Identification of an early subset of cerebellar nuclei neurons in mice

Reviewer #1 (Public Review):

Summary:

The authors are interested in the developmental origin of the neurons of the cerebellar nuclei. They identify a population of neurons with a specific complement of markers originating in a distinct location from where cerebellar nuclear precursor cells have been thought to originate that show distinct developmental properties. The cerebellar nuclei have been well studied in recent years to understand their development through an evolutionary lens, which supports the importance of this study. The discovery of a new germinal zone giving rise to a new population of CN neurons is an exciting finding, and it enriches our understanding of cerebellar development, which has previously been quite straightforward, where cerebellar inhibitory cells arise from the ventricular zone and the excitatory cells arise from the rhombic lip.

Strengths:

One of the strengths of the manuscript is that the authors use a wide range of technical approaches, including transgenic mice that allow them to disentangle the influence of distinct developmental organizers such at ATOH.
Their finding of a novel germinal zone and a novel population of CN neurons is important for developmental neuroscientists, cerebellar neuroscientists.

Weaknesses:

One important question raised by this work is what do these newly identified cells eventually become in the adult cerebellum. Are they excitatory or inhibitory? Do they correspond to a novel cell type or perhaps one of the cell classes that have been recently identified in the cerebellum (e.g. Fujita et al., eLife, 2020)? Understanding this would significantly bolster the impact of this manuscript.

The major weakness of the manuscript is that it is written for a very specialized reader who has a strong background in cerebellar development, making it hard to read for eLife's general audience. It's challenging to follow the logic of some of the experiments as well as to contextualize these findings in the field of cerebellar development.

Reviewer #2 (Public Review):

Summary:

Canonically cerebellar neurons are derived from 2 primary germinal zones within the anterior hindbrain (dorsal rhombomere 1). This manuscript identifies an important, previously underappreciated origin for a subset of early cerebellar nuclei neurons - likely the mesencephalon. This is an exciting finding.

Strengths:

The authors have identified a novel early population of cerebellar neurons with likely novel origin in the midbrain. They have used multiple assays to support their conclusions, including immunohistochemistry and in situ analyses of a number of markers of this population which appear to stream from the midbrain into the dorsal anterior cerebellar anlage.

The inclusion of Otx2-GFP short term lineage analyses and analysis of Atoh1 -/- animals also provide considerable support for the midbrain origin of these neurons as streams of cells seem to emanate from the midbrain. However, without live imaging there remains the possibility that these streams of cells are not actually migrating and rather, gene expression is changing in static cells. Hence the authors have conducted midbrain diI labelling experiments of short term and long term cultured embryos showing di-labelled cells in the developing cerebellum. These studies confirm migration of cells from the midbrain into the early cerebellum.

The authors have appropriately responded to review issues, replacing panels in figures and updating legends and text. They have also appropriately noted the limitations of their work.

Author response:

The following is the authors’ response to the original reviews.

Reviewer #1 (Recommendations For The Authors):

Additional experiments to characterize what this novel cell type becomes in older animals would be ideal to strengthen the manuscript, but the authors should at least address this in the Discussion.

The manuscript could be significantly improved if the authors included, for example, a timeline and/or cartoon contextualizing these cells relative to the formation of other CN neurons and their locations, perhaps as a summary figure at the end. Furthermore, the logic of each figure could be enhanced if the authors graphically show - again, perhaps with a schematic/cartoon - the question being tested for each figure. Furthermore, making the figure titles less descriptive and more explanatory would also help a reader follow the logic of the experiments.

These are indeed valid and important questions for our research, and understanding the distribution, fate, and connectivity of this new cell type in the cerebellar nuclei postnatally is a focus of ongoing investigation in our lab. To address these questions, we are currently utilizing SNCA-GFP mice, a project led by a PhD student in my lab. While this work will be the subject of a full-length research paper, we do add a sentence to the paper concerning a recent report about the presence of SNCA neurons in the adult CN. We have included a reference to the postnatal expression of SNCA (“In adult mice, postnatal expression of SNCA has been reported in medial CN neurons. PMID: 32639229”.) on page 8 of our manuscript (highlighted in yellow). In addition, we have included a cartoon as a summary figure (Fig. 9) illustrating the origin of cerebellar nuclei from the caudal and rostral ends in both Atoh1+/+ and Atoh1-/- mice. Thank you once again, we have revised and improved the Fig. titles accordingly.

Reviewer #2 (Recommendations For The Authors):

Figure 3:

(1) If most SNCA+ cells are OTX2+ based on the IHCs, why are there so many SNCA+ Otx2- cells in the sort?

In each group, 350,000 cells were sorted. Due to the relatively small population size of this subset of cerebellar nuclei neurons, the sorting procedure could not perfectly mirror our immunohistochemistry results. In each group, 350,000 cells were sorted. Due to the relatively small population size of this subset of cerebellar nuclei neurons, the sorting procedure could not perfectly mirror our immunohistochemistry results. However, it is noteworthy that a portion of sorted cells expressed SNCA or Otx2 while a smaller population co-expressed both Otx2 and SNCA in the cerebellar primordium.

(2) Panel 3F: FACS graphs - the resolution of the figures is too poor on the PDF to read any of the text of these graphs. What are the axes?

We thank the reviewer for this comment. In the revision a high resolution of the FACS graph has replaced the lower quality graph in panel 3F. This clearly identifies the axes and text for this panel.

Figure 4:

(1) Arrowheads are making a subset of + cerebellar cells -Why? Not defined in the legend.

