently, same-sex social networks often serve as
a source of protection and social support that
are vital for health and survival, and these
social engagements can facilitate coordination
and collaboration for shared goals (31,32).
However, the neural mechanisms through
which sociosexual preference (social prefer-
ence to male or female conspecifics) can be
determined—and how this preference is mod-
ulated by the external environment—remain
virtually unknown.
Switching of sociosexual preference under
survival stress
To investigate whether and how the sociosex-
ual preference can be altered in the context of
survival threats, we conducted a three-chamber
social interaction assay to assess sex preference
in the absence or presence of trimethylthiazoline
(TMT), a potent stressful odor cue linked with
survivalthreatsforrodents(33,34). Consistent
with a previous report (33), male mice displayed
a pronounced preference for female mice (Fig.
1A and fig. S1A), which conforms to the innate
requirements in mating and reproduction. This
preference was dramatically reversed upon ex-
posure to the TMT stressor, shown as a decrease
in the time interacting with female mice and an
increased visiting time with male mice (Fig. 1A
and fig. S1A). Consequently, the proportion of
time allocated to male interaction increased
from ~34% under control conditions to ~56%
when they were exposed to TMT (Fig. 1A). In
parallel with this behavioral shift, the female-
bedding preference for male mice was also
reversed in the presence of TMT (Fig. 1B and
fig. S1B). To confirm that the enhanced male
interaction was due to preference alteration
between different sexes rather than an aver-
sion to females under TMT conditions, we con-
ducted similar behavior tests to compare the
preference for a conspecific partner versus a
toymouse.Wefoundthatmalemiceshoweda
clear preference for conspecific partners, re-
gardless of the sex of the social partner, and
this preference was further increased by the
exposure to the TMT stressor (fig. S1, C and D),
supporting the idea that male mice switch to
male preference when confronted with TMT
exposure. This phenomenon suggests that the
environmental context exerts a strong impact
on social preferences and underscores the dy-
namic nature of sociosexual preference in re-
sponse to innate requirements and external
survival threats.
To ascertain that the observed changes in
social preference were attributable to the so-
cial context of survival stress rather than the
specific odorant stimulation, we carried out
a fear conditioning (FC) test associated with
visual and auditory cues and assessed the sex
preference of male mice (Fig. 1C). Consistently,
male mice also exhibited female social prefer-
ence under the control condition with no elec-
tric shock during the preconditioning procedure
(Fig. 1D and fig. S1E). By contrast, when sub-
jected to the visual contextual FC conditions,
the preconditioned male mice allocated more
time to interacting with male mice (Fig. 1D).
Similarly, in male animals, the preference for
female mice was also switched to male pref-
erence in auditory-cued FC tests (Fig. 1E and
fig. S1F). These results suggest that the socio-
sexual preference of male mice can be modu-
lated by survival stress through various sensory
pathways such as olfaction (TMT), vision (con-
textual FC), or auditory (cued FC).
We next investigated the social preference of
female mice when exposed to survival stressors.
Female mice also showed preference for female
encounters under control conditions (Fig. 1F
and fig. S1G). However, TMT exposure increased
the interaction time with male mice and re-
duced the time spent with female counterparts
(Fig. 1F and fig. S1G). Consequently, the propor-
tion of social time that female mice spent with
males increased substantially in the presence
of TMT (Fig. 1F). Similarly, the female-bedding
preference of female mice was also reversed to
male-bedding preference upon TMT exposure
(Fig. 1G and fig. S1H). Such changes were not
RESEARCH
1
Department of Neurology, the First Affiliated Hospital,
Neuroscience Research Center, Key Laboratory of Biomedical
Information Engineering of Ministry of Education, School of
Life Science and Technology, Xi’an Jiaotong University, Xi’an,
China.
2
Key Laboratory of Medical Electrophysiology, Ministry
of Education of China, Collaborative Innovation Center for
Prevention and Treatment of Cardiovascular Disease, and
the Institute of Cardiovascular Research, Southwest Medical
University, Luzhou, Sichuan, China.
3
Department of
Neurosurgery, the Affiliated Hospital of Southwest Medical
University, Luzhou, Sichuan, China.
4
College of Agriculture
and Biology, Liaocheng University, Liaocheng, China.
5
Solomon H. Snyder Department of Neuroscience, Johns
Hopkins University School of Medicine, Baltimore, MD, USA.
6
Department of Acupuncture, Massage and Rehabilitation,
Shaanxi Provincial Hospital of Chinese Medicine, Xi’an,
China.
7
Department of Neurology, the Second Affiliated
Hospital of Xi’an Jiaotong University, Xi’an, China.
*Corresponding author. Email: changhewang@xjtu.edu.cn (C.W.);
hdxu@pku.edu.cn (H.X.); kxj335@163.com (X.K.);
zhangchx999@163.com (C.Z.)
†These authors contributed equally to this work.
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 1of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
due to an aversion to females, because TMT
exposure also increased the social preference
of female mice for female partners compared
with toy mice (fig. S1, I and J). Furthermore,
fear-conditioned female mice spent more time
with male mice and less time with female
ones, as compared with the control group that
did not receive shock conditioning, in both the
contextualandcuedFCtests(Fig.1,HtoJ,
and fig. S1, K and L). Collectively, these results
provide compelling evidence that both males
and females shift their social preference toward
male interactions when confronted with sur-
vival stressors.
VTA
DA
neuron is essential for the switching
of sociosexual preference
Where and how is the sociosexual preference
encoded in the brain? The mesolimbic dopa-
mine (DA) pathway originating from the VTA
hasbeenimplicatedinbothsocialinteraction
(34–36) and sexual motivation (21,37–39). This
pathway is activated when individuals experi-
ence incentives (both positive and negative),
rewards, or the anticipation of rewards from
social engagements (14,40,41). Moreover, DA
release exhibits sexual dimorphism during
mating and aggressive behaviors (21), suggest-
ing a potential role in modulating sociosexual
preferences. Therefore, we next investigated
whether and how mesolimbic DA pathways
are functionally involved in the threat-induced
shifts in male preference by using c-Fos ex-
pression as a biomarker of neural activity. As
anticipated, both male and female mice spent
more time approaching and interacting with
the caged stranger male mice when exposed to
TMT (Fig. 2, A and E, and fig. S2, A and D).
Concurrently, the number of c-Fos expressing
neurons increased greatly in the VTA in both
sexes (Fig. 2, B, D, F, and H), indicating the
increased neural activityinthisregionupon
TMT exposure. Immunostaining analysis showed
that most of c-Fos positive neurons (~84.6% in
males and ~82.6% in females) colocalized with
tyrosine hydroxylase (TH), a specific molecular
marker for DA neurons (fig. S2, C and F). These
findings indicate that dopaminergic neurons
in the VTA (VTA
DA
neurons)mayplayacru-
cial role in the modulation of sex preference
during social interaction in both male and
female animals. In male mice exposed to TMT
during social interaction, there was also a pro-
nounced increase of c-Fos expression in the
mPOA (Fig. 2, C and D), a downstream brain
region of VTA
DA
neurons implicated in mating,
parental support, and defensive and aggres-
sive behaviors (42–48). This increase was not
observed in other DA downstream regions,
such as the NAc, mPFC, or basolateral amyg-
dala (BLA) (Fig. 2, C and D, and fig. S2B). By
contrast, in female animals that were exposed
to TMT during social activities, the increase of
c-Fos expression was specifically observed in
the NAc, a central hub in the DA reward sys-
tem (49,50), but not in the mPOA, mPFC, or
BLA (Fig. 2, G and H, and fig. S2E). Specifi-
cally, the increased c-Fos staining was mainly
located in the core region of the NAc, although
both the core and the shell regions are rele-
vant to social reward and social interaction
(21,51). These sexually dimorphic changes in
Fig. 1. Sexually dimorphic
changes of sociosexual prefer-
ence under survival stress.
(A)(Left)Schematicdiagram
showing the social preference test,
and (right) statistics of social
interaction time for female or
male encounters and percentage
of male-interaction time of male
mice in the absence [control
(Ctrl)] or presence (TMT) of the
stressor TMT. (B) Similar to
(A), except that bedding preference
test was performed. (C) Schematic
of acquisition and extinction of
conditioned fear (FC) through
contextual or tone-based conditioning.
(Dand E) Statistics of social time
(left) and male social percentage
(right) in contextual (D) and
cued (E) FC tests. (Fto J)As
shownin(A)to(E),exceptthat
female mice were used. Data
are presented as mean ± SEM.
Two-way analysis of variance
(ANOVA) for comparison of time
spent on male and female mice
between Ctrl and TMT groups;
unpaired Student’sttest for male
percentage. *P<0.05,**P<0.01,
***P<0.001.
Ctrl FC
0
20
40
60
80
100
Male percentage(
%)
Ctrl FC
0
20
40
60
80
100
Male percentage (%)
Crtl TMT
0
20
40
60
80
100
Male percentage (%)
Crtl TMT
0
20
40
60
80
Malepe
rcentage (%)
Ctrl FC
0
100
200
300
400
Time (s)
ns
Ctrl FC
0
20
40
60
80
Male percentage (%)
Ctrl TMT
0
20
40
60
80
100
Male percentage(
%)
Crtl TMT
0
20
40
60
80
100
Malepercentage(%)
Ton e
E-shock
Training
ASocial preference BBedding preference
FSocial preference GBedding preference
Male
bedding
Female
bedding
Male
bedding
Female
bedding
Ton e
E-
shock
Training
DContextual FC ECued FC
Contextual FC
Cued FC
IContextual FC JCued FC
Male
mouse
Female
mouse
CFC
H FC Contextual FC
Cued FC
Male mice
Female mice
Male
mouse
Female
mouse
18 19 13 16
12 14
Ctrl FC
0
100
200
300
400
Time (s)
17 20
98
911
88
99
TMT TMT
TMT TMT
o
e
e
n
o
Ctrl TMT
0
100
200
300
400
Time (s)
Ctrl TMT
0
100
200
300
400
Time (s)
Ctrl TMT
0
100
200
300
400
500
Time(
s)
female
male
female
male
Ctrl TMT
0
100
200
300
400
500
Time (s)
female
male
female
male
Ctrl FC
0
100
200
300
400
500
Time (s)
female
male
female
male
female
male
Ctrl FC
0
100
200
300
400
500
Time (s)
ns
Ctrl FC
0
20
40
60
80
100
Male percentage (%)
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 2of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
c-Fos expression patterns in response to TMT
exposure suggest that the mesolimbic DA path-
ways may be differentially involved in the mod-
ulation of sociosexual preference under survival
threats in both sexes.
To substantiate the functional changes of
VTA
DA
neurons, we injected TH-Cre and Cre-
dependent GCaMP (a genetically encoded Ca
2+
indicator)–expressing viruses into the VTA of
male mice for the specific expression of GCaMP
in VTA
DA
neurons. Fiber photometry recording
of Ca
2+
signals was conducted to monitor the
activity of VTA
DA
neurons during an open-field
social interaction test involving mouse partners
of both sexes (Fig. 3, A and F, and fig. S3, A and
B). Interactions with either sex of mice induced
Ca
2+
transients in VTA
DA
neurons of male mice.
In general, interaction with female animals pro-
voked a more pronounced Ca
2+
signal com-
pared with that elicited by male interactions
(Fig. 3B and fig. S3, C and D). However, re-
peated socialization with the same mouse led
to a decrease in Ca
2+
signals (fig. S3, C and
D). Along with the reversal of sex preference,
the male interaction–coupled Ca
2+
signals in
male mice increased greatly in the presence
of TMT in the testing apparatus, exhibiting
larger amplitude and area under the curve
(AUC) values than those under control con-
ditions (Fig. 3, C to E).
We next performed similar experiments with
female mice (Fig. 3F and fig. S3, E and F).
