Brain atlas for glycoprotein hormone receptors at single-transcript level

There is increasing evidence that anterior pituitary hormones, traditionally thought to have unitary functions in regulating single endocrine targets, act on multiple somatic tissues, such as bone, fat, and liver. There is also emerging evidence for anterior pituitary hormone action on brain receptors in mediating central neural and peripheral somatic functions. Here, we have created the most comprehensive neuroanatomical atlas on the expression of TSHR, LHCGR, and FSHR. We have used RNAscope, a technology that allows the detection of mRNA at single-transcript level, together with protein level validation, to document Tshr expression in 173 and Fshr expression in 353 brain regions, nuclei and subnuclei identified using the Atlas for the Mouse Brain in Stereotaxic Coordinates. We also identified Lhcgr transcripts in 401 brain regions, nuclei and subnuclei. Complementarily, we used ViewRNA, another single-transcript detection technology, to establish the expression of FSHR in human brain samples, where transcripts were co-localized in MALAT1-positive neurons. In addition, we show high expression for all three receptors in the ventricular region—with yet unknown functions. Intriguingly, Tshr and Fshr expression in the ependymal layer of the third ventricle was similar to that of the thyroid follicular cells and testicular Sertoli cells, respectively. In contrast, Fshr was localized to NeuN-positive neurons in the granular layer of the dentate gyrus in murine and human brain—both are Alzheimer’s disease-vulnerable regions. Our atlas thus provides a vital resource for scientists to explore the link between the stimulation or inactivation of brain glycoprotein hormone receptors on somatic function. New actionable pathways for human disease may be unmasked through further studies.


Introduction
There is increasing evidence that pituitary hormones traditionally thought of as 'pure' regulators of single physiological processes affect multiple bodily systems, either directly or via actions on brain receptors (Zaidi et al., 2018;Abe et al., 2003). We established, for the first time, a direct action of thyroid-stimulating hormone (TSH) on bone and found that TSH receptor (TSHR) haploinsufficiency causes profound bone loss in mice (Abe et al., 2003). We also found that follicle-stimulating hormone (FSH), hitherto thought to solely regulate gonadal function, displayed direct effects on the skeleton to cause bone loss (Sun et al., 2006), and on fat cells, to cause adipogenesis and body fat accumulation . Likewise, we showed that hormones from the posterior pituitary, namely, oxytocin and vasopressin, displayed direct, but opposing, skeletal actions-effects that may relate to the pathogenesis of bone loss in pregnancy and lactation, and in chronic hyponatremia, respectively Sun et al., 2016;Tamma et al., 2009;Tamma et al., 2013). To add to this complexity, and in addition to the poorly recognized ubiquity of pituitary hormone receptors, the ligands themselves, or their variants, are expressed widely. We find the expression of a TSHβ variant (TSHβv) in bone marrow macrophages, while oxytocin is expressed by both osteoblasts and osteoclasts (Colaianni et al., 2011;Colaianni et al., 2012;Baliram et al., 2013;Baliram et al., 2016). These studies have together shifted the paradigm from established unitary functions of pituitary hormones to an evolving array of yet unrecognized roles of physiological and pathophysiological importance.
There is a compelling body of literature to support the expression of oxytocin receptors in various brain regions, and their function in regulating peripheral actions, such as social behavior and satiety Bale et al., 2001). However, there is relatively scant information on the expression and, importantly, function of the anterior pituitary glycoprotein hormone family of receptors, namely, FSHR, TSHR, and luteinizing hormone/human chorionic gonadotropin receptor (LHCGR). Discrete sites of the rat, mouse, and human brain express receptors for these hormones, with several studies pointing to their relationship to neural functions, such as cognition, learning, neuronal plasticity, and sensory perception, as well as to neuropsychiatric disorders, including affective disorders and neurodegeneration (Crisanti et al., 2001;Emanuele et al., 1985;Lei et al., 1993;Luan et al., 2020;Bi et al., 2020;Blair et al., 2019;Apaja et al., 2004;Naicker and Naidoo, 2018; Table 1). In the light of such discoveries, the link between the stimulation of these receptors in the brain and the regulation of peripheral physiological processes needs further investigation.