The population of cells indicated by the arrowheads are now defined in the legend. We have added the statement “Examples of Otx2 expressing cells are indicated by arrowheads in panels B, D, E, and F.”

(2) The orientation of panels E and F is unclear - please provide low mag panel insets.

An orientation marker (ie, (r-c and d-v; rostral caudal and dorsal ventral, respectively)) has been added to panel A, which applies to all panels, including panels E and F. Furthermore, the isthmus is noted with an “i” to provide further orientation.

(3) G - and throughout the paper - whisker plots (not simple box plots) are required. Also, it is unclear from the methods how Otx2+ cells were counted - how many embryos/age? The description of 10 sections across 3 slides is incomplete. Are these cells distributed equally across the mediolateral axis of the anlage? Where are comparable M/L sections compared across ages? Is the increase in # across time because these cells are proliferative or are more migrating into the anlage?

The plot has been replaced with whisker plots. A more detailed description of the Method used has been on page 15; “To assess the number of OTX2-positive cells, we conducted immunohistochemistry (IHC) labeling on slides containing serial sections from embryonic days 12, 13, 14, and 15 (n=3 at each timepoint). Under the microscope, we systematically counted OTX2-positive cells within the cerebellar primordium. This analysis encompassed a minimum of 10 sections, spread across at least 3 slides, ensuring comprehensive coverage of OTX2 expression along the mediolateral axis of the cerebellar primordium. For each slide, the counts of OTX2-positive cells from all sections were cumulatively calculated to determine the total number of positive cells per slide. Subsequently, statistical analysis was employed to compare the results obtained different developmental time points.”

Figure 5:

The use of confocal microscopy creates clear data re Otx2-GFP expression, but I cannot understand the origin of the panels. How do they relate to E/F and H/I? Different sections?

In Figure 5, panels A-D display Otx2 expressing cells in the cerebellar primordium of Otx2-GFP transgenic mice, whereas panels E-J depict RNAscope fluorescence in situ hybridization (FISH) for the Otx2 probe in wild type mice. These represent complementary approaches to map Otx2+ cells in the developing cerebellum. This is made clear in a revised legend in Fig 5.

Figure 6:

The justification for the in-culture experiments, particularly the long (4 and 21DIV) times is unclear and needs to be strengthened or the in vitro data should be removed.

Thank you for the respected reviewer’s comment. The E-H panels, show the co-expression of SNCA and p75NTR, highlight a significant role in the differentiation of specific neuronal populations during development. These findings validate our previous results (PMID: 31509576) and are consistent with the results of our current study. Therefore, we have chosen to keep these panels. However, in line with the suggestion from the reviewer, we have removed panels I-L from Fig. 6.

Figure 7:

SNCA expression in panels A and G is not specific nor is the Otx2 staining in panel B making the data in panels C and I uninterpretable and these panels need to be replaced. The Meis2 data however is much better and I agree this data shows that the dorsal RL-derived cells are deleted in Atoh1-/- while the SNCA+ cells remain. This is strong data supporting the dual origins of NTZ.

Thank you for the points, Panel A and G have been replaced with high-resolution images. In addition, panels A-C have been carefully cropped to enhance focus on the NTZ area, to improve the quality and visibility of panels. To enhance clarity, we have included a summary fig. 9 for clarification.

Figure 8:

The diI experiments are a key addition to this paper and clearly show the direct movement of some cells from the mesencephalon into the developing cerebellum, but data presentation must be considerably strengthened.

(1) What is the inset in panel A? Low mag of embryo? Perhaps conversion of image to PDF degraded resolution - add a description in the legend. Arrowhead and arrow identities are reversed in the legend. The arrow points to the isthmus.

Thank you for the comment, for clarification we have included information in the Fig. legend (highlighted in yellow). In addition, the issues with the arrows have been addressed and corrected.

(2) Panels B and C are also shown in Supplementary Figure 2 with arrows indicating rostral and caudal movement - these arrows need to be added here. There is no need to replicate these same panels in the supplement.

Thanks, arrows have been added in panels B, C of Fig. 8.

(3) The text states that "almost all DiI cells migrated caudally into the cerebellum" and refers to Figure 8E and Suppementl 3 but there is no evidence/support shown for this, just a few + cells in 8E and some very difficult-to-see positive cells in sections in Supplement E-F. Given the importance of this data, I am surprised that the authors chose bright field/phase microscopy to show this. This section's data is not convincing data at all. I find it very difficult to see specific staining. These panels must be improved. This is key data for paper conclusions.

These are valid points, and we acknowledge that this experiment alone may not provide conclusive evidence regarding the subset of CN originating from mesencephalon. At this stage of the study, we do not claim definitively that the SNCA/OTX2/MEIS2 positive cells originate from the mesencephalon. As stated in our manuscript, "In conclusion, our study indicates that the SNCA+/ OTX2+/ MEIS2+/ p75NTR+/ LMX1A- rostroventral subset of CN neurons do 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." We have also discussed embryonic culture approaches in the manuscript, which could involve the use of other agents such as plasmid/viral vectors, hinting at the possibility of origin from the mesencephalon. While tracing the origin from the mesencephalon in vivo and in vitro is promising and on our to-do list, the data will not be available for this manuscript. To prevent confusion, we have eliminated redundant panels of Fig. 8 with Supplementary Fig. 2 and 3. However, if the reviewer deems it necessary to remove these panels, we are prepared to do so.