Female interactions elicited stronger Ca
2+
sig-
nals in VTA
DA
neurons in female mice than
those coupled with male interactions, although
repeated socialization with the same sex led to a
substantial decrease in Ca
2+
signals (fig. S3, G
and H). However, the larger Ca
2+
signals during
female interactions were substantially reversed
by TMT exposure (Fig. 3, G and H). Both the
amplitude and AUC values of Ca
2+
signals trig-
geredbymaleinteractionswerehigherthan
those coupled to female interactions in the
presence of TMT stressor (Fig. 3, I and J). These
results suggest the critical involvement of VTA
DA
neurons in shaping sociosexual preference of
both sexes and indicate that the hyperactivity
of VTA
DA
neurons may be functionally involved
in driving the preference shifts under condi-
tions of survival stress.
To confirm the critical roles of VTA
DA
neu-
rons in male preference under TMT threat,
TH-Cre and Cre-dependent hM3Dq-expressing
viruses (with the Cre-dependent mCherry-
expressing virus as a control) were stereo-
taxically coinjected into the VTA of male mice
(Fig. 4A and fig. S4, A and B) for the chem-
ogenetic manipulation of VTA
DA
neurons. Com-
pared with the control group, chemogenetic
activation (hM3Dq) of VTA
DA
neurons through
the intraperitoneal (i.p.) injection (0.5 mg/kg)
of clozapine N-oxide (CNO) increased the so-
cial time of male mice with male counterparts
and reduced that with females, thereby re-
sulting in male social preference (Fig. 4B and
fig. S4C). Consistent with this, the chemo-
genetic activation group spent more time in
male bedding and less time in female bed-
ding (Fig. 4C and fig. S4D). In female mice,
chemogenetic activation of VTA
DA
neurons
also led to an increase in social interaction with
male mice and a pronounced reduction in in-
teraction with female mice (Fig. 4, D and E,
and fig. S4, E to G), as well as similar bedding
preferences (Fig. 4F and fig. S4H). To achieve a
more specific genetic manipulation of VTA
DA
neurons, we conducted similar experiments
by using DAT-Cre transgenic mice (fig. S5, A
and E) instead of the coinjection with the
TH-Cre virus (52). Similarly, chemogenetic
activation of VTA
DA
neurons through the i.p. in -
jection of CNO switched the social preference
from females to male in both sexes, whereas
saline administration had no effect on social
preference (fig. S5, B to D, and F to H). Con-
currently, chemogenetic inhibition of VTA
DA
neurons in both male and female DAT-Cre
mice abolished their male preference in the
presence of TMT stressor (fig. S6). Collectively,
these results validate the critical roles of VTA
DA
neurons in mediating the switching of social
preference in both sexes.
Competition between VTA
DA
-mPOA and
VTA
DA
-NAc projections dictates the
sociosexual preference of male mice
Given that both female and male social in-
teractions induced the hyperactivity of VTA
DA
neurons in both sexes (Fig. 3) and that TMT
exposure led to male preference accompanied
by distinct c-Fos expression patterns in down-
stream brain regions (Fig. 2), we hypothesized
that sexually dimorphic alterations in DA cir-
cuits might be responsible for the shift in sex
Ctrl TMT
0
100
200
300
Social time (s)
Ctrl TMT
0
100
200
300
400
Social time (s)
Ec-Fos
Female mice
VTA: c -Fos
TMT
Ctrl
TMT
NAc: c-Fos
Ctrl
TMT
mPOA: c-Fos
TMT
Ctrl
Ac-Fos
Male mice
VTA: c-Fos
TMT
Ctrl
TMT
TMT
Ctrl
12 12
12 12
mPOA: c-Fos
TMT
Ctrl
B
CD
F
GH
NAc: c-Fos
0 50 100 150 200
BLA
mPFC
mPOA
NAc
VTA
c-fos+ cells/slice
Control (6)
TMT (6)
ns
ns
ns
0 50 100 150 200
BLA
mPFC
mPOA
NAc
VTA
c-fos+ cells/slice
Control (6)
TMT (6)
ns
ns
ns
Fig. 2. Sexually dimorphic changes of DA system under survival stress. (A) (Left) Schematic showing
the open-field social test, and (right) statistics of social interaction time of male mice in the absence
(Ctrl) or presence of TMT. (Bto D) Representative micrographs, and statistics of c-Fos–positive neurons in
the VTA, NAc, mPOA, mPFC, and BLA from animals as shown in (A). Scale bars, 500 mm. (Eto H)As
shown in (A) to (D), except that female mice were used. Scale bars, 500 mm. Data are presented as
mean ± SEM. Unpaired Student’sttest for social time for (A) and (E) and c-Fos staining for (D) and (H).
ns (not significant), P> 0.05; *P< 0.05; ***P< 0.001.
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 3of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
preference triggered by survival threats. Thus,
we next investigated whether and how the
VTA
DA
-NAc (53,54)andVTA
DA
-mPOA (42)path-
ways are involved. Double retrograde-virus
tracing strategy was used to separately label
VTA
DA
neurons [blue fluorescent protein
(BFP)] projecting to the NAc core and those
[green fluorescent protein (GFP)] projecting
to the mPOA (Fig. 5A and fig. S7, A and B). As
expected, the retrograde viruses from the NAc
(BFP) and the mPOA (GFP) mostly colocalized
with TH (Fig. 5A). However, there was mini-
mal colocalization between the two retrograde
viruses (Fig. 5A), suggesting that the NAc-
projecting and mPOA-projecting VTA
DA
neu-
rons represent distinct subpopulations.
To further investigate how the two DA cir-
cuits (VTA
DA
→mPOA and VTA
DA
→NAc) are
functionally involved in the switching of sex
preference in the context of survival threats, a
retrograde-tracing viral vector expressing the
red protein calcium indicator (jRGECO1) was
injected into the NAc core, whereas a viral
vector carrying the GCaMP was injected in
the mPOA of male mice (Fig. 5B and fig. S7C).
Dual-color fiber photometry recordings of
jRGECO1 and GCaMP were conducted to simul-
taneously monitor Ca
2+
signals in these two
subpopulations of VTA
DA
neurons (Fig. 5B). We
found that the mPOA-projecting VTA
DA
neu-
rons exhibited a stronger response to male in-
teractions, which was further strengthened in
the presence of TMT. Both the peak and AUC
values of the male interaction–coupled Ca
2+
sig-
nals in mPOA-projecting VTA
DA
neurons showed
a substantial increase in the TMT context (Fig.
5, C to G). Although DA projections from the
hypothalamus to the mPOA have been re-
ported to be involved in male-female mating
behaviors (44), the mPOA-projecting VTA
DA
neurons showed only a minimal response to
female interaction either under control con-
ditions or after TMT exposure (Fig. 5, C to G).
By contrast, the NAc-projecting VTA
DA
neu-
rons were only responsive to female interac-
tion, which was markedly suppressed by TMT
exposure (Fig. 5, C, and H to K). This is con-
sistentwiththefindingthatonlyaslightDA
signal can be observed in the NAc during male-
male social approach that is not followed by
attacking or mounting (21). These findings
suggest that the VTA
DA
-mPOA projection is
associated with male social preference, poten-
tially being activated under conditions of survival
threats. Conversely, the VTA
DA
-NAc projection
appears to underlie female preference in male
mice, which is attenuated by the presence of
survival stressors.
On the basis of these observations, we pro-
posedthatthebalancebetweentheVTA
DA
-
mPOA and VTA
DA
-NAc projections determines
the sex preference of male individuals during
social interactions. To test this hypothesis, we
next evaluated the mPOA dominance index, cal-
culated using the formula D¼FmPOA FNAc
ðÞ=
FmPOA þð FNAc Þ100%;where Frepresents the
10
20
-5
0
5
10
048
Time (s)
10
20
01020
0
20
40
60
80
100
Cumulative distribution (%)
0 5 10 15 20
0
20
40
60
80
100
Cumulative distribution (%)
0246810
0
20
40
60
80
100
Cumulative distribution (%)
0 5 10 15
0
20
40
60
80
100
Cumulative distribution (%)
-2 0 2 4 6 8
-2
0
2
4
6
-2 0 2 4 6 8
-2
0
2
4
6
-2 0 2 4 6 8
-2
0
2
4
6
-202468
-2
0
2
4
6
TH-Cre
DIO-GCamp
VTA
C57
G Social interaction without TMT
B Social interaction without TMT
A Ca2+ signal
C Social interaction with TMT
F Ca2+ signal
Male mice
Female mice
D
H Social interaction with TMT
I
Trials
E
J
Time (s)
Time (s)
dF/F (%)
dF/F (%)
dF/F (%)
dF/F (%)
Time (s)
Time (s)
TH-Cre
DIO-GCamp
VTA
C57
(18)
(35)
(33)
(27)
(17)
(12)
(24)
(20)
Time (s)
Time (s)
Time (s)
Time (s)
female
male
female
male
female
male
female
male
4
8
12
048
Time (s)
5
10
15
-5
0
5
10
10
20
30
0
10
20
048
Time (s)
10
048
Time (s)
10
20
30
10
20
-10
0
10
20
Ctrl TMT
-20
0
20
40
60
AUC
Ctrl TMT
0
5
10
15
20
25
Peak dF/F (%)
female
male
Ctrl TMT
0
5
10
15
Peak dF/F (%)
Ctrl TMT
-20
-10
0
10
20
30
40
AUC
Fig. 3. Association of VTA
DA
neural activity with the switching of sociosexual preference. (Ato E) Schematic, heatmaps, averaged traces, and statistics of
photometric Ca
2+
recordings of VTA
DA
neurons of male mice during social interactions in the absence or presence of TMT. (Fto J) As shown in (A) to (E), except that
female mice were used. Data are presented as mean ± SEM. Two-way ANOVA for (D), (E), (I), and (J); Kolmogorov-Smirnov test for peak distribution analysis of Ca
2+
signals in (B), (C), (G), and (H). *P< 0.05, **P< 0.01, ***P< 0.001.
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 4of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
normalized Ca
2+
signal DF=F0
ðÞas recorded
above. When the mPOA dominance index
was plotted against the interaction time with
either males or females, we found a positive
correlation with male-interaction time and a
negative correlation with female-interaction
time (Fig. 5L). The fitted lines intersected
near the equilibrium point between the two
projections (Fig. 5L), confirming our hypoth-
esis that the confrontation between mPOA
and NAc projections likely dictates the sex
preference in male mice. For detailed anal-
ysis, we further divided the plots into two
functional zones, in which the NAc projec-
tion was dominant in zone I and the mPOA
projection was dominant in zone II (Fig. 5M).
Consistent with the notion that TMT exposure
enhances the activity of mPOA-projecting VTA
DA
neurons, social interaction events under TMT
exposure were predominantly distributed in
zone II (Fig. 5, M and N). Male-interaction
events under control conditions were evenly
distributed but shifted predominantly to zone
II upon TMT exposure (Fig. 5M). By contrast,
female-interaction events were primarily found
in zone I but also switched to zone II in the
presence of TMT (Fig. 5M). Overall, male inter-
actions were predominant in zone II, whereas
female-interaction events were dominant in
zone I (Fig. 5, N and O). The social time with
males was much shorter than that with females
in zone I, but this was reversed in zone II (Fig. 5,
M and P), further confirming the correlation
between the two DA neuron populations and
sex preferences. Collectively, these findings
strongly support the concept that the balance
between VTA
DA
-mPOA and VTA
DA
-NAc pro-
jections is a pivotal determinant of sex prefer-
ence in male individuals, especially when they
are confronted with conflicts between innate
requirements and external threats.
To further confirm the sexually dimorphic
functions of the DA circuits in sex preference,
TH-Cre and Cre-dependent hM3Dq-expressing
viral vectors were coinjected into the VTA, and a
cannula was implanted in the mPOA for the
local application of CNO (Fig. 6A and fig. S8A).
