Sustained correction of hippocampal neurogenic and cognitive deficits after a brief treatment by Nutlin-3 in a mouse model of fragile X syndrome

Background Fragile X syndrome (FXS), the most prevalent inherited intellectual disability and one of the most common monogenic forms of autism, is caused by a loss of fragile X messenger ribonucleoprotein 1 (FMR1). We have previously shown that FMR1 represses the levels and activities of ubiquitin ligase MDM2 in young adult FMR1-deficient mice, and treatment by a MDM2 inhibitor Nutlin-3 rescues both hippocampal neurogenic and cognitive deficits in FMR1-deficient mice when analyzed shortly after the administration. However, it is unknown whether Nutlin-3 treatment can have long-lasting therapeutic effects. Methods We treated 2-month-old young adult FMR1-deficient mice with Nutlin-3 for 10 days and then assessed the persistent effect of Nutlin-3 on both cognitive functions and adult neurogenesis when mice were 6-month-old mature adults. To investigate the mechanisms underlying the persistent effects of Nutlin-3, we analyzed the proliferation and differentiation of neural stem/progenitor cells isolated from these mice and assessed the transcriptome of the hippocampal tissues of treated mice. Results We found that transient treatment with Nutlin-3 of 2-month-old young adult FMR1-deficient mice prevents the emergence of neurogenic and cognitive deficits in mature adult FXS mice at 6 months of age. We further found that the long-lasting restoration of neurogenesis and cognitive function might not be mediated by changing intrinsic properties of adult neural stem cells. Transcriptomic analysis of the hippocampal tissue demonstrated that transient Nultin-3 treatment leads to significant expression changes in genes related to the extracellular matrix, secreted factors, and cell membrane proteins in the FMR1-deficient hippocampus. Conclusions Our data indicates that transient Nutlin-3 treatment in young adults leads to long-lasting neurogenic and behavioral changes likely through modulating adult neurogenic niche that impact adult neural stem cells. Our results demonstrate that cognitive impairments in FXS may be prevented by an early intervention through Nutlin-3 treatment. Supplementary Information The online version contains supplementary material available at 10.1186/s12916-022-02370-9.


Background
Fragile X Syndrome (FXS) is the most common cause of inherited intellectual disability with prevalence rates estimated to be 1:5000 in males and 1:8000 in females [1]. FXS is one of the most common single-gene causes of Open Access *Correspondence: Xinyu.zhao@wisc.edu 1 Waisman Center, University of Wisconsin-Madison, Madison, WI 53705, USA Full list of author information is available at the end of the article autism spectrum disorder (ASD), with approximately half of male FXS patients being clinically diagnosed with ASD [2,3]. FXS is mainly caused by an expansion of trinucleotide repeats (CGG) to over 200 repeats in the 5′ untranslated region of the fragile X messenger ribonucleoprotein 1 (FMR1) gene which leads to transcriptional silencing of the gene with a subsequent reduction or absence of FMR1 (also known as FMRP, fragile X mental retardation protein) [4,5]. FMR1 is a polyribosome-associated, brain-enriched, RNA-binding protein (RBP) that selectively targets specific mRNAs and regulates their translation, transport, stability, and nonsense-mediated RNA decay [4][5][6][7][8][9]. In addition, it has been shown that FMR1 is involved in histone modification and chromatin remodeling [10]. Hence, FMR1 is a multifunctional protein that could be involved in diverse biological processes. FMR1 deficiency has been associated with numerous co-occurring conditions including, but not limited to, intellectual and emotional disabilities ranging from learning problems to mental retardation, and mood instability to autism [11]. A better understanding of the neurobiology and pathophysiology of FXS, together with advances in FXS animal models, has paved a way for the development of numerous targeted treatments for FXS [12]. FMR1 is a multifunctional protein that regulates the expression of a large number of direct and indirect targets [8,13,14]. Despite the rich literature aiming to investigate short-term therapeutic outcomes of treatments, very few studies have evaluated the potential long-lasting rescue effects of these treatments [12,15]. Promising data from a recent study show that impairment of cognitive repertoire in FXS could be sustainably prevented by short-term pharmacological interventions [12,16]. The long-lasting restoration of cognitive deficits of a rat model of FXS used in this study [12] is associated with sustained rescue of both synaptic plasticity and altered protein synthesis. These results have raised the question whether other reported interventions have the potential to exert long-lasting therapeutic effects.
Even though FMR1 is highly expressed in neurons, other cells have been implicated in FXS as well [17,18]. Studies from our and other laboratories have shown that FMR1 regulates adult hippocampal neurogenesis [19][20][21][22][23][24][25][26]. Neurogenesis continuously occurs in at least two specific regions of the adult mammalian brain: the subventricular zone (SVZ) of the lateral ventricles and the subgranular zone (SGZ) of the dentate gyrus (DG) in the hippocampus [27]. Adult hippocampal neurogenesis is a multi-stage process, encompassing a number of developmental phases [27]. Activated neural stem cells (NSCs) or radial glia-like cells (RGLs) generate intermediate neural progenitors (NPs) that subsequently differentiate into neuroblasts, immature neurons, and mature granule neurons (GCs) that finally integrate into existing circuits [28]. Adult hippocampal neurogenesis is implicated in many functional processes such as learning, memory, plasticity, and mood regulation [27] and is impaired in a number of neurological conditions including FXS [19][20][21][22][23][24][25][26]. Therefore, interventions aimed at regulating adult neurogenesis are being evaluated as potential therapeutic strategies [29]. We have previously shown that the absence of FMR1 leads to increased NSC proliferation but reduced neuronal differentiation in young adult (2-month-old) Fmr1 knockout (Fmr1 KO) mice but reduced NSC proliferation and reduced neuronal differentiation in mature adult (6-month-old) Fmr1 KO mice [20,21]. At molecular levels, FMR1-deficient NSCs have elevated mouse double minute 2 (MDM2) protein levels and activities throughout adult ages. Treatment with Nutlin-3, a compound used for cancer clinical trial, specifically inhibits the interaction between MDM2 and its target proteins tumor protein P53 (TP53) and histone deacetylase 1 (HDAC1) and rescues both adult hippocampal neurogenic and behavioral deficits in both young adult and mature adult FXS mice. However, it remains unknown whether Nutlin-3 can have long-lasting effects in FXS mice, which is a key question for therapeutic development.
In this study, we investigated whether Nutlin-3 treatment had a sustained impact on neurogenesis and cognitive behaviors of FXS mice. We discovered that a transient treatment of young adult mice with Nutlin-3 led to a long-lasting effect in both hippocampal neurogenesis and cognitive tasks in adult FXS mice. To our surprise, we found that the long-lasting effect of Nutlin-3 might be mediated by regulating the gene expression of the adult stem cell niche rather than modulating the intrinsic properties of adult NSCs.

