Cerebellar Purkinje cell stripe patterns reveal a differential vulnerability and resistance to cell loss during normal aging in mice

Reviewer #1 (Public review):

Summary:

In this study, Donofrio et al. investigated cerebellar Purkinje cell (PC) degeneration during normal aging using both mouse and human samples. They found that PC loss followed a stripe pattern rather than occurring randomly. Although this pattern resembled the pattern of zebrin II expression in the anterior cerebellum, the overall pattern was different from zebrin II expression. Surviving PCs exhibited severe degeneration, including thickened axons, axonal torpedoes, and shrunken dendrites. These structural changes were accompanied by functional deficits in motor coordination and tremor. Understanding why certain PC subpopulations are more vulnerable than others may provide insight into regional susceptibility (or resilience) to aging and inform potential therapeutic strategies for age-related neurological disorders. Overall, the findings are novel and significant, supported by compelling evidence from structural and functional analyses. However, I have several concerns about the results and hope that my comments will help improve the clarity and impact of this paper.

Strengths:

The cerebellum is often overlooked in aging research, despite its increasingly recognized role in motor and non-motor functions. This study, which examines the pattern of PC loss during normal aging, offers a new perspective on the aging process.

The finding that PC loss follows a stripe pattern is a major conceptual advance, challenging the previous assumption that PC loss occurs uniformly in the cerebellum.

The analyses using wholemount immunohistochemistry, PC-specific reporter mice, and light-sheet imaging of cleared brain tissue are meticulous. By visualizing PCs in three dimensions, this study provides strong evidence for the patterned loss of PCs across different cerebellar subdivisions during aging.

The inclusion of human samples along with the animal model strengthens the impact and translational relevance of these findings.

The data are clearly presented, and the manuscript is very well written.

Weaknesses:

While the authors have largely ruled out zebrin II as the key protein underlying PC vulnerability or resistance to age-related loss, the molecular basis of this phenomenon remains unidentified. This reviewer acknowledges the complexity of this investigation and considers it a minor issue, as the manuscript thoughtfully discusses the gap and highlights it as a future direction.

In cases where no PC loss is observed in aged mice (Figure 1F), it is unclear whether these PCs undergo morphological degeneration, such as thickened axons and shrunken dendrites. Further characterization of these resilient PCs would help understand why the aged mice without PC loss still exhibit motor deficits (Figure 7).

The histologic analysis is based on mice with different genetic backgrounds. For example, the PC-specific reporter mice include two strains: Pcp2-Cre; Ai32 and Pcp2-Cre; Ai40D. These genetic variations may contribute to the heterogeneity of PC loss (Figure 1). To improve clarity, please add the genetic background details to Table 1.

Please indicate from which lobule in the anterior or posterior human cerebellum the images in Figure 8 were taken.

Reviewer #2 (Public review):

Summary:

The cerebellum is known to be vulnerable to aging, yet specific cell type vulnerability remains understudied. This important study convincingly demonstrates that the normal aged mouse cerebellum exhibits Purkinje cell loss, and that the vulnerable PCs to age are arranged on the basis of the known zebrin stripe pattern that represents a particular subtype of the PCs. Although the patterns of PC loss were analyzed qualitatively, the phenotype is robust enough to clearly appreciate that PC loss occurs predominantly in zebrin-negative regions when combined with zebrin immunohistochemistry. Interestingly, the authors demonstrate that this phenotype appears stochastically even within the inbred C57BL/6J mouse strain examined, though the mechanisms behind this individual variability remain unexplored. In contrast to the expectation that the PC loss could account for age-related motor decline, the authors did not find any correlation between them. While the authors attempt to draw parallels with normal human aging, the human phenotypes have not been conclusively shown to match those in mice beyond the occurrence of potentially age-related PC loss. Future studies should investigate why this PC loss phenotype occurs stochastically across the population and whether these findings parallel human cerebellar aging.

Strength:

(1) Banding pattern of PC loss is very clearly demonstrated by combining immunostaining for zebrin.

(2) A critical methodological concern that a standard PC marker, calbindin, could be compromised in aging has been addressed by performing control experiments with appropriate counterstaining.

(3) Parallels with neurodegenerative phenotype would be helpful to understand the mechanisms of PC loss in the future.

Weakness:

(1) Limited strain diversity: The study exclusively uses C57BL/6J mice despite known genetic and motor differences even the closely related strains like C57BL/6N.

(2) No correlation quantified between the degree of PC loss, aging, and motor performance.

(3) It has not been demonstrated whether the neurodegenerative changes are indeed observed in zebrin-negative PCs.

(4) The mechanisms of why only a subset of mice show PC loss remain unexplored and not discussed.