We found that chemogenetic activation of
VTA
DA
terminals in the mPOA increased the
social time of male mice with males and de-
creased social time spent with females (Fig. 6B
and fig. S8B), resulting in the pronounced shift
in social preference from female to male en-
counters (Fig. 6B). Similar results were also
observed in the bedding preference tests (Fig.
6C). Consistently, the TMT-induced male pref-
erence of male mice was diminished upon the
chemogenetic inhibition of VTA
DA
terminals
in the mPOA (Fig. 6, D to F, and fig. S8, C and
D). These results demonstrate the critical roles
of VTA
DA
-mPOA transmission in mediating
the social preference for male conspecifics in
male mice. By contrast, chemogenetic inhibi-
tion of VTA
DA
-NAc core transmission led to the
shift of social preference from female to male,
whereas chemogenetic activation of this path-
way reversed the male preference to females
in male animals in the presence of TMT (Fig.
6, G to L, and fig. S9). These results show that
the dynamic interplay between the VTA
DA
-NAc
and VTA
DA
-mPOA pathways, representing the
ongoing balance between innate requirements
and external threats, determines the socio-
sexual preference of male mice.
Association of VTA
DA
-NAc projection with the
sociosexual preference of females
We next investigated neural mechanisms un-
derlying the social preference of female mice
with dual retrograde-tracing strategies. Sim-
ilarly to what we observed in male mice, the
retrograde viruses from the NAc (BFP) and
the mPOA (GFP) both colocalized with TH in
the VTA but showed minimal colocalization
between BFP and GFP (fig. S10A), validating
the distinct subpopulations of VTA
DA
neurons
that project to the NAc and the mPOA. To
verify the functional roles of each DA pro-
jection in the sex preference of female mice,
retrograde-tracing viral vectors expressing
jRGECO1 and GCaMP were injected into the
NAc and the mPOA, respectively, followed by
dual Ca
2+
recordings in the VTA (Fig. 7A and
fig. S10B). Differently from what we observed
in male mice, mPOA-projecting VTA
DA
neu-
rons in female mice showed no response to
male interactions, irrespective of whether TMT
was absent or present (Fig. 7, B to E, and fig.
S10C). By contrast, NAc-projecting VTA
DA
neu-
rons displayed higher Ca
2+
signals during so-
cial interactions with females, and this was
reversed in the presence of TMT (Fig. 7, B, and
F to H). Specifically, the amplitudes and AUC
values of the Ca
2+
signals during female inter-
action decreased, whereas those during male
interaction increased greatly upon TMT expo-
sure (Fig. 7, F to H, and fig. S10, D and E). Both
female and male interactions correlated pos-
itively with the Ca
2+
signals of NAc-projecting
VTA
DA
neurons. However, TMT exposure re-
sulted in an opposite distribution of these social
interaction events (Fig. 7, I and J). To quan-
tify these changes, the distribution region was
bifurcated into zone I and zone II by using the
median percentile line of Ca
2+
signals DF=F0
ðÞ.
Same-sex interaction events of female mice
mainly localized in zone II with larger DF=F0
values and social interaction time under con-
trol conditions, whereas these same-sex social
events were predominantly situated in zone I
in the context of TMT (Fig. 7I). By contrast,
opposite-sex interaction events were primar-
ily localized in zone I and shifted to zone II
Fig. 4. VTA
DA
neuron
excitation facilitates
male preference in both
sexes. (Ato C)Schematic
of virus injection, repre-
sentative micrograph
showing the colocaliza-
tion of mCherry and
TH-staining in the VTA,
and statistics of social
preference and bedding
preference tests after
the chemogenetic acti-
vation of VTA
DA
neu-
rons. Scale bar, 100 mm.
(Dto F) As shown in
(A) to (C), except that
female mice were used.
Data are presented
as mean ± SEM. Two-
way ANOVA for compar-
ison of time spent on male and female mice between Ctrl and TMT group; unpaired Student’st-test for male percentage. *P<0.05,**P< 0.01, ***P<0.001.
Crtl
hM3Dq
0
20
40
60
80
Male percentage (%)
Ctrl
hM3Dq
0
20
40
60
80
100
Male percentage (%)
Ctrl
hM3Dq
0
20
40
60
80
100
Male percentage (%)
Ctrl
hM3Dq
0
20
40
60
80
Male percentage (%)
BSocial preference CBedding preference
DESocial preference FBedding preference
Female mice
A
TH-Cre
DIO-hM3Dq
C57
VTA
TH-Cre
DIO-hM3Dq
C57
VTA
7716 13
11 11 13 13
VTA: TH/hM3Dq
VTA: TH/hM3Dq
100 µm
female
male
female
male
Ctrl hM3Dq
0
100
200
300
400
Time (s)
Ctrl hM3Dq
0
100
200
300
400
500
Time (s)
Ctrl hM3Dq
0
100
200
300
400
Time (s)
Ctrl hM3Dq
0
100
200
300
400
Time (s)
100 µm
Male mice
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 5of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
0
4
8
12
16
-100 -50 0 50 100
R² = 0.6362
R² = 0.6019
0
0
5
10
10
15
15
-100
-100 0
100
100
5
10
15
20
-2 0 2 4 6 8
Time (s)
5
10
15
5
10
15
20
-202468
Time (s)
5
10
15
048
-2
0
2
4
6
048
-2
0
2
4
6
-202468
Time (s)
10
20
30
40
10
-202468
Time (s)
10
20
30
40
10
048
-2
0
2
4
6
048
-2
0
2
4
6
VTA
mPOA
NAc
dF/F (%)
(41) (59)
A
Dwithout TMT Ewith TMT F
Iwith TMT J
Male mice
VTA: 2/R-GFP / 2/R-BFP/ TH
BC
Ca2+ signals Ctrl
TMT
with female with male
2/Retro-BFP 2/R-BFP 2/R-GFP
Merge
100 μm
G
K
Trials
LM
2/Retro-GFP
Trials
Trials
Zone I Zone II
Ctrl (23) Ctrl (15)
TMT (18) TMT (44)
Social time (s)
Social time (s)
Hwithout TMT
Trials
NOP
Zone I II Zone I II
50 s
5%
Ctrl
TMT
VTADA→mPOA
VTADA→NAc
mPOA dominance index
mPOA dominance index
Time (s) Time (s)
Time (s) Time (s)
dF/F (%)
dF/F (%)
dF/F (%)
VTA
2/R-jRGECO1
2/R-GCaMP
mPOA
NAc
24
8
17
VTADA→NAc
VTADA→mPOA
VTADA→NAc
VTADA→mPOA
51
BFP GFP
0
50
100 with TH
with BFP
% of colocalization
with GFP
Alone
-5
0
5
10
15
female(23)
male(15)
female(18)
male(44
female(23)
male(15)
female(18
male(44)
-5
0
5
10
15
Ctrl TMT
-20
0
20
40
60
AUC
Ctrl TMT
0
5
10
15
20
Peak dF/F (%)
Ctrl TMT
0
5
10
15
20
25
Peak dF/F (%)
Ctrl TMT
-20
0
20
40
60
AUC
0
50
100
female
male
% of social interaction
Ctrl TMT
0
50
100
Zone I
Zone II
%of distribution
0
5
10
15
20
Time (s)
female
male
Fig. 5. The balance between VTA
DA
-mPOA and VTA
DA
-NAc projections
determines sexual preference of males. (A) Schematic of retrograde virus
injection in the NAc (carrying BFP) and the mPOA (carrying GFP) separately
(left), representative micrographs (middle), and statistics showing distinct
subpopulations of VTA
DA
neurons projecting to the NAc and the mPOA in male
mice (right). Scale bar, 100 mm. (Band C) Schematic of retrograde virus injection
in the NAc (carrying jRGECO1) and the mPOA (carrying GCaMP), and
representative traces of Ca
2+
signals (DF/F
0
) of NAc-projecting (red) and
mPOA-projecting (green) VTA
DA
neurons in social-interacting male mice in the
absence (Ctrl) or presence of TMT. The enlarged insets of the social interaction–
coupled Ca
2+
signals are shown on the right. (Dto G) Heatmaps, averaged
traces, and statistics of peak and AUC values of social interaction–coupled
Ca
2+
signals of mPOA-projecting VTA
DA
neurons in male mice. (Hto K)As
shown in (D) to (G), except that Ca
2+
signals of NAc-projecting VTA
DA
neurons
were recorded. (L) Pearson correlation analysis between social time (single
interacting events) and the mPOA dominance index [calculated with the formula
of Ca
2+
signals as below: (VTA
DA
→mPOA –VTA
DA
→NAc)/(VTA
DA
→mPOA +
VTA
DA
→NAc) × 100%] in male mice. (M) The distribution zones, as shown in
(L), were divided into zone I and zone II by the equilibrium line of the mPOA
dominance index. (Nto P) Statistics of social events (Ctrl versus TMT; female-
versus male-interaction events; female- versus male-interaction time) located
in zone I and zone II. Data are presented as mean ± SEM. Two-way ANOVA
for (F) to (K), and (P); Fisher’s exact test for (A), (N), and (O). *P< 0.05, **P<
0.01, ***P< 0.001.
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 6of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
upon TMT exposure (Fig. 7J). These findings
suggest that the higher excitability of the VTA
DA
-
NAc projection is underlying the preference
switch of female mice, transitioning from same-
sex interaction in control states to opposite-
sex interaction when confronted with survival
threats.
Firing patterns of VTA
DA
-NAc projection
determine the sociosexual preference
of females
Why is the hyperactivity of VTA
DA
-NAc pro-
jection positively correlated with both female
(in the control state) and male (under survival
threats) interactions? Because VTA
DA
neurons
exhibit two primary modes of firing activities,
including the sustained low-frequency (typically
~5-Hz) tonic firing and the high-frequency
(>10-Hz, usually <50-Hz) phasic firing with
three to five spikes per burst (55–57), we spec-
ulated that the alteration between phasic ver-
sus tonic firing patterns may be responsible
for the switching of social preference of fe-
male individuals. To test this hypothesis, a TH-
driven GCaMP-expressing adeno-associated
virus (AAV) vector was injected into the VTA
and an optical fiber was implanted in the NAc
for the specific recording of Ca
2+
signals of
VTA
DA
projections in this area (Fig. 8A and
fig. S11, A and B). Although both female and
male interactions were correlated with Ca
2+
transients of VTA
DA
terminals in the NAc, fe-
male interaction–coupled Ca
2+
transients were
much stronger and had faster kinetics, indi-
cative of the transient high-frequency or phasic-
like action potential (AP) firings of VTA
DA
neu-
rons. However, the amplitude and AUC values
of these Ca
2+
transients were substantially re-
duced in the presence of TMT (Fig. 8, B and
C). By contrast, male interactions were linked
to Ca
2+
signals with much slower kinetics, re-
flective of the sustained low-frequency or tonic-
like AP firings of VTA
DA
neurons. These signals
were increased upon exposure to TMT (Fig. 8,
B and C). Consistent with these observations,
male interaction–coupled Ca
2+
transients dis-
played a smaller rise rate and a larger half-
height duration compared with those occurring
during female interactions, irrespective of the
presence or absence of TMT (Fig. 8D). These
findings collectively suggest that the firing-
patternalterationofVTA
DA
terminals in the
NAc aligns with the sociosexual preference of
female mice. On the basis of these results, we
propose that phasic firing promotes female
preference, whereas the survival threat of TMT
exposure shifts the behavior to male preference
Fig. 6. VTA
DA
-NAc and
VTA
DA
-mPOA path-
ways modulate sex
preference in opposite
directions in male
mice. (Ato C) Sche-
matic of virus injection
and cannula implanta-
tion, representative
micrograph showing the
colocalization of mCherry
and TH-staining in the
VTA, and statistics
of social preference and
bedding preference
tests after the chemo-
genetic activation of
VTA
DA
-mPOA projection
in male mice. Scale bar,
100 mm. (Dto F)As
shownin(A)to(C),except
that chemogenetic
inhibition of the VTA
DA
-
mPOA projection was
used in male mice
in the presence of TMT.
