Introduction

In the mouse, the cerebellar primordium emerges at approximately embryonic day (E) 7–E8 as a neuroepithelial swelling on the rostral lip of the fourth ventricle, which is part of the alar plate of the metencephalon (rhombomere-1) (Kebschull et al., 2023, Goldowitz and Hamre, 1998, Sotelo, 2004, Wang and Zoghbi, 2001a, Marzban et al., 2015). The cerebellar primordium is rostrally limited by the isthmus, the mesencephalon-rhombencephalon (rhombomere-1, r1) or midbrain-hindbrain boundary. Orthodenticle homeobox 2 (Otx2) and gastrulation brain homeobox 2 (Gbx2) are among the earliest genes expressed in the neuroectoderm, dividing it into anterior and posterior domains with a border that marks the isthmus. Otx2 is required for the development of the forebrain and midbrain, and Gbx2 for the anterior hindbrain (Millet et al., 1999). The isthmus, at the junction of Otx and Gbx expressing areas, is a multi-signaling center regulating genes essential for the early development of both the mesencephalon and cerebellar primordium (Joyner et al., 2000, Wurst and Bally-Cuif, 2001).

The cerebellar primordium contains two distinct germinal zones, which are responsible for cerebellar neurogenesis and gliogenesis; the ventrally-located ventricular zone (VZ) and dorsally-located rhombic lip (Fink et al., 2006). Early studies on the development of CN indicated that both glutamatergic and GABAergic CN neurons originate from the cerebellar VZ (Voogd and Glickstein, 1998). Further studies highlighted the involvement of the rhombic lip as an origin for glutamatergic CN neurons (between E9-E12) (Fink et al., 2006, Wang and Zoghbi, 2001a, Green and Wingate, 2014, Machold and Fishell, 2005, Kebschull et al., 2023). LIM homeobox transcription factor 1 alpha (LMX1A) regulates the development of glutamatergic CN neurons originating from the rhombic lip. It is also expressed in the nuclear transitory zone (NTZ) and is considered to be a marker for the majority of rhombic lip-derived CN neurons in addition to a subset of LMX1A positive cells that do not originate from the rhombic lip migratory stream (Chizhikov et al., 2006, Chizhikov et al., 2010).

In this report, we provide evidence that a subset of CN neurons expressing OTX2, SNCA, MEIS2, and p75NTR originate from the rostral end of the cerebellar primordium and are positioned in the rostroventral region of the NTZ during early cerebellar development. Our data suggest that the mesencephalon could represent a third germinal zone and, as such, the origin of previously unrecognized neurons that contribute to CN formation and development.

Results

In mouse embryos, the earliest neuronal populations, which are located at the rostral end of the cerebellar primordium, are immunopositive for NAA 3A10, a marker typically expressed in neuronal somata and axons (Marzban et al., 2008, Marzban et al., 2019) (Fig. 1). To further explore these cells in the rostral end of the cerebellar primordium, we used an antibody to detect SNCA, which is expressed in CN neurons during embryogenesis (Zhong et al., 2010). In sagittal sections of E9 cerebellar primordium, we identified SNCA⁺ neurons at the rostral end of the NTZ (Fig. 1E-F and Fig. Suppl. 4). To determine the location of the SNCA⁺ cells and compare it to that of the rhombic lip-derived LMX1A⁺ CN neurons (E10-12) in the NTZ, double immunofluorescence labeling of SNCA and LMX1A was performed on sagittal sections of E12 cerebellar primordium (Fig. 2 A-D, medial section and Fig. E-H, lateral section). This shows that an LMX1A⁺ population of CN neurons, originating from the rhombic lip, either flanks SNCA⁺ neurons in the medial section (Fig. 2A-D) or is located ventrally in the lateral section (Fig. 2E-H) in the NTZ. To further explore the characteristics of SNCA⁺ cells, double immunofluorescence labeling with OTX2 and SNCA was performed during early cerebellar development. OTX2 is highly expressed in the mesencephalon and its caudal limit coincides with the boundary of the rhombencephalon (i.e., the isthmus) (Kurokawa et al., 2004). Interestingly, results from double immunofluorescence experiments in sagittal sections of E12 cerebella showed that OTX2 was highly expressed in the mesencephalon (Fig. 3A, B), and was co-expressed with SNCA⁺ cells in the rostral region of the cerebellar primordium (Fig. 3A-E). Next, we used quantitative flow cytometry analysis to determine whether OTX2⁺ cells co-express SNCA and/or LMX1A (Fig. 3F, G). An increase in the number of OTX2 and SNCA positive cells was observed at E14 in comparison to E12 (*p<0.05). The majority of neural cells were not SNCA⁺/LMX1A⁺, consistent with our IHC findings, or SNCA⁺/OTX2⁺, with the exception of a few cells that did show co-labeling. Of note, more SNCA⁺/OTX2⁺ than SNCA⁺/LMX1A⁺ cells were detected at E12 compared to E14.

