Enriched environment ameliorates adult hippocampal neurogenesis deficits in Tcf4 haploinsufficient mice

Background Transcription factor 4 (TCF4) has been linked to human neurodevelopmental disorders such as intellectual disability, Pitt-Hopkins Syndrome (PTHS), autism, and schizophrenia. Recent work demonstrated that TCF4 participates in the control of a wide range of neurodevelopmental processes in mammalian nervous system development including neural precursor proliferation, timing of differentiation, migration, dendritogenesis and synapse formation. TCF4 is highly expressed in the adult hippocampal dentate gyrus – one of the few brain regions where neural stem / progenitor cells generate new functional neurons throughout life. Results We here investigated whether TCF4 haploinsufficiency, which in humans causes non-syndromic forms of intellectual disability and PTHS, affects adult hippocampal neurogenesis, a process that is essential for hippocampal plasticity in rodents and potentially in humans. Young adult Tcf4 heterozygote knockout mice showed a major reduction in the level of adult hippocampal neurogenesis, which was at least in part caused by lower stem/progenitor cell numbers and impaired maturation and survival of adult-generated neurons. Interestingly, housing in an enriched environment was sufficient to enhance maturation and survival of new neurons and to substantially augment neurogenesis levels in Tcf4 heterozygote knockout mice. Conclusion The present findings indicate that haploinsufficiency for the intellectual disability- and PTHS-linked transcription factor TCF4 not only affects embryonic neurodevelopment but impedes neurogenesis in the hippocampus of adult mice. These findings suggest that TCF4 haploinsufficiency may have a negative impact on hippocampal function throughout adulthood by impeding hippocampal neurogenesis.


Introduction
The transcription factor 4 (TCF4, Gene ID: 6925) forms together with its paralogues TCF3 and TCF12 the class I basic Helix-Loop-Helix (bHLH) subgroup of transcription factors [1]. TCF4 has received growing attention following the discovery that loss-of-function mutations and single nucleotide polymorphisms are linked to neuropsychiatric disease. In humans, the TCF4 gene consists of 20 exons, 18 of which are protein-coding. Heterozygote lossof-function mutations of TCF4 before exon 7 have been linked to non-specific intellectual disability [2][3][4], while heterozygote disrupting mutations after exon 9 have been causally linked to Pitt-Hopkins Syndrome (PTHS, MIM #610954), a neurodevelopmental disorder characterized by distinctive facial features, moderate to severe intellectual disability, autistic behavior, intermittent breathing abnormalities, seizures [5][6][7]. Moreover, SNPs in non-coding regions of the TCF4 gene are associated with an increased risk of schizophrenia and autism [8][9][10][11].
TCF4 is highly expressed in the adult brain [13], and is thought to have critical function in the regulation of neural plasticity [21][22][23]. High levels of TCF4 expression are observed in the dentate gyrus of the hippocampal formation [13]. The dentate gyrus is one of the few regions of the CNS, where neural stem cells give rise to new neurons throughout adulthood. Adult hippocampal neurogenesis has been unequivocally demonstrated in rodents and non-human primates [24]. Studies using different methodologies to detect adult-born neurons provided strong evidence for the existence of adult hippocampal neurogenesis in humans [25][26][27]. Nevertheless, the notion of adult neurogenesis in humans remains contested as a recent study failed to detect the expression of markers indicative of neurogenesis in the adult human hippocampus [28]. In rodents, adult hippocampal neurogenesis is critical for the regulation of anxiety and depression-like behaviour as well as for hippocampusdependent learning and memory [29]. Notably, impaired adult hippocampal neurogenesis was found to contribute to cognitive deficits in preclinical models for autismspectrum disorders and intellectual disability [30][31][32][33].
Here, we sought to determine whether Tcf4 haploinsufficiency, which has been associated with autism and intellectual disability in humans, affects adult hippocampal neurogenesis. Analysis of Tcf4 heterozygote knockout mice revealed that Tcf4 haploinsufficiency is associated with a smaller size of the hippocampal neural stem/progenitor cell pool and impaired maturation and survival of adult-born dentate granule neurons. Interestingly, long-term housing in an enriched environment enhanced survival of adult-born dentate granule neurons and substantially increased adult hippocampal neurogenesis levels, raising the interesting possibility that in mice behavioural modifications during adulthood can ameliorate a subset of neural deficits caused by TCF4 haploinsufficiency.

Tcf4 haploinsufficiency leads to proliferation deficits in adult neurogenesis
We first performed immunohistochemical analysis against TCF4 and select stage-specific markers to confirm the notion that TCF4 is expressed in the adult hippocampal neurogenic lineage. Indeed, TCF4 co-localized with the radial glia like marker NESTIN (Fig. 1a), with MCM2, a marker for proliferating precursor cells (Fig. 1b), with DCX, a marker for immature neurons (Fig. 1c) and with CALBINDIN, a marker for mature granule cells (Fig. 1d), indicating that TCF4 is expressed during all stages of adult hippocampal neurogenesis. In line with the central function of TCF4 in hippocampal development and our previous report, Tcf4 haploinsufficient mice (Tcf-4Het) showed a significantly reduced volume of dentate gyrus granule cell layer [Volume in µm 3 : control 5.77 × 10 8 ± 3.58 × 10 7 ; Tcf4Het 4.51 × 10 8 ± 2.41 × 10 7 ; p-value = 0.012 (Fig. 2a)] [13].