(Gto L)Asshownin
(A) to (F), except that
chemogenetic manipula-
tion was conducted
on the VTA
DA
-NAc
projection in male mice.
Data are presented
as mean ± SEM. Two-
way ANOVA for social
time and unpaired
Student’sttest for
male percentage. *P<
0.05, **P< 0.01,
***P< 0.001.
Male mice
TH-Cre
DIO-hM3Dq
mPOA VTA
CNO
NAc VTA
CNO
TH-Cre
DIO-hM3Dq
with TMT
AVTADA → mPOA
VTADA → NAc
C Bedding preference
100 µm
VTA: TH/hM3Dq
BSocial preference
TH-Cre
DIO-hM4Di
mPOA VTA
CNO
100 µm
VTA: TH/hM4Di
DF Bedding preference
ESocial preference
NAc VTA
CNO
TH-Cre
DIO-hM4Di
−
−
GI Bedding preference
HSocial preference
JL Bedding preference
KSocial preference
100 µm
VTA: TH/hM3Dq
VTA: TH/hM4Di
100 µm
with TMT
Ctrl hM3Dq
0
50
100
150
200
250
Time (s)
Ctrl hM4Di
0
100
200
300
400
Time (s)
Ctrl hM3Dq
0
100
200
300
400
Time (s)
ns
ns
Ctrl hM4Di
0
100
200
300
400
Time (s)
Ctrl hM4Di
0
100
200
300
400
Time (s)
ns
Crtl
hM4Di
0
20
40
60
80
Male percentage (%)
Crtl
hM4Di
0
20
40
60
80
Male percentage (%)
15
16
13 17
Ctrl hM3Dq
0
100
200
300
400
500
Time (s)
female
male
Crtl
hM3Dq
0
20
40
60
80
Male percentage (%)
Crtl
hM3Dq
0
20
40
60
80
Male percentage (%)
10 10 12 12
CrtlhM4Di
0
20
40
60
80
Male percentage (%)
CrtlhM4Di
0
20
40
60
80
Male percentage (%)
88
99
Ctrl
hM3Dq
0
20
40
60
80
Male percentage (%)
Crtl
hM3Dq
0
20
40
60
80
Male percentage (%)
Ctrl hM3Dq
0
100
200
300
400
Time (s)
ns
10 12 10 12
Ctrl hM4Di
0
100
200
300
400
Time (s)
−
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 7of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
through the inhibition of phasic firing and
the facilitation of tonic firing of these NAc-
projecting VTA
DA
neurons.
To test whether the firing pattern of VTA
DA
neurons is a decisive factor in dictating the
sex preference of female individuals, TH-Cre
and Cre-dependent ChR2-expressing viral vec-
tors were coinjected into the VTA, and a pair
of optical fibers were implanted in the NAc
to selectively elicit DA terminals with either
phasic (20-Hz, five pulses, 1-s burst interval)
or tonic (5-Hz) light stimulation (Fig. 8E and
fig. S11, C and D). We found that, in control
conditions, phasic stimulation of DA termi-
nals in the NAc resulted in an increase in
social time of female mice spent with females
and a decrease in time spent with male ani-
mals, leading to a more pronounced female
preference (Fig. 8F and fig. S11E). Similarly,
phasic activation of VTA
DA
-NAc projections
in female mice also facilitated the preference
for female bedding (fig. S11F). Conversely,
mice that received tonic stimulation spent
more time with males and less time with fe-
males, exhibiting increased male preference
in the social preference test (Fig. 8G and fig.
S11G). A similar phenomenon was observed
in the bedding preference test (fig. S11H). These
results confirmed that the firing-pattern alter-
ation of NAc-projecting VTA
DA
neurons deter-
minesthesexpreferenceoffemalemiceduring
social interaction, during which phasic firing
promotes female preference and tonic firing
facilitates male preference.
Although tonic versus phasic firing patterns
of VTA
DA
neurons have been reported to be
functionally involved in motivation behavior
andrewardprediction(58–61), how the firing-
pattern alteration determines social decision-
making remains to be fully elucidated. To
determine whether the altered DA release is
underlying sex preference governed by dif-
ferent firing patterns, we performed electro-
chemical carbon-fiber electrode (CFE)–based
amperometric recordings (62–64)inresponse
to phasic or tonic optogenetic stimuli and found
that repetitive phasic stimulation induced the
release of larger DA transients, characterized by
Fig. 7. Positive associ-
ation of VTA
DA
-NAc
projection with both
male and female
interactions of female
mice. (Aand B) Sche-
matic of retrograde
virus injection in the
NAc (carrying jRGECO1)
and the mPOA (carrying
GCaMP), and represen-
tative traces of Ca
2+
signals (DF/F
0
)ofNAc-
projecting (red) and
mPOA-projecting (green)
VTA
DA
neurons in social-
interacting female mice
in the absence (Ctrl) or
presence of TMT (B).
(Cto E) Heatmaps,
averaged traces, and
statistics of social inter-
action–coupled Ca
2+
signals of mPOA-projecting
VTA
DA
neurons in female
mice. (Fto H)Asshownin
(C) to (E), except that
social interaction–coupled
Ca
2+
signals of NAc-
projecting VTA
DA
neurons
were recorded. (I)Pearson
correlation analysis
between social time of
female interacting events
and the Ca
2+
signals of
NAc-projecting VTA
DA
neu-
rons, zone division by the
50% percentile of Ca
2+
signals, and statistics of
social events (social events
and interacting time of
single event in Ctrl versus
TMT) located in zone I
and zone II. (J)Asshown
in (I), except that male
interacting events were used for analysis. Data are presented as mean ± SEM. Two-way ANOVA for (E) and (H) and social time analysis in (I) and (J); Fisher’sexact
test for social events distribution in (I) and (J). *P<0.05,**P< 0.01, ***P<0.001.
R² = 0.6279
0
4
8
12
16
0 6 12 18
R² = 0.6125
0
4
8
12
04812
04 8
Time (s)
10
20
30
10
-2 0 2 4 6 8
Time (s)
10
20
30
10
20
30
10
20
30
048
Time (s)
10
20
30
III
0
5
10
15
Social time (s)
Ctrl TMT
048
Time (s)
-2
0
2
4
6
048
Time (s)
-2
0
2
4
6
048
Time (s)
-2
0
2
4
6
048
Time (s)
-2
0
2
4
6
10
04 8
Time (s)
10
20
30
Female mice
H
Ctrl
TMT
with female with male
Trials
Trials
C Social without TMT D Social with TMT E
Trials
Trials
A B
Ca2+ signals
VTADA→NAc VTA DA→mPOA
Zone I Zone II
Social time (s)
Peak dF/F (%)
Ctrl(39) TMT (16) Ctrl (32) TMT(32)
Social time (s)
Peak dF/F (%)
Zone I Zone II
F Social without TMT G Social with TMT
I Female interaction J Male interaction
dF/F (%)
dF/F (%)
dF/F (%)
dF/F (%)
Zone
Zone
VTA
2/R-jRGECO1
2/R-GCaMP
mPOA
NAc
(32)
(39) (32)
(16)
(32)
(39)
(32)
(16)
25 7
7
25
15 24
11 5
VTADA→NAc
VTADA→mPOA
VTADA→NAc
VTADA→mPOA
-5
0
5
10
15
-5
0
5
10
15
female
male
female
male
female
male
female
male
Ctrl TMT
0.0
0.5
1.0
1.5
2.0
2.5
Peak dF/F (%)
Ctrl TMT
0
5
10
15
20
Peak dF/F (%)
Ctrl TMT
0
50
100
Zone I
Zone II
%of distribution
Ctrl TMT
0
50
100
Zone I
Zone II
%of distribution
III
0
5
10
15
Social time (s)
Ctrl TMT
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 8of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
robust periodic fluctuations (Fig. 8H and fig.
S11I). By contrast, the same number of AP-like
tonic stimuli induced a lower sustained release
of DA (Fig. 8H and fig. S11J), reflecting the steady
and continuous DA release associated with ton ic
firing. Consistent with this, phasic stimulus led
to a much higher local DA concentration and
thus more total DA release compared with that
resulting from the tonic stimulus (Fig. 8, H and
I). Thus, the firing pattern alteration of VTA
DA
neurons determines the sex preference of female
mice most probably through the modulation of
local DA fluctuations.
Considering that DA transmission in the
NAc is mediated via either the inhibitory D2R
Fig. 8. The firing pattern alteration of
VTA
DA
-NAc projection determines
sex preference of females. (A) Sche-
matics of virus (TH-GCaMP) injection
in the VTA and Ca
2+
recordings in
the NAc of the social-interacting female
mice. (Bto D) Heatmaps, averaged
traces, and statistics of social interac-
tion–coupled Ca
2+
signals, including the
peak, AUC, rise time, and half-height
duration (HHD) of VTA
DA
terminals in
the NAc of female mice. (E) Schematics
of virus (TH-Cre and DIO-ChR2)
injection in the VTA and optical stimuli
[light stimulation (L-stim), phasic or
tonic] in the NAc. (Fand G) Statistics of
social preference test of female mice
after phasic or tonic activation of
VTA
DA
terminals in the NAc. (H) CFE
amperometry recordings of DA release
with the calibrated heatmaps of DA
concentration, and statistics of peak
values of DA release in response to phasic
or tonic L-stim. (I) Cumulative traces,
and statistics of total DA release in the
NAc in response to phasic or tonic
L-stim as shown in (H). (J) Schematic
showing the proposed DA transmission in
response to phasic or tonic AP firing of
DA terminals, and representative AP
traces of D1R and D2R MSNs in the NAc.
(Kand L) Statistics of AP frequency
and frequency changes of D1R and D2R
MSNs in response to phasic or tonic
L-stim. (M) Schematic of virus injection
(D1R-Cre and DIO-ChR2) and optical fiber
implantation in the NAc, and statistics
of social interaction time of female mice
after the L-stim (470 nm, 5 mW, 5-ms
pulse at 20 Hz). (N) As shown in (M),
except that optogenetic inhibition
of D2R MSNs was used and the 580-nm
L-stim (10 mW, continuous) was
applied. (O) Working model showing the
sexual dimorphic DA circuits involved in
sociosexual preference. The balance
between VTA
DA
-NAc (female preference)
and VTA
DA
-mPOA (male preference,
dominates when confronted with survival
threats) projections determines the
social decisions of male individuals,
whereas the firing-pattern alteration of
VTA
DA
terminals and thus the biased DA
transmission (phasic firing–facilitated
D1R transmission mediates female preference, and tonic firing–dominated D2R transmission mediates male preference) in the NAc is utilized by females to determine
their sociosexual preference. Data are presented as mean ± SEM. Two-way ANOVA for (C) to (G) and (K) to (N), unpaired Student’sttest for (H) and (I) and male percentage
in (F) and (G), and paired Student’sttest for L-stim effects in (K). *P<0.05,**P< 0.01, ***P<0.001.
