Skip to main content
NCBI home page
Journal of Advanced Research logoLink to Journal of Advanced Research
. 2025 Dec 11;87:279–291. doi: 10.1016/j.jare.2025.12.014

Essential function of Nr4a2 in subicular development and social novelty

Xi-Yue Liu a,1, Yun-Chao Tao a,1, Qiong Zhang b,1, Wei-Tang Liu b,c, Li Zhao a, Zhi-Bin Hu b, Bing-Yao Zhou a, Ze-Xuan Li b, Cong-Cong Qi b, Lei Zhang d,e, Jia-Yin Chen b, Ning-Ning Song b,⁎, Yu-Qiang Ding a,b,c,f,⁎⁎
PMCID: PMC13539151  PMID: 41390117

Graphical abstract

graphic file with name ga1.webp

Open in a new tab

Keywords: Nr4a2, Subiculum proper, Retrosplenial cortex, Social novelty

Highlights

  • •

    Nr4a2 is required for the specification and maintenance of subicular neurons.

  • •

    Nr4a2 regulates subicular connectivity with retrosplenial cortex.

  • •

    Loss of Nr4a2 impairs social novelty in mice.

Abstract

Introduction

The subicular regions are a major output hub of the hippocampus. Recent studies have shown that subicular neurons are highly diverse in their gene expression patterns, neural circuitry, and electrophysiological properties. However, the developmental transcriptional program underlying their specification remains poorly understood.

Objectives

To delineate how the orphan nuclear receptor Nr4a2 orchestrates both the developmental specification and functional maintenance of subicular neurons, and to uncover the underlying molecular mechanisms.

Methods

Nr4a2 conditional knockout (CKO) mice and adult subicular-specific AAV-Nr4a2 CKO mice were comprehensively profiled using RNA-seq, CUT&Tag, immunohistochemistry, slice electrophysiology, and behavioural assays to elucidate the developmental and functional sequelae of Nr4a2 ablation.

Results

In this study, using conditional genetic models targeting the orphan nuclear receptor Nr4a2, we demonstrate that Nr4a2 is essential for both the acquisition and maintenance of the molecular identity of subiculum proper (Sub) neurons, as well as for establishing its connectivity with the retrosplenial cortex (Rsp) during embryonic development and adulthood. Loss of Nr4a2 in subicular regions results in deficits in social novelty in mice. Furthermore, we show that Nr4a2 can directly bind to the promoters of Sub-enriched genes and promote their transcription in vivo.

Conclusion

Together, these findings reveal a pivotal role for Nr4a2 in establishing the genetic identity and connectivity of Sub neurons and suggest that its dysfunction may contribute to hippocampus-related neurodevelopmental disorders.

Introduction

The subicular regions serve as the principal output areas of the hippocampal formation [[1], [2], [3], [4]], and comprise the subiculum proper (Sub) and the prosubiculum (PS) [4,5]. The Sub and PS occupy distinct anatomical locations within the subicular regions, characterized by precise molecular boundaries along the dorsoventral axisv [6]. Recent studies have revealed that subicular neurons exhibit high heterogeneity in transcriptomic taxonomy, circuit connections, and firing patterns [7,8]. Due to differences in axonal guidance molecules, Sub and PS establish connections with distinct brain regions [6]. Sub receives afferents from distal CA1 and projects to the Rsp, mammillary body and other regions [9,10]. On the other hand, PS receives afferents from proximal CA1 and projects to regions such as the lateral septal nucleus and other specific nuclei [11,12]. The high heterogeneity of Sub and PS determines their diverse functions in various behaviors [6,8,13]. Notably, a novel subicular neuron type has been identified by scRNA-Seq, which is located in the polymorphic layer (po) of both Sub and PS [6]. This subclass displays a distinctive morphology and unique firing properties, differing from the pyramidal neurons in Sub and PS [14]. To date, research on hippocampal development has primarily focused on the CA1-4 regions [[15], [16], [17]]. However, the molecular mechanisms regulating the development of subicular regions, particularly in the specification of their subclasses, remain largely unclear.

Nr4a2 (also known as Nurr1), a member of the nuclear orphan receptor Nr4a family [18], is highly and specifically expressed in the subicular regions, as we reported previously [19]. Furthermore, we demonstrated that co-electroporation of Nr4a2 and Ctip2 into the embryonic cortex can partially reprogram cortical neurons toward a Sub-like neuronal identity [19]. Given that Nr4a2 plays critical roles in neuronal development, including in dopaminergic neurons and claustral neurons [[20], [21], [22], [23]], as well as cortical layer 6b neurons and other neuronal types [[24], [25], [26]], more evidence should be identified on whether Nr4a2 is essential in specifying neuron fate or maintaining functions within the subicular regions.

To investigate the roles of Nr4a2 in the subicular regions, we deleted Nr4a2 from embryonic stages by generating Nr4a2 conditional knockout mice (Nr4a2 CKO) or in adulthood using AAV-mediated targeting (AAV-Nr4a2 CKO). Our findings demonstrate that Nr4a2 is a critical transcription factor selectively required for establishing the molecular identity and circuit integration of Sub-neurons. Embryonic deletion of Nr4a2 impairs the integrity of the Sub-Rsp projection pathway and markedly diminishes social novelty. Interestingly, the maintenance of the molecular identity in the mature brain also depends on Nr4a2, and defective social novelty is also present in AAV-Nr4a2 CKO mice. These results highlight the essential role of developmentally regulated by Nr4a2 in specifying Sub-neuron subtypes and shaping their circuit-specific functional outputs, offering novel insights into the developmental mechanisms underlying hippocampal circuit formation and its relevance to social behaviors.

Materials and methods

Animals

Nr4a2flox/flox mice were generated as previously described [24]. To conditionally knock out the Nr4a2 gene in the forebrain, Nr4a2flox/flox mice were crossed with Emx1-Cre mice [43], and Nr4a2 CKO (Emx1-Cre; Nr4a2flox/flox) mice were obtained. Other genotypes (i.e., Nr4a2flox/flox and Nr4a2flox/+) from the littermates were used as control mice. In addition, to achieve subicular regions-specific knockout in the adult mice, we injected AAV-Cre virus into the subicular regions of Nr4a2flox/flox mice, generating AAV-Nr4a2 CKO mice. The morning on which vaginal plugs were observed was considered embryonic day 0.5 (E0.5), and the day on which pups were born was designated as postnatal day 0 (P0). Mice older than 8 weeks were considered adults. Animal care practices and all experiments were reviewed and approved by the Laboratory Animal Ethics Committee of Fudan University, China.

Stereotactic injections and viruses

Mice were anesthetized with 1.5 %-2.5 % isoflurane and secured in a stereotaxic frame. For subicular injections, the virus was injected bilaterally at AP −3.30 mm, ML ± 1.75 mm, DV −1.80 mm. Depending on the experiment, we used AAV2/9-hSyn-Cre-EGFP-WPRE (AAV-Cre), AAV2/9-hSyn-EGFP-WPRE (AAV-EGFP),AAV2/9-hSyn-mCherry-WPRE (AAV-mCherry), AAV2/9-hSyn-hChR2(H134R)-mCherry-WPRE-pA, and AAV2/9-hSyn-DIO-hChR2(H134R)-mCherry-WPRE-pA. For Rsp injections, AAV2/R-hSyn-EGFP-WPRE (AAV-Retro-EGFP) was infused unilaterally at AP −3.30 mm, ML + 0.20 mm, DV −1.00 mm. All AAV viruses were purchased from BrainVTA. Virus-injected mice were used for immunohistochemical staining and behavioral tests 4 weeks after surgery.

