Endosomal dysfunction contributes to cerebellar deficits in spinocerebellar ataxia type 6

Joint Public Review:

Summary:
Cook, Watt, and colleagues previously reported that a mouse model of Spinocerebellar ataxia type 6 (SCA6) displayed defects in BDNF and TrkB levels at an early disease stage. Moreover, they have shown that one month of exercise elevated cerebellar BDNF expression and improved ataxia and cerebellar Purkinje cell firing rate deficits. In the current work, they attempt to define the mechanism underlying the pathophysiological changes occurring in SCA6. For this, they carried out RNA sequencing of cerebellar vermis tissue in 12-month-old SCA6 mice, a time when the disease is already at an advanced stage, and identified widespread dysregulation of many genes involved in the endo-lysosomal system. Focusing on BDNF/TrkB expression, localization, and signaling they found that, in 7-8 month-old SCA6 mice early endosomes are enlarged and accumulate BDNF and TrkB in Purkinje cells. Curiously, TrkB appears to be reduced in the recycling endosomes compartment, despite the fact that recycling endosomes are morphologically normal in SCA6. In addition, the authors describe a reduction in the Late endosomes in SCA6 Purkinje cells associated with reduced BDNF levels and a probable deficit in late endosome maturation.

Strengths:
The article is well written, and the findings are relevant for the neuropathology of different neurodegenerative diseases where dysfunction of early endosomes is observed. The authors have provided a detailed analysis of the endo-lysosomal system in SCA6 mice. They have shown that TrkB recycling to the cell membrane in recycling endosomes is reduced, and the late endosome transport of BDNF for degradation is impaired. The findings will be crucial in understanding underlying pathology. Lastly, the deficits in early endosomes are rescued by chronic administration of 7,8-DHF.

Weaknesses:
The specificity of BDNF and TrkB immunostaining requires additional controls, as it has been very difficult to detect immunostaining of BDNF. In addition, in many of the figures, the background or outside of Purkinje cell boundaries also exhibits a positive signal.

One important concern about the conclusions is that the RNAseq experiment was conducted in 12-month-old SCA6 mice suggesting that the defects in the endo-lysosomal system may be caused by other pathophysiological events and, likewise, the impairment in BDNF signaling may also be indirect, as also noted by the authors. Indeed, Purkinje cells in SCA6 mice have an impaired ability to degrade other endocytosed cargo beyond BDNF and TrkB, most likely because of trafficking deficits that result in a disruption in the transport of cargo to the lysosomes and lysosomal dysfunction. Moreover, the beneficial effects of 7,8-DHF treatment on motor coordination may be caused by 7,8-DHF properties other than the putative agonist role on TrkB. Indeed, many reservations have been raised about using 7,8-DHF as an agonist of TrkB activity. Several studies have now debunked (Todd et al. PlosONE 2014, PMID: 24503862; Boltaev et al. Sci Signal 2017, PMID: 28831019) or at the very least questioned (Lowe D, Science 2017: see Discussion: https://www.science.org/content/blog-post/those-compounds-aren-t-what-you-think-they-are Wang et al. Cell 2022 PMID: 34963057). Another interpretation is that 7,8-DHF possesses antioxidant activity and neuroprotection against cytotoxicity in HT-22 and PC12 cells, both of which do not express TrkB (Chen et al. Neurosci Lett 201, PMID: 21651962; Han et al. Neurochem Int. 2014, PMID: 24220540). Thus, while this flavonoid may have a beneficial effect on the pathophysiology of SCA6, it is most unlikely that mechanistically this occurs through a TrkB agonistic effect considering the potent anti-oxidant and anti-inflammatory roles of flavonoids in neurodegenerative diseases (Jones et al. Trends Pharmacol Sci 2012, PMID: 22980637).

Author Response

We would like to thank the reviewers for their careful reading of the manuscript and for the positive feedback and constructive criticism that they have provided. We intend to incorporate this feedback into an improved and updated version of the manuscript. We will address the reviewer comments point by point when we submit an updated version but for now we would like to discuss the major points that we intend to address.

The first concern raised by the reviewers related to the specificity of the BDNF and TrkB staining. We agree that this is an important concern. We tested several antibodies and staining protocols and found that the optimal protocol involved the antibody used in this paper (abcam ab108319), in combination with a heat induced epitope retrieval (HIER) step. Together, this gave robust staining of BDNF in cerebellar tissue and the results of quantification of the staining were in agreement with a BDNF ELISA that we carried out to measure levels of BDNF in the cerebellar vermis of WT and SCA6 mice (Cook et al., 2022). We outline the epitope retrieval method briefly in the methods section of this manuscript but in a revised version we will include further details and data showing the troubleshooting and validation experiments that we have conducted.

Another concern raised by the reviewers is that 7,8-DHF may not be acting as a TrkB agonist. There has been controversy over the mechanism of action of 7,8-DHF and we welcome the opportunity to discuss the issue further in the present manuscript. We have some evidence that 7,8-DHF is acting via TrkB in this case, as we had previously shown that 7,8-DHF administration to SCA6 mice leads to increased cerebellar TrkB levels and phosphorylation of Akt, an activation event known to be downstream of TrkB (Cook et al., 2022). This implicates TrkB in the mechanism of rescue in this case, but we have not demonstrated this directly. We acknowledge that 7,8-DHF could be acting via a different mechanism, such as anti-oxidant or anti-inflammatory effects. This would be interesting and could be followed up on in the future, potentially providing further insights into the pathophysiology of SCA6. We plan to revise the manuscript and provide additional discussion of the potential mechanism of action of 7,8-DHF. Despite this uncertainty, we believe that the finding that 7,8-DHF rescues early endosome abnormalities is a valuable addition to the paper. Whatever the mechanism of 7,8-DHF, this compound holds promise for potential treatment of SCA6.

With further staining experiments and addition of information to the text, we feel confident that we can address the concerns of the reviewers and that an updated version will strengthen our manuscript and thereby provide valuable insight into the pathophysiology and potential treatment of SCA6.