Here, we use RNAscope-a cutting-edge technology that detects single RNA transcripts-to create the most comprehensive atlas of glycoprotein hormone receptors in mouse brain. This compendium of glycoprotein hormone receptors in concrete brain regions and subregions at a single-transcript level should allow investigators to study both peripheral and central effects of the activation of individual receptors in health and disease. Our identification of brain nuclei with the highest density for each receptor should also create a new way forward in understanding the functional engagement of receptor-bearing nuclei within a large-scale functional network.

Results
Very little is known about the function(s) of anterior pituitary hormone receptors in the brain, except for isolated studies showing a relationship with cognition and affect ( Table 1). We therefore used RNAscope to map the expression of Tshr, Lhcgr, and Fshr in the mouse brain; immunofluorescence and qPCR to provide confirmatory evidence for Tshr and Fshr expression; and ViewRNA and qPCR to examine for FSHR expression in AD-vulnerable regions of the human brain. RNAscope, which allows the detection of single transcripts, uses ~20 pairs of transcript-specific double Z-probes to hybridize 10-µm-thick whole-brain sections. Preamplifiers first hybridize to the ~28-bp binding site formed by each double Z-probe; amplifiers then bind to the multiple binding sites on each preamplifier; and finally, labeled probes containing a fluorescent molecule bind to multiple sites of each amplifier. RNAscope data was quantified on sections from coded mice. Each section was viewed and analyzed using CaseViewer 2.4 (3DHISTECH, Budapest, Hungary) or QuPath v.0.2.3 (University of Edinburgh, UK). The Atlas for the Mouse Brain in Stereotaxic Coordinates (Paxinos and Franklin, 2007) was used to identify every nucleus or subnucleus in which we manually counted Tshr, Lhcgr, or Fshr transcripts in every tenth section using a tag feature. Repeat counting of the same section agreed within <2%. Receptor density was calculated by dividing transcript count by the total area (µm 2 , ImageJ) of each region, nucleus or subnucleus. Photomicrographs were prepared using Photoshop CS5.1 (Adobe) only to adjust brightness, contrast, and sharpness, remove artifacts (i.e., obscuring bubbles), and make composite plates.
Tshr was detected bilaterally in 173 brain nuclei and subnuclei, in the following descending order of transcript densities: ventricular region, olfactory bulb, forebrain, hypothalamus, medulla, cerebellum, midbrain and pons, cerebral cortex, hippocampus, and thalamus ( Figure 1A, Figure 1source data 1). Importantly, thyroid glands from Tshr -/mice did not show a signal, proving probe specificity ( Figure 1B). Tshr expression in pooled brain samples was confirmed by qPCR ( Figure 1C, Figure 1-source data 2). The hypothalamus and hippocampus expressed Tshr, with hypothalamic expression being considerably higher (p<0.01) in females than in males. Furthermore, within other regions of the brain, highest Tshr densities were as follows: ependymal layer of the third ventricle (slightly higher than the thyroid follicular cells); VTT in the olfactory bulb; HDB in the forebrain; MTu in the hypothalamus; SoIV in the medulla; PFI in the cerebellum; LDTg in midbrain and pons; DP in the cerebral cortex; DG in hippocampus; and PPT in the thalamus ( Figure 1D, Figure 1-source data  For purposes of replicability, we employed a complementary approach to study brain Tshr expression-the Tshr-deficient mouse-in which exon 1 of the Tshr gene is replaced by a Gfp cassette. This reporter strategy allows for the in vivo display of Tshr locations using GFP immunoreactivity (GFP-ir) as a surrogate for Tshr expression (Abe et al., 2003). Of note is that the Tshr +/-(haploinsufficient) mouse has one Tshr allele intact with normal thyroid function but expresses GFP in lieu of one lost allele. In contrast, the Tshr +/+ mouse does not express GFP-ir because both Tshr copies are intact and are therefore our negative control.
Consistent with our RNAscope finding, profound GFP-ir was noted in the ependymal region of the third ventricle, mostly in NeuN-negative cells, but with some neuronal localization ( Figure 1E). The SVZ of the lateral ventricles, and the SI, and dorsal and ventral BNST of the forebrain also showed GFPir, but immunoreactivity was much lower than the ependymal layer of the third ventricle ( Figure 1F). In all, while there was overall concordance between the two methodologies for high Tshr-expressing areas, GFP-ir was not detected in a number of Tshr-positive regions. This latter discrepancy most likely reflects the grossly lower sensitivity of immunohistochemical detection.