Immunohistochemistry and in situ hybridization

Immunohistochemistry and BrdU labeling were performed as previously described [24]. The following primary antibodies were used: mouse anti-Nr4a2 (1:500; Abcam ab41917, RRID: AB_776887), rabbit anti-Nr4a2 (1:500; Proteintech 10975–2-AP, RRID: AB_2153760), rat anti-Ctip2 (1:500; Abcam ab18465, RRID: AB_2064130), mouse anti-Tle4 (1:300; Santa Cruz sc-365406, RRID: AB_10841582), rabbit anti-mCherry (1:800; Oasis Biofarm OB-PRB013, RRID: AB_2934227), Guinea pig anti-EGFP (1:500; Oasis Biofarm OB-PGP007, RRID: AB_2934228), Mouse anti-BrdU (1:1000; Oasis Biofarm OB-MMS018, RRID: AB_2938682), Rabbit anti-Gfap (1:500; Oasis Biofarm OB-PRB005, RRID: AB_2938935), Rabbit anti-Iba1 (1:500; Oasis Biofarm OB-PRB029, RRID: AB_2934252), and Rabbit anti-Olig2 (1:500; Oasis Biofarm OB-PRB009, RRID: AB_2934240).

In situ hybridization (ISH) used digoxigenin UTP-labeled probes, performed as previously described [24]. Cbln2, Epha5, Epha6, Fn1, Gad65, Gad67, Gfra1, Lxn, Nnat, Nr4a2, Ntng2, Nts, Rxfp1, S100a10, Slc17a6, Slc17a7 and Trib2 mRNA probes were constructed according to the Allen Brain Atlas.

Bulk RNA-seq

The subicular regions of P7 Nr4a2 CKO and littermate control mice were dissected (N = 5). RNA was extracted with TRIzol reagent and assessed for integrity using the Agilent Bioanalyzer. Libraries were prepared with the NEBNext Ultra RNA Library Prep Kit and sequenced on an Illumina platform (150 bp paired-end). Reads were processed with Cutadapt, aligned to the mouse mm39 genome using HISAT2, and quantified with featureCounts. Differential expression was analyzed using DESeq2.

CUT&Tag-seq

The CUT&Tag assay was performed using P7 subicular regions dissected from C57 mice. CUT&Tag was performed as previously described [44] with minor modifications.

Cells were incubated with Concanavalin A-coated beads and antibodies, followed by tagmentation with a pA-Tn5 transposome. DNA was extracted and libraries were prepared. Sequencing was performed on an Illumina NovaSeq 6000 platform. Reads were aligned to the mouse genome (mm39), and peaks were called and visualized using MACS2 and IGV.

Dual-luciferase reporter assay

Promoter fragments containing putative Nr4a2-binding sites were cloned into the pGL4.10 firefly luciferase reporter vector. HEK293T cells (5 × 105 per well) were transfected in 48-well plates using Lipo8000. Each well was co-transfected with 120 ng of pGL4-promoter construct (or empty vector), 5 ng of Renilla luciferase plasmid, and 120 ng of Nr4a2 overexpression plasmid (or empty vector). Luciferase activity was measured 36 h post-transfection using the Dual-Luciferase Reporter Assay System (Promega). Firefly luciferase signals were normalized to Renilla luciferase signals. Each condition was tested in quadruplicate wells, and all experiments were independently repeated at least three times.

Promoter regions of Fn1, Nts, and Slc17a6 were amplified by PCR and cloned into the pGL4.10 vector. Primers used:

Fn1-F: GGGACTGAAAGAGAAAACTTTCT.

Fn1-R: AGAGGCTTATATGGGACAGTAC.

Nts-F: AGAGAATCTGTTTGGTCTTCAGT.

Nts-R: GGCTTTCAAAGTGAGTGAACTGT.

Slc17a6-F: AGTCAACTGGGATGGAGCTTAAG.

Slc17a6-R: TAGGGCTCTTCAGGAACGGA.

Behavioral tests

Adult male mice were tested during the light phase in a sound-attenuated room. They were habituated for 30 min before each assay, provided at least one day of rest between assays, and all experiments were conducted in a blinded manner relative to the group. Behavior was video tracked using EthoVision XT 17 (Noldus) unless otherwise noted.

Three-chamber test

The apparatus is a rectangular plexiglass box (90 cm × 50 cm × 30 cm) divided into three equal chambers. After a 10-minute habituation, mice were tested in two consecutive 10-minute sessions. Session 1 (sociability): a stranger mouse (S1) was placed in one wired cage, and the other cage remained empty. Session 2 (social novelty): the familiar mouse S1 stayed in its cage, while a novel stranger (S2) was introduced into the previously empty cage, allowing the subject to choose between the familiar and the novel conspecific. Close interaction time was recorded when test mice sniffed within 2 cm of the cages. Time spent in each chamber and close interaction were quantified. The social preference index was defined as (S1 − E) / (S1 + E), and the social novelty preference index was calculated as (S2 − S1) / (S1 + S2), where S1 represents the time spent close to Stranger 1, S2 to Stranger 2, and E to the empty cage.

Novel object recognition test

The Novel object recognition test was conducted over two days. On day 1, mice were habituated to the empty testing arena for 30 min. On day 2, two identical objects were placed in the testing arena. Mice were then allowed to explore these objects for 10 min (training phase). One hour later, one familiar object was replaced with a novel one, and mice were tested for 5 min (test phase). Exploration time was recorded when mice interacted with the objects, including touching or sniffing within 2  cm of them. The discrimination index (DI) was calculated as (T novel − T familiar) / (T novel + T familiar), where T novel and T familiar represent the exploration times for the novel and familiar objects, respectively.

Open field test

The open field test was performed in a 40 cm × 40 cm arena enclosed by 30 cm-high plexiglass walls. Mice were allowed to freely explore the arena for 60 min, during which total traveling distance and ambulatory time were quantified using the SuperFlex system.

Elevated plus maze

The elevated plus maze comprised two open arms (30 cm × 5 cm), two closed arms (30 cm × 5 cm × 15 cm), and a central square (5 cm × 5 cm), all positioned 40 cm above the floor. Each mouse was placed in the central area facing an open arm at the beginning of the trial and monitored for 5 min. The duration spent in the open arms and the number of open-arm entries were measured. An entry was defined as the mouse fully entering an arm with all four paws.

Light-dark box test

The light–dark box consisted of a 50 cm × 30 cm × 30 cm chamber divided into a covered dark compartment (one-third of the area) and an uncovered light compartment (two-thirds of the area), connected by a 5 cm × 5 cm floor-level opening. At the start of each session, a mouse was placed in the center of the dark compartment and allowed to explore for 5 min. The time spent in the light compartment and the number of transitions from dark to light were recorded.

Tail suspension test

Mice were acclimated in the testing room for 30 min. A piece of tape was attached to the distal 1/4 of the tail to suspend the mouse from the apparatus, positioning it at the center of the observation zone. Immobility time, an index of depression-like behavior, was recorded over 5 min. The analysis area was defined to include the mouse’s movement without hand interference.

Forced swim test

Mice were acclimated in the testing room for 30 min before being individually placed in a 25  cm-high cylinder filled with 15  cm of water at 23–25 °C. The test lasted 6 min, with the first 1 min for habituation and the last 5 min for recording immobility, an indicator of depression-like behavior. After the test, mice were dried, kept warm, and fresh water was used between trials to avoid interference.

Patch-clamp recordings

Acute coronal brain slices (300 µm) were prepared using an N-methyl-D-glucamine (NMDG)-based protective recovery protocol. Slices recovered for 10–12 min in oxygenated NMDG solution at 34 °C, and were then transferred to oxygenated ACSF at room temperature for 1 h before recordings.

Excitatory synaptic currents were recorded under voltage-clamp mode at a holding potential of −70 mV. Miniature excitatory postsynaptic currents (mEPSCs) were recorded from layer II/III pyramidal neurons of the Rsp using K-gluconate filled electrodes in the presence of TTX (1 µM) and picrotoxin (50 µM) to block action potentials and inhibitory transmission. Optogenetically-evoked EPSCs were elicited by blue-light pulses (473 nm, 2 ms, 2–5 mW/mm2) generated by a laser system (ThinkerTech, China) and delivered through an optical fiber. Light output and timing were controlled with pClamp 10 (Molecular Devices, USA) via a Digidata 1550B digitizer (Axon, USA). Optical stimulation targeted axon terminals of subicular neurons projecting to the Rsp, and evoked EPSCs were recorded from Rsp layer II/III pyramidal neurons. Signals were amplified with a MultiClamp 700B amplifier (Axon, USA), digitized at 10 kHz, and low pass filtered at 1 kHz. Offline analysis of mEPSCs and evoked EPSCs was performed in Clampfit 10, and statistical analyses and plotting were carried out in GraphPad Prism 10.