(5) Linkages with normal human aging and cerebellar function are not well supported. While motor behavioral assays captured phenotypes that mimic aged people, correlation with PC loss is demonstrated to be absent in mice. It remains unclear whether this PC loss phenomenon is universal or specific to a particular individual; and whether specific to a human PC subtype.

(6) Analyses in the paraflocculus are currently not easy to understand. This lobule has heterogeneous PC subtypes, developmentally or molecularly. Zebrin-weak and Zebrin-intense PCs are known to be arranged in stripes, which resembles the pattern of developmentally defined PC subsets (Fujita et al., 2014, Plos one; Fujita et al., 2012, J Neurosci). In the data presented, it is hard to appreciate whether the viewing angle is consistent relative to the angle of the paraflocculus. This may be a limitation of the analysis of the paraflocculus in general, that the orientation of this lobule is so susceptible to fixation and dissection. Discrepancy between PC loss stripe and zebrin pattern may be an overstatement, because appropriate analyses on the paraflocculus would require a rigorously standardized analytic method.

Reviewer #3 (Public review):

Summary:

Donofrio et al. report a new observation that in normal aging mice, anti-calbindin wholemount staining and coronal immunohistochemistry in the cerebellum often show a sagittally patterned loss of Purkinje cells with age. The authors address a central concern that calbindin antibody staining alone is not sufficient to definitively assess Purkinje cell loss, and corroborate their antibody staining data with transgenic Pcp2-CRE x flox-GFP reporter mice and Neutral Red staining. The authors then investigate whether this patterned Purkinje loss correlates with the known parasagittal expression of zebrin-II, finding a strong but imperfect correlation with zebrin-II antibody staining. They next draw a connection between this age-related Purkinje loss to the age-related decline in motor function in mice, with a trending but non-significant statistical association between the severity/patterning of Purkinje loss and motor phenotypes within cohorts of aged mice. Finally, the authors look at post-mortem human cerebellar tissues from deceased healthy donors between 21 and 74 years of age, finding a positive correlation between Purkinje degeneration and age, but with unknown spatial patterning.

Strengths:

The conclusions drawn from this study are well supported by the data provided. The authors highlight several examples of parasagittal patterning of Purkinje cell degeneration in disease, and show that proper methodologies must be used to account for these patterns to avoid highly variable data in the sagittal plane. The authors aptly point out that additional work is needed to investigate the spatial patterns of Purkinje cell loss in the human cerebellum.

Weaknesses:

(1) In Figure 3, the authors use Pcp2-CRE mice to drive GFP expression in Purkinje cells in order to avoid the confounding variable of loss of calbindin expression in aging Purkinje cells. The authors go on to say, "we argue that calbindin expression alone is not a reliable, sufficient indicator of Purkinje cell loss". However, in Figure 4, the authors return to calbindin staining alone to assess the correlation of Purkinje cell loss with zebrin-II expression. Could the authors comment on why zebrin-II co-staining experiments were not performed in GFP reporter mice to avoid potential confounds of calbindin expression? Without this experiment, should readers accept the data presented in Figure 4 as a "reliable, sufficient indicator of Purkinje cell loss", given the author's prior claim?

(2) Throughout the manuscript, there is a considerable reliance on the authors' interpretation of imaging data with no accompanying quantification (categorization of "striped" or "non-striped" PC loss, correlation of GFP/calbindin/zebrin-II staining, etc.). While this may be difficult to obtain, the results would be much stronger with a quantitative approach to support the stated categorizations/observations.

Author response:

We thank all three reviewers for providing excellent suggestions that we feel will enhance the clarity and impact of our manuscript. When we submit the revised manuscript, we plan to respond to each comment and provide additional data and discussion points as requested. Below, we include an outline of the main points that we intend to address.

(1) Reviewers 1 and 2 both suggested investigating degenerative changes in Purkinje cells that are more resistant to age-related loss. We will look for hallmarks of neurodegeneration, such as shrunken dendrites and axonal swellings, in two areas: surviving Purkinje cells adjacent to stripes of cell loss, and the Purkinje cells in aged mice without Purkinje cell loss.

(2) We agree with Reviewer 2’s point that our manuscript would benefit from discussion of the differences in vulnerability between individual mice. Therefore, we will elaborate upon possible reasons why some aged mice are more resistant to age-related Purkinje cell loss than others.

(3) We will take Reviewer 3’s suggestion to perform zebrin II co-staining in our GFP reporter mice, given our findings that calbindin staining can be unreliable in this context. 4) We appreciate Reviewer 3’s comment that quantification would support the observations made in our study. To provide quantitative evidence for our categorization of mice with striped and non-striped Purkinje cell loss, we will measure the gaps (or lack thereof) between Purkinje cell bodies in the anterior zone.

(4) We will also incorporate several minor but important changes suggested by all three reviewers.

Thank you to the reviewers and editors for taking the time and effort to review our manuscript.