Ctrl ChR2
0
100
200
300
Time(s)
Crtl ChR2
0
20
40
60
80
Malepercentage (%)
Ctrl ChR2
0
100
200
300
Time(s)
Crtl ChR2
0
20
40
60
80
100
Male percentage (%)
048
Time (s)
0
4
8
048
Time (s)
0
4
8
048
Time (s)
10
-5
0
5
10
10
048
Time (s)
10
20
30
10
20
10
10
10
20
30
10
20
NAc
D2R-Cre
DIO-NpHR
C57
L-Stim
NAc VTA
Peak
C57
E Social preference
TH-Cre
DIO-ChR2
C57
L-Stim
(tonic or phasic)
Tonic stim (5 Hz)
Phasic stim (20 Hz, 5 pluses)
AB
F
Rise rate
Social interaction without TMT
C
D
Female mice
G
H
Trials
Trials
Social interaction with TMT
AUC
HHD
Ca2+
recording
Ca2+ signals
dF/F (%)
dF/F (%)
Phasic firing
Tonic firing 1 s
0
120
J
nM
Ton ic
Phasic
I
2 s
60 nmol
Tonic
Phasic
DA (nmol)
8 s
60 mV
Tonic Phasic
D1R
D2R
Total release
K
1 s
D1R
D2R
Tonic
Phasic
DAergic
VTA NAc
D1R
D2R
mPOA
NAc
NAc
Male mice
Female mice
Male
VTA
Female
Male
Female
Socio sexual preference
NAc VTA
TH-GCaMP
NAc
D1R-Cre
DIO-ChR2
C57
L-Stim
MNO
D1R D2R
L
Loff
Lon
Ton i c Phasic To n ic Phasic Tonic Phasic
(18)
(17)
(33)
(26)
88 99
15
15 15 15
22 22
27
27
8888
Tonic
Phasic
Ctrl ChR2
0
100
200
300
Time (s)
Ctrl NpHR
0
50
100
150
200
250
Time(s)
-10
0
10
20
female
male
female
male
Ctrl TMT
0
20
40
60
Peak / rise time
ns
ns
Ctrl TMT
0
1
2
3
4
5
Half-heigth Duration(s)
Ctrl TMT
-40
-20
0
20
40
60
AUC
Ctrl TMT
0
1
2
3
4
5
Half-heigthwidth(s)
0
50
100
150
200
DA concentration (nM)
0
50
100
150
200
250
0
1
2
3
4
5
Frequency (Hz)
ns
0
1
2
3
4
Frequency (Hz)
-200
0
200
400
600
Frequencychanges(%)
D1R (22)
D2R (27)
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 9of14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
(type 2 DA receptor) or the excitatory D1R
(type 1 DA receptor) and that the D2R has a
~10-fold-higher binding affinity to DA than
does the D1R (65,66), we propose that differ-
ent DA fluctuations might determine social
decisions through their distinct modulation
of downstream D1R and D2R medium spiny
neurons (MSNs) in the NAc. To test this idea,
we conducted whole-cell patch clamp AP re-
cordings of D1R and D2R MSNs in response
to the optogenetic activation of VTA
DA
ter-
minals in the NAc (Fig. 8J). We found that a
phasic stimulus, but not a tonic stimulus, led
to the robust increase in AP frequency of D1R
MSNs (Fig. 8, J and K). Conversely, tonic
stimuli resulted in a much more pronounced
inhibition of D2R MSNs compared with phasic
stimuli (Fig. 8, J and K). Thus, D1R MSNs are
specifically responsive to the transient, large
DA fluctuations associated with phasic firing of
DA terminals, whereas D2R MSNs are more
sensitive to the sustained DA release resulting
from prolonged tonic firing of DA terminals
(Fig. 8L).
To further confirm that the two types of
MSNs are sufficient to mediate the sociosexual
preference of female mice, we used an optoge-
netic strategy (coinjection of D1R/D2R-Cre
and Cre-dependent ChR2-expressing viral vec-
tors) to selectively manipulate D1R and D2R
MSNs in the NAc (Fig. 8, M and N, and fig. S12,
A and D). Optogenetic excitation of D1R MSNs,
which mimics the phasic firing of DA termi-
nals, further increased social time with females,
resulting in decreased preference for males by
female animals (Fig. 8M and fig. S12B). By
contrast, optogenetic inhibition of D2R MSNs,
which mimics the tonic firing of DA terminals,
led to t he switch of fe ma le animal s’social
preference from female to male (Fig. 8N and
fig. S12E). Similar results were obtained in the
bedding preference test (fig. S12, C and F).
Additionally, similar results were achieved when
D2R-Cre viral vectors were replaced with a virus
carrying A2A-Cre (fig. S12, G to I). Collectively,
these findings suggest that females make the
social decision throughthebalancebetween
phasic firing–facilitated DA-D1R transmission
and tonic firing–dominated DA-D2R trans-
mission in the NAc.
Discussion
Social interaction between different sexes is a
fundamental aspect of life for most species
because it plays an essential role in repro-
duction, survival, and evolution. In this study,
we found that males prefer social interaction
with females, which is probably driven by the
innate needs of mating and reproduction. Fe-
males also show a preference for female en-
counters under control conditions. This has
beenobservedinbothrodents(33)andhu-
mans (such as bosom friends) (67,68), which
might stem from the long-term evolution of
the same-sex social collaborations in activities
such as maternal support, emotional sustenance,
and consolation. However, in the presence of
survival stress or in a hostile or challenging
environment, both sexes shift their preference
toward male interactions, through which they
may reach a maximal benefit from social pro-
tection. Thus, sociosexual preference is an adap-
tive response to the specific environment in
which they live, and this preference can be con-
vergently refined by the internal-drive condi-
tion and external-environment context to ensure
survival and species continuation in rapidly
changing environments.
Although critical roles of the mesolimbic DA
system in reward, motivation, and social inter-
action have been well-established (14,40,41),
its involvement in shaping sociosexual pref-
erence remains largely unknown. In the pre-
sent work, we demonstrated a strong link
between the excitability of VTA
DA
neurons
and the shift in sex preference under survival
stress. Our findings define the mesolimbic DA
system as a key pivotal hub for sex preference
in social decision-making, which may explain
how animals or humans can make appropriate
social decisions to get the maximal benefits,
especially when they encounter conflicts be-
tween innate requirements and external sur-
vival threats.
Male and female individuals utilize very
different schemes for the neural coding of so-
cial preferences. The competition between dif-
ferent VTA
DA
-projecting circuits was used by
males to encode their sociosexual preference;
the VTA
DA
-NAc projection is predominant
under control conditions to promote female
preference, whereas the activities of VTA
DA
-
mPOA projection can be enhanced by survival
threats to facilitate social interaction with male
individuals (Fig. 8O). These findings support
the notion that the VTA
DA
-NAc reward pathway
may play crucial roles in the innate require-
ments of males for mating and reproduction,
whereas the defensive VTA
DA
-mPOA pathway
can be activated by survival threats and thus
leads to male preference, which may contrib-
ute to forming a stronger protective network
for individual survival and population conti-
nuity. By contrast, the firing-pattern altera-
tion of the VTA
DA
-NAc projection is utilized
by females to determine their sex preference.
Phasic firing–like excitation of VTA
DA
neu-
rons promotes female preference through the
facilitated D1R transmission in the NAc; how-
ever, environmental threats lead to male pref-
erence through the enhanced tonic firing–like
excitation of DA neurons and thus the pre-
dominant DA-D2R transmission in the NAc
(Fig. 8O). Thus, these findings not only elu-
cidate the sex-binary properties of VTA
DA
neurons in terms of neural circuit integration
and firing-pattern alteration but also high-
light critical roles of this sexually dimorphic
DA system in mediating social preference shifts
in response to survival threats; they also pro-
vide valuable insights into the neural mech-
anisms that govern social decision-making and
its modulation by environmental factors.
How could the dimorphic DA circuits con-
tribute to the determination of social prefer-
ence? Although male and female individuals
utilize different strategies to encode their so-
cial decisions on sex selectivity, their prefer-
ence can be similarly modulated by different
survival stressors even when those stressors
are transduced via distinct sensory pathways
such as olfaction (TMT), vision (contextual
FC), and auditory (cued FC). How these sen-
sory inputs converge onto specific subpopula-
tions of VTA
DA
neurons and how their firing
patterns can be modified by external survival
stresses deserve systematic investigations in
the future. It would also be intriguing to ex-
plore whether the VTA
DA
-NAc and VTA
DA
-mPOA
projections are selectively activated in other
species such as nonhuman primates and hu-
mans and how this selectivity may contribute
to their sociosexual preferences. Further stud-
ies are needed to examine the extent to which
these projections are involved in other aspects
of social decisions, such as social avoidance,
social preference, social support, and social
hierarchy. It would also be interesting to test
whether and how sex preference can be mod-
ulated by other physiological conditions or by
external conditions such as extreme climate
changes, food deprivation, and social stress
exposure (69).
Taken together, our study reveals that both
sexes exhibit female preference during social
interaction but switch to male preference un-
der survival threats, indicating an adaptive
shift in social decision-making. However, they
utilize very different schemes for the neural
coding of social preference. This study not only
establishes critical roles of DA transmission
in social decision-making and defines the
sexually dimorphic DA circuit mechanisms
underlying sex preference but also provides
a conceptual framework for understanding
how social decisions can be convergently de-
termined by the balance between innate re-
quirements and external survival threats.
Materials and methods
Animals
Adult (∼3 months) male and female (non-estrus)
C57/BL6 mice were from Charles River (Beijing).
DAT-CretransgenicmicewerefromJackson
lab (JAX: 006660). Female mice used in this
work were all non-estrous virgin females. The
animals were housed in the animal facility
withamaximumoffivemicepercageundera
12-hour light-dark cycle at a consistent tem-
perature of 22° ± 2°C and were provided with
food and water ad libitum. The use and care
of animals were approved and directed by
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 10 of 14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
the Animal Care and Use Committee of Xi’an
Jiaotong University.
Stereotaxic surgery
Stereotaxic surgery was performed as previ-
ously described (14,62,63).Themicewerean-
esthetized with avertin (1.5% tribromoethanol,
2.5% tertiary butanol, and 0.9% NaCl; 0.1 ml/
5 g body weight). Mice were fixed in a stereo-
taxic frame (RWD Instruments, China) for
precise surgical manipulation. The skull was
exposedbymakingamidlineincision,and
cleaned with hydrogen peroxide to expose the
bregma and lambda for head position adjust-
ment. A cranial drill (RWD Instruments, China)
was used to create small cranial windows for
the implantation of guide cannulas (RWD In-
struments, China), optical fibers (Inper, China),
or for virus injection. For virus injection, the
virus (each 200 to 300 nl) was injected into the
target area at 100 nl min
−1
through a glass pi-
pette. The pipettes were left in place for 10 min
post-injection. After surgery, the skin was su-
tured, and animals were placed on a heating
plate (KEL-2000, Nanjing, China) at 37°C for
recovery. Three weeks later, when required,
mice were anesthetized and fixed on a stereo-
taxic frame for the implantation of the guide
cannulas or optical fibers using similar stereo-
taxic procedures.
Slice preparation
The general methods for preparing NAc slices
were similar to those described previously
(62,70). Mice were anaesthetized with tribromo-
ethanol via intraperitoneal injection (0.2 ml/
10 g). Then, the whole brain was quickly re-
movedintoice-coldcuttingsolution(inmM):
110 C
5
H
14
ClNO, 2.5 KCl, 0.5 CaCl
2
,7.0MgCl
2
,
1.3 NaH
2
PO
4
, 25 NaHCO
3
, 10 D-(+)-glucose,
saturated with 95% O
2
and 5% CO
2
. Addi-
tional 1.3 L-ascorbate (C
6
H
8
O
6
), 0.6 Na-pyruvate
(C
3
H
3
NaO
3
) were included for whole-cell patch
clamp recordings. Then, horizontal slices con-
taining the NAc were cut at 300 mm using a
vibratome (Leica VT 1200s). The slices were
transferred into artificial cerebrospinal fluid
(aCSF, in mM): 125 NaCl, 2.5 KCl, 2.0 CaCl
2
,
1.3 MgCl
2
,1.3NaH
2
PO
4
,25NaHCO
3
,10D-(+)-
glucose, saturated with 95% O
2
/5% CO
2
(addi-
tional 1.3 C
6
H
8
O
6
,0.6C
3
H
3
NaO
3
for whole-cell
patch clamp recording) and incubated at 37°C
for 30 min, followed by an additional 30 min
at room temperature before recording.