For intrinsic excitability assessment, neurons were recorded in current-clamp mode. A series of depolarizing current steps (−100 pA to 500 pA, 20 pA increments, 1000 ms duration) was applied to evaluate firing properties. Input resistance was measured immediately after whole-cell access. To determine the action potential threshold, a 1000-ms ramp current (−100 pA to 500 pA) was injected, and the threshold was defined as the membrane potential at which dV/dt first exceeded 10 V/s. Data were acquired and processed using pClamp 10 and Clampfit 10, and quantitative analysis was performed in GraphPad Prism 10.

Quantification and statistical analysis

Experimental data in this study were obtained from at least three independent biological replicates. Mouse brains were sectioned into 25  μm slices, with five sets of sections prepared per group. For cell counting, at least three representative sections of the subicular region were selected in each mouse, based on the adult mouse brain atlas by George Paxinos and Keith B. J. Franklin. Specifically, sections containing the dorsal subicular region were taken at the levels between Bregma −3.10  mm and −3.80  mm, and those containing the ventral subicular regions were selected between Bregma −3.90  mm and −4.30  mm. For quantification of axonal fluorescence intensity in the Rsp and retrogradely labeled cell numbers in the subicular regions, brain sections exhibiting fluorescence within the anatomical range of Bregma −3.20 mm to −3.50  mm were selected. The mean cell count across these sections was used as the statistical value for each animal. N denotes the number of mice, and n indicates the number of neurons analyzed.

All statistical analyses were performed in GraphPad Prism 10. Data normality was assessed with the Shapiro-Wilk test. Comparisons between two independent groups were conducted with a two-tailed unpaired t-test when the data were normally distributed and with the Mann-Whitney U test when they were not. For paired samples, normally distributed differences were analyzed with a two-tailed paired Student’s t-test; otherwise, the Wilcoxon signed-rank test was used. For the two-way comparison between the two groups, a two-way repeated measures ANOVA was employed. Data are shown as mean ± SEM. Statistical significance is indicated by *p < 0.05, **p < 0.01, and ***p < 0.001. Detailed statistical methods and results for each experiment are provided in Supplementary Table 1.

All experiments were performed under double-blind conditions. During behavioral testing, virus injection, immunohistochemistry, and electrophysiological recording, both the experimenters and data analysts were blinded to the genotype and treatment group of each animal. Group identities were revealed only after data collection and statistical analysis were completed.

Results

Nr4a2 is specifically required for the establishment of Sub-neurons' molecular identity

In our previous study, Nr4a2 was found to be highly expressed in the subicular regions, and misexpression of Nr4a2 and Ctip2 induces cortical neurons to ectopically express specific genes of Sub-neuron [19]. To investigate the role of Nr4a2 in the development of subicular regions, we generated Nr4a2 CKO (Emx1-Cre; Nr4a2flox/flox) mice (Fig. 1A, B). No apparent alterations in cellular architecture were observed in the subicular regions of Nr4a2 CKO mice compared with control mice, as shown by the data from NeuN immunostaining and Nissl staining (Fig. 1B, C) as well as quantitative counts of neuron numbers (Fig. 1C). In addition, Ctip2, which colocalized largely with Nr4a2 in the subicular regions, was not obviously altered in Nr4a2 CKO mice relative to control mice (Fig. 1D).

Fig. 1.

Fig. 1

Open in a new tab

Nr4a2 is selectively required for the molecular identity of Sub-neurons. (A) Schematic of Nr4a2 conditional-knockout (CKO) mouse line. (B) Colocalization of Nr4a2 (red) and NeuN (green) in the subicular neurons of Nr4a2 CKO and control mice at P7. (C) Nissl staining shows no alterations in cellular architecture or cell number in the subicular regions of Nr4a2 CKO mice relative to control mice at adulthood (N = 3). (D) Double immunostaining for Nr4a2 (red) and Ctip2 (green) shows the deletion of Nr4a2, mutual colocalization of Nr4a2 and Ctip2, and numbers of Ctip2+ neurons are not altered in the subicular regions of Nr4a2 CKO mice compared with control mice at P7 (N = 3). (E) Coronal sections are dissected for bulk RNA-seq (up). Volcano plot (down) showing the Sub-enriched genes that are significantly downregulated in Nr4a2 CKO mice at P7. (N = 5 for each; p < 0.05, |log2FC| > 1). (F) Expression of Fn1 and Nts is significantly downregulated to an undetectable extent in the Sub-neurons of Nr4a2 CKO mice at adulthood, as shown by in situ hybridization and quantification (N = 3). (G) Expression of PS-enriched genes Ntng2 and Nnat is not significantly altered in Nr4a2 CKO mice compared with control mice at adulthood (N = 3). (H) Expression of po-enriched genes Cbln2 and Tle4 is not significantly affected either in Nr4a2 CKO mice compared with control mice at adulthood (N = 3). N, number of mice. Statistical significance is indicated by *p < 0.05, **p < 0.01 and ***p < 0.001. N, number of mice; n, number of sections per mouse. See Materials and Methods, as well as Supplementary Table 1 for detailed statistical analysis. Sub, subiculum proper; PS, prosubiculum; po, polymorphic layer.

We next dissociated the subicular regions of P7 Nr4a2 CKO and control mice and performed RNA sequencing. A total of 131 downregulated and 145 upregulated genes were found in the Nr4a2 CKO group at P7 (Fig. 1E). Based on anatomical location and distinct transcriptomic profile, the subicular neurons are subdivided into three major clusters: Sub-, PS- and po-neurons [6]. Interestingly, our bulk RNA-seq data showed that those genes specifically expressed in Sub-neurons, including Fn1, Nts, Gfra1, Rxfp1 and Trib2 were significantly downregulated, while PS-enriched genes (Nnat, Ntng2) and po-enriched genes (Cbln2, Tle4) remained unchanged in Nr4a2 CKO mice compared with control mice (Fig. 1E). These alterations were confirmed through in situ hybridization or immunostaining at the cellular level in adult mice. That is, signals for Fn1, Nts, Gfra1, Trib2, Lxn and Rxfp1 were hardly detected in the Sub region, while those for Nnat, Ntng2 and S100a10 in the PS region and Cbln2 and Tle4 in the po-region were unchanged in Nr4a2 CKO mice compared with control mice (Fig. 1F-H, Fig. 3C and Supplementary Fig. 1A). Furthermore, the selective changes in gene expression were also observed in the ventral subicular regions, as shown by remarkable reduction of Nts in the Sub-neurons, and unchanged expression of Ntng2 in the PS-neurons and Tle4 in the po-neurons (Supplementary Fig. 1B). Taken together, these findings demonstrated that the deletion of Nr4a2 results in a selective impairment of gene expression in the Sub-neurons with no effects on the PS- and po-neurons.

Fig. 3.

Fig. 3

Open in a new tab

Nr4a2 is required for the initiation of Sub-gene expression. (A) Expression of Fn1 is initiated in the Sub of control mice at E18.5 but is undetectable in Nr4a2 CKO mice at the same stage. Quantification is shown to the right of each panel (N = 3). (B) Expression of Gfra1 begins in the Sub of control mice at P0 but is absent in Nr4a2 CKO mice at the corresponding age. Quantification is shown to the right of each panel (N = 3). (C) Rxfp1 is expressed in the Sub of control mice at P7, but very weak if any is observed in Nr4a2 CKO at this stage. Quantification is shown to the right of each panel (N = 3). (D) At E17.5, Nts (purple) expression is confined to the po-layer and shows no detectable difference between control and Nr4a2 CKO mice. Tle4 (brown) marks the po layer. (E) Nts expression in the po-layer is not altered, but is dramatically reduced in Sub-neurons of Nr4a2 CKO mice compared with control mice at P7. Quantification showing a significant reduction of Nts+ neurons in the Sub, with no significant change in the po-layer in Nr4a2 CKO mice at P7 (N = 3). Tle4 (brown) marks the po layer. N, number of mice. Statistical significance is indicated by ***p < 0.001. See Materials and Methods, as well as Supplementary Table 1 for detailed statistical analysis. po, polymorphic layer; Sub, subiculum proper.