There is evidence that high LH levels in postmenopausal women correlate with a higher incidence of Alzheimer's disease (AD) (Henderson et al., 1994;Rocca et al., 2007); LHβ transgenic mice are cognitively impaired Casadesus et al., 2007;LH receptors (LHCGR) are present in the hippocampus (Rao, 2017;Liu et al., 2007); and hCG induces cognitive deficits in rodents (Berry et al., 2008;Barron et al., 2010). Thus, we mapped Lhcgr in mouse brain to document expression in 401 brain nuclei and subnuclei. Probe specificity was established by a positive signal in testicular Leydig cells, and with an absent signal in juxtaposed Sertoli cells (Figure 2A). Notably similar to Tshr transcripts, the ventricular regions displayed the highest transcript density ( Figure 2B, Figure 2-source data 1). Among the brain divisions, the densities were as follows: OV in the ventricular region; SFO in the forebrain; PFI in the cerebellum; MiA in the olfactory bulb; SCO in the thalamus; PMD in the hypothalamus; MVPO in the medulla; DT in midbrain and pons; GrDG in the hippocampus; and SL in the cerebral cortex ( Figure 2C, Figure 2-source data 1). Raw transcript counts in each region and representative micrographs are shown in We recently reported the expression of FSHR in mouse, rat, and human brains, particularly in AD-vulnerable regions, including hippocampus and cortex (Xiong et al., 2022). We also found that FSH exacerbated AD-like neuropathology and cognitive decline in 3xTg, APP/PS1, and APP-KI mice, while the inhibition of FSH action rescued this phenotype. Most notably, shRNA-mediated knockdown of the Fshr in the hippocampus prevented the onset of AD-like features (Xiong et al., 2022).
Here, using RNAscope, we report the expression of Fshr at the single-transcript resolution in 353 brain nuclei and subnuclei-and suggest that FSHR in the brain may have roles beyond cognition. Probe specificity was established by a positive signal in testicular Sertoli cells, and an absent signal in set at p<0.05. (D) Tshr transcript density in nuclei and subnuclei of the ventricular regions, olfactory bulb, forebrain, hypothalamus, medulla, cerebellum, midbrain and pons, cerebral cortex, hippocampus, and thalamus. (E) Abundant GFP immunofluorescence (green) was detected in the ependymal layer of the third ventricle in Tshr +/heterozygous mice, in which a GFP cassette replaced exon 1 of the Tshr gene. This GFP signal was absent in Tshr +/+ mice. (F) GFP immunofluorescence was also detected in the subventricular zone (SVZ) of the lateral ventricle, and substantia innominata (SI) and dorsal and ventral bed nucleus of stria terminalis (BNST) in the forebrain of the Tshr +/mice. Sections were co-stained with DAPI (blue) and a neuronal marker, NeuN (red). Scale bar: 100 µm.
The online version of this article includes the following source data and figure supplement(s) for figure 1: Source data 1. Tshr density in brain regions, nuclei, and subnuclei.
Source data 2. Tshr mRNA expression levels in mouse tissues (qPCR).       juxtaposed Leydig cells and in the testes of Fshr -/mice-as negative controls ( Figure 3A). Immunofluorescence confirmed the expression of FSHR in NeuN-positive neurons, but not in GFAP-positive glial cells or IBA1-positive microglia ( Figure 3B). Fshr transcript density was highest in the ventricular region, followed, in descending order, by the cerebellum, olfactory bulb, hippocampus, cerebral cortex, medulla, midbrain and pons, forebrain, thalamus, and hypothalamus ( Figure 3C, Figure 3-source data 1). Within each region, respectively, the highest transcript densities were as follows: ependymal layer of the third ventricle (slightly higher than the testicular Sertoli cells); PFI in the cerebellum; GrA in the olfactory bulb; GrDG in the hippocampus; AIV in the cerebral cortex; RMg in the medulla; MHb in the thalamus; IPDL in midbrain and pons; aci in the forebrain; and ArcL in the hypothalamus ( Figure 3D We used ViewRNA to examine the expression of FSHR transcripts in specific regions of the human brain ( Figure 4A). Expression was noted in neuronal cells co-expressing the noncoding RNA MALAT1 in the GrDG-consistent with the RNAscope data in mouse brain-and in the parahippocampal cortex. This latter data is consistent with FSHR expression in a population of excitatory glutamatergic neurons noted in human brain by 10× single-cell RNA-seq (Allen Brain Atlas). Affymetrix microarray analysis confirmed FSHR expression in the frontal, cingulate, temporal, parietal, and occipital subregions of human cortex in postmortem normal and AD brains ( Figure 4B, Figure 4-source data 1). Interestingly, FSHR expression trended to be higher in the frontal cortex of the AD brains compared to that of unaffected brains (p=0.060). In all, the data suggest that, beyond a primary role in regulating cognition, brain FSHR may have a wider role in the central regulation.