Whole-cell patch clamp recording
Whole-cell patch clamp recordings were per-
formed as described previously (62,63). Neu-
rons in brain slices were visualized under an
upright microscope (BX51W1, Olympus) using
a 60× water-immersion lens, along with an
infrared-differential interference contrast and
an infrared camera (IR-2000) connected to
the video monitor. Slices were perfused with
aCSF at a rate of 1.5 ml/min. Somatic patch-
clamp signals (action potentials, APs) were
recorded under whole-cell configuration using
an EPC10/2 amplifier under the control of
PatchMaster version 2X90.2 software (HEKA
Electronik, Germany). Signals were digitized
at 20 kHz and low-pass filtered at 2.9 kHz.
The pipettes were pulled with a PC-10 glass
micropipette puller (Narishige, Japan) to achieve
resistances of 3 to 5 MW.Theintracellular
pipette solution contained (in mM) 120 K
+
-
Glucose, 5 NaCl, 10 HEPES, 1 MgCl
2
,0.2EGTA,
2 MgATP, 0.1 Na
3
GFP, 10 phosphocreatine,
with pH adjusted to 7.2, and 305 mOsm. Optical
stimulation (589 nm, continuous; or 488 nm
at 5 Hz) was delivered using laser generators
(Beijing Viasho Technology, China) through
an optical fiber (200 mm in diameter) posi-
tioned ~200 mm from the slice surface. Drugs
were delivered with a multichannel perfusion
system (World Precision Instruments, Sarasota).
All recordings were made at room temper-
ature. Data were analyzed with Igor Pro 6.22
software (Wavemetrics, Lake Oswego, OR), with
series conductance and membrane conduct-
ance monitored to ensure the seal condition of
patch-clamp recordings.
Amperometric DA recording in brain slices
Amperometric recordings in slices were made
as described previously with small modifica-
tions (62–64). CFEs with a diameter of 7 mm
which had an exposed sensor tip of ~200 mm
were used to measure DA release. The exposed
sensor tip was completely inserted into the
tissue of NAc slices and a holding potential of
780 mV was applied using an EPC10/2 amplifier
under the control of Pulse software PatchMaster
v2x90.2 (HEKA Electronic, Germany). DA re-
lease in the NAc was evoked by using 470-nm
yellow light with specific stimulation patterns
to mimic phasic (20-Hz, five spikes/burst, 1-s
burst-interval) or tonic (5-Hz, sustained spikes)
AP firings. The amperometric current (I
amp
)
was low-pass filtered at 100 Hz and digitized
at 3.13 kHz. Off-line data analysis was per-
formed using Igor software (WaveMetrix).
Optogenetic and chemogenetic manipulation
in vivo
To perform chemogenetic manipulation (62,71),
pAAV-EF1a-DIO-hM3D(Gq)-EGFP-WPRE (AAV2/
9, 1.0 × 1 0
12
vg ml
−1
)orpAAV-EF1a-DIO-hM4D
(Gi)-EGFP-WPRE (AAV2/9, 1.0 × 10
12
vg ml
−1
)
mixed with rAAV-TH-NLS-CRE-WPRE-hGH pA
(AAV2/9, 1.0 × 10
12
vg ml
−1
) were injected into
the VTA of male/female mice. After a recovery
period of 3 weeks, guide cannulas (internal
diameter of 1.25 mm) were implanted into the
NAc or mPOA for the local application of CNO,
and the mice were allowed to recover for 1 week
prior to behavioral experiments. A stainless-steel
injector attached to a 10-mlsyringeandaninfu-
sion pump was inserted into the guide cannula.
CNO (3 mM, 0.5 ml) was infused at a rate of 100 nl
min
−1
under the control of a micro-injection
pump (R462, RWD Instruments, China). The
injector was slowly withdrawn 2 min after the
infusion, and behavioral tests were performed
5 min after the microinjection.
For the optogenetic manipulation (45,63,72),
rAAV-EF1a-DIO-hChR2(H134R)-EYFP-WPRE-
hGH pA (AAV2/9, 1.0 × 10
12
vg ml
−1
)mixedwith
rAAV-TH-NLS-CRE-WPRE-hGH pA (AAV2/9,
1.0 × 10
12
vg ml
−1
) were injected into the VTA
of female mice. Three weeks later, an optical
fiber was implanted in the NAc for the de-
livery of light stimuli. The mice were allowed
to recover for 1 week prior to behavioral tests.
The delivery of a train of stimuli with 470-nm
blue light (20 Hz, five spikes/burst, 1-s burst-
interval; or 5 Hz, sustained spikes) was con-
trolledbyaMaster-8pulsestimulator(Hangzhou
Newton Technology, China).
Optical fiber photometry
The Ca
2+
transient signal was measured using a
custom-built fiber photometry system (QAXK-
FPS-SS-LED-OG, ThinkerTech, China) as de-
scribed previously (62,73). To record the somatic
Ca
2+
signal of VTA
DA
neurons, rAAV-EF1a-DIO-
GCaMp6s-WPRE-hGH pA (PT-0017, BrainVTA,
China) mixed with rAAV-TH-NLS-CRE-WPRE-
hGH pA (PT-0179, BrainVTA, China) were in-
jected into the VTA and an optical fiber was
implanted in the VTA. For simultaneous record-
ing of the somatic Ca
2+
signals of NAc- and
mPOA-projecting VTA
DA
neurons, rAAV-hSyn-
GCaMp6s-WPRE-hGH polyA, 2/Retro (PT-0145,
BrainVTA, China) was injected into the mPOA
and rAAV-hSyn-jRGECO1a-WPRE (PT-6767,
BrainVTA, Wuhan, China) was injected into
the NAc. An optical fiber (Inper, Hangzhou,
China) was implanted into the VTA to record
the somatic Ca
2+
signals in VTA
DA
neurons 3
weeks after virus injection. To record the Ca
2+
signal of VTA
DA
terminals in the NAc, we in-
jected rAAV-EF1a-DIO-GCaMp6s-WPRE-hGH
pA mixed with rAAV-TH-NLS-CRE-WPRE-hGH
pA into the VTA and implanted an optical fiber
in the NAc. The mice were allowed 1 week recov-
ery befo re Ca
2+
signal recording.
During the recording test, a female stranger
mousewasputinameshcageonthecornerof
an open field, and a male stranger mouse was
placed on the diagonal opposite side. The test
mouse was placed in the center of the open
fieldandallowedtoexplore freely. Social in-
teraction included both nose-to-nose interac-
tion and body-sniffing (tail, body, and paw) of
the caged mice by the testing mice. The time
points that the experiment mice interacted
with female/male caged mice were manually
recorded. The 470-nm and 580-nm laser beams
were reflected off a dichroic mirror and focused
with an objective lens to excite the calcium in-
dicators GCaMP6s and jRGECO1a, respec-
tively. The Ca
2+
signals were collected with a
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 11 of 14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
photomultiplier tube and converted into elec-
trical signals using customized acquisition
software written in NI Lab View (National
Instruments). The changes of Ca
2+
signals were
evaluated using DF=Fvalues, where DF=F¼
FFmean
ðÞ
Fmean 100%.Fmean was the average fluo-
rescence intensity (baseline) before laser stim-
ulation. Recording results were analyzed with
custom MATLAB (MathWorks, Natick, MA,
USA) MAT files produced by Thinker Biotech.
Immunochemistry
Immunohistochemistry was applied as de-
scribed previously (62,63,74). Briefly, mice
were anesthetized with avertin and perfused
with ice-cold phosphate-buffered saline (PBS)
containing 4% paraformaldehyde (PFA) for
tissue fixation. The brain was removed and
immersed in 4% PFA at 4°C for 24 hours. After
dehydration in 10, 20, and 30% sucrose, a
series of 30-mm coronal slices across the VTA/
NAc/mPOA were cut using a Thermo cryostat
(RWD instruments, China). The slices were
rinsed three times with PBS and permeabilized
with 0.3% Triton X-100 in PBS containing
2% BSA for 5 min at room temperature. After
blocking with 10% FBS in PBS for 2 hours, the
slices were incubated with primary antibodies
at 4°C overnight. The following primary anti-
bodies were used as indicated: anti-tyrosine
hydroxylase (AB1542, Sigma-Aldrich), and anti-
cFos (SYS-226-308, SYSY). The slices were washed
three times with blocking solution and then
incubated with secondary antibodies (Rabbit
Anti-Guinea Pig IgG (HL) - Alexa Fluor 594,
M213619S, Abmart; Alexa Fluor 488 donkey
anti-rabbit IgG (H+L), A-21206; Alexa Fluor 647
goat anti-rabbit IgG (H+L), A-21070, Invitrogen)
for 1 hour. Finally, slices were mounted on slides
with DAPI/Antifade solution (S7113, Sigma)
and fluorescence images were captured using a
Leica TCS SP8 STED microscope (Leica, Germany).
Images were analyzed with ImageJ (National
Institutes of Health, Bethesda, MD).
Behavioral tests
The procedures for behavioral tests were as
described previously (62,74). All behavioral
tests were conducted at the early stage of dark
phase (20:00 to 22:00). The animals were trans-
ported into the testing room in a holding cage to
acclimate for at least 1 hour before testing. Their
behaviors were monitored using a video track-
ing system (SMART, Panlab, USA) and later
analyzed with video analysis software. The
testing room was kept at a low light (~20 lux)
to minimize anxiety. To avoid the adaptation
effects and experience-dependent responses,
we used different populations of mice for dif-
ferent behavioral assays.
Social/bedding preference test
The social preference test was performed in a
Plexiglas rectangular box (60 cm by 40 cm by
22 cm) with three interconnected chambers of
equal size as described previously (75). The test
mouse was placed in the center chamber and
allowed to explore the box for 10 min with both
gates to the side chambers closed. This design
helped in minimizing the influence of prior
exposure to the test room, and it also helped in
establishing a consistent environment for all
the behavioral tests. A female stranger mouse
was introduced in a mesh cage in one side
chamber, and a male stranger mouse was
placed in a similar mesh cage in the other side
chamber. When applicable, TMT was placed at
the central region 30 cm above the apparatus
to serve as a survival stressor. The test mouse
was then allowed to freely explore the three
interconnected chambers for a total of 10 min.
Thetimethetestmousespentinteractingwith
each sex (5 cm nearby the mesh cages) was au-
tomatically recorded by using the SMART soft-
ware (Panlab, USA). The social preference index
(male percentage) was calculated by dividing
the sniffing time with the male stranger by the
total sniffing time with both mice, as previously
described (76).
Similar to social preference, in the bedding
preference test, the bedding from female or
male strangers was used instead of the caged
mice in each side of the chamber. The test
mouse was then allowed to freely explore the
three interconnected chambers for 10 min.
The time the test mouse spent sniffing and ap-
proaching the male or female bedding materials
was recorded. The bedding preference index
(male percentage) was calculated by dividing
the sniffing time with the male bedding by
the total sniffing time with the bedding on
both sides.
Cued and contextual fear conditoning test
The cued and contextual test was performed as
described previously (77). The pre-conditioning
was conducted in an experiment chamber (30 cm
by 30 cm by 40 cm) with an electrified grid floor.
Thetestingmousewasplacedinthechamber
and allowed to explore freely for 1 min. Then
the mouse was exposed to a neutral stimulus
(tone, for 2 s), which is followed by a paired
aversive stimulus (electric shock, 0.3 mA for
1 s, twice in 2 min). The experimental chamber
used in training procedure was used for the
contextual FC test to ensure exactly the same
context. A female stranger mouse was put in a
mesh cage in one corner and a male stranger
mouse was caged on the diagonal opposite
side. The fear-conditioned mouse was then
put into the experimental chamber and allowed
toexplorefreelyfor10min.ThecuedFCtestwas
performed in a three-chamber box, in which
femaleandmalestrangermicewereplacedin
thesidecages,andthefear-conditionedmouse
was placed in the middle cage and allowed to
explore the three chambers in free for 10 min.
During the exploration period, the tone was
periodically introduced once per minute. The
time the testing mice spent interacting with
each sex (5 cm nearby the mesh cages) was
automatically recorded by using the SMART
software (Panlab, USA), and the social prefer-
ence index (male percentage) was calculated by
dividing the sniffing time with the male stranger
by the total sniffing time with both mice.