Nr4a2 is expressed in the three types of subicular neurons

Selective regulation of the development of the subicular regions by Nr4a2 prompted us to explore the neuroanatomical characteristics of Nr4a2 in Sub-, PS- and po-neurons. We performed coimmunostaining of Nr4a2 with specific subicular subclass markers, Fn1 and Nts (Sub-neurons), Ntng2 and Nnat (PS-neurons), and Cbln2 and Tle4 (po-neurons) (Fig. 2A-F). In the subicular regions, approximately 50 % of Nr4a2+ cells co-expressed Fn1, 20 % of Nr4a2+ cells co-expressed Nnat, and 30 % of Nr4a2+ cells co-expressed Tle4 (Fig. 2B, D, F). From another perspective, almost all Fn1+ Sub-neurons and Nnat+ PS-neurons expressed Nr4a2, and about 70 % of Tle4+ po-neurons were Nr4a2 positive. In the ventral subicular regions, Nr4a2 was likewise expressed in three distinct neuronal populations (Supplementary Fig. 2H-J). Besides, Nr4a2 did not colocalize with glial cell markers, including Gfap, Iba1 or Olig2 in the subicular regions (Supplementary Fig. 2A-C). Next, co-labeling of Nr4a2 with excitatory neuron markers (Slc17a7 and Slc17a6) and inhibitory neuron markers (Gad65 and Gad67) were performed. Almost all of the Nr4a2+ neurons expressed Slc17a7 and Slc17a6, and none of them overlapped with Gad65 and Gad67 (Supplementary Fig. 2D-G). Thus, Nr4a2 is expressed in the three types of subicular neurons with excitatory identity.

Fig. 2.

Fig. 2

Open in a new tab

Nr4a2 is expressed in the three types of subicular neurons. (A, B) Immunostaining of Nr4a2 (brown) combined with in situ hybridization of Sub-enriched genes Nts (purple, A) and Fn1 (purple, B) and quantification of neurons expressing both Fn1 and Nr4a2 at P7 (N = 3). (C, D) Immunostaining of Nr4a2 (brown) combined with in situ hybridization of PS-enriched genes Ntng2 (purple, C) and Nnat (purple, D) and quantification of neurons expressing both Nnat and Nr4a2 at P7 (N = 3). (E, F) Immunostaining of Nr4a2 (brown, E) combined with in situ hybridization of po-enriched gene Cbln2 (purple, E), in situ hybridization of Nr4a2 (purple, F) with immunostaining of po-enriched gene Tle4 (brown, F), and quantification of neurons expressing both Tle4 and Nr4a2 at P7 (N = 3). (G) In situ hybridization shows that the initiation of Nr4a2 expression in the subicular regions at E14.5 and E15.5. (H) Immunofluorescence staining of Nr4a2 (red) and BrdU (green) showing that Nr4a2 is expressed in postmitotic cells at E16.5. (I) Persistent Nr4a2 expression in the subicular regions at P0 and P60. (J, K) Colocalization of BrdU (green) and Nr4a2 (red) in the subicular regions of P7 mice with a single pulse of BrdU injection in pregnant mice at E12.5, E13.5, E14.5, or E15.5 (K). The quantification of the proportion of Nr4a2+ neurons co-labeled with BrdU (Nr4a2+BrdU+/Nr4a2+) (J) (N = 3). N, number of mice. CA1, cornu ammonis 1; DG, dentate gyrus; Hb, habenula; Hip, hippocampus; LV, lateral ventricle; Ncx, neocortex; po, polymorphic layer; PS, prosubiculum; Rsp, retrosplenial cortex; Sub, subiculum proper; SubC, subicular region.

In addition, we examined the expression of Nr4a2 during the embryonic stages. Nr4a2 mRNA was initially detected in the subicular regions at E15.5 (Fig. 2G), followed by the appearance of Nr4a2 protein at E16.5 (Fig. 2H). Nr4a2 was expressed in postmitotic neurons, as evidenced by the absence of Nr4a2+ cells co-labeled with BrdU at E16.5 (Fig. 2H). This expression persisted from the early postnatal period through adulthood (Fig. 2I). Investigation into the timing of neurogenesis in Nr4a2+ subicular neurons involved a single BrdU injection from E12.5 to E15.5, with mice sacrificed at P7. The proportion of Nr4a2+/BrdU+ neurons gradually increased from E12.5, peaking at E14.5 and then sharply declining at E15.5 (Fig. 2J, K), indicating peak generation of Nr4a2+ subicular neurons at E14.5. In summary, our results demonstrated persistent Nr4a2 expression in the three major neuronal subtypes of the subicular regions from embryonic stages to adulthood.

Nr4a2 deletion disrupts the initiation of Sub-specific genes

Having found the loss of Sub-specific genes in Sub-neurons in the postnatal brain, we set out to explore if Nr4a2 is required for their initiation during the developmental stages. Fn1 expression was first observed at E18.5 in the Sub of control embryos, while no Fn1 expression was detected in Nr4a2 CKO embryos (Fig. 3A). Gfra1 exhibited initiation of expression at P0 in control mice but was absent in Nr4a2 CKO mice (Fig. 3B). In contrast, Rxfp1, which showed robust expression at P7 in the Sub of control mice, was significantly diminished in Nr4a2 CKO mice (Fig. 3C). Both the representative images and statistical analyses consistently demonstrate that the initiation of Sub-specific gene expression is significantly impaired in the absence of Nr4a2.

Nts, which also served as a marker for Sub-neurons, displayed a dynamic expression pattern during the subicular development. Nts was predominantly expressed in po-neurons by E17.5, as indicated by coexpression of Nts with Tle4, a marker of po-neurons (Fig. 3D), while its expression expanded into the Sub region by P7 (Fig. 3E). In Nr4a2 CKO mice, Nts expression in the po-layer remained unaffected at these time points (Fig. 3D, E), but its expression in the Sub region was detected to be reduced at P7 (Fig. 3E) and undetectable in adulthood (Fig. 1F). These results indicated that Nr4a2 is required for the initiation of Sub-neuron-specific genes during embryonic and postnatal development.

Nr4a2 CKO mice show impaired Sub-Rsp circuit and social novelty deficits

To investigate the impact of Nr4a2 loss on subicular output, we injected anterograde virus (AAV-mCherry) into the subicular region to label axonal targets in Nr4a2 CKO mice. Strong terminal labeling was observed in layers II/III of the control mice Rsp (Fig. 4A), the primary target of the subicular regions [6]. In contrast, a drastic reduction in mCherry-labeled axons was noted in the Rsp of Nr4a2 CKO mice (Fig. 4A), along with a decrease in labeled axons along the Sub-Rsp pathway (Fig. 4B). It should be noted that the terminal labeling in the main downstream targets of the other two subicular subdivisions [6] were unchanged in Nr4a2 CKO mice: projections from the PS to the lateral septum was comparable to control mice, as did those from the po neurons to the anterior ventral and anterior medial thalamic nuclei (Supplementary Fig. 3A, B). Subsequent injection of retrograde virus (AAV-retro-EGFP) into the Rsp revealed a restricted distribution of many EGFP-labeled neurons in the Sub of control mice (Fig. 4C), most of which were Nr4a2+ (Fig. 4D). In marked contrast, Nr4a2 CKO mice exhibited few EGFP-labeled neurons in the Sub (Fig. 4C, D). Through RNA-seq, some key genes involved in axonal guidance such as Fn1, Robo3, Ephb2, Epha3 and Epha5 [[27], [28], [29], [30], [31]] showed significant downregulation (Fig. 4E), potentially contributing to axonal connection defects in Nr4a2 CKO mice. We further examined the expression of axonal guidance genes and found that in addition to Fn1, Epha5 was also highly and specifically enriched in the Sub, and both of them were dramatically reduced in the Sub of Nr4a2 CKO mice (Fig. 1F and Fig. 4F). In contrast, Ntng2 and Epha6, which are expressed in the PS, showed no significant change in expression (Fig. 1G and Supplementary Fig. 3C). Consistent with the loss of molecular identity of Sub-neurons, its efferent connection to the Rsp is selectively defective in the absence of Nr4a2.

