The Effects of Atoh8 on Postnatal Murine Neurogenesis

Neural precursor cells, which proliferate in the ventricular zone, give rise to a wide variety of neurons and glial cells [1, 2]. The process of neurogenesis consists of two sequential steps: the initial neuronal fate determination and the subsequent neuronal differentiation/maturation. It has been shown that both steps of neurogenesis are controlled by multiple basic helix-loop-helix (bHLH) transcription factors [1]. In addition, the rate of neurogenesis is highest in neonates at 55% of total neuron formation and falls to 2% by adulthood [3‒7]. These results suggest that Atoh8 affects the proliferation and neuroblast formation in the brain and does not influence their migration per se. However, other bHLH transcription factors are likely to modulate cell fate independently of migration [8].

On a cellular level, neurogenesis in higher vertebrates is driven by neuroblasts, which arise from the asymmetric division of stem cells along the lateral ventricles and the adjacent subventricular zone (SVZ) [9, 10]. Thereby, two sequential steps, the initial neuronal fate determination and the subsequent neuronal differentiation/maturation, are characteristic processes of neurogenesis. In mice, newly formed neuroblasts express doublecortin (DCX) within a few hours for up to 4 weeks, allowing their detection [11]. From the SVZ, neuroblasts proliferate and migrate in chains through the rostral migratory stream (RMS) to the olfactory bulb (OB), where they differentiate into granular cells (GCs) and periglomerular cells (PGCs) [12, 13].

In this context, during the process of development, multiple bHLH transcription factors play a pivotal role in regulating the expression of genes that are critical for determining cell fate and promoting differentiation in neurogenesis [14‒16]. These factors can act either as transcriptional activators or repressors [17, 18].

The mouse atonal homolog 8 (Math6, Atoh8), a bHLH transcription factor, was initially discovered in the context of murine neurogenesis [14]. It was found to be expressed in the early neuronal precursor cells located in the ventricular zone and later in a specific subset of differentiating and mature neurons [14]. Interestingly, in the developing retina, Atoh8 was demonstrated to induce neurogenesis while simultaneously repressing gliogenesis [14]. This suggests that Atoh8 is essential not only for determining the fate of neuronal precursor cells but also may regulate the functioning of mature neurons. Furthermore, Atoh8 is activated by bone morphogenetic protein 14, which in turn acts as a modulator of retinal neuron differentiation. The loss of Atoh8 was observed to inhibit the differentiation of retinal stem cells, further emphasizing its role in promoting neuronal differentiation [19].

In summary, Atoh8, as a bHLH transcription factor, plays a crucial role in neurogenesis and retinal development. Its expression in both early neuronal precursor cells and mature neurons suggests that it has dual functions in determining cell fate and regulating the functioning of neurons. The findings in mice also support Atoh8’s importance as a regulator of retinal neuron differentiation, further influenced by BMP14 signaling.

In this study, we investigate the distribution of neuroblasts in Atoh8 knockout based on alteration in the number of neuroblasts, migration, and proliferation. In addition, the distribution of neurons in the OB was analyzed. Anti-neuronal-nuclei (NeuN) protein, a neuron-specific marker, is expressed exclusively in nervous tissue and is generally used to study postmitotic neurons and differentiated cells [20]. Calretinin, a Ca²⁺-binding protein, is involved in calcium signaling and is also highly expressed in neurons, including within the retina [21].

The studies were performed under the terms of the German animal protection law. Breeding and genotyping of Atoh8^−/− strain was performed as previously described [22]. We used male homozygous M6KO mice as well as wild-type (WT) mice at postnatal days 10, 21, and 56 (P10, P21, and P56). The use of only male mice in the study was primarily driven by animal husbandry considerations. An additional factor was the ethical and political discourse surrounding animal welfare in scientific research. Including female mice would have effectively doubled the required number of animals. Given the heightened public and political focus on minimizing animal use in research, any increase in animal numbers needs rigorous justification. Although understanding sex as a biological variable is valuable, the researchers chose to prioritize animal welfare and thus included only male mice. The researchers acknowledge the limitations this may impose on the study’s generalizability and recommend further research including both sexes when possible and ethically justifiable.

To perform morphometric analysis of brain tissue, we anesthetize mice with ketamine (80 mg/kg) and xylazine (10 mg/kg) and perfuse transcardially with 50 mL 4% paraformaldehyde, followed by postfixation for 2 days. We analyzed three mutants and three WT mice for each investigated age.