Fig. 1.

Fig. 2.

Fig. 3.

IHC performed on sagittal sections of E12 to E15 cerebella showed that OTX2⁺ cells are present at the rostral end of the cerebellar primordium (Fig. 4 A-F). To evaluate the number of OTX2⁺ cells in the developing rostral cerebellum, the total number of OTX2⁺ cells in the cerebellar primordium was counted in slides from E12 to E15 (Fig. 4G). A comparison between these embryonic stages indicated a slight increase in the number of OTX2+ cells from E12-E15. Significant differences were observed between E12 and E14 (**p<0.01), E12 and E15 (***p<0.001), as well as E13 and E15 (#p<0.05).

Fig. 4.

To confirm the presence of OTX2⁺ cells in the rostral end of the cerebellar primordium, we used Otx2-GFP transgenic mice and performed double immunofluorescence labeling with OTX2 and GFP on E13 sagittal sections (Fig. 5). Results showed that an extension of OTX2-GFP positive cells, highly prevalent in the mesencephalon, crossed the isthmus and terminated in the rostral end of the cerebellar primordium in the NTZ (Fig. 5B, C, and D). To further validate our observations, we employed RNAscope ISH at E12 to detect the presence of Otx2 mRNA in the cerebellar primordium (Fig. 5 E-J). Otx2 mRNA was highly expressed in the mesencephalon and caudally extended to the rostral cerebellar primordium in the NTZ (Fig. 5 E-J and Fig. Suppl. 4).

Fig. 5.

To evaluate whether OTX2⁺ cells are present in the absence of Snca gene expression, we used Pap^-/-; Snca^-/- mice. OTX2 immunoperoxidase staining showed that despite the lack of Snca expression, OTX2⁺ cells still terminated in the rostral cerebellar primordium (Fig. Suppl. 1). Nerve growth factor receptor (p75NTR; Ngfr), which regulates neuronal proliferation and differentiation (Jiang et al., 2008, Bernabeu and Longo, 2010, Dechant and Barde, 2002), has also been shown to be expressed in the rostral end of the cerebellar primordium in Pap^-/-; Snca^-/- mice (Rahimi-Balaei et al., 2019b). To determine whether SNCA⁺ neurons are p75NTR immunopositive, double immunolabeling with SNCA and p75NTR was performed. Section IHC revealed that SNCA⁺ cells in the NTZ exhibited cell membrane expression of p75NTR (Fig. 6A–D). This data was further confirmed by Western blot analysis of SNCA and p75NTR protein expression in embryonic cerebellar lysates (E11, 13 and 15; Fig. 6d). In addition, to conclusively determine that p75NTR is expressed in the membranes of SNCA⁺ cells, we utilized primary dissociated cultures of cerebellum at E10, DIV 4 (Fig. 6E-H). IHC demonstrated that p75NTR was indeed expressed in the membranes of SNCA⁺ cells (with punctate appearance, Fig. 6E–H (Dechant and Barde, 2002)). In order to confirm the presence of p75NTR in neuronal membranes, NAA 3A10, as a neuronal marker, was utilized at DIV 21 (Fig. 6I-L); this implies that SNCA⁺ cells constitute a subset of NAA expressing neurons immunopositive for p75NTR.