Study design
The purpose of this study was to investigate the long-lasting effect of Nutlin-3 treatment on impaired neurogenesis and behavioral deficits in adult FXS mice. In addition, we aimed to determine the potential mechanisms underlying the sustained rescue effect by Nutlin-3. Based on our publications and power analysis, at least three biological replicates were used for each in vitro or in vivo biochemical and histological analysis, whereas a sample size of 9 to 21 per group was used for behavioral testing. The NSPCs used for in vitro analyses were isolated from three pairs of Fmr1 KO and WT littermates born to different parents, and NSPCs isolated from each animal were considered as a biological replicate. For drug treatment, animals were randomly assigned to treatment arms with approximately equivalent numbers in each group.
All cell counting and behavioral analyses were performed by experimenters who were blind to the identity of the samples.

Animal studies
All animal procedures were performed according to the protocols approved by the University of Wisconsin-Madison Institutional Animal Care and Use Committee. All mice were of C57B/L6 genetic background. The crossing and genotyping of these mice were carried out as described previously [20,21]. Briefly, the Fmr1 KO;Nestin-GFP mice (Fmr1 −/y ;Nestin-GFP) were created by crossing female Fmr1 heterozygous KO mice (Fmr1 +/− ) [30] with homozygous Nestin-GFP transgenic males [31]. The generation of FMR1 inducible conditional mutant mice (Fmr1 loxP/y ;Nestin-CreER T2 ;Rosa26-tdT or cKO;Cre;Ai14 ) and tamoxifen injections to induce recombination were performed as described [22]. To induce recombination, mice (6-7 week old) received tamoxifen (160 mg/kg; Sigma-Aldrich) daily for 5 days as described [22]. Nutlin-3 (10 mg/kg) was dissolved in dimethyl sulfoxide, given to 2-month-old mice through intraperitoneal injections every other day for five injections, and sacrificed at 4 months after the last injection. For in vivo differentiation analysis, mice also received four BrdU injections (100 mg/kg) within 12 h at 1 month before sacrifice.

Tissue preparation and immunohistochemistry
Brain tissue processing and histological analysis of mouse brains were performed as described in our publications [19][20][21][22]32]. Briefly, mice were euthanized by intraperitoneal injection of a mixture of ketamine/xylazine/ acepromazine followed by transcardiac perfusion with saline and then 4% paraformaldehyde (PFA). The brains were dissected out, post-fixed overnight in 4% PFA, and equilibrated in 30% sucrose. The brain sections of 40 μm thickness were generated using a sliding microtome and stored in a − 20 °C freezer as floating sections in cryoprotectant solution (glycerol, ethylene glycol, and 0.1 M phosphate buffer (pH 7.4), 1:1:2 by volume). We performed immunohistological analysis on 1-in-6 serial floating brain sections (240 μm apart). After staining with primary and fluorescent secondary antibodies, the sections were counterstained with DAPI (1:1000; Roche Applied Science) and then mounted, coverslipped, and maintained at 4 °C in the dark until analysis.