Fig. 4.

Fig. 4

Open in a new tab

Nr4a2 CKO mice exhibit a disrupted Sub-Rsp circuit accompanied by deficient social novelty. (A) AAV-mCherry (virus) is stereotaxically injected into the subicular regions, and it results in many mCherry-labeled axons in the Rsp of control mice. Still, few axons are observed in the Rsp of Nr4a2 CKO mice. Insets show magnified views of mCherry-labeled axons in the Rsp. (B) Axons along the pathway from the Sub to the Rsp are also greatly reduced in Nr4a2 CKO mice compared with control mice (N = 3). (C) Injection of AAV-retro-EGFP (virus) into the Rsp to retrogradely label projecting neurons in the Sub. Insets show many retrogradely-labeled (EGFP) neurons in the Sub of control mice, but much fewer in Nr4a2 CKO mice. (D) Note that the vast majority of retrogradely-labeled neurons (green) in the Sub are immunostained with the Nr4a2 antibody (red) in control mice, whereas the number of GFP-labeled neurons detected in Nr4a2 CKO mice is markedly reduced compared to control (N = 4). (E) Heatmap showing downregulation of axonal guidance-related genes in Nr4a2 CKO mice. (F) The axon guidance molecule Epha5 shows markedly reduced expression in the Sub of Nr4a2 CKO mice compared with control mice. (G) Schematic of mEPSC recording from Rsp layer II/III neurons. (H) Representative traces of mEPSCs recorded from layer II/III neurons in the Rsp of control (blue) and Nr4a2 CKO (orange) mice. (I) Cumulative distributions and bar graphs for mEPSC frequency and amplitude of Rsp neurons show a reduced frequency in Nr4a2 CKO mice (n = 11 neurons from N = 3 control mice, and n = 13 neurons from N = 3 Nr4a2 CKO mice). Each dot represents an individual neuron. (J) In situ hybridization of Slc17a6 shows a marked reduction in the Sub of Nr4a2 CKO mice compared to control mice at adulthood. Quantification is shown on the right of each panel (N = 3). (K) Schematic of action potential recordings from Sub neurons. (L) Input resistance of Sub neurons. Nr4a2 CKO neurons display significantly higher input resistance compared with control mice (n = 36 neurons from N = 5 control mice, and n = 35 neurons from N = 5 Nr4a2 CKO mice). Each dot represents an individual neuron. (M) Firing frequency current relationship of Sub neurons. Sub neurons in Nr4a2 CKO mice exhibit a significantly reduced firing rate in response to step current injections compared with control mice (n = 24 neurons from N = 5 control mice, and n = 41 neurons from N = 5 Nr4a2 CKO mice). (N) Action potential threshold of Sub neurons. Sub neurons in Nr4a2 CKO mice exhibit a significantly elevated action potential threshold compared with control mice, indicating reduced intrinsic excitability (n = 23 neurons from N = 5 control mice, and n = 47 neurons from N = 5 Nr4a2 CKO mice). Each dot represents an individual neuron. (O) Schematic and heat maps of the three-chamber test. Both control and Nr4a2 CKO mice exhibited normal sociability during the first phase. However, in the social novelty phase, Nr4a2 CKO mice fail to show a preference for the novel mouse (S2), indicating impaired social novelty. (P) Three-chamber test assesses sociability and social novelty in control (N = 18) and Nr4a2 CKO (N = 16) mice. Statistical analysis shows normal sociability in both groups but impaired social novelty recognition in Nr4a2 CKO mice, as indicated by reduced preference for the novel mouse (S2) and a lower social novelty preference index. Statistical significance is indicated by *p < 0.05, **p < 0.01 and ***p < 0.001. N, number of mice; n, number of neurons. See Materials and Methods, as well as Supplementary Table 1 for detailed statistical analysis.

The defective axonal connections from the Sub to Rsp might have resulted in dysfunction of Rsp neurons and also behavioral consequences. To investigate this, we first recorded mEPSCs of layer II/III neurons in the Rsp (Fig. 4G). In Nr4a2 CKO mice, mEPSC frequency was significantly reduced, whereas the amplitude remained unchanged (Fig. 4H, I). The gene Slc17a6, encoding vGluT2 (vesicular glutamate transporter 2), was found to be specifically expressed in the Sub neurons of control mice (Fig. 4J). In Nr4a2 CKO mice, there was a marked reduction in its expression, potentially also contributing to the observed mEPSC defects in the Rsp (Fig. 4J). Besides, Nr4a2-deficient subicular neurons exhibited a higher input resistance, an elevated action potential threshold, and a reduced firing frequency in whole-cell patch-recordings, indicating a decrease in intrinsic neuronal excitability (Fig. 4K-N). In conclusion, the loss of Nr4a2 not only alters the intrinsic electrophysiological properties of subicular neurons but also selectively impairs their outputs and functional connectivity from the Sub to the Rsp.

The Sub-Rsp circuitry plays a crucial role in social behaviors [32]. To investigate the behavioral consequences of abnormal Sub-Rsp connectivity in Nr4a2 CKO mice, social behaviors were assessed. In the three-chamber test, Nr4a2 CKO mice exhibited normal sociability compared to control mice (Fig. 4O, P). However, during the social novelty phase, while control mice displayed a preference for the novel stranger mouse, Nr4a2 CKO mice spent a similar amount of time with both the stranger and the familiar mouse (Fig. 4O, P), indicating impaired social novelty in Nr4a2 CKO mice. These behavioral deficits suggested that Nr4a2-dependent molecular identity and efferent connection of the Sub neurons were required for social behavior.

Adult deletion of Nr4a2 also compromises social novelty

The expression of Nr4a2 in the subicular regions persists into adulthood (Fig. 2I). To investigate its role in the adult subicular regions and to rule out the possibility that behavioral abnormalities were caused by neurodevelopmental defects due to embryonic Nr4a2 deletion. AAV-Cre virus was stereotactically injected into the subicular regions in Nr4a2flox/flox adult mice to create AAV-Nr4a2 CKO mice, with AAV-EGFP-injected Nr4a2flox/flox mice as controls (AAV-control mice) (Fig. 5A). Nr4a2 expression was markedly reduced in the subicular regions in AAV-Nr4a2 CKO mice (Fig. 5B). Similar to Nr4a2 CKO mice with embryonic Nr4a2 deletion, specific Sub-genes such as Fn1, Nts, and Slc17a6 were significantly decreased in AAV-Nr4a2 CKO mice (Fig. 5B). In contrast, the expression levels of Tle4 which is po-specific gene (Supplementary Fig. 4A) and Nnat which is PS-specific gene (Supplementary Fig. 4B) remained unchanged. These results indicate that Nr4a2 deletion in adulthood also selectively affects Sub neurons, without altering the molecular identity of adjacent regions. Simililarly, electrophysiological recordings of mEPSCs in the Rsp revealed a significant reduction in mEPSC frequency in AAV-Nr4a2 CKO mice (Fig. 5C-E), suggesting defective Sub-Rsp connectivity. Furthermore, light stimulation of Sub-derived axon terminals elicited significantly smaller evoked EPSCs in Rsp neurons of Nr4a2-deficient mice in optogenetically-evoked EPSC recording (Fig. 5F, G), indicating impaired synaptic transmission from the Sub to the Rsp.