Discussion
The past decade has witnessed the unraveling of nontraditional physiological actions of anterior pituitary glycoprotein hormones, and hence, the unmasking of functional receptors in bone, fat, brain, and immune cells, among other organs (Zaidi et al., 2018;Sun et al., 2006;Liu et al., 2017;Liu et al., 2015;Williams, 2011;Sun et al., 2020;Fields and Shemesh, 2004). We report here for the first time that Tshr, Lhcgr, and Fshr are expressed in multiple brain regions. The data provide new insights into the distributed central neural network of anterior pituitary hormone receptors, particularly in relation to their role in regulating the somatic tissue function. Specifically, we find a surprising and striking overlap in central neural distribution of the three receptors-with highest transcript densities in the ventricular regions. Furthermore, at least for the TSHR and FSHR, expression levels in ependymal layer of the third ventricle was similar to that of the thyroid follicular cells and testicular Sertoli cells, respectively. Albeit intriguing, this may suggest a primary role for these receptors in central neural regulation.
Among 173 Tshr-positive brain regions, subregions, and nuclei, the ependymal layer of the third ventricle displayed the highest Tshr transcript number and density. This region is juxtaposed to the anterior pituitary that produces TSH in response to hypothalamic TRH. Furthermore, TSH has been reported to be expressed in the hypothalamus (DeVito et al., 1986;Hojvat et al., 1983). It is therefore possible that a yet uncharacterized central TSH-TSHR feedback circuit may directly regulate the hypothalamic-pituitary-thyroid axis, thought solely to be controlled by thyroid hormones. To add to this complexity, thyroxine-to-triiodothyronine conversion occurs in tanycytes (Fonseca et al., 2013), which calls into question whether central TSH actions regulate thyroid hormone metabolism in these The online version of this article includes the following source data and figure supplement(s) for figure 2: Source data 1. Lhcgr density in brain regions, nuclei, and subnuclei.      cells and/or directly modulate hypothalamic TRH neuronal projections. Interestingly, it has been shown that Tshr expression is not different between young and old mice (Kerp et al., 2019). However, there is conflicting evidence for the expression of TSH with age-with evidence of no difference between 6-, 15-, and 22-month-old mice (Wang et al., 2019), but a 44% increase in the old rat compared with the young rat (Miler et al., 2019). The forebrain and olfactory bulb also displayed abundant Tshr transcripts, with the highest density in the nucleus of the horizontal limb of the diagonal band (HDB) of the forebrain and ventral tenia tecta (VTT) of the olfactory bulb. These regions are involved, respectively, in learning and odor processing (Shiotani et al., 2020;Cleland and Linster, 2019;McNamara et al., 2004;Chaves-Coira et al., 2018;Zhan et al., 2013). In the hypothalamus, the highest density was found in medial tuberal nucleus (MTu), which controls ingestive behaviors and metabolism (Luo et al., 2018). Finally, we found density in the testis and various brain regions detected by RNAscope. (D) Fshr transcript density in nuclei and subnuclei of the ventricular regions, cerebellum, olfactory bulb, hippocampus, cerebral cortex, medulla, midbrain and pons, forebrain, thalamus, and hypothalamus.
The online version of this article includes the following source data and figure supplement(s) for figure 3: Source data 1. Fshr density in brain regions, nuclei and subnuclei.     more recently that the modulation of TSHRs in the bed nucleus of the stria terminalis (BNST), which receives direct afferents from the MTu (Dong and Swanson, 2006), influences anxiety responses, suggesting that TSHR signaling might, in fact, mediate psychosocial behaviors.
