Article activity feed

  1. Evaluation Summary:

    Mutations in MINAR2 causes deafness in human and mice. Loss of function of Minar2 in mice causes a reduction of stereocilia and subsequent hair cell degeneration but the underlying mechanism is unclear. This zebrafish study demonstrated that the Minar2 protein interacts with cholesterol and is localized to the stereocilia of hair cells. The loss of Minar2 reduces cholesterol enrichment in the stereocilia with concomitant accumulation in lysosomes. Thus, this study provides the mechanistic insight of Minar2 and the first glimpse at the importance of cholesterol homeostasis in hair cell function.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 agreed to share their name with the authors.)

  2. Reviewer #1 (Public Review):

    In this paper, Gao et al report that Kiaa1024L/Minar2 causes hearing loss in mice and in zebrafish. The animal studies are well executed. Mechanistically, the authors claim that Kiaa1024L/Minar2 is responsible for the enrichment of an accessible pool of cholesterol in the hair bundle membrane. Increasing cholesterol levels rescues hair cell defects whereas decreasing cholesterol aggravates the problem.

    Unfortunately, the mechanistic arm of this study doesn't go beyond this correlation. The characterization of cholesterol levels and pools is not rigorous and it is unclear why cholesterol matters for hearing.

  3. Reviewer #2 (Public Review):

    Gao and colleagues examine the roles of the zebrafish kiaa1024L/minar2 gene, previously identified to cause hearing loss in mice, in hair cells of the lateral line system, and in the inner ear. A very recent paper, published after the submission of this manuscript, also identifies mutations in this gene underlying hearing loss in human patients. However the cellular functions of this gene remained unknown.

    In the present work, the authors demonstrate that minar2 is necessary for hair cell function and hearing in zebrafish by generating mutations in the gene using CRISPR/Cas9. They show localization of GFP-tagged Minar2 protein to hair cell stereocilia. They provide evidence that Minar2 interacts with cholesterol through conserved binding domains, that cholesterol is also enriched to the stereocilia compartment, and that the loss of minar2 reduces this enrichment with concomitant accumulation in lysosomes. Moreover, drugs that decrease cholesterol exacerbate some aspects of the minar2 mutant phenotype, while drugs that promote cholesterol accumulation rescue mutant phenotypes including some hearing restoration.

    The work has several strengths. The authors take a multifaceted approach to the problem: using behavioral and electrophysiological methods to assess hearing, using both in vivo and cell culture methods to probe protein function, computational analysis to develop hypotheses about function, and pharmacological interventions to test their hypotheses.

    Some minor weaknesses detract from the work. In particular additional details are needed in the methods section so that image analysis is reproducible. Nevertheless, this is a major contribution to the field and is likely to have a significant impact.

  4. Reviewer #3 (Public Review):

    In the manuscript by Gao et al, the authors were trying to achieve an understanding of how Kiaa1024L/Minar2 is necessary for hearing in vertebrates. It is known that the Kiaa1024L/Minar2 mutation causes deafness in mice but not much beyond that is known.

    - In this manuscript, they were successful in making two zebrafish mutant zebrafish strains in the Kiaa1024L/Minar2 gene using Crispr/Cas9. The mutant(s) has defects in hearing (using the C-start assay and determining thresholds) and reduced hair cell numbers in the ear (phalloidin labeling to determine hair cell density in utricle and saccule) and the lateral line (including using the AM1-43 assay). From these data, they demonstrate that hair cells are defective in these mutants.

    - The authors show that Lamp1-GFP labeled lysosomes change in size in the minar2fs139 mutant. In addition, they show that GFP-Minar2 localizes to lysosomal membranes in cultured cells (human and monkey).

    - They performed primary amino acid sequence analysis on Minar2 and showed that it contained a putative CSD of caveolin, which is known to interact with cholesterol. They then show that when Minar2 is expressed in cells in culture, there is an increase in cholesterol detection in the region that contained Minar2, supporting the idea that cholesterol interacts with Minar2.

    The experiments in figure 5 seem to show that lowering cholesterol levels using pharmacology exacerbates hair cell defects in a minar2 mutant.

    1. The authors attempt to show localization (Fig 2 A and B) of Minar2 to the stereocilia and the apical region of hair cells using GFP-MINAR2 fusion protein expression in hair cells of transgenic animals. Although this is a typical way of demonstrating localization, it is usually used to validate location after a similar pattern has been shown using an antibody (usually in mice.) So, special precautions must be taken when interpreting this kind of transgenic data. According to the authors, GFP-MINAR2 localized to the stereocilia and the apical region of hair cells. This needs to be validated by some other means. I can also see the localization of the green signal at the basolateral area of the cells in Fig 2a. Moreover, it's important to note that other mislocalized fusion proteins localize to the apical region of hair cells.

    2. Figure 2C and D. The defects in the hair bundles are plausible but not convincing. Electron microscopy should be used to validate. Also, are hair bundle defects seen in the neuromast? EM would be easier to do there.

    3. Fig 1A do prim 1- and prim 2-derived neuromasts express minar2? Do anterior neuromasts express minar2?

    4. It's my impression that the authors don't take into account that there is much more plasma membrane in the stereocilia than in the basolateral membrane. So, this statement, "These data suggested that there are high levels of accessible cholesterol located to the stereocilia membranes, while the accessible cholesterol levels are marked lower in the basolateral membranes in the hair cells" based on Figure 4 needs to be reconsidered. The authors need to show that the little reporter that is present in the basolateral membrane is not equal to the reporter present in a single sheet of the plasma membrane in a stereocilium. I can see basolateral labeling in the lateral line hair cells.

    5. It's not clear if there is a paralog of the Kiaa1024L/Minar2 gene.