In vivo cell quantification
Quantitative analyses of adult neurogenesis were carried out using an unbiased stereology method through the use of a Stereo Investigator software (MBF Biosciences) as described [20,21,33]. Briefly, Z-stack images (2-μm interval) were acquired using an AxioImager Z2 ApoTome confocal microscope (Plan-APOCHOROMAT, 20X, numerical aperture = 0.8; Zeiss). The measured thickness of the sections was ~ 30 μm. The cell numbers were quantified by random sampling one in six coronal serial sections (240 μm apart) encompassing the entire hippocampus, with 3-μm guard zones on each side. Schaeffer's coefficient of error (CE) < 0.1 was required for each type of cell quantification. The experimenter was blinded to the identity of the samples. The total numbers of GFP+GFAP+ cells in the dentate gyrus of each animal were counted. Then, the percentage of activated and proliferating NSCs was determined by co-localizing of GFP+GFAP cells with the cell cycle marker, MCM2. Cell lineage analysis was performed as described [20]. Briefly, at least 100 BrdU + cells in the Z-stacks of each mouse were randomly selected, and their co-localization with cell-lineage marker, NeuN, was determined using Stereo Investigator.

Novel location recognition test
This test measures spatial memory through an evaluation of the ability of mice to recognize the new location of a familiar object with respect to spatial cues. The experimental procedure was carried out as described previously [20,21]. Briefly, mice were handled for approximately 5 min a day for a maximum of 5 days prior to the experiment. Testing consisted of five 6-min trials, with a 3-min intertrial interval between each trial. All procedures were conducted during the light cycle of the animal between 10 a.m. and 4 p.m. Before the trial session, mice were brought into the testing room and were allowed to acclimate for at least 30 min. During the intertrial interval, the mouse was placed in a holding cage, which remained inside the testing room. In the first trial (pre-exposure), each mouse was placed individually in the center of the otherwise empty open arena (38.5 cm long × 38.5 cm wide, and 25.5-cm-high walls) for 6 min. For the next three trials (sample trials 13), two identical objects were placed equidistantly from the arena wall in the corners against the wall with the colored decal. Tape the objects to the floor of the arena. Then, each mouse was placed individually in the center of the arena and was allowed to explore for 6 min. At the end of the trial, the mouse was removed and returned to the home cages for 3 min. In the last trial (test), one of the objects was moved to a novel location, and the mouse was allowed to explore the objects for 6 min, and the total time spent exploring each object was measured. During the test phase, the exploration time was defined as any investigative behavior (i.e., head orientation, climbing on, sniffing occurring within < 1.0 cm) or other motivated direct contact occurring with each object. To control for possible odor cues, objects were cleaned with 70% ethanol solution at the end of each trial, and the floor of the arena was wiped down to eliminate possible scent/trail markers. During the test phase, two objects were wiped down prior to testing so that the objects would all have the same odor. Based on a previous study [34], the discrimination index was calculated as the percentage of the time spent investigating the object in the new location minus the percentage of time spent investigating the object in the old location: discrimination index = (novel location exploration time/ total exploration time × 100) − (old location exploration time/total exploration time × 100). A higher discrimination index is considered to reflect greater memory retention for the novel location object. All experiments were videotaped and scored by scientists who were blinded to the experimental conditions to ensure accuracy.

Novel object recognition test
This test is based on the natural propensity of rodents to preferentially explore the novel objects over familiar ones. The experimental procedure was carried out as described previously [20,21]. Briefly, mice were handled for approximately 5 min a day for a maximum of 5 days prior to the experiment. The test was conducted during the light cycle of the animal between 10 a.m. and 4 p.m. Before the trial or test phase, mice were brought into the testing room and were allowed to acclimate for at least 30 min. On the first day, mice were habituated for 10 min to the V-maze, made out of black Plexiglas with two corridors (30 cm long × 4.5 cm wide, and 15-cm-high walls) set at a 90° angle, in which the task was performed. On the second day, mice were put back in the maze for 10 min, and two identical objects were presented. Twentyfour hours later, one of the familiar objects was replaced with a novel object, and the mice were again placed in the maze and were allowed to explore for 10 min, and the total time spent exploring each of the two objects (novel and familiar) was measured. During the test phase, the novel and familiar objects were wiped down prior to testing so that the objects would all have the same odor, and the exploration time was defined as the orientation of the nose to the object at a distance of less than 2 cm. The discrimination index was calculated as the difference between the percentages of the time spent investigating the novel object and the time spent investigating the familiar objects: discrimination index = (novel object exploration time/total exploration time × 100) − (familiar object exploration time/total exploration time × 100). A higher discrimination index is considered to reflect greater memory retention for the novel object. All experiments were videotaped and scored by scientists who were blinded to experimental conditions to ensure accuracy.

Adult NSPC isolation and analyses
NSPCs were isolated from the pooled DG tissue dissected from two 6-month-old male mice using our published method [21,35]. NSPCs were cultured as described previously [35]. The proliferation and differentiation of NPCs were analyzed as described [20]. We used only early passage cells (between passages 4 and 10) and only the same passage numbers of wild-type and Fmr1 KO cells for data collection. For each experiment, triplicate wells of cells (experimental replicates) were analyzed, and the results were averaged as one data point (n = 1). At least three independent biological replicates (cells isolated from different animals) were used (n = 3) for statistical analyses.