Discussion
Loss-of-function mutations in the bHLH transcription factor TCF4 are linked to neurodevelopmental disorders such as intellectual disability and Pitt-Hopkins Syndrome. Here, we analysed a Tcf4 heterozygote knockout mouse model to begin to shed light on the dependency of adult neurogenesis in the hippocampal dentate gyrus on Tcf4 gene dosage. Our analyses confirm that constitutive Tcf4 haploinsufficiency is associated with a reduced dentate gyrus size and reveal a profound reduction in the production of new dentate granule neurons during adulthood. We found that Tcf4 haploinsufficiency reduced the proliferative activity in the adult dentate gyrus. Given that Tcf4Het mice showed a 30% reduction in radial-glia like stem / progenitor cells and that the reduction in proliferation (about 20% less BrdU incorporating cells) was in a comparable range, it seems most likely that the proliferation deficit was to a large part the consequence of a smaller radial-glia like stem / progenitor cell pool. It will be interesting to determine, why the Tcf4 haploinsufficient dentate gyrus harbours a smaller stem cell population. Radial-glia like stem / progenitor cells in the murine dentate gyrus are derived from neural precursor cells in the mouse dentate neuroepithelium, which migrate into the primitive dentate region and enter quiescence around postnatal day 7. TCF4 is highly expressed in the dentate neuroepithelium and the developing hippocampus [12,13]. Considering recent findings that TCF4 dosage affects proliferation of embryonic neural precursor cells [14][15][16] and cell migration in the developing CNS [14,17] including the migration of hippocampal neural progenitors [12], it is tempting to speculate that a combination of proliferation and migration defects of neural precursors contribute to the decreased size of the neural stem/progenitor cell pool in the adult dentate gyrus. TCF4 and the related E-protein TCF3 inhibit neural stem cell differentiation and cell cycle exit in the adult subventricular zone [37] raising the alternative possibility that Tcf4 haploinsufficiency results in accelerated stem cell depletion due to premature neuronal fate commitment.
Tcf4 haploinsufficiency had a profound impact on survival of adult generated cells. It is possible that TCF4 directly regulates the expression of, e.g., anti-apoptotic pathways in the adult neurogenic lineage. Adult-born neuron survival is highly dependent on maturation and synaptic integration [38][39][40]. Recent studies reported that loss-of-Tcf4 causes delayed maturation of embryonically and early postnatally born neurons, impairs dendrite and synapse formation in the developing cortex [17,41], and decreases spine density of mature cortical and hippocampal neurons [42]. When analysing the marker profile of 4-week old adult-born neurons, we found that    How TCF4 regulates maturation of neurons on the molecular level remains to be determined. E-box proteins such as TCF4, TCF3 and TCF12 function as transcription factors as homodimers or through formation of heterodimers with bHLH class II transcription factors such as the proneural transcription factors Neu-rog1 and 2 and the transcription factors of the NeuroD family [1]. NeuroD1 and NeuroD2 have been implied in the maturation of adult-born hippocampal neurons [43,44] and it will be interesting to test whether their interaction with TCF4 is required for dentate granule neuron maturation.
Given our primary goal to gain first insight how Tcf4 haploinsufficiency, which has been causally linked to neurodevelopmental disorders such as Pitt-Hopkins Syndrome and intellectual disability, affects adult neurogenesis, we analysed constitutive heterozygote Tcf4 knockout mice. TCF4 is broadly expressed and is critical for the development of a number of neural and non-neural cell types [18,[45][46][47][48]. It therefore remains to be determined how decreased TCF4 dosage in different cell populations contributes to the impairment in adult hippocampal neurogenesis.
We made the interesting observation that long-term exposure to an enriched environment substantially increased the generation of new neurons with a mature marker profile, indicating that behavioural modifications and environmental stimulation may ameliorate TCF4 dosage-dependent defects. Exposure to an enriched environment promotes hippocampal network activity and stimulates adult-born neuron survival and maturation [35,49]. Interestingly, recent work demonstrated that the function of TCF4 is neuronal activity dependent [18,50,51] raising the intriguing possibility that enriched environment ameliorated hippocampal neurogenesis deficits through modulation of TCF4 activity. Previous studies showed that TCF4 dosage affects hippocampus-dependent behaviour [23,51]. It will be interesting to determine whether deficits in adult neurogenesis contribute to hippocampal dysfunction in Tcf4 haploinsufficient mice and whether behavioural modifications such as enriched environment can ameliorate Tcf4 haploinsufficiency associated hippocampus-dependent cognitive deficits in adult mice.