Fig. 5.

Fig. 5

Open in a new tab

Adult selective deletion of Nr4a2 also impairs gene expression in the Sub-neurons and leads to social novelty deficits. (A) Schematic showing bilateral injection of AAV-hSyn-EGFP (AAV-control) or AAV-hSyn-Cre-EGFP (AAV-Nr4a2 CKO) into the Sub of adult Nr4a2flox/flox mice. (B) Immunofluorescence shows the loss of Nr4a2 (red) in Cre (EGFP)-expressing neurons (green) in the Sub of Nr4a2flox/flox mice (AAV-Nr4a2 CKO). In situ hybridization shows the loss of expression of Fn1, Nts and Slc17a6 in AAV-Nr4a2 CKO mice at adulthood. (C) Schematic of mEPSC recordings from Rsp layer II/III neurons. (D) Representative image and traces of mEPSCs recorded from layer II/III Rsp neurons in AAV-control (black) and AAV-Nr4a2 CKO mice (green). (E) Cumulative distributions and bar graphs show a significant decrease in mEPSC frequency but unchanged amplitude in AAV-Nr4a2 CKO neurons (n = 34 neurons from N = 5 AAV-control mice, and n = 25 neurons from N = 5 AAV-Nr4a2 CKO mice). Each dot represents an individual neuron. (F) Representative image showing the recording electrode positioned in layers II/III of the Rsp for evoked EPSC recordings following optogenetic stimulation of subicular axon terminals projecting to the Rsp. (G) Representative traces and quantification of evoked EPSCs recorded from Rsp layer II/III neurons. It exhibits a significant reduction in evoked EPSC amplitude in Rsp of AAV-Nr4a2 CKO mice compared with AAV-control mice, indicating impaired synaptic transmission from the Sub to the Rsp (n = 31 neurons from N = 5 AAV-control mice, and n = 22 neurons from N = 5 AAV-Nr4a2 CKO mice). Each dot represents an individual neuron. (H, I) Heatmaps and quantification indicate that both AAV-control (N = 12) and AAV-Nr4a2 CKO mice (N = 14) show a preference for Stranger (S1) other than empty cup in the three-chamber test. However, AAV-Nr4a2 CKO mice fail to show a preference for Stranger 2 (S2) over the familiar Stranger 1 (S1), indicating impaired social novelty recognition. (J) In the novel object recognition test, AAV-Nr4a2 CKO mice exhibit intact recognition memory (N = 14 for AAV-control mice and N = 12 for AAV-Nr4a2 CKO mice). N, number of mice; n, number of neurons. Statistical significance is indicated by **p < 0.01 and ***p < 0.001. See Materials and Methods, as well as Supplementary Table 1 for detailed statistical analysis.

To find out the possible functional consequence of adult deletion of Nr4a2 in the Sub region, behavioral tests were conducted on AAV-Nr4a2 CKO mice and AAV-control mice. The three-chamber test (Fig. 5H) revealed no difference in the sociability test phase (Fig. 5I), but unlike AAV-control mice, the AAV-Nr4a2 CKO mice showed no preference for the stranger mouse 2 during the social novelty phase (Fig. 5I). In the novel object recognition test, however, AAV-Nr4a2 CKO mice displayed a discrimination index (DI) comparable to that of AAV-control mice, showing normal discrimination ability (Fig. 5J). On the other hand, we assessed anxiety-like behaviors and found that AAV-Nr4a2 CKO mice showed no significant changes in the elevated plus maze or light–dark box test compared with AAV-control mice (Supplementary Fig. 4C, D). In addition, these mice showed no abnormalities in locomotor activity in the open field test (Supplementary Fig. 4E), nor did they display significant alterations in depression-like behaviors in the tail suspension or forced swim tests (Supplementary Fig. 4F, G). These findings suggested that the adult specific deletion of Nr4a2 in the Sub region selectively impairs social novelty recognition, while leaving general sociability and other behaviors unaffected.

Nr4a2-mediated mechanisms underlying Sub-neurons specification and social novelty

To further investigate how Nr4a2 regulates the expression of Sub-region-specific genes at the molecular level, we performed CUT&Tag-seq and observed strong Nr4a2 binding peaks in the promoter regions of Fn1, Nts, and Slc17a6 (Fig. 6A). We then conducted a dual-luciferase reporter assay in HEK293T cells by co-transfecting an Nr4a2 overexpression construct with reporter plasmids containing the promoter regions of these genes. Nr4a2 significantly enhanced the luciferase activity driven by each promoter, indicating its ability to activate their transcription (Fig. 6B). Taken together, these findings suggested that Nr4a2 directly activates the transcription of Sub-specific genes by binding to their promoters, thereby preserving the integrity of the Sub-Rsp circuit, which is essential for the formation of social novelty in mice (Fig. 6C).

Fig. 6.

Fig. 6

Open in a new tab

The molecular mechanism of Nr4a2 governs Sub-neuron molecular identity and social novelty. (A) IGV tracks from CUT&Tag analysis showing Nr4a2 enrichment at the promoter of Fn1, Nts, and Slc17a6. Yellow bars highlight the signal at Fn1, Nts, and Slc17a6 gene promoter. The normalized Nr4a2 CUT&Tag signal is obtained from two independent biological replicates. (B) Nr4a2-activated transcription from the Fn1, Nts, and Slc17a6 promoters in a dual-luciferase assay. Each data point represents an independent biological replicate. (C) A proposed model of how Nr4a2 regulates subicular neurons specification and social novelty. Statistical significance is indicated by **p < 0.01 and ***p < 0.001. See Materials and Methods, as well as Supplementary Table 1 for detailed statistical analysis.

Discussion

In this study, we highlighted the critical roles of Nr4a2 in the development of Sub-neurons, including the initiation and maintenance of gene expression as well as its circuits with other brain regions. The defective output to the Rsp may lead to deficient social behaviors in mice.

The subicular regions are a key component of the hippocampal formation [33,34]. Single-cell RNA sequencing (scRNA-seq) revealed that subicular neurons consist of discrete subclasses with distinct gene expression profiles [6,7], and there are three major subclasses defined based on their location: the Sub-pyramidal neurons, the PS-pyramidal neurons, and the po-ovoid neurons [6]. The transcription factor Nr4a2 is expressed in all three subclasses (Fig. 2), prompting us to investigate its role in their development. Our previous study found that Nr4a2, when co-expressed with Ctip2, induces Sub-specific gene expression ectopically [19]. In this study, we found that Nr4a2 uniquely regulates the development of Sub-pyramidal subclass, while that of the PS-pyramidal neurons and po-ovoid neurons remains intact after the deletion of Nr4a2. In Sub-pyramidal neurons, gene expression patterns showed dramatic changes due to deficiencies in gene initiation (Fig. 3) and axonal projections to the Rsp (Fig. 4). Moreover, the deletion of Nr4a2 in adulthood also impairs the maintenance of these Sub-specific genes (Fig. 5). These results indicated that Nr4a2 played a crucial role in both the initiation and maintenance of Sub-enriched genes. The positive regulation of Sub-enriched genes by Nr4a2 is very likely achieved by direct binding to the promoter of these genes (Fig. 6). Although Nr4a2 is expressed across the three subtypes, it has a unique regulatory role merely in the development of Sub-neurons, implying that additional transcriptional regulatory genes are involved. Therefore, further experiments are needed to elucidate the distinct roles of Nr4a2 in the three subclasses of subicular neurons.