RNA isolation and RNA-seq
Freshly dissected hippocampal tissue was immediately frozen on dry ice. For RNA isolation, TRIzol was added to frozen tissue followed by homogenization using a Polytron (PT1200, Kinematica, Inc., Bohemia, NY, USA) as described [36]. RNA was isolated from TRIzol samples using the TRIzol Reagent following the manufacturer's instructions. RNA quality assessment, library construction, library quality control, and sequencing were performed by Novogene Bioinformatics Institute (Sacramento, CA, USA). Briefly, the quality, size, and concentration of the isolated RNA were analyzed using agarose gel electrophoresis, nanodrop, and an Agilent 2100 Bioanalyzer. Twelve cDNA libraries were constructed with three biological replicates for each condition: WT-Veh, KO-Veh, WT-Nut3, and KO-Nut3. Messenger RNA was purified from total RNA using poly-T oligo-attached magnetic beads. After fragmentation, the first-strand cDNA was synthesized using random hexamer primers followed by the second strand cDNA synthesis. The library was ready after end repair, A-tailing, adapter ligation, size selection, amplification, and purification (Novogene, Sacramento, CA, USA). Library quality was assessed by Qubit 2.0, Agilent 2100, and qPCR. The libraries were clustered and sequenced on an Illumina Novaseq 6000 on an S4 flow cell. The 150-bp paired-end reads were generated after clustering of the index-coded samples. About 20-30 million reads were obtained for each sample.

Bioinformatics analysis
FastQC was used to perform a quality check of .fastq reads. Paired-end reads were mapped to the reference genome (mm10) using STAR (Additional file 1: STAR parameters). The raw count matrix was normalized by correcting for library size using the DESeq2 R package (Additional file 2: Code 2). The differential expression analysis was performed by using the DEseq2 R package (Additional file 2: Code 2). Adjusted P-values < 0.05 were used as cutoffs for differential expression. GO term enrichment was analyzed using Enricher (https:// maaya nlab. cloud/ Enric hr) [37] and plotted using GOplot [38]. Differential expression was visualized by volcano plots (Additional file 3: Code 3). Transcription factor enrichment analysis and network were performed using ChEA3 (https:// maaya nlab. cloud/ chea3) [39], and the average integrated ranks across all libraries RNA-seq data have been deposited to Gene Expression Omnibus (GEO: GSE198403).

Real-time PCR assay
Real-time PCR was performed using standard methods as described [20]. The first-strand cDNA was generated by reverse transcription with Oligo (dT) primer (Roche). To quantify the mRNA levels using real-time PCR, aliquots of first-stranded cDNA were amplified with genespecific primers and Power SYBR Green PCR Master Mix  Table S1.

Statistical analysis
All experiments were randomized and blinded to scientists who performed the quantification. Statistical analysis was performed using ANOVA and Student's t-test, unless specified, with the GraphPad Prism software 9. Two-tailed and unpaired t-test was used to compare two conditions. Two-way ANOVA with Tukey's post hoc analysis was used for analyzing multiple groups. All data were shown as mean with standard error of the mean (mean ± SEM). Probabilities of P < 0.05 were considered as significant.

Transient Nutlin-3 treatment has a long-lasting rescue effect on impaired hippocampal neurogenesis in Fmr1 KO mice
We have previously shown that young adult (2-monthold) Fmr1 KO mice exhibited elevated NSC activation and impaired neurogenesis, and mature adult (6-monthold) Fmr1 KO mice exhibited reduced NSC activation and impaired neurogenesis in the hippocampus, which can be normalized to WT levels immediately after a 10-day treatment by a specific MDM2 inhibitor Nutlin-3 [20,21]. To assess the potential of MDM2 inhibition as a therapeutic treatment, a critical question remained whether the Nutlin-3 treatment has long-lasting rescue effects. We thus decided to investigate whether a transient Nutlin-3 treatment could have a persistent therapeutic effect on NSC activation and adult neurogenesis in FXS mouse models.
We crossed Fmr1 mutant mice with Nestin-GFP (green fluorescent protein) mice in which GFP expression is driven by the promotor of a neural stem and progenitor cell marker NESTIN to create the Fmr1 KO;Nestin-GFP double transgenic mice as described previously [20]. We treated 2-month-old Fmr1 KO (Fmr1 −/y ;Nestin-GFP) and littermate wild-type (Fmr1 +/y ;Nestin-GFP) mice with either vehicle or Nutlin-3 (10 mg/kg) every other day over 10 days (total 5 injections) as we have done previously [20,21] and analyzed them at 4 months after the last injections, when the mice were 6 months old (Fig. 1a). Glial fibrillary acidic protein (GFAP) is a radial glia marker expressed in both quiescent and activated adult hippocampal NSCs (Fig. 1b) [40]. To determine the activation of NSCs, we used the cell cycle marker minichromosome maintenance complex component 2 (MCM2; Fig. 1b). We quantified the percentage of activated (GFP + GFAP + MCM2 + ) NSCs over total (GFP + GFAP + ) NSCs in Fmr1 KO and WT mice. We found that Fmr1 KO mice treated with vehicle at 2 months of age exhibited reduced NSC activation at 6 months of age compared to WT with the same vehicle treatment (Fig. 1c), which is consistent with our previous finding on 6-month-old Fmr1 KO mice [21]. Similarly to what we have published before [20,21], Nutlin-3 treatment had no significant effect on WT mice (Fig. 1c). In contrast, Fmr1 KO mice treated with Nutlin-3 at 2 months of age showed no significant difference in NSC activation at 6 months of age compared to WT mice treated with either vehicle or Nutlin-3 (Fig. 1c). Therefore, a 10-day transient Nutlin-3 treatment of young adult Fmr1 KO mice has a long-lasting rescue effect on adult hippocampal NSC activation, which persists for at least 4 months.
We then assessed whether the therapeutic effect of Nutlin-3 on neuronal differentiation [20,21] could persist long after treatment. Thus, we injected 2-month-old Fmr1 KO (Fmr1 −/y ) mice and WT (Fmr1 +/y ) littermates with Nutlin-3 as described above (Fig. 1a) [20]. At 5 months of age, the mice received four injections of a synthetic thymidine analog bromodeoxyuridine (BrdU) over a 12-h period to pulse label proliferating NSCs and progenitors in the adult DG and were sacrificed at 4 weeks after BrdU injections (6 months of age) for differentiation analysis (Fig. 1a) [20]. To identify the fate of the BrdU-labeled NSPCs, we performed co-immunostaining using antibodies against mature neuronal marker NeuN (neuronal nuclei antigen or RBFOX3) and BrdU and quantified the percentage of neuronal differentiation (BrdU + NeuN + /BrdU + ) (Fig. 1d). We found that Fmr1 KO mice treated with vehicle showed a significant reduction in neuronal differentiation compared to WT counterparts treated with either vehicle or Nutlin-3 (Fig. 1e), which is consistent with what has been reported previously [21]. In contrast, 6-month-old Fmr1 KO mice treated with Nutlin-3 at 2 months of age showed elevated neuronal differentiation levels comparable to that of the WT mice (Fig. 1e). Nutlin-3 administration did not show significant effects on neurogenesis of WT mice (Fig. 1e). Therefore, transient treatment of Fmr1 KO mice with Nutlin-3 at young adult ages could prevent impairment of neuronal differentiation in mature adult mice. In summary, our findings have revealed a long-lasting therapeutic effect of Nutlin-3 on impaired NSC activation and neuronal differentiation in a FXS mouse model.