Fig. 6.

Supplementary Figure 1.

To determine whether OTX2 and SNCA immunopositive cells exist in the absence of rhombic lip-derived CN neurons, we used Atoh1^-/- mice (Fig. 7 A-C). Immunostaining revealed the presence of SNCA⁺ and OTX2⁺ cells at the rostral end of the cerebellar primordium in these animals. To validate our findings, we evaluated MEIS2, a transcription factor typically expressed in the NTZ, in Atoh1^+/+ and Atoh1^-/- sagittal sections of the cerebellar primordium at E12. Notably, our results show the presence of two distinct sets of MEIS2⁺ cells in the NTZ of Atoh1^+/+ embryos: rhombic lip-derived CN neurons located in the caudodorsal region, which do not express SNCA (MEIS2⁺/SNCA^-), and a subset of MEIS2⁺/SNCA⁺ CN neurons situated in the rostroventral region of the NTZ (Fig. 7D-I). In Atoh1^-/- sections, MEIS2⁺/SNCA⁺ cells were still present in the rostroventral region of the NTZ, whereas the subpopulation of MEIS2⁺/SNCA^- rhombic lip-derived CN neurons was absent in the caudodorsal region of the NTZ (Fig. 7 J-O).

Fig. 7.

To explore if the rostroventral subset of CN neurons derived from an external cerebellar germinal zone, possibly the mesencephalon, we utilized FAST DiI as a neuronal tracer to mark cells within a potential source region. FAST DiI was applied to the dorsum of the caudal mesencephalon at E9, where it was maintained for 4 days (Fig. 8 A, a). At DIV 4, cells stained with DiI in the dorsum of the mesencephalon were present in both rostral (Fig. 8B) and caudal (Fig. 8C) directions. Sections from the cerebellar primordium revealed the presence of DiI-labeled cells within the cultured embryo (Fig. Suppl. 2). To investigate the earliest DiI-positive cell population in the mesencephalon and avoid unwanted cell staining due to long DiI exposure, we focused on early cerebellar development at E9; it has been reported that one of the premier neuronal populations in the CNS is present at this time in the mesencephalon and projects caudally (Stainier and Gilbert, 1990, Easter et al., 1993). To determine if an early generation of mesencephalic DiI-labeled cells is present within the cerebellar primordium, we limited cellular DiI exposure to only 24 hours (Fig. 8 D, d), after which it was removed (Fig. Suppl. 3). Intriguingly, almost all DiI⁺ cells migrated caudally and presented in the cerebellar primordium (Fig. 8E, e and Fig. Suppl. 3). DiI staining was clear and strong in cells as shown in Fig. 8 E, e and Fig. Suppl. 3G’(arrowhead); it should be noted, however, that some DiI staining was evident in other cells in the rostral end of the cerebellar primordium, although the signal appeared relatively weak (Fig. Suppl. 3E-G”).

Fig. 8.

Supplementary Figure 2.

Supplementary Figure 3.

Discussion

Previously, it was commonly held that all CN neurons had a singular origin in the ventricular zone (Pierce, 1975, Miale and Sidman, 1961). However, recent discoveries have challenged this notion, revealing a dual origin for CN neurons. Some originate from the ventricular zone, while others arise from the rhombic lip (Wang and Zoghbi, 2001b, Wang et al., 2005, Ben-Arie et al., 1997). Genetic fate mapping further suggested that most glutamatergic CN projection neurons may arise from the rhombic lip (Ben-Arie et al., 1997, Machold and Fishell, 2005, Fink et al., 2006). Transcription factor expression patterns indicate that CN neurons migrate from the rhombic lip to the NTZ through a subpial stream pathway, while sequentially expressing the genes Lmx1a, Pax6, Tbr2, and Tbr1 (Fink et al., 2006).