After postfixation, specimens were incubated in cryoprotection solution (30% sucrose in PBS) for 2 days. Tissue were embedded and oriented in tissue freezing medium (Tissue-Tek O.C.T., Sakura Finetek, Umkirch, Germany) and were frozen and cut with a cryotome (Leica CM 3050 S, Wetzlar, Germany). Twenty-micrometer-thick sagittal sections were collected with sterile slides wetted with DEPC water and dried. Slides were heated in the microwave oven for 5 min at 1,000 W and 15 min 300 W with antigen retrieval solution containing 19 mm citrate buffer. Afterward, slides were washed and blocked with 3% bovine serum albumin for 20 min. For neuronal staining, we incubated slides with the primary antibodies anti-DCX (ab18723, Abcam, 1:1,000), anti-NeuN (ab177487, Abcam, 1:2,000), and anti-calretinin (ab92341, Abcam, 1:1,500) and for staining of proliferation activity, anti-phospho histone H3 (06–570, Merck, 1:1,000) overnight. Blocking the endogenous peroxidase was applied with 0.3% H₂O₂ for 15 min. We labeled for neuronal marker a secondary horseradish peroxidase-linked anti-rabbit (P0488, DakoCytomation, 1:200) 2 h and anti-rabbit immunoglobulins/biotin (0432, DakoCytomation, 1:200) for labeling anti-phospho histone H3, followed by the Avidin/Biotin Blocking Kit (Vector Laboratories). The development of the staining was generated using the diaminobenzidine chromogen (NeuN = 7 min, calretinin = 15 min, DCX = 30 min, phospho-histone H3 [PHH3] = 10 min). The slides were covered with mounting medium (Aquatex, Merck).

Slides were scanned with ×20 magnification of an Axio Scan.Z1 slide scanner microscope (Zeiss), and the regions of interest were isolated with ZEN light blue version 3.1 (Zeiss). PHH3-positive cells were counted manually and calculated by cells/mm². For further morphometric analysis, ImageJ (National Institutes of Health, Bethesda, MD, USA) was used to isolate the DAB staining. Images were formatted to 8-bit images, expressing pixels between 0 (black) and 255 (white). Next, the background coloration was subtracted. For this, the mean optical density (0–255) in the corpus callosum of randomly selected sections was measured, and the mean value was subtracted from all sections. This was done individually for the three antibody stains. Now, all areas to be examined were marked manually, and the mean optical density was determined. Individual measurement points were considered in all analyses to ensure accuracy and robustness. The normality assumption was evaluated using the Shapiro-Wilk test. If normality was confirmed (p > 0.05), a Student’s t-test was performed to compare means between groups. Conversely, in cases where normality was not met, the Mann-Whitney U test was employed as a nonparametric alternative. A significance level of *α = 0.05, **α = 0.01, **α = 0.001 was used to determine statistical significance.

Adequate neurogenesis is characterized by four basic principles (neurogenesis, proliferation, migration, and differentiation/integration) [23].

First, DCX-positive neuroblasts (Fig. 1a), which develop from stem cells in the SVZ and migrate through the RMS (Fig. 1c) to the OB, were detected in Atoh8^−/− mice. Only at P21, significantly altered optical density of DCX-positive neuroblasts were measured compared to controls (67.1 ± 4.5 S.E.M, p = 0.009, vs. 91.1 ± 7.8 S.E.M, p = 0.009) in the SVZ (Fig. 1b). At P10 and P56, no differences were observed. In the RMS (Fig. 1c), at all three stages, no difference of optical density of DCX-positive neuroblasts were measured (Fig. 1d). Through the RMS, neuroblasts migrate into to OB. Here, no statistically significant differences in the amount of DCX-positive neuroblasts were measured at all three stages examined (Fig. 2).

Second, the proliferation rate, measured by PHH3-positive cells in the SVZ, showed alterations in Atoh8^−/− mice compared to the control in the SVZ (Fig. 3a) and RMS (Fig. 3c) at different stages of age (P10, P21, P56). In the SVZ, at P10, Atoh8^−/− mice showed significantly decreased PHH3-positive cells (54.6 cells/mm² ± 16.4 S.E.M, p = 0.001, vs. 211.9cells/mm² ± 15.4 S.E.M, p = <0.001) (Fig. 3b). In P21, significantly less PHH3-positive cells was observed (59.9 cells/mm² ± 7.8 S.E.M, p = <0.001, vs. 125.3 cells/mm² ± 11.5S.E.M, p = <0.001). At P56, no significant differences of PHH3-positive cells were detected.

In the RMS, a pattern of disturbed proliferation rate can be observed (Fig. 3c). Interestingly, similar results in alterations in the proliferation rate were detected at P10 (83.5 cells/mm² ± 18.9 S.E.M, p = 0.039, vs. 144.0 cells/mm² ± 26.0 S.E.M, p = 0.039) and P21 (39.7 cells/mm² ± 18.3 S.E.M, p = 0.005, vs. 127.9 cells/mm² ± 24.0 S.E.M, p = 0.005), but they were not as highly significant as those observed in the SVZ. However, at P56, no differences in PHH3-positive neuroblasts were found (Fig. 3d).