Conclusion
Our findings suggest that in rodents Tcf4 haploinsufficiency may have a continuous negative impact on hippocampal function by perturbing the physiological formation of new neurons in the adult dentate gyrus. Moreover, our findings raise the interesting possibility that behavioural interventions may allow to ameliorate a subset of Tcf4 haploinsufficiency associated neural deficits during adulthood.

Animals and Ethics Statement
All animal experiments were conducted in accordance with the European Communities Council Directive (86/609/EEC) and received ethical approval by the committee for Animal Research of the Bavarian State authorities. The generation of the knockout allele has been described previously [13]. All animals-except for the animals in the enriched environment experimentswere housed in standard cages (size: 37 × 21 × 15 cm) with 3-4 mice per cage under a 12 h light/dark cycle with unlimited access to water and standard rodent food. Mice were housed in the animal facilities of the Helmholtz Center Munich and the Friedrich-Alexander-Universität Erlangen. Animal care was in accordance with institutional guidelines.
Genotyping of the mice was done using PCR and the following primers: Tcf4Hetfwd MutTCG TGG TAT CGT TAT GCG CC. fwd WTCCG ATG ACA GTG ATG ATG GT. revAAG TTA AGC TGA AGT AAA TAC CCA CA. lacZ fwdATC ACG ACG CGC TGT ATC. lacZ revACA TCG GGC AAA TAA TAT CG.

BrdU injections and Enriched environment
At the age of eight weeks intraperitoneal 5-bromo-2′deoxyuridine (BrdU) injections were performed twice a day on three consecutive days (0.1 mg BrdU/g body weight, per dose). Animals were sacrificed either three hours (6 control and 6 haploinsufficient mice) after last injection or after additional four weeks (9 control and 5 haploinsufficient mice). An additional cohort of eight week old Tcf4 haploinsufficient mice (5 animals) was placed in an enriched environment (EE cage size: 60 × 26 × 33 cm, containing running wheels, toys, tunnels and nest materials). Mice were injected twice intraperitoneal with BrdU on three consecutive days (0.1 mg BrdU/g body weight, per dose). Animals were sacrificed four weeks after the last injection.

Tissue preparation
Mice were killed using CO 2 and perfused transcardially with PBS for 5 min (20 ml/min), followed by fixation with 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB), pH 7.4, for 5 min. Brains were postfixed overnight in 4% PFA at 4 °C, followed by dehydration in 30% sucrose in 0.1 M PBS at 4 °C. Brains were cut coronally at 40 µm, using a sliding microtome. Sections were stored at − 20 °C.

Immunohistochemistry
Free-floating sections were rinsed six times for 10 min in Tris-buffered saline (TBS: 1 M Tris-HCL, pH 7.5/0.9% NaCl), and incubated for 72 h at 4 °C with primary antibodies (Table 1) in blocking solution, containing 0.25% Triton X-100 and 3% donkey serum in TBS. Sections were rinsed in TBS six times for 10 min, incubated overnight at 4 °C with fluorochrome-labelled secondary antibodies ( Table 2) diluted in blocking solution, and rinsed in TBS three times for 10 min. Nuclei were counterstained with DAPI (1:10,000 in 1xTBS) for 10 min, followed by three rinses in TBS for 10 min. Sections were mounted on slides and covered with Aqua-Poly/Mount (Polysciences).
For BrdU stainings, slices were first stained for all antigens of interest except for BrdU. Slices were then postfixed in 4% PFA for 10 min at room temperature. Sections were rinsed three times in TBS, incubated in 2 N HCl for 10 min at 37 °C. After two rinses in 0.1 M borate buffer, sections were washed three times with PBS. Detection of BrdU immunoreactivity was conducted as described above.

Imaging and quantification
For volume, BrdU, and MCM2 quantification, fluorescence signal was detected with an AF6000 Modular Systems Leica fluorescent microscope and documented with a SPOT-CCD camera and the Leica software LAS AF (Version 2.6.0.7266; Leica Microsystems, Wetzlar Germany). For co-localization analyses, fluorescence signal was detected using a Zeiss LSM 780 confocal microscope with four lasers (405, 488, 550, and 633 nm) and × 25 and × 40 objective lens. Images were processed using ImageJ.
For each animal, a series of every 12th section of the dentate gyrus was selected. Volume measurements were performed with ImageJ by tracing the granular zone of dentate gyrus. For BrdU and MCM2 quantification, cells in the granule cell layer and contiguous subgranular zone were counted (52). For co-localization analyses, all BrdU + cells within the granule cell layer and contiguous subgranular zone in at least one section were analysed for expression of DCX or PROX1. A minimum 50 cells per animal were analysed per marker and animal. Statistical analysis was performed with GraphPad Prism (Graphpad Software Inc.), using unpaired two tailed t-test. The data are expressed as mean values ± SEM. Significant differences were assumed at a level of p < 0.05.

Antibodies
See Tables 1 and 2.