In this study, we characterized a subset of CN neurons that do not originate from the cerebellar primordium germinal zones during early cerebellar development. We demonstrated a newly identified subset of CN neurons, which are SNCA⁺/ OTX2⁺/ MEIS2⁺/ p75NTR⁺/ LMX1A^-, in the rostroventral region of the NTZ in the rostral end of the cerebellar primordium. This suggests the existence of a new germinal zone during cerebellar neurogenesis.

The exact origin of the SNCA⁺/ OTX2⁺/ MEIS2⁺/ p75NTR⁺/ LMX1A^- CN neurons is currently unclear; however, they do not originate from the rhombic lip or ventricular zones. Rhombic lip-derived Tbr1⁺/Lmx1a⁺ CN neurons are born at E9, but do not reach the NTZ until ∼E11 (Kebschull et al., 2023, Machold and Fishell, 2005, Wang et al., 2005, Manto et al., 2012, Fink et al., 2006, Rahimi-Balaei et al., 2018). Our results revealed that SNCA⁺ cells are a group of differentiating neurons (NAA 3A10⁺) present in the NTZ at E9, before the arrival of any neurons that originate from the rhombic lip. The majority of SNCA⁺ CN neurons are not LMX1A⁺; however, a small proportion of cells seems to exhibit co-expression. This suggests that in the early stages of CN neurogenesis, the pattern of protein expression in SNCA⁺ neurons may be changing, possibly with respective down-and upregulation of Snca and Lmx1a⁺. Conversely, this may not occur at all, and ‘co-labeling’ could have been observed due to overlapping cells. A similar display of SNCA expression has recently been demonstrated in oligodendrocyte development and maturation (Djelloul et al., 2015). Our findings indicated that CN neuron somata are highly positive for SNCA till E14, after which SNCA expression is refined to PCs at P0 (Zhong et al., 2010) and axonal projections, indicating that SNCA⁺ cells at the rostral end of the cerebellar primordium are associated with CN establishment in NTZ. Our results from studies using Pap^-/- ; Snca^-/- mice showed that lack of SNCA during cerebellar development did not affect the rostroventral subset of CN neurons (Rahimi-Balaei et al., 2019a), which still expressed OTX2 and terminated in the rostral cerebellar primordium.

Furthermore, our observations reveal the expression of MEIS2 in the NTZ (Fig. 7 and (Morales and Hatten, 2006)), along with several other NTZ markers such as POU3F1, BRN2, IRX3, and MEIS1 as documented in (Kebschull et al., 2023), c-Ret and Tlx3 (Fig. Suppl. 4), indicating the existence of discrete subpopulations of precursors within the CN. MEIS2 (homeobox transcription regulator) is expressed in the mesencephalic alar plate of both mice (Cecconi et al., 1997) and chicks (Agoston and Schulte, 2009, Bobak et al., 2009). MEIS2 plays a crucial role in the development of cranial nerves and craniofacial structures (Machon et al., 2015). It has been suggested that mesencephalon development is regulated by Meis2 cross-talk with Otx2 (Agoston and Schulte, 2009). Notably, the expression of MEIS2 in CN neurons is remarkably distinct, allowing for a clear demarcation of the caudodorsal (SNCA^-/MEIS2⁺) and rostroventral (SNCA⁺/MEIS2⁺) regions of the NTZ, whether originating from the rhombic lip or not (Fig. 7D-I). CN neurons derived from the rhombic lip require ATOH1 expression, as they are absent in an Atoh1-null mutant (Wang et al., 2005, Machold and Fishell, 2005, Zordan et al., 2008, Florio et al., 2012, Ben-Arie et al., 1997). In the Atoh1-null cerebellar primordium, the caudodorsal (SNCA^-/MEIS2⁺) subset of CN is absent in the NTZ, while the rostroventral (SNCA⁺/MEIS2⁺) population is still present, indicating that the Meis2⁺ rostroventral subset of the CN in the NTZ develops independently from Atoh1.