Transient Nutlin-3 treatment has an endured corrective effect on cognitive deficit in FXS mouse models
We have previously shown that Fmr1 KO mice exhibited deficits in hippocampus-dependent cognitive functions [20][21][22]41] and Nutlin-3 treatment reversed impaired spatial learning assessed by a novel location recognition test (NLR) and defective cognitive function assessed by a novel object recognition (NOR) test at 1 month after treatment [20,21]. Since our current study has revealed a persistent therapeutic effect of Nutlin-3 on impaired NSC activation and neurogenesis in Fmr1 KO mice (Fig. 1), we decided to investigate whether the Nutlin-3-dependent restoration of cognitive deficit is also long-lasting.
First, to confirm that selective deletion of FMR1 from NSCs in young adult mice leads to long-lasting impaired performances on hippocampus-dependent learning tasks, we generated tamoxifen-inducible Fmr1 conditional knockout (cKO;CreER T2 ;Ai14) triple transgenic mice by crossing Fmr1-floxed (Fmr1 f/f or cKO) mice with inducible Nestin promotor-driven Cre transgenic mice (Nes-CreER T2 ) and Rosa26-STOP-tdTomato (Ai14) reporter mice as described previously [20] (Additional file 9: Fig. S1a). We found that targeted deletion of FMR1 from NSCs and their progenies at 2 months of age led to learning deficits at 6 months of age (Additional file 9: Fig. S1b-d), which corroborated our previous findings [20,22]. More importantly, transient treatment of cKO;CreER T2 ;Ai14 mice with Nutlin-3 at 2 months of age led to the restoration of cognitive function when assessed at 6 months of age (Additional file 9: Fig. S1b-d). Therefore, transient Nutlin-3 treatment in young adult mice with selective deletion of FMR1 from adult new neurons has a long-lasting effect.
We then treated 2-month-old Fmr1 KO mice and their WT littermates with either vehicle or Nutlin-3 and analyzed their behaviors 4 months later (Fig. 2a). Consistent with our previous findings, Fmr1 KO mice treated with vehicle exhibited impaired performance in spatial learning on the NLR test and defective learning on the NOR test (Fig. 2b-e) [21]. In contrast, Nutlin-3 administration rescued the impaired performances of Fmr1 KO mice in both NLR (Fig. 2c) and NOR (Fig. 2e) to the wildtype levels without significant effect on wild-type mice (Fig. 2c, e). Therefore, a transient Nutlin-3 treatment of FMR1-deficient mice at young adulthood could rescue impaired cognitive performance for at least 4 months.