Third, neuroblasts that have reached the OB integrate into the GC layer and glomerular layer and differentiate into GCs and periglomerular cells. Some subpopulations of these cells express NeuN (Fig. 4a) and calretinin (Fig. 4c). Hence, the differentiation of neuroblasts and the differentiation of neurons can be determined by the optical density of NeuN- and calretinin-positive neurons. Based on this, at P10, significantly reduced optical density of NeuN-positive neurons (Fig. 4b) (27.0 ± 6.0 S.E.M, p = <0.001, vs. 58.2 ± 4.8 S.E.M, p =<0.001) as well as significantly reduced optical density of calretinin-positive neurons are observed (Fig. 4d) (54.3 ± 13.0 S.E.M, p = 0.025, vs. 89.3 ± 10.1 S.E.M, p = 0.025).

bHLH transcription factors have been identified to be integral to multiple developmental processes. In neurogenesis, several bHLH transcription factors (Mash1, Math1, NeuroD, and neurogenin) have been described [24]. In this study, we present alterations in the postnatal neurogenesis and proliferation of neuroblasts in between Atoh8^−/− mice and WT mice in different brain areas. Atoh8 is an embryonic transcriptional activator, crucial for stem cell maintenance and differentiation [25, 26]. Additionally, Atoh8 affects cell proliferation of various organ and skeletal muscle during development [27, 28]. Overexpressing Atoh8 induces neurogenesis and inhibits gliogenesis in the retina [14], contrasting with our observations in the brains of Atoh8 mutant mice. In addition to the low proliferation rate in Atoh8 mutants, the neuroblasts may have adopted the cell fate of glial cells, but this hypothesis still needs to be confirmed by further studies. There were significant differences in the number of neuroblasts in the SVZ in juvenile mice between adult Atoh8 mutants and WT mice. Since neurogenesis is mostly complete in the adult animal, no differences between control and Atoh8 mutant could be detected. Interestingly, a significantly lower amount of NeuN- and calretinin-positive neurons between mutant mice and WT was detectable in the main OB at the neonatal stage.

Besides Atoh8, several other well-established proneural factors have been described in the context of neurogenesis in the brain. Of them, NeuroD is required for postnatal neurogenesis in the hippocampus and cerebellum [29]. In addition, the bHLH transcription factors Math 5 and Math1 play major roles in the neurogenesis of retinal ganglion cells as well as in the development of sensory inner ear hair cells, Merkel cells, GCs in the cerebellum, and dorsal spinal cord interneurons [30‒32].

There is evidence that Atoh8 promotes neuronal versus glia cell fate determination in the nervous system [14, 33‒35]. Neuronal progenitor cells use the proneural bHLH transcription factors to preserve the neuronal cell fate, with gliogenesis being inhibited. Thus, the cell fate decision between neuronal and glial fate can be influenced by a loss of function of the proneural transcription factors [34].

Astrocytes stimulate proliferation and influence neural fate of neural stem cells, by forming a glial tube in the RMS. In addition, they direct migration, maturation, and integration of neural progenitor cells by creating a microenvironment with growth factors and both soluble and membrane-bound factors that lead to the stimulation of neurogenesis [36]. The formation of the glial tube and a possible proliferation of glial cells by the Atoh8 mutants could be related to the small difference in neuroblasts in the RMS and OB. To confirm this hypothesis, further experiments to characterize the distribution pattern of glia cells in Atoh8 mutants are required.

So far, there is no study on whether the neuroblasts and later the interneurons in the OB arise from Atoh8-expressing progenitor cells in the VZ. Furthermore, to generate a better understanding, further studies of the effects of Atoh8 in other neurogenic niches such as the hippocampus and the development of Purkinje cells in the cerebellum are required.

We would like to express our gratitude to Prof. Dr. Beate Brand-Saberi for the resources she provided and her helpful expertise. We would like to thank Prof. Dr. Stefan Wiese for his helpful advice during the implementation of the research project. We are also grateful for the professional laboratory support provided by Ms. Ute Neubacher, Ms. Swantje Wulf, Dr. Urs Kindler, and Mr. Boris Burr.

The animal study was reviewed and approved by LANUV (Landesamt für Umweltschutz, Naturschutz, and Verbraucherschutz; Nordrhein-Westfalen, D-45659, Recklinghausen, Germany), Approval No. 81-02.04.2019.A451. The study was supervised by the Animal Welfare Commission of the Ruhr-University Bochum.

The authors have no conflicts of interest to declare.

Darius Saberi received a scholarship for the clinical scientist program of the SFB Transregio 274 program 2023. Additionally, this work was funded by the Stem Cell Medicine program scholarship from the Academy of Ruhr University Bochum, awarded to Morris Gellisch.

D.C., M.G., and G.M.-P. participated in the research design. G.M.-P. supervised and coordinated the study. D.C. carried out the experiments. M.G. and D.C. analyzed the data and drafted the manuscript supported by D.S. D.C. and D.S. participated in the experimental design and procedure. G.M.-P., M.G., and D.S. revised the manuscript. All the authors read and approved the final version of the manuscript.

Dilek Cuhalik and Morris Gellisch contributed equally to this work and share first authorship.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.