Supplementary Figure 4.

What are the potential origins of SNCA⁺/ OTX2⁺/ MEIS2⁺/ p75NTR⁺/ LMX1A^- cells in the NTZ during the early stages of cerebellar development? Our study suggests that the origin of these cells may be guided by a continuous flow of rostroventral cells toward the rostral end (mesencephalon), and caudodorsal cells toward the caudal end (rhombic lip) of the cerebellar primordium. It is well known that Otx2 is required for the development of the fore- and midbrain, while Gbx2 is necessary for anterior hindbrain development (Simeone et al., 1992, Alvarado-Mallart, 2005, Li and Joyner, 2001, Joyner et al., 2000). The expression of OTX2 in the rostral and GBX2 in the caudal neural tube, is considered limited to the mesencephalon/metencephalon boundary (isthmus) (Alvarado-Mallart, 2005). However, a study by Martinez et al. on chick/quail chimeras determined that the rostral portion of the cerebellar primordium is located more rostrally in the so-called ‘mesencephalic’ alar plate (Martinez and Alvarado-Mallart, 1989). Recently, Isl1 positive cells in the anterior cerebellum were shown to exhibit residual OTX2 protein activity (Wizeman et al., 2019). Several other studies have shown that rostral cerebellar development is associated with factors expressed in the caudal mesencephalon, such as engrailed family En1 and -2 (Joyner et al., 1991, Hanks et al., 1995, Millen et al., 1995), and Acp2 (Bailey et al., 2013, Bailey et al., 2014), which are regulated by the multi-signaling center known as the isthmic organizer (Martinez et al., 2013). Furthermore, by employing embryonic cultures in conjunction with DiI labeling, we determined that the source of the rostroventral subset of CN neurons appears to be an external cerebellar germinal zone, possibly originating from the mesencephalon, which is in line with other studies (Nichols and Bruce, 2006, Wizeman et al., 2019). While our current study hints at the possibility of the caudal mesencephalon serving as a novel extrinsic germinal zone for cerebellar primordium, further investigation is required. Genetic inducible fate mapping and long-term follow-up studies are essential to gain a comprehensive understanding of the origin, fate, and connectivity of as well as the role played by the SNCA⁺/ OTX2⁺/ MEIS2⁺/ p75NTR⁺/ LMX1A^- rostroventral subset of CN neurons in the development of the cerebellum.

In conclusion, our study indicates that the SNCA⁺/ OTX2⁺/ MEIS2⁺/ p75NTR⁺/ LMX1A^- rostroventral subset of CN neurons does not originate from the well-known distinct germinative zones of the cerebellar primordium. Instead, our findings suggest the existence of a previously unidentified extrinsic germinal zone, potentially the mesencephalon.

Material and methods

Animal maintenance

All animal procedures were performed in accordance with institutional regulations and the Guide to the Care and Use of Experimental Animals from the Canadian Council for Animal Care. For this study, we used embryos from 47 CD1 timed-pregnant mice at embryonic day (E) 9 to E18 (total number of embryos was 409). Our approach to sample size was to include enough samples to reach a sufficient level, which was determined empirically (in most of the experiments, n was equal to 9).Timed-pregnant, prostatic acid phosphatase (PAP) mutant (Pap KO) (Zylka et al., 2008, Hokin and Hokin, 1959) mice were used at E12, because they do not express Snca (Pap^-/- ;Snca^-/-) and are a valuable experimental tool to assess whether SNCA is required in the development of the CN neurons (Rahimi-Balaei et al., 2019b). In addition, we studied GFP-tagged Otx2 mouse embryos (from Thomas Lamonerie lab at Université Côte d’Azur), a reporter mouse line in which the Otx2 protein is fused to the fluorescence protein GFP (Otx2^Otx2-GFP/+; (Fossat et al., 2007)). The animal procedures related to GFP-tagged Otx2 mouse embryos were in accordance with Université Côte d’Azur Institutional Animal Care and Use Committee guidelines. Atoh1 knockout embryos were provided by Dr. Huda Zoghbi at the Baylor College of Medicine. The mice were bred, phenotyped, and genotyped (by the Zoghbi lab) according to the previously described protocol (Wang et al., 2005). Animal procedures involving Atoh1 knockout mice were in accordance with The Baylor college of Medicine Institutional Animal Care and Use Committee guidelines.