Statistics
All experiments were performed meticulously
with at least three biological repeats and con-
ductedinarandomizedordertoavoidcon-
founding factors. The behavioral tests were
applied and analyzed in a double-blind man-
ner. Appropriate sample sizes were chosen
based on data variability and the literature.
No samples or animals that provided success-
ful measurements were excluded from anal-
ysis. The data were presented as the mean ±
SEM, and statistical comparisons were made
using the two-tailed unpaired Student’sttest,
Kolmogorov-Smirnov test, two-way ANOVA
(followed by Sidak’s multiple comparisons),
or Pearson correlation analysis as indicated.
Data normality was assessed by using the
Shapiro-Wilk test, and the variance equality
was determined with Levene’s test. All statis-
tical analyses were performed using GraphPad
Prism 8.0, and significant differences were ac-
cepted at P< 0.05. Numbers of mice, slices, or
cells analyzed are indicated in the figures.
REFERENCES AND NOTES
1. J. J. Walsh, D. J. Christoffel, R. C. Malenka, Neural circuits
regulating prosocial behaviors. Neuropsychopharmacology 48,
79–89 (2023). doi: 10.1038/s41386-022-01348-8;
pmid: 35701550
2. Z. Guo et al., Neural dynamics in the limbic system during male
social behaviors. Neuron 111, 3288–3306.e4 (2023).
doi: 10.1016/j.neuron.2023.07.011; pmid: 37586365
3. J. T. Kwon et al., An amygdala circuit that suppresses social
engagement. Nature 593,114–118 (2021). doi: 10.1038/
s41586-021-03413-6; pmid: 33790466
4. L. Kingsbury et al., Correlated Neural Activity and Encoding of
Behavior across Brains of Socially Interacting Animals. Cell
178, 429–446.e16 (2019). doi: 10.1016/j.cell.2019.05.022;
pmid: 31230711
5. T. H. Miller et al., Social Status-Dependent Shift in Neural
Circuit Activation Affects Decision Making. J. Neurosci.
37, 2137–2148 (2017). doi: 10.1523/JNEUROSCI.1548-16.2017;
pmid: 28093472
6. L. A. O’Connell, H. A. Hofmann, Evolution of a vertebrate
social decision-making network. Science 336, 1154–1157
(2012). doi: 10.1126/science.1218889; pmid: 22654056
7. M. Liu, D. W. Kim, H. Zeng, D. J. Anderson, Make war
not love: The neural substrate underlying a state-dependent
switch in female social behavior. Neuron 110, 841–856.e6
(2022). doi: 10.1016/j.neuron.2021.12.002; pmid: 34982958
8. C. Zhang et al., Dynamics of a disinhibitory prefrontal
microcircuit in controlling social competition. Neuron 110,
516–531.e6 (2022). doi: 10.1016/j.neuron.2021.10.034;
pmid: 34793692
9. D. Lee, H. Seo, Neural Basis of Strategic Decision
Making. Trends Neurosci. 39,40–48 (2016). doi: 10.1016/
j.tins.2015.11.002; pmid: 26688301
10. J. LeDoux, N. D. Daw, Surviving threats: Neural circuit and
computational implications of a new taxonomy of defensive
behaviour. Nat. Rev. Neurosci. 19, 269–282 (2018).
doi: 10.1038/nrn.2018.22; pmid: 29593300
11. G. K. Adams, K. K. Watson, J. Pearson, M. L. Platt, Neuroethology
of decision-making. Curr. Opin. Neurobiol. 22, 982–989 (2012).
doi: 10.1016/j.conb.2012.07.009; pmid: 22902613
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 12 of 14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
12. I. Carta, C. H. Chen, A. L. Schott, S. Dorizan, K. Khodakhah,
Cerebellar modulation of the reward circuitry and social
behavior. Science 363, eaav0581 (2019). doi: 10.1126/
science.aav0581; pmid: 30655412
13. L. A. Gunaydin et al., Natural neural projection dynamics
underlying social behavior. Cell 157, 1535–1551 (2014).
doi: 10.1016/j.cell.2014.05.017; pmid: 24949967
14. C. Solié, B. Girard, B. Righetti, M. Tapparel, C. Bellone, VTA
dopamine neuron activity encodes social interaction and
promotes reinforcement learning through social prediction
error. Nat. Neurosci. 25,86–97 (2022). doi: 10.1038/s41593-
021-00972-9; pmid: 34857949
15. V. Rappeneau, F. Castillo Díaz, Convergence of oxytocin
and dopamine signalling in neuronal circuits: Insights into
the neurobiology of social interactions across species.
Neurosci. Biobehav. Rev. 161, 105675 (2024). doi: 10.1016/
j.neubiorev.2024.105675; pmid: 38608828
16. W. C. Huang, A. Zucca, J. Levy, D. T. Page, Social Behavior
Is Modulated by Valence-Encoding mPFC-Amygdala
Sub-circuitry. Cell Rep. 32, 107899 (2020). doi: 10.1016/
j.celrep.2020.107899; pmid: 32668253
17. P. Gangopadhyay, M. Chawla, O. Dal Monte, S. W. C. Chang,
Prefrontal-amygdala circuits in social decision-making.
Nat. Neurosci. 24,5–18 (2021). doi: 10.1038/s41593-020-
00738-9; pmid: 33169032
18. J. K. Rilling, A. G. Sanfey, The neuroscience of social decision-
making. Annu. Rev. Psychol. 62,23–48 (2011). doi: 10.1146/
annurev.psych.121208.131647; pmid: 20822437
19. R. K. Hu et al., An amygdala-to-hypothalamus circuit for social
reward. Nat. Neurosci. 24, 831–842 (2021). doi: 10.1038/
s41593-021-00828-2; pmid: 33820999
20. O. Dal Monte, C. C. J. Chu, N. A. Fagan, S. W. C. Chang,
Specialized medial prefrontal-amygdala coordination in
other-regarding decision preference. Nat. Neurosci. 23,
565–574 (2020). doi: 10.1038/s41593-020-0593-y;
pmid: 32094970
21. B. Dai et al., Responses and functions of dopamine in nucleus
accumbens core during social behaviors. Cell Rep. 40, 111246
(2022). doi: 10.1016/j.celrep.2022.111246; pmid: 36001967
22. H. Seo, D. Lee, Neural basis of learning and preference during
social decision-making. Curr. Opin. Neurobiol. 22, 990–995
(2012). doi: 10.1016/j.conb.2012.05.010; pmid: 22704796
23. S. Jang et al., Impact of the circadian nuclear receptor
REV-ERBain dorsal raphe 5-HT neurons on social interaction
behavior, especially social preference. Exp. Mol. Med. 55,
1806–1819 (2023). doi: 10.1038/s12276-023-01052-7;
pmid: 37537215
24. M. M. Rogers-Carter et al., Insular cortex mediates approach
and avoidance responses to social affective stimuli. Nat.
Neurosci. 21, 404–414 (2018). doi: 10.1038/s41593-018-0071-y;
pmid: 29379116
25. A. Nair et al., An approximate line attractor in the hypothalamus
encodes an aggressive state. Cell 186,178–193.e15 (2023).
doi: 10.1016/j.cell.2022.11.027; pmid: 36608653
26. T. Karigo et al., Distinct hypothalamic control of same- and
opposite-sex mounting behaviour in mice. Nature 589,
258–263 (2021). doi: 10.1038/s41586-020-2995-0;
pmid: 33268894
27. X. Jiang, Y. Pan, Neural Control of Action Selection Among
Innate Behaviors. Neurosci. Bull. 38, 1541–1558 (2022).
doi: 10.1007/s12264-022-00886-x; pmid: 35633465
28. S. Fang et al., Sexually dimorphic control of affective
state processing and empathic behaviors. Neuron 112,
1498–1517.e8 (2024). doi: 10.1016/j.neuron.2024.02.001;
pmid: 38430912
29. P. Chen, W. Hong, Neural Circuit Mechanisms of Social
Behavior. Neuron 98,16–30 (2018). doi: 10.1016/
j.neuron.2018.02.026; pmid: 29621486
30. S. W. Newman, The medial extended amygdala in male
reproductive behavior. A node in the mammalian social
behavior network. Ann. N. Y. Acad. Sci. 877, 242–257 (1999).
doi: 10.1111/j.1749-6632.1999.tb09271.x; pmid: 10415653
31. L. Kamas, A. Preston, Can social preferences explain gender
differences in economic behavior? J. Econ. Behav. Organ. 116,
525–539 (2015). doi: 10.1016/j.jebo.2015.05.017
32. E. H. M. Sterck, D. P. Watts, C. P. van Schaik, The Evolution of
Female Social Relationships in Nonhuman Primates. Behav.
Ecol. Sociobiol. 41, 291–309 (1997). doi: 10.1007/
s002650050390
33. Y. Liu et al., Molecular regulation of sexual preference
revealed by genetic studies of 5-HT in the brains of male mice.
Nature 472,95–99 (2011). doi: 10.1038/nature09822;
pmid: 21441904
34. J. Wang et al., Basal forebrain mediates prosocial behavior
via disinhibition of midbrain dopamine neurons. Proc. Natl.
Acad. Sci. U.S.A. 118, e2019295118 (2021). doi: 10.1073/
pnas.2019295118; pmid: 33563763
35. Y. Beny-Shefer et al., Nucleus Accumbens Dopamine Signaling
Regulates Sexual Preference for Females in Male Mice. Cell
Rep. 21, 3079–3088 (2017). doi: 10.1016/j.celrep.2017.11.062;
pmid: 29241537
36. E. S. Bromberg-Martin, M. Matsumoto, O. Hikosaka, Dopamine
in motivational control: Rewarding, aversive, and alerting.
Neuron 68, 815–834 (2010). doi: 10.1016/j.neuron.2010.11.022;
pmid: 21144997
37. F. Sun et al., A Genetically Encoded Fluorescent Sensor
Enables Rapid and Specific Detection of Dopamine in Flies,
Fish, and Mice. Cell 174, 481–496.e19 (2018). doi: 10.1016/
j.cell.2018.06.042; pmid: 30007419
38. J. B. Becker, C. N. Rudick, W. J. Jenkins, The role of dopamine
in the nucleus accumbens and striatum during sexual
behavior in the female rat. J. Neurosci. 21, 3236–3241 (2001).
doi: 10.1523/JNEUROSCI.21-09-03236.2001; pmid: 11312308
39. W. J. Jenkins, J. B. Becker, Dynamic increases in dopamine
during paced copulation in the female rat. Eur. J. Neurosci. 18,
1997–2001 (2003). doi: 10.1046/j.1460-9568.2003.02923.x;
pmid: 14622232
40. J. S. Flannery et al., Developmental Changes in Habenular and
Striatal Social Reinforcement Responsivity Across Adolescence
Linked With Substance Use. Biol. Psychiatry 94, 888–897
(2023). doi: 10.1016/j.biopsych.2023.04.018; pmid: 37120062
41. D. W. Bayless et al., A neural circuit for male sexual behavior
and reward. Cell 186, 3862–3881.e28 (2023). doi: 10.1016/
j.cell.2023.07.021; pmid: 37572660
42. S. M. Miller, J. S. Lonstein, Dopaminergic projections to the
medial preoptic area of postpartum rats. Neuroscience 159,
1384–1396 (2009). doi: 10.1016/j.neuroscience.2009.01.060;
pmid: 19409227
43. S. X. Zhang et al., Hypothalamic dopamine neurons
motivate mating through persistent cAMP signalling. Nature
597, 245–249 (2021). doi: 10.1038/s41586-021-03845-0;
pmid: 34433964
44. J. Kohl et al., Functional circuit architecture underlying
parental behaviour. Nature 556, 326–331 (2018). doi: 10.1038/
s41586-018-0027-0; pmid: 29643503
45. G.-W. Zhang et al., Medial preoptic area antagonistically
mediates stress-induced anxiety and parental behavior.