The subicular regions serve as principal output nodes of the hippocampal formation [35], and one of their key targets is layers II/III of the Rsp [36,37]. This connection provides a direct link between the hippocampus and Rsp, conveying hippocampal signals to this cortical area [36,38]. When Nr4a2 is deleted from excitatory subicular neurons during embryonic development, a reduction in the intrinsic excitability of Sub neurons and a complete loss of their axonal projections to the Rsp are present, and consequently, these are very like lead to decreased frequency of mEPSCs recorded in the Rsp (Fig. 4), indicating impaired synaptic connectivity from the Sub to the Rsp in Nr4a2 CKO mice. Recent studies have highlighted the crucial roles of Rsp in social behaviors [[39], [40], [41], [42]], and in support of this Nr4a2 CKO mice show impaired social novelty. Since Nr4a2 is deleted in multiple cortical regions including layer 6b and claustral neuron [23,24], the social novelty deficit cannot simply be attributed to dysfunction of Sub-Rsp pathway However, this possibility was excluded based on the behavioral data from AAV-Nr4a2 CKO mice, in which the deletion of Nr4a2 in the Sub was achieved via stereotactic AAV-Cre injection in Nr4a2flox/flox mice, AAV-Nr4a2 CKO mice show reduced Slc17a6 expression in the Sub neurons, lowered mEPSC frequency and attenuated optogenetically-evoked EPSCs in the Rsp neurons (Fig. 5), collectively demonstrating a pronounced functional deficit of the Sub-Rsp circuit and supporting the interpretation that the social behavior abnormalities in AAV-Nr4a2 CKO mice are caused by dysfunction of the Sub-Rsp pathway. In conclusion, Nr4a2 plays a significant and selective role in the functional connectivity between the Sub and Rsp across developmental stages, and its disruption results in social-behavioral dysfunctions. In summary, we found that the transcription factor Nr4a2 plays fundamental roles in the development of the Sub-neurons and the integrity of Sub-Rsp connection. We propose that the loss of Nr4a2 during embryonic development or in adulthood not only affects the molecular identity of Sub-neurons but also impairs their neuronal network with the Rsp, ultimately affecting social behaviors.

Conclusion

Nr4a2 is essential for both establishing and maintaining the molecular identity of subicular pyramidal neurons and their projections to the retrosplenial cortex. Embryonic loss of Nr4a2 prevents activation of Sub-specific genes and axonal growth of Sub to Rsp, whereas adult deletion of Nr4a2 disrupts a similar gene program. Finally, the mice without Nr4a2 in Sub show defects in social novelty, positioning Nr4a2 and the Sub-Rsp pathway as key targets for treating disorders with social deficits.

Ethics approval

All animal experiments were conducted in accordance with institutional and national guidelines and were approved by the Laboratory Animal Ethics Committee of Fudan University (approval number: DSF-2020-041). No human participants or human-derived materials were involved in this study.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the STI2030-Major Projects (2022ZD0204900), Natural Science Foundation of Shanghai (22ZR1466300; 23ZR1413500), National Natural Science Foundation of China (82401384, 32271072), Collaborative Innovation Program of Shanghai Municipal Health Commission (2020CXJQ01), Shanghai Sailing Program of the Shanghai Municipal Science and Technology Program (23YF1407500), Shanghai Blue Cross Brain Hospital Co., Ltd., Shanghai Tongji University Education Development Foundation, Shanghai Municipal Science and Technology Major Project (2018SHZDZX01) and Zhangjiang (ZJ) Lab.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jare.2025.12.014.

Contributor Information

Ning-Ning Song, Email: songnn@fudan.edu.cn.

Yu-Qiang Ding, Email: dingyuqiang@vip.163.com.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary Data 1
mmc1.docx (24KB, docx)
Supplementary Data 2
mmc2.docx (3.7MB, docx)