All timed-pregnant CD1 mice were obtained from the Central Animal Care Service, University of Manitoba. Animals were housed at room temperature and relative humidity (18–20°C, 50–60%) on a light:dark cycle (12:12 h) with free access to food and water. The embryonic age was determined by referencing the morning after the detection of a vaginal plug following overnight mating, considered to be E 0.5. CD1 timed-pregnant mice were anesthetized at E9, 10, 11, 12, 13, 14, 15, or 18 (+ 0.5) (n=47) using 40% isoflurane (USP, Baxter Co. Mississauga, Ontario, Canada), after which embryos were removed and prepared for Western blotting, embryonic culture, or flow cytometry analysis (FCM) or fixed in 4% paraformaldehyde (PFA) for immunohistochemistry (IHC) or in situ hybridization (ISH).

Section immunohistochemistry

Cryostat sections (20 μm) of 4% PFA fixed samples were utilized for IHC as described in our previous studies (Bailey et al., 2014, Bailey et al., 2013). Antibody dilutions were used as follows: α-synuclein 1:500 (sc-69977, Santa Cruz), p75NTR 1:1000 (8238, Cell Signaling), LMX1A 1:500 (AB10533, EMD Millipore Corporation), OTX2 1:1000 (ab114138, Abcam), NAA 1:500 (3A10, Developmental Studies Hybridoma Bank), Meis2 1:500 (ab244267, Abcam), and GFP 1:1000 (1020, Aves Labs). Fluorescent detection was performed using antibodies as follows: Streptavidin Alexa Fluor® 488 conjugate, Alexa Fluor® 568 Goat Anti-Rabbit IgG (H+L), Alexa Fluor® 488 Chicken Anti-Mouse IgG (H+L), Alexa Fluor® 488 Chicken Anti-Rabbit IgG (H + L), and Alexa Fluor® 568 Goat Anti-Mouse IgG (H+L) (S-11223, A-11036, A21200, A21441, A11004, respectively, from Life Technologies), AlexaFluro® 568 Donkey anti-goat (A-11057, invitrogen), and AlexaFluro® 647 Donkey anti-mouse (A-31571, Invitrogen), all at 1:1000. Detection of peroxidase IHC was performed as described previously (Rahimi Balaei et al., 2016, Bailey et al., 2014, Bailey et al., 2013) using HRP conjugated goat anti-rabbit IgG and goat anti-mouse IgG (H+L) antibodies (EMD Millipore Corporation, 12-348 and AP308P, respectively), both at 1:500, and developed with 3,3’-diaminobenzidine solution (DAB, Sigma, St. Louis MO, USA).