Transient treatment with Nutlin-3 does not have a persistent effect on intrinsic properties of adult neural stem/progenitor cells
We next sought to reveal the molecular mechanisms that are associated with Nutlin-3-induced enduring rescue of impaired hippocampal neurogenesis and related cognitive functions. Adult NSCs and adult hippocampal neurogenesis are regulated by both intrinsic and extrinsic factors [42]. To determine whether Nutlin-3 treatment acted through the NSC intrinsic pathways, we decided to analyze the neural stem/progenitor cells (NSPCs) isolated from the hippocampus of 6-month-old Fmr1 KO and littermate WT mice treated with either vehicle or Nutlin-3 at 2 months of age (Fig. 3a) using our published methods [35]. We used BrdU pulse labeling to assess NSPC proliferation and found that Fmr1 KO NSPCs exhibited a reduced BrdU incorporation rate compared to WT NSPCs (Fig. 3b, d), consistent with our published results on NSPCs isolated from 6-month-old Fmr1 KO mice [21]. Surprisingly, Nutlin-3 treatment did not rescue the impaired proliferation of Fmr1 KO cells (Fig. 3d). We then assessed NPSC neuronal differentiation using an antibody for immature neuron, βIII-tubulin (Tuj1), and found that NSPCs isolated from Fmr1 KO mice at 4 months after either Nutlin-3 or vehicle injection exhibited similarly reduced neuronal differentiation (Fig. 3c,  e). We have previously shown that NSCs isolated from 6-month-old mice exhibited elevated levels of phosphorylated MDM2 (P-MDM2, the active form of MDM2), EP300, acetylated Histone H3, and HDAC1, a target of MDM2 in 6-month-old mice [21], but not protein levels of P53, a target of MDM2 in NSCs in 2-month-old mice [20]. We therefore assessed the levels of these proteins and found that NSPCs isolated from KO mice exhibited increased P-MDM2 and EP300 protein levels, elevated H3 acetylation, and reduced HDAC1 levels, without significant change in P53 levels ( Fig. 3f-o), consistent with our published results [21]. More importantly, treatment of Nutlin-3 at 2 months of age did not alter the levels of these proteins in the NSPCs isolated from 6-month-old mice ( Fig. 3f-o). Therefore, these data suggest that the long-lasting effect of Nutlin-3 on hippocampal NSPCs might not lead to changes in the intrinsic properties of NSPCs.

Transient Nutlin-3 treatment leads to significant and specific gene expression changes in Fmr1 KO hippocampus
Since transient Nutlin-3 treatment did not have a significant rescue effect on NSPCs isolated 4 months later, we reckoned that Nutlin-3 might exert its long-lasting impact on hippocampal neurogenesis through modulating stem cell niche in the hippocampus. Indeed, Nutlin-3 treatment had a persistent effect on reducing H3 acetylation in hippocampal and cortex tissue of Fmr1-KO mice (Additional file 9: Fig. S2). To explore the potential regulatory mechanisms, we injected 2-month-old Fmr1 KO and WT control mice with Nutlin-3 or vehicle and harvested the hippocampi at 4 months post-treatment for transcriptomic analysis in triplicates (Fig. 4a). Total sequencing reads generated for each sample were between 21 and 26 million (21 × 10 6 < TRs < 26 × 10 6 ) (Additional file 5: Table S2). More than 94% of reads were uniquely mapped to the mouse genome, which corresponds to more than 25 × 10 3 genes (Additional files 5 and 6: Tables S2 and S3). We evaluated the distribution of read counts across the samples and found that the overall density distribution of raw log intensities exhibited a highly consistent pattern (Additional file 9: Fig. S3). Next, we performed differential expression analysis to identify differentially regulated genes (DEGs) among the four experimental groups. Adjusted P-values < 0.05 were used as cutoffs for the differential expression. We identified no DEG between vehicle-treated and Nutlin-3-treated WT (WT-Veh vs WT-Nut3) mice, consistent with a lack of effect by Nutlin-3 on WT mice shown in our published results [20,21] and current neurogenic and behavioral data (Figs. 1, 2, 3, and 4b-d; Additional file 9: Fig. S4, Additional file 7: Table S4). We identified 21 DEGs between vehicle-treated Fmr1 KO (KO-Veh) and WT mice (WT-Veh), of which 13 DEGs were downregulated and 8 genes were upregulated ( Fig. 4d; Additional file 9: Fig. S4, Additional file 7: Table S4). Surprisingly, we found that the most significant gene expression changes were between vehicle-treated and Nulin-3treated Fmr1 KO mice (KO-Veh vs KO-Nut3) (Fig. 4c,d; Additional file 7: Table S4). Out of a total of 237 DEGs (KO-Veh vs KO-Nut3), 6 genes were downregulated and 231 genes were upregulated in Nutlin-3-treated KO mice (Fig. 4d,e; Additional file 7: Table S4). Only 1 DEG, gene Gm21887 or Erdr1 (erythroid differentiation regulator 1), was shared between these two groups of DEGs and was downregulated in KO-Veh both groups ( Fig. 4c; Additional file 7: Table S4). Therefore, transient Nutlin-3-triggered significant gene expression changes in KO mice but did not make gene expression in KO mice to be more comparable to that in WT mice.
To understand the biological significance of DEGs found in Fmr1 KO mice treated with Nutlin-3, we performed Gene Ontology (GO) analysis using three categories of term analysis: biological pathway, cell component, and molecular function (Additional file 8: Table S5). We generated circle plots to demonstrate the specific enrichment and the directionality of the gene expression changes within each GO category (Fig. 5a). The DEGs were generally enriched for extracellular matrix, cell membrane proteins, and secreted factors ( Fig. 5a; Additional file 8: Table S5) known to be the key components of the adult neurogenic niche [43][44][45][46], including the well-known bone morphogenetic protein (BMP) and transforming growth factor beta (TGFß) signaling pathway [47] and insulin-like growth factor 2 (IGF2) pathway [48]. In each GO category, there was a robust upregulation of the DEGs for the enriched terms (Fig. 5a). Because the enriched terms have shown strong potential in stem cell regulation through modulating stem cell niche [43][44][45][46], we next selected a number of candidate DEGs from each group and validated their differential expression in KO-Nut3 compared to KO-Veh using quantitative polymerase chain reaction (qPCR) analysis. Administration of Nutlin-3 in Fmr1 KO mice led to significant changes in the expression levels of genes associated to extracellular matrix (Col8a1 and Timp3), cell membrane (Aqp1 and Tmem72), and secreted factors (Angptl2, Enpp2, BMP6, and Igf2) and the only DEG shared between the two groups (Erdr1), compared to WT littermates ( Fig. 5b-k).
To further explore how these gene expression changes might have happened, we next performed transcription factor (TF) target enrichment analysis to identify potential upstream TFs responsible for observed changes in the gene expression of Nutlin-3-treated Fmr1 KO mice. Our analysis showed that the top TFs are mainly involved in stem cell fate specification (such as MEOX1, MEOX2, PRRX1, BNC2, SOX18, and TWIST1) [49][50][51][52][53] and/or extracellular matrix organization (such as HEYL, TBX18, PRRX1, PRRX2, and TCF21 ) [54][55][56] (Fig. 6a; Additional file 9: Fig. S5). We then assessed the relationship among our top TFs using a published TF network. We found that our top TFs related to Nutlin-3 treatment in KO mice showed a high level of interactions (Fig. 6b). Together, our transcriptomic analysis of the hippocampal tissue supports that the long-lasting rescue effects of Nutlin-3 treatment on impaired adult neurogenesis and dependent cognitive functions of Fmr1 KO mice might be through modulating the adult NSC niche.