While LH has a key role in reproduction and sexual development, we found 401 brain regions, subregions, and nuclei expressing Lhcgr. There were nominal differences in Lhcgr expression in many brain regions, but the ventricles stood out as having the highest Lhcgr density. Two regions deserve special mention. The Lhcgr-rich mitral cell layer of the accessory olfactory bulb (MiA) has a known role in scent communication during mating (Gildersleeve et al., 2012;Lydell and Doty, 1972;Huck and Banks, 1984;Singh and Bronstad, 2001). A growing body of evidence suggests that men are attracted to cues of impending ovulation in women, raising an intriguing question on whether cycling hormones affect men's attraction and sexual behavior (Gildersleeve et al., 2012;Singh and Bronstad, 2001). The broader question is whether LH surges in women during cycling may, in fact, alter male sexual behavior through central mechanisms. Second, a high Lhcgr density in the subfornical organ (SFO) of the forebrain was surprising. SFO sends efferent projections to the organum vasculosum of the lamina terminalis (OVLT) (Miselis, 1981;Lind, 1986), which is surrounded by GnRH neurons and contains estrogen receptors (ESR) (Low, 2016). We therefore speculate that circumventricular interactions between LHCGR, LH, GnRH, and ESR underpin the central regulation of reproduction.
RNAscope revealed 353 Fshr-expressing brain regions, subregions, and nuclei. Highest expression was noted in the ependymal layer, not surprisingly given its anatomical proximity to the anterior pituitary gland where FSH is produced in response to hypothalamic gonadotropin-releasing hormone (GnRH). The functional significance of Fshr expressed in the cerebellum, particularly in the paraflocculus (PFI), is yet unknown. However, other Fshr-high subregions, including the granular cell layer of the accessory olfactory bulb (GrA), granular layer of the dentate gyrus (GrDG), and agranular insular cortex (AIV), have known associations with odor processing, learning, memory formation, and anticipation of reward (Eichenbaum, 2001;Nagayama et al., 2014;Kesner and Gilbert, 2007). It is possible that the anosmia of Kallman syndrome, with unclear etiology, may arise from a dysfunctional FSHR-olfaction circuitry. We also find that inactivation of the hippocampal Fshr blunts the cognitive impairment and AD-like neuropathology induced by ovariectomy in 3xTg mice. This data, together with gain-and loss-of-function studies, suggests that hippocampal and cortical FSHR could represent therapeutic targets for AD.
In all, our results provide compelling evidence for multiple central nodes being targets of the anterior pituitary glycoprotein hormones-a paradigm shift that does not conform with the dogma that pituitary hormones are solely master regulators of single bodily processes. Through the intercession of emerging technologies, we compiled the most complete atlas of glycoprotein hormone receptor distribution in the brain at a single-transcript resolution. In addition, we have identified brain sites with the highest transcript expression and density, findings that are imperative toward a better understanding of the neuroanatomical and functional basis of pituitary hormone signaling in the brain. This understanding should provide the foundation for innovative pharmacological interventions for a range of human diseases, wherein direct actions of pituitary hormones have been implicated, importantly, AD.

Mice
We used Tshr +/-(strain #004858, Jackson Laboratory), Lhcgr -/-(strain #027102, Jackson Laboratory), Fshr -/mice (Dierich et al., 1998), and their wild-type littermates in this study. Adult male mice (~3-4-month-old) were housed in a 12 hr:12 hr light:dark cycle at 22 ± 2°C with ad libitum access to water and regular chow. All procedures were approved by the Mount Sinai Institutional Animal Care and Use Committee (approval number IACUC-2018-0047) and are in accordance with Public Health Service and United States Department of Agriculture guidelines.

RNAscope
Mouse brain tissue was collected for RNAscope. Briefly, mice were anesthetized with isoflurane (2-3% in oxygen; Baxter Healthcare, Deerfield, IL) and transcardially perfused with 0.9% heparinized saline followed by 4% paraformaldehyde (PFA). Brains were extracted and post-fixed in 4% PFA for 24 hr, dehydrated, and embedded into paraffin. Coronal sections were cut at 5 μm, with every tenth section mounted onto ~20 slides with 2-6 sections on each slide. This method allowed to cover the entire brain and eliminate the likelihood of counting the same transcript twice. Sections were air-dried overnight at room temperature and stored at 4°C until required.
Slides were baked at 60°C for 1 hr, deparaffinized, incubated with hydrogen peroxide for 10 min at room temperature, pretreated with Target Retrieval Reagent for 20 min at 100°C and with Protease III for 30 min at 40°C. Probe hybridization and signal amplification were performed as per the manufacturer's instructions for chromogenic assays.