Primary dissociated cerebellum cell cultures

Primary cerebellar cultures were prepared from E10 CD1 mice, and maintained for varying ‘days in vitro’ (DIV; 1, 2, 3, 5, and 8), according to published methods (Shabanipour et al., 2019). Briefly, the entire cerebellum was removed from each embryo and immediately placed into ice cold Ca²⁺/Mg²⁺-free Hank’s balanced salt solution (HBSS) containing gentamicin (10 μg/ml) and glucose (6 mM). Next, cerebella were incubated at 34°C for 12 min in HBSS containing 0.1% trypsin. After washing, the cerebella were gently triturated with a Pasteur transfer pipette in HBSS containing DNase I (5 U/ml) and 12 mM MgSO₄ until the cell mass was no longer visible. Cells were collected by centrifugation (1,200 rpm, 4°C for 5 min) and re-suspended in seeding medium (Dulbecco’s modified Eagle’s medium and F12, 1:1) supplemented with putrescine (100 μM), sodium selenite (30 nM), L-glutamine (1.4 mM), gentamicin (5 μg/ml), and 10% heat-inactivated fetal bovine serum. Cells were seeded on poly-L-ornithine coated glass coverslips (12 mm) at a density of 5×10⁶ cells/ml, and each coverslip was placed into an individual well of a 24-well plate. After 6–8 h incubation in a CO₂ incubator (95% humidity, 37°C, 5% CO₂) to accomplish cell adherence, 500 μl of culture medium supplemented with transferrin (200 μg/ml), insulin (20 μg/ml), progesterone (40 nM), and triiodothyronine (0.5 ng/ml) was added to each culture well. After 7 days, half of the medium in each dish was replaced with fresh medium that was additionally supplemented with cytosine arabinoside (4 μM) and bovine serum albumin (100 μg/ml) (Bailey et al., 2013, Marzban and Hawkes, 2007, Marzban et al., 2003).

Embryonic cultures and DiI labeling of cells within the mesencephalon

Embryonic cultures were prepared from E9 and E10 CD1 timed-pregnant mice, and maintained for various DIV (4 and 6). Each embryo was removed from the amniotic sac and immediately placed into ice cold Ca²⁺/Mg²⁺-free HBSS containing gentamicin (10 μg/ml) and glucose (6 mM). Embryos were placed into 24-well plates in culture medium plus 10% fetal bovine serum and incubated in a CO₂ incubator (95% humidity, 37°C, 5% CO₂) (Marzban and Hawkes, 2007). During incubation, embryos were monitored every 6 h to evaluate the heartbeat (as an indicator of survival). On DIV 4 or 6, each well was fixed with 4% PFA and prepared for whole mount IHC.

For neuronal tracing and labeling, we used FAST DiI crystal (FAST DiI™ solid; DiIΔ9,12-C18(3), CBS (1,1’-Dilinoleyl-3,3,3’,3’-Tetramethylindocarbocyanine, 4-Chlorobenzenesulfonate, D7756, Fisher Scientific). Briefly, FAST DiI was inserted into the mesencephalon at E9 using a sharp-ended needle (30g). After insertion of FAST DiI, images were captured by a stereomicroscope to assess the location of DiI at DIV 0. After placing the embryos into 24-well plates in culture medium, they were monitored every 6 hours and fixed with 4% PFA on the desired day. Next, whole mount IHC with neurofilament-associated antigen (NAA 3A10) was performed to visualize neural fiber growth, followed by sectioning and imaging of the DiI positive cells in the mesencephalon and cerebellar primordium.

Fluorescent in situ hybridization (FISH)-RNAscope

All of the ISH experiments were carried out using E12 CD1 mice, in which the cerebellar primordium is well established, using RNAscope ACD HybEZ™ II Hybridization System and RNAscope^® Multiplex Fluorescent Reagent Kit v2 (Advanced Cell Diagnostics, Hayward, CA, USA). Briefly, embryos were fixed in 10% (vol/vol) neutral buffered formalin at room temperature for 24 h, dehydrated, and embedded in paraffin. Tissue sections cut at 10 μm thickness were processed for RNA in situ detection according to the manufacturer’s user manual. The sequence of the probe used in this study was Mm-Otx2-C3 (444381, ACD); cyanine 3 (NEL744E001KT; TSA^® Plus, Perkin Elmers, Waltham, MA, USA) was used as the fluorophore.