Discussion
This study sought to test the hypothesis that transient therapeutic intervention can produce long-lasting beneficial effects on cognitive functions in a mouse model of FXS. Our results demonstrate that a transient Nutlin-3 treatment of young adults for 10 days restored impaired hippocampal neurogenesis and related cognitive abilities in Fmr1 KO mouse for at least 4 months after treatment cessation. Together with our publications [20,21], these findings indicate that not only brief Nutlin-3 treatment rescues the neurogenic and cognitive deficits in adult FXS mice, but also these beneficial effects are sustained long after the end of treatment. Our data also suggest that Nutlin-3 treatment during the early adulthood time window might establish the normal adult NSC niche required for intact neurogenesis and cognitive performances in the absence of FMR1.
Numerous therapeutic alternatives including newly developed compounds or repurposed drugs have been proposed for FXS [20,57]. There are many advantages of drug repurposing in the treatment of disease, including shortening the time frame and reducing the cost associated with new drug development [58]. When assessing the feasibility of initiating treatments, an obvious concern is the resulting toxicity from long-term administration. For these reasons, there has been extensive interest in the possibility of repurposing drugs with potentially long-lasting therapeutic effects. However, only very few studies have assessed the persistent effect of treatment long after treatments are stopped. In a recent study, the minocycline treatment effect has lasted for 4 weeks in young FXS mice but not in adult FXS mice [15]. In another study, transient treatment of FXS rats with lovastatin at 4 weeks of age for 5 weeks prevented the emergence of cognitive deficits in object-place recognition and object-place-context recognition [12]. The authors show that the corrective effect has been sustained for at least 3 months (the last time point tested) after treatment termination, and the observed restoration of normal cognitive function is associated with sustained rescue of both synaptic plasticity and altered protein synthesis [12]. One promising candidate for drug repurposing is a group of MDM2 inhibitors, and its prototype is Nutlin-3. Nutlin-3 is a small molecule that specifically inhibits MDM2, an E3 ubiquitin ligase, and the best known MDM2 targets are tumor suppressor TP53; therefore, Nutlin-3 and its derivative have been worked on extensively and used in clinical trials for cancer treatment [59]. Our lab has found that, in adult NSPCs, FMR1 directly regulates the expression levels and activities of MDM2, which targets TP53 and HDAC1 [20,21]. Our published studies have shown that Nutlin-3 administration at a dosage significantly lower than those used for cancer treatment rescues impaired hippocampal neurogenesis and cognitive functions in both 2-month-old young adult FXS mice and 6-month-old mature adult FXS mice analyzed shortly after the treatment [20,21]. However, the long-lasting effect of Nutlin-3 was unknown. Our current study has addressed this important question and taken one step further to potential therapeutic applications of MDM2 inhibition for the treatment of FXS. Understanding the molecular mechanism underlying drug action is important for both therapeutic application and improvement of drug development. To investigate the mechanism underlying the long-lasting effect of Nutlin-3, we first determined whether this effect was due to the persistent changes in intrinsic properties of NSCs by using primary NSPCs isolated from Fmr1 KO or WT mouse hippocampus. We have previously shown that NSPCs isolated from 2-month-old Fmr1 KO hippocampus had reduced TP53 gene expression, increased proliferation, and reduced neuronal differentiation, which can be corrected by Nutlin-3 treatment [20]. TP53 gene encodes a transcription factor TP53 regulating a network of target genes that play roles in various cellular processes including but limited to apoptosis, cell cycle arrest, genomic integrity, metabolism, redox biology, and stemness [60]. Tp53 binds DNA in a sequence-specific manner and recruits transcriptional machinery components to activate or suppress the expression of a network of target genes [61]. TP53 has also been shown to regulate gene expression through epigenetic mechanisms [62][63][64], which may lead to long-lasting alteration in the gene expression. We therefore hypothesized that Nutlin-3 treatment may exert a sustained therapeutic effect on the FXS mouse model through modulating epigenetic pathways in NSPCs. To our surprise, our results