Following RNAscope assay, the slides were scanned at ×20 magnification and the digital image analysis was successfully validated using the CaseViewer 2.4 (3DHISTECH) software. The same software was employed to capture and prepare images for the figures in the article. Detection of Tshr-, Lhcgr-, and Fshr-positive cells was also performed using the QuPath-0.2.3 (University of Edinburgh, UK) software based on receptor intensity thresholds, size, and shape.

Histology and immunofluorescence
Heterozygous Tshr +/in which a GFP cassette replaced exon 1 of the Tshr gene and their Tshr +/+ littermates were euthanized with carbon dioxide and perfused transcardially with 0.9% heparinized saline followed by 4% PFA in 0.1 M phosphate-buffered saline (PBS; pH 7.4). Brains were collected and post-fixed in the same fixative overnight at 4°C, then transferred to a 30% sucrose solution in 0.1 M PBS with 0.1% sodium azide and stored at 4°C until they were sectioned on a freezing stage sliding microtome at 30 μm. Sections were stored in 0.1 M PBS solution with 0.1% sodium azide until processed for double immunofluorescence.
For the double-label fluorescent immunohistochemistry, free-floating brain sections were rinsed in 0.1 M PBS (2 × 15 min), followed by a 30 min blocking in 3% normal horse serum (Vector Laboratories, Burlingame, CA) and 0.3% Triton X-100 in 0.1 M PBS. Sections were incubated with a mixture of primary rabbit anti-GFP antibody (1:500; Cat# SP3005P, OriGene, Rockville, MD) and mouse anti-NeuN antibody (1:1000; Cat# ab104224, Abcam, Cambridge, MA) for 18 hr. Sections were then incubated with the secondary donkey anti-rabbit Alexa 488 (1:700; Cat# 711-545-152, Jackson Immu-noResearch, West Grove, PA) and donkey anti-mouse DyLight 594 (1:700; Cat# DK-2594, Vector Laboratories) antibodies in 0.1 M PBS for 3 hr at room temperature. For immunohistochemical controls, the primary antibody was either omitted or pre-adsorbed with the immunizing peptide overnight at 4°C, resulting in no immunoreactive staining. In addition, we expectedly did not detect GFP immunoreactivity (-ir) in the Tshr +/+ littermates as the Tshr gene was intact and did not express GFP. Sections were mounted onto slides (Superfrost Plus) and cover-slipped using ProLong Gold Antifade Reagent (Life Technologies, Grand Island, NY). All steps were performed at room temperature.

Microarray analysis
Affymetrix Human Genome U133 Plus 2.0 Array data for FSHR expression in the frontal, cingulate, temporal, parietal, and occipital cortex from both AD and non-AD human brains were curated from a previously published dataset (GEO accession #GSE84422; Wang et al., 2016).

Quantitative PCR
For quantitative RT-PCR performed on homogenates of brain tissues, total RNA from the hypothalamus and the hippocampus isolated from five Tshr +/+ mice was extracted using an RNeasy Mini kit (QIAGEN) as per the manufacturer's protocol. Thyroid and liver tissues were used as positive and negative controls, respectively. RNA was treated with DNAse I (Invitrogen), and reverse-transcribed using the SuperScript II Reverse Transcriptase (Thermo Fisher Scientific). qPCR was performed with a QuantStudio 7 Real-Time PCR system (Applied Biosystems). PCR reaction mix consisted of first-strand cDNA template, exon-spanning primer pairs, and SYBR Green PCR master mix (Thermo Fisher Scientific). Expression of the selected targets was compared to that of a panel of normalizing genes (Rps11, Tubg1, and Gapdh) measured on the same sample in parallel on the same plate, giving a Ct difference (ΔCt) for the normalizing gene minus the test gene. Relative expression levels were calculated by 2 -ΔΔCt using thyroid as the reference tissue.

Quantitation, validation, and statistical analysis
Immunofluorescent images were viewed and captured using ×10 or ×20 objectives with an Observer. Z1 fluorescence microscope (Carl Zeiss, Germany) with appropriate filters for Alexa 488, Cy3, and DAPI. The captured GFP and NeuN images were evaluated and overlaid using AxioVision v.4.8 software (Carl Zeiss, Germany) and ImageJ (NIH, Bethesda, MD).