Western blot analyses

Equal amounts of total protein were separated by SDS/PAGE in 10–15% precast gels (Bio-Rad, Hercules, CA, USA) and transferred onto PVDF-membranes. For Western blot analysis, membranes were blocked in 5% nonfat dry milk (NFDM) in TBS containing 0.02% Tween 20 (TBST) for 1hour at RT and subsequently incubated overnight at 4°C with primary antibodies [α-synuclein (sc-69977, Santa Cruz, 1:2000) or p75NTR (8238, Cell Signaling, 1:1000), all dilutions in blocking buffer]. After washing with TBST, membranes were exposed to secondary antibodies [HRP conjugated goat anti-mouse IgG (AP308P, Millipore, 1:6000) or HRP conjugated goat anti-rabbit IgG (12-348, Millipore, 1:6000)]. Bands were visualized using the enhanced chemiluminescence (ECL) protocol on scientific imaging film. All bands were normalized to β-actin expression.

Counting of OTX2 positive cells

To assess the number of OTX2 positive cells, slides with sections of E12, 13, 14 and 15 were labeled for OTX2 using IHC. Under the microscope, OTX2 positive cells in the cerebellar primordium were counted in each section of at least 3 slides (minimum 10 sections). The number of positive cells from all sections from the same slide was combined to determine total positive cell number per individual slide; results from all slides were compared by statistical analysis.

Flow cytometry analysis

After dissociation (see primary dissociated cerebellum cell culture section), cells were fixed in 4% PFA. Each group consisted of 350,000 cells and all procedures before sorting were performed at room temperature. Primary antibodies used: α-synuclein 1:500 (sc-69977, Santa Cruz), LMX1A 1:1000 (AB10533, EMD Millipore Corporation), and OTX2 1:500 (ab114138, Abcam). Fluorescent detection was performed using antibodies as follows: Alexa Fluor 488 Chicken Anti-Rabbit IgG (H / L) and Alexa Fluor 568 Goat Anti-Mouse IgG (H+L) (A21441 and A11004, respectively, from Life Technologies), both at 1:750. Cells were pelleted by centrifugation, washed, and resuspended in 200 µL staining buffer (1X PBS contains 1% FBS and 1mM EDTA). Data were acquired on a CytoFlex-LX flow cytometer (Beckman Coulter) equipped with 355, 375, 405, 488, 561, 638 and 808 Laser lines using CytExpert software, and analyzed with Flow Jo (version 10, Treestar, San Carlos, CA) at the University of Manitoba flow cytometry core facility. Cellular debris was excluded using forward light scatter/side scatter plot.

Imaging and figure preparation

For bright field microscopy, images were captured using a Zeiss Axio Imager M2 microscope and subsequently analyzed with Zeiss Microscope Software (Zen Image Analyses software; Zeiss, Toronto, ON, Canada). For fluorescence microscopy of entire cerebellar sections, a Zeiss Lumar V12 Fluorescence stereomicroscope (Zeiss, Toronto, ON, Canada) equipped with a camera was used; captured images were analyzed using Zen software. For high magnification fluorescence microscopy, a Zeiss Z1 and Z2 Imager and a Zeiss LSM 700 confocal microscope (Zeiss, Toronto, ON, Canada) equipped with camera and Zen software were used to capture and analyze images. Images were cropped, corrected for brightness and contrast, and assembled into montages using Adobe Photoshop CS5 Version 12.

Statistical analysis

Data on OTX2 positive cell counts are presented as the mean ± standard error of mean (SEM) from n separate experiments. Statistical significance of differences was evaluated by One-Way ANOVA followed by a Tukey’s multiple comparison test. Differences were considered statistically significant when P<0.05. All statistical analyses were performed using GraphPad Prism 6 software for Windows.

Acknowledgements

The authors would like to acknowledge Dr. Christine Zhang and Mr. Gerald Stelmack for their technical assistance, and Science Impact (Winnipeg, Canada) for (post-)editing the manuscript.

Competing interest

The authors have no conflicts of interest.

Funding

This study was supported by grants from the Natural Sciences and Engineering Research Council (HM: NSERC Discovery Grant # RGPIN-2018-06040), The Children’s Hospital Research Institute of Manitoba (HM: CHRIM Grant # 320035), Research Manitoba Tri-Council Bridge Funding Program (HM: Grant # 47955), and University Collaborative Research Program (UCRP).