REFERENCES

  • 1.Swanson L.W., Cowan W.M. Hippocampo-hypothalamic connections: origin in subicular cortex. Not Ammon’s Horn Science. 1975;189:303–304. doi: 10.1126/science.49928. [DOI] [PubMed] [Google Scholar]
  • 2.Rosene D.L., van Hoesen G.W. Hippocampal efferents reach widespread areas of cerebral cortex and amygdala in the Rhesus Monkey. Science, New Series. 1977;198:315–317. doi: 10.1126/science.410102. [DOI] [PubMed] [Google Scholar]
  • 3.O’Mara S. Controlling hippocampal output: the central role of subiculum in hippocampal information processing. Behav Brain Res. 2006;174:304–312. doi: 10.1016/j.bbr.2006.08.018. [DOI] [PubMed] [Google Scholar]
  • 4.Ding S. Comparative anatomy of the prosubiculum, subiculum, presubiculum, postsubiculum, and parasubiculum in human, monkey, and rodent. J of Comparative Neurology. 2013;521:4145–4162. doi: 10.1002/cne.23416. [DOI] [PubMed] [Google Scholar]
  • 5.Bienkowski M.S., Sepehrband F., Kurniawan N.D., Stanis J., Korobkova L., Khanjani N., et al. Homologous laminar organization of the mouse and human subiculum. Sci Rep. 2021;11:3729. doi: 10.1038/s41598-021-81362-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Ding S.-L., Yao Z., Hirokawa K.E., Nguyen T.N., Graybuck L.T., Fong O., et al. Distinct Transcriptomic Cell Types and Neural Circuits of the Subiculum and Prosubiculum along the Dorsal-Ventral Axis. Cell Rep. 2020;31 doi: 10.1016/j.celrep.2020.107648. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Cembrowski M.S., Wang L., Lemire A.L., Copeland M., DiLisio S.F., Clements J., et al. The subiculum is a patchwork of discrete subregions. Elife. 2018;7 doi: 10.7554/eLife.37701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Cembrowski M.S., Phillips M.G., DiLisio S.F., Shields B.C., Winnubst J., Chandrashekar J., et al. Dissociable Structural and Functional Hippocampal Outputs via Distinct Subiculum Cell classes. Cell. 2018;173:1280–1292.e18. doi: 10.1016/j.cell.2018.03.031. [DOI] [PubMed] [Google Scholar]
  • 9.Kinnavane L., Vann S.D., Nelson A.J.D., O’Mara S.M., Aggleton J.P. Collateral Projections Innervate the Mammillary Bodies and Retrosplenial Cortex. A New Category of Hippocampal Cells eNeuro. 2018;5 doi: 10.1523/ENEURO.0383-17.2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Sun Y., Jin S., Lin X., Chen L., Qiao X., Jiang L., et al. CA1-projecting subiculum neurons facilitate object-place learning. Nat Neurosci. 2019;22:1857–1870. doi: 10.1038/s41593-019-0496-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Pederick D.T., Lui J.H., Gingrich E.C., Xu C., Wagner M.J., Liu Y., et al. Reciprocal repulsions instruct the precise assembly of parallel hippocampal networks. Science. 2021;372:1068–1073. doi: 10.1126/science.abg1774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Honda Y., Shimokawa T., Matsuda S., Kobayashi Y., Moriya-Ito K. Hippocampal Connectivity of the Presubiculum in the Common Marmoset (Callithrix jacchus) Front Neural Circuits. 2022;16 doi: 10.3389/fncir.2022.863478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Bienkowski M.S., Bowman I., Song M.Y., Gou L., Ard T., Cotter K., et al. Integration of gene expression and brain-wide connectivity reveals the multiscale organization of mouse hippocampal networks. Nat Neurosci. 2018;21:1628–1643. doi: 10.1038/s41593-018-0241-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Kinman A.I., Merryweather D.N., Erwin S.R., Campbell R.E., Sullivan K.E., Kraus L., et al. Atypical hippocampal excitatory neurons express and govern object memory. Nat Commun. 2025;16:1195. doi: 10.1038/s41467-025-56260-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Soltesz I., Losonczy A. CA1 pyramidal cell diversity enabling parallel information processing in the hippocampus. Nat Neurosci. 2018;21:484–493. doi: 10.1038/s41593-018-0118-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Cossart R., Khazipov R. How development sculpts hippocampal circuits and function. Physiol Rev. 2022;102:343–378. doi: 10.1152/physrev.00044.2020. [DOI] [PubMed] [Google Scholar]
  • 17.Dudek S.M., Alexander G.M., Farris S. Rediscovering area CA2: unique properties and functions. Nat Rev Neurosci. 2016;17:89–102. doi: 10.1038/nrn.2015.22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Jakaria Md., Haque M.E., Cho D.-Y., Azam S., Kim I.-S., Choi D.-K. Molecular Insights into NR4A2(Nurr1): an Emerging Target for Neuroprotective Therapy against Neuroinflammation and Neuronal Cell Death. Mol Neurobiol. 2019;56:5799–5814. doi: 10.1007/s12035-019-1487-4. [DOI] [PubMed] [Google Scholar]
  • 19.Zhang L., Song N.-N., Zhang Q., Mei W.-Y., He C.-H., Ma P., et al. Satb2 is required for the regionalization of retrosplenial cortex. Cell Death Differ. 2020;27:1604–1617. doi: 10.1038/s41418-019-0443-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Saucedo-Cardenas O., Quintana-Hau J.D., Le W.-D., Smidt M.P., Cox J.J., Mayo F.D., et al. Nurr1 is Essential for the Induction of the Dopaminergic Phenotype and the Survival of Ventral Mesencephalic late Dopaminergic Precursor Neurons. PNAS. 1998;95:4013–4018. doi: 10.1073/pnas.95.7.4013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Kadkhodaei B., Ito T., Joodmardi E., Mattsson B., Rouillard C., Carta M., et al. Nurr1 is Required for Maintenance of Maturing and Adult Midbrain Dopamine Neurons. J Neurosci. 2009;29:15923–15932. doi: 10.1523/JNEUROSCI.3910-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Luo G.R., Chen Y., Li X.P., Liu T.X., Le W.D. Nr4a2 is essential for the differentiation of dopaminergic neurons during zebrafish embryogenesis. Mol Cell Neurosci. 2008;39:202–210. doi: 10.1016/j.mcn.2008.06.010. [DOI] [PubMed] [Google Scholar]
  • 23.Mantas I., Flais I., Masarapu Y., Ionescu T., Frapard S., Jung F., et al. Claustrum and dorsal endopiriform cortex complex cell-identity is determined by Nurr1 and regulates hallucinogenic-like states in mice. Nat Commun. 2024;15:8176. doi: 10.1038/s41467-024-52429-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zhao L., Tao Y.-C., Hu L., Liu X.-Y., Zhang Q., Zhang L., et al. Satb2 and Nr4a2 are required for the differentiation of cortical layer 6b. Cell Death Discov. 2025;11:126. doi: 10.1038/s41420-025-02402-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Quina L.A., Wang S., Ng L., Turner E.E. Brn3a and Nurr1 Mediate a Gene Regulatory Pathway for Habenula Development. J Neurosci. 2009;29:14309–14322. doi: 10.1523/JNEUROSCI.2430-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Jiang H., Xiang M. Subtype Specification of GABAergic Amacrine Cells by the Orphan Nuclear Receptor Nr4a2/Nurr1. J Neurosci. 2009;29:10449–10459. doi: 10.1523/JNEUROSCI.3048-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.In: Current Topics in Developmental Biology Elsevier. 2021:147–196. doi: 10.1016/bs.ctdb.2020.11.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Kuhn T.B., Schmidt M.F., Kater S.B. Laminin and fibronectin guideposts signal sustained but opposite effects to passing growth cones. Neuron. 1995;14:275–285. doi: 10.1016/0896-6273(95)90285-6. [DOI] [PubMed] [Google Scholar]
  • 29.Cramer K.S., Cerretti D.P., Siddiqui S.A. EphB2 regulates axonal growth at the midline in the developing auditory brainstem. Dev Biol. 2006;295:76–89. doi: 10.1016/j.ydbio.2006.03.010. [DOI] [PubMed] [Google Scholar]
  • 30.Vaidya A., Pniak A., Lemke G., Brown A. EphA3 Null Mutants do not Demonstrate Motor Axon Guidance Defects. Mol Cell Biol. 2003;23:8092–8098. doi: 10.1128/MCB.23.22.8092-8098.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Cooper M.A., Crockett D.P., Nowakowski R.S., Gale N.W., Zhou R. Distribution of EphA5 receptor protein in the developing and adult mouse nervous system. J of Comparative Neurology. 2009;514:310–328. doi: 10.1002/cne.22030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Li H., Zhao Z., Jiang S., Wu H. Brain circuits that regulate social behavior. Mol Psychiatry. 2025;30:3240–3256. doi: 10.1038/s41380-025-03037-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.O’Mara S.M., Commins S., Anderson M., Gigg J. The subiculum: a review of form, physiology and function. Prog Neurobiol. 2001;64:129–155. doi: 10.1016/s0301-0082(00)00054-x. [DOI] [PubMed] [Google Scholar]
  • 34.Peitz K., Bittner N., Heim S., Caspers S. Bilingualism and “brain reserve” in subregions of the hippocampal formation. GeroScience. 2025;47:4935–4954. doi: 10.1007/s11357-025-01639-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Lee S.-M., Seol J.-M., Lee I. Subicular neurons represent multiple variables of a hippocampal-dependent task by using theta rhythm. PLoS Biol. 2022;20 doi: 10.1371/journal.pbio.3001546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Nitzan N., McKenzie S., Beed P., English D.F., Oldani S., Tukker J.J., et al. Propagation of hippocampal ripples to the neocortex by way of a subiculum-retrosplenial pathway. Nat Commun. 2020;11:1947. doi: 10.1038/s41467-020-15787-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Gao M., Noguchi A., Ikegaya Y. The subiculum sensitizes retrosplenial cortex layer 2/3 pyramidal neurons. J Physiol. 2021;599:3151–3167. doi: 10.1113/JP281152. [DOI] [PubMed] [Google Scholar]
  • 38.Yamawaki N., Corcoran K.A., Guedea A.L., Shepherd G.M.G., Radulovic J. Differential Contributions of Glutamatergic Hippocampal→Retrosplenial Cortical Projections to the Formation and Persistence of Context Memories. Cereb Cortex. 2019;29:2728–2736. doi: 10.1093/cercor/bhy142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Shi Y., Yan J., Xu X., Qiu Z. Gating of Social Behavior by Inhibitory Inputs from Hippocampal CA1 to Retrosplenial Agranular Cortex. Neurosci Bull. 2024;40:1635–1648. doi: 10.1007/s12264-023-01172-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Yang K., Shi Y., Du X., Wang J., Zhang Y., Shan S., et al. SENP1 in the retrosplenial agranular cortex regulates core autistic-like symptoms in mice. Cell Rep. 2021;37 doi: 10.1016/j.celrep.2021.109939. [DOI] [PubMed] [Google Scholar]
  • 41.Garrido D., Beretta S., Grabrucker S., Bauer H.F., Bayer D., Sala C., et al. Shank2/3 double knockout-based screening of cortical subregions links the retrosplenial area to the loss of social memory in autism spectrum disorders. Mol Psychiatry. 2022;27:4994–5006. doi: 10.1038/s41380-022-01756-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Ding C., Zhou W., Shi Y., Shan S., Yuan Y., Zhang Y., et al. Srcap haploinsufficiency induced autistic-like behaviors in mice through disruption of Satb2 expression. Cell Rep. 2024;43 doi: 10.1016/j.celrep.2024.114231. [DOI] [PubMed] [Google Scholar]
  • 43.Gorski J.A., Talley T., Qiu M., Puelles L., Rubenstein J.L.R., Jones K.R. Cortical Excitatory Neurons and Glia, but not GABAergic Neurons, are Produced in the Emx1-Expressing Lineage. J Neurosci. 2002;22:6309–6314. doi: 10.1523/JNEUROSCI.22-15-06309.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Kaya-Okur H.S., Wu S.J., Codomo C.A., Pledger E.S., Bryson T.D., Henikoff J.G., et al. CUT&Tag for efficient epigenomic profiling of small samples and single cells. Nat Commun. 2019;10:1930. doi: 10.1038/s41467-019-09982-5. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data 1
mmc1.docx (24KB, docx)
Supplementary Data 2
mmc2.docx (3.7MB, docx)

Articles from Journal of Advanced Research are provided here courtesy of Elsevier

RESOURCES