Nat. Neurosci. 24, 516–528 (2021). doi: 10.1038/s41593-020-
00784-3; pmid: 33526942
46. K. Fukumitsu et al., Amylin-Calcitonin receptor signaling in the
medial preoptic area mediates affiliative social behaviors in
female mice. Nat. Commun. 13, 709 (2022). doi: 10.1038/
s41467-022-28131-z; pmid: 35136064
47. L. Mei, R. Yan, L. Yin, R. M. Sullivan, D. Lin, Antagonistic
circuits mediating infanticide and maternal care in female
mice. Nature 618, 1006–1016 (2023). doi: 10.1038/s41586-
023-06147-9; pmid: 37286598
48. S. C. Motta et al., Dissecting the brain’s fear system reveals
the hypothalamus is critical for responding in subordinate
conspecific intruders. Proc. Natl. Acad. Sci. U.S.A. 106,
4870–4875 (2009). doi: 10.1073/pnas.0900939106;
pmid: 19273843
49. N. F. Parker et al., Reward and choice encoding in terminals
of midbrain dopamine neurons depends on striatal target.
Nat. Neurosci. 19, 845–854 (2016). doi: 10.1038/nn.4287;
pmid: 27110917
50. A. Mohebi et al., Dissociable dopamine dynamics for learning
and motivation. Nature 570,65–70 (2019). doi: 10.1038/
s41586-019-1235-y; pmid: 31118513
51. Q. Shan, Y. Hu, S. Chen, Y. Tian, Nucleus accumbens
dichotomically controls social dominance in male mice.
Neuropsychopharmacology 47, 776–787 (2022). doi: 10.1038/
s41386-021-01220-1; pmid: 34750567
52. G. D. Stuber, A. M. Stamatakis, P. A. Kantak, Considerations
when using cre-driver rodent lines for studying ventral
tegmental area circuitry. Neuron 85, 439–445 (2015).
doi: 10.1016/j.neuron.2014.12.034; pmid: 25611514
53. K. T. Beier et al., Topological Organization of Ventral Tegmental
Area Connectivity Revealed by Viral-Genetic Dissection of
Input-Output Relations. Cell Rep. 26, 159–167.e6 (2019).
doi: 10.1016/j.celrep.2018.12.040; pmid: 30605672
54. K. T. Beier et al., Circuit Architecture of VTA Dopamine Neurons
Revealed by Systematic Input-Output Mapping.Ce ll 162,622–634
(2015). doi: 10.1016/j.cell.2015.07.015;pmid:26232228
55. A. A. Grace, S. B. Floresco, Y. Goto, D. J. Lodge, Regulation of
firing of dopaminergic neurons and control of goal-directed
behaviors. Trends Neurosci. 30, 220–227 (2007). doi: 10.1016/
j.tins.2007.03.003; pmid: 17400299
56. D. Chaudhury et al., Rapid regulation of depression-related
behaviours by control of midbrain dopamine neurons.
Nature 493, 532–536 (2013). doi: 10.1038/nature11713;
pmid: 23235832
57. J. K. Dreyer, K. F. Herrik, R. W. Berg, J. D. Hounsgaard,
Influence of phasic and tonic dopamine release on receptor
activation. J. Neurosci. 30, 14273–14283 (2010). doi: 10.1523/
JNEUROSCI.1894-10.2010; pmid: 20962248
58. J. D. Berke, What does dopamine mean? Nat. Neurosci. 21,
787–793 (2018). doi: 10.1038/s41593-018-0152-y;
pmid: 29760524
59. S. Sarno et al., Dopamine firing plays a dual role in coding
reward prediction errors and signaling motivation in a working
memory task. Proc. Natl. Acad. Sci. U.S.A. 119, e2113311119
(2022). doi: 10.1073/pnas.2113311119; pmid: 34992139
60. Y. Wang, O. Toyoshima, J. Kunimatsu, H. Yamada,
M. Matsumoto, Tonic firing mode of midbrain dopamine
neurons continuously tracks reward values changing
moment-by-moment. eLife 10, e63166 (2021). doi: 10.7554/
eLife.63166; pmid: 33689680
61. A. C. Krok et al., Intrinsic dopamine and acetylcholine
dynamics in the striatum of mice. Nature 621, 543–549
(2023). doi: 10.1038/s41586-023-05995-9; pmid: 37558873
62. Q. Song et al., An ACC-VTA-ACC positive-feedback loop
mediates the persistence of neuropathic pain and emotional
consequences. Nat. Neurosci. 27, 272–285 (2024).
doi: 10.1038/s41593-023-01519-w; pmid: 38172439
63. C. Wang et al., Synaptotagmin-11 is a critical mediator of
parkin-linked neurotoxicity and Parkinson’s disease-like
pathology. Nat. Commun. 9, 81 (2018). doi: 10.1038/s41467-
017-02593-y; pmid: 29311685
64. L. Wang et al., Cocaine induces locomotor sensitization
through a dopamine-dependent VTA-mPFC-FrA cortico-cortical
pathway in male mice. Nat. Commun. 14, 1568 (2023).
doi: 10.1038/s41467-023-37045-3; pmid: 36944634
65. L. Hunger, A. Kumar, R. Schmidt, Abundance Compensates
Kinetics: Similar Effect of Dopamine Signals on D1 and D2
Receptor Populations. J. Neurosci. 40, 2868–2881 (2020).
doi: 10.1523/JNEUROSCI.1951-19.2019; pmid: 32071139
66. J. C. Martel, S. Gatti McArthur, Dopamine Receptor Subtypes,
Physiology and Pharmacology: New Ligands and Concepts
in Schizophrenia. Front. Pharmacol. 11, 1003 (2020).
doi: 10.3389/fphar.2020.01003; pmid: 32765257
67. A. Bedrov, S. L. Gable, Thriving together: The benefits of
women’s social ties for physical, psychological and
relationship health. Philos. Trans. R. Soc. London Ser. B 378,
20210441 (2023). doi: 10.1098/rstb.2021.0441;
pmid: 36440568
68. A. J. Rose, K. D. Rudolph, A review of sex differences in peer
relationship processes: Potential trade-offs for the emotional
and behavioral development of girls and boys. Psychol. Bull.
132,98–131 (2006). doi: 10.1037/0033-2909.132.1.98;
pmid: 16435959
69. S. Ayash et al., Fear circuit-based neurobehavioral signatures
mirror resilience to chronic social stress in mouse. Proc. Natl.
Acad. Sci. U.S.A. 120, e2205576120 (2023). doi: 10.1073/
pnas.2205576120; pmid: 37068238
70. Q. Song et al., Selective Phosphorylation of AMPA Receptor
Contributes to the Network of Long-Term Potentiation in
the Anterior Cingulate Cortex. J. Neurosci. 37, 8534–8548
(2017). doi: 10.1523/JNEUROSCI.0925-17.2017;
pmid: 28765333
71. P. C. Keyes et al., Orchestrating Opiate-Associated Memories
in Thalamic Circuits. Neuron 107, 1113–1123.e4 (2020).
doi: 10.1016/j.neuron.2020.06.028; pmid: 32679036
72. S. Lin et al., Somatostatin-Positive Neurons in the Rostral Zona
Incerta Modulate Innate Fear-Induced Defensive Response in
Mice. Neurosci. Bull. 39, 245–260 (2023). doi: 10.1007/
s12264-022-00958-y; pmid: 36260252
73. X. Zhu et al., Distinct thalamocortical circuits underlie allodynia
induced by tissue injury and by depression-like states. Nat.
Neurosci. 24, 542–553 (2021). doi: 10.1038/s41593-021-
00811-x; pmid: 33686297
74. M. Li et al., Impaired D2 receptor-dependent dopaminergic
transmission in prefrontal cortex of awake mouse model of
Parkinson’s disease. Brain 142, 3099–3115 (2019).
doi: 10.1093/brain/awz243; pmid: 31504219
75. B. Rein, K. Ma, Z. Yan, A standardized social preference
protocol for measuring social deficits in mouse models of
autism. Nat. Protoc. 15, 3464–3477 (2020). doi: 10.1038/
s41596-020-0382-9; pmid: 32895524
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 13 of 14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
76. Z. Dong et al., CUL3 Deficiency Causes Social Deficits and
Anxiety-like Behaviors by Impairing Excitation-Inhibition
Balance through the Promotion of Cap-Dependent
Translation. Neuron 105, 475–490.e6 (2020). doi: 10.1016/
j.neuron.2019.10.035; pmid: 31780330
77. P. Curzon, N. R. Rustay, K. E. Browman, in Methods of Behavior
Analysis in Neuroscience, J. J. Buccafusco, Ed. (CRC Press/
Taylor & Francis, 2009), pp. 19–37.
ACKNOWL EDGME NTS
We thank S. Qiu (Arizona State University, USA), B. Li (Sun Yat-Sen
University, China), and W. Wang (The Fourth Military Medical
University, China) for reading the manuscript; Y. Hao (Core
Facilities Sharing Platform, Xi’an Jiaotong University, China) for
assistance with confocal imaging; and E. Wang (High School
Affiliated to Xi’an Jiaotong University, China) for artwork. Funding:
This work was supported by the National Natural Science
Foundation of China (32171233 and 31670843 to C.W., 81901308
to H.X., 81974203 to X.K., 32300819 to Y.C., 82030007 to
C. Zhang, 32400650 to R.H., and 32000704 to Q.S.), the
Natural Science Foundation of Shaanxi Province of China (2023-
ZDLSF-23, 2021TD-37, and 2019JC-07 to C.W.; 2020JQ-029
to Q.S.; 2023-JC-QN-0236 to N.D.; 2024JC-YBMS-146 to H.X.;
JC-YBQN-0172 to C. Zheng; and 2024JC-YBMS-141 and
2023SYJ09 to R.H.), the Sichuan Science and Technology Program
(2022YFS0615 and 2024ZYD0077 to X.K.), the China Postdoctoral
Science Foundation (2018M640972 to Q.S.; 2022M712543 to N.D.;
and 2024M752559, 2024T170724, and GZC20232111 to Y.C.),
the Shaanxi Postdoc Funding (2023BSHTBZZ15 to H.X.,
2023BSHYDZZ39 to Y.C., and 2023BSHEDZZ67 to C.Zhe), and
the Luzhou Science and technology Program (2024LZXNYDJ002
to X.K.). Author contributions: C.W., X.K., H.X., C.Zha., and
A.W. conceived the study and designed the experiments with help
of Q.Q. A.W., A.Z., C.Zhe., N.D., X.C., X.D., S.Z., X.L., J.J., Y.Q.,
Y.Y., Y.G., Bi.W., J.H., J.Y., W.L., K.H., Y.L., F.M., R.W., M.S.,
Bo.W., Y.Z., Y.C., Q.S., and R.H. performed the experiments and
analyses. C.W., H.X., X.K., C.Zha., A.W., and N.G. wrote the
manuscript. All authors reviewed the manuscript and approved
its submission. Competing interests: The authors declare
that they have no competing interests. There is no consultation,
paid or unpaid, and related patent utilized or applied in or based
on this work. Data and materials availability: The data that
support the findings of this study are available in the main
manuscript or the supplementary materials. Source data are
provided with this paper. License information: Copyright ©
2025 the authors, some rights reserved; exclusive licensee
American Association for the Advancement of Science. No c laim
to original US government works. https://www.science.org/
about/science-licenses-journal-article-reuse
SUPPLEMENTARY MATERIALS
science.org/doi/10.1126/science.adq7001
Figs. S1 to S12
MDAR Reproducibility Checklist
Submitted 27 May 2024; accepted 8 November 2024
10.1126/science.adq7001
RESEARCH |RESEARCH ARTICLE
Wei et al., Science 387, eadq7001 (2025) 10 January 2025 14 of 14
Downloaded from https://www.science.org at Xi'an Jiaotong University on January 09, 2025