indicate that transient Nutlin-3 treatment did not lead to persistent corrections in active MDM2 levels nor proliferation and differentiation of NSPCs isolated from 6-month-old Fmr1 KO mice. This suggests that, unlike the immediate response to Nutlin-3 treatment, the long-lasting therapeutic effect of Nutlin-3 on neurogenesis might not act mainly through modulating NSPC intrinsic properties [20,21]. Because  neurogenesis is regulated by both NSC intrinsic pathways and extrinsic stem cell niche [65], we performed gene expression profile analysis of Fmr1 KO and WT hippocampal tissue. Nutlin-3 treatment exerted minimal change in the gene expression profile of WT hippocampus which is supported by our previous findings [20,21]. On the other hand, Nultlin-3-treated KO mice mounted persistent and significant gene expression changes compared to vehicletreated KO mice and WT mice. However, Nutlin-3 treatment did not make KO more similar to WT in the gene expression profile. This observation suggests that the lowdosage Nutlin-3 we used was effective in rebalancing cellular pathways in the absence of FMR1 but had no significant effect on WT because the cellular pathways were well regulated by the presence of FMR1. Among DEGs, we found mRNAs of proteins associated to the extracellular matrix, cell membrane, and secreted factors, many of which have been shown to regulate adult neurogenesis [44][45][46]. For example, genes in TGFß and BMP signaling are upregulated in Nutlin-3-treated KO mice, which we have confirmed using qPCR. It has been shown that TGFß and BMP activation in adult NSC niche can activate adult neurogenesis [66,67]. In addition, Igf2 mRNA expression levels were significantly higher in the Nutlin-3-treated KO hippocampus compared to either vehicle-treated KO hippocampus or WT mice. IGF2 has also been shown to promote adult NSC proliferation and neurogenesis [48]. It is possible that the Nutlin-3 has an effect on both niche cells and NSCs, but the long-lasting effects of Nutlin-3 on NSCs can only be observed in the presence of niche cells, which explains why we saw rescue in mice with NSC-targeted FMR1 deletion. Furthermore, since Nutlin-3 was given systemically, we do not rule out that the Nutlin-3 effect might in part act through modulating peripheral systems.
One potential limitation of this study is that we have not defined whether an age range or a critical period exists for the initial Nutlin-3 treatment to achieve the long-lasting effectiveness of Nutlin-3 treatment. Future experiments on Nutlin-3 administration time at younger or older ages than 2 months should be considered. In addition, we showed that the beneficial effects of Nutlin-3 on impaired neurogenesis and cognition of FXS mice sustained for at least 4 months. Whether the effect lasts for a longer period or even for the rest of the animal's life will need to be addressed in future studies. Furthermore, we have assessed adult neurogenesis-dependent behaviors. It is possible that Nutlin-3 also improves other aspects of behavioral deficits in FXS mice which is independent of adult neurogenesis; therefore, it will be beneficial to assess whether the beneficial effects of Nutlin-3 can be generalized to other forms of cognitive and behavior functions found in FXS. Although our transcriptomic analysis has provided an important clue for the long-term effect of Nutlin-3 treatment, a comprehensive assessment of the gene expression and epigenetic profiles of neurogenic niche will be needed to fully understand the molecular basis of the persistent effect of Nutlin-3. Finally, humans and rodents have distinct physiology, lifespan, and genetics [68]. Whether Nutlin-3 treatment can have therapeutic effects on human FXS remains unknown. Although we observed that the effect of Nutlin-3 lasted 4 months which is a long time for mice with a lifespan of 2 years, human has a much longer lifespan and different metabolic rate. Translating our results into human treatment will need substantial further investigations.

Conclusions
In summary, our findings indicate that a brief Nutlin-3 treatment of young adult FXS mice has a long-lasting therapeutic effect on both neurogenesis and behaviors and that the sustained beneficial effect might be exerted through modulating the adult NSC niche. Our observation strengthens the idea that Nutlin-3 is one of the ideal candidates for optimal therapy with minimal toxicity in a targeted therapeutic approach. The findings provide proof-of-concept evidence that FXS, and perhaps neurodevelopmental disorders more generally, may be amenable to transient, early intervention to permanently restore normal cognitive functions.