Galectin and Myc enable cochlear progenitor expansion in vitro and in vivo

Reviewer #1 (Public review):

Summary:

The overall aims of this study are a bit unclear. The first experiments use organoids derived from cochlear GER cells in combination with single-cell RNA-seq to try to identify factors that might be important in the initiation of cellular proliferation, although the definition of proliferation is a bit loose and includes the number of organoids, the size of organoids, cell viability, and/or expression of Mki67.

Based on those results, the authors chose to focus on galectins 1 and 3 and Myc. The reasoning for these choices is a bit unclear, as their ranks in the DE gene list are 51 and 67, and the fold change for each is less than 2. Regardless, the subsequent experiments use inhibitors to examine the effects of galectins and Myc on proliferation of organoids. The results of these experiments do show an effect for inhibition of Lgals1 and Myc, although not Lgals3, but it was unclear whether the effects of these factors on growth could be separated from toxicity treatment, as both OTC008 and 10058-F4 seemed to lead to cell death.

Next, overexpression of Lgals1, 3 and Myc was actuated in organoids using AAV viruses. The results do show an effect on proliferation, but the results are confusing in that the mRNA expression profiles for two of the transgenes are markedly different in terms of timing, which would not be predicted based on similarities in the constructs. Also, while showing comparable results in some assays, the Myc vector is apparently toxic, killing ~25% of the cells by D9 even though mRNA levels are steady between D5 and D9 in those cells.

Finally, an in vivo model is used to kill several different types of cochlear cells followed by inhibition of Lgals1. The results of these experiments show a strong inhibition of expression of Ki67 following treatment with OTX008, which is intriguing. However, OTX008 was administered IP, and it does not appear that the ability of OTX008 to cross the blood-labyrinth or even blood-brain barrier has been examined. So it isn't clear whether the results of these experiments indicate a direct or indirect role for OTX008 and galectin-1 in cochlear proliferation. These issues need to be addressed.

Strengths:

The results present evidence for potential roles for galectins and myc in the modulation of proliferation of cochlear GER cells. In vitro and in vivo approaches are combined with single-cell profiling to provide a comprehensive analysis.

Weaknesses:

(1) Multiple transgenic mouse lines are used in this study, but there are no citations as to where these lines came from, how they were validated, and, for some inducible Cre lines, when the injections of tamoxifen were made.

(2) Sixty-four organoids were formed per well, but from an average of how many seeded single GER cells? This is not clear (page 5, third paragraph).

(3) Page 6: Why was cluster 7 grouped with clusters 1,2 and 3? Most cluster 7 cells are from D1.

(3) In Figure 3A, there does not appear to be a correlation between expression of either galectin-1 or galectin-3 and expression of Mki67, which I would expect would be predicted if these markers play a role in proliferation.

(4) Figure 3C: A more direct way to examine this would be immunofluorescence for galectin-1 and galectin-3 on cochlear tissue. This would also indicate whether galectin expression correlates with the Sox2+/Fgfr3- population of GER cells.

(5) For the data shown in Figure 4, what were the experimental conditions? In particular, how long in culture? One interpretation of the data in 4B and E is a decreased increase in the number of organoids, but an alternative is that the treatments are toxic and the organoids are dying. Based on a comparison with the results for myc inhibition, isn't cell toxicity in response to treatment with OTX008 or GB1107 the more likely explanation?

(6) I think the data in Figure 5 show that the inhibitor experiment demonstrates that the inhibitors, or their targets, are required for organoid survival, as the number of organoids drops to 0, which must be below the starting value.

(7) It is suggested (page 11, third paragraph) that galectins and myc could be linked or independent effectors of organoids. But couldn't this be tested by combining the inhibitors in the same experiment?

(8) On page 12, it seems AAV infection of the target cell population prevented organoid formation? This could be a major concern. If nothing else, doesn't this suggest that the effects observed in these experiments might be a result of induced organoid formation from other cochlear duct cells? Also, was expression of the transgenes (Lgals or Myc) confirmed in a cell type that is normally negative for those genes?

(9) The data in Figure 6C are confusing. The rate of mRNA expression from the AAV transgene should be comparable regardless of the construct given that the promoter is the same. But the results suggest a significant difference in the behavior of the two vectors, with Myc levels reaching a 15-fold increase in just three days while the Lgals vector is at only half that level after 7 days.

(10) An increase that is not significant is not an increase and should not be described as one (page 13 in the first paragraph).

(11) In the AAV-Myc experiments, the overall level of mRNA for Mki67 on D9 is comparable to that in the AAV-lgals1 AAV (Figure 6B), but 25% of the cells are dead (page 13, first paragraph)? Similarly, in Figures 6E and 6F, the number of organoids in the AAV-Myc samples is significantly larger than in either control or Lgals, but are most of those cells dead, then?

(12) Regarding the isolation process in Figure 7A, I am concerned this will also isolate cells from the stria vascularis? Do they retain a greater potential for growth that might lead to their predominance in the growth assay?

(13) Was the Ki67creERT2 used to label a subset of cells for FACS (page 14)? If not, why was this included? If so, when was the induction made? And doesn't this bias the selection to cells that were proliferating at the time of the induction?

(14) It is stated that "proliferation is most active at P4 with robust cycling of cells observed in the lateral GER". But then on the following page (page 15), it's stated that the single cell data indicates essentially no proliferating cells in the control, even though there are a lot of lateral GER cells. Can the authors give an explanation for this discrepancy?

(15) In the first figures in the study, the isolation approach collected lateral GER cells and identified Lgals and Myc as important for organoid expansion (page 15). In Figure 8, there appears to be no change in Lgals or Myc expression in lateral GER cells in response to the damage. Instead, it is medial GER cells that appear to have increased Lgals1 and Myc. And from Figure 8H, are those increases significant?

(16) A quick search of the literature suggests that there is no evidence that OTX008 can cross the blood-labyrinth or blood-brain barrier (page 15). Was this examined by the authors?

Reviewer #2 (Public review):

Summary:

The study uncovers novel factors driving proliferation in the greater epithelial ridge (GER), a proliferative tissue in the neonatal cochlea that may hold important clues on the quest for hair cell regeneration via proliferative means in the adult cochlea.

Strengths:

The strengths include the use of both cochlear organoids and in vivo mouse models combined with pharmacological and genetic approaches to inhibit or overexpress proliferative targets identified in the RNA-Seq analysis from the FACS-sorted GER cells. The genetic and pharmacologic manipulation experiments are very strong and convincingly demonstrate that galectins 1 and 4 and Myc are necessary (and in some cases sufficient) to drive cell proliferation in cochlear tissue. However, the real treasure trove is the carefully generated RNA-Seq dataset itself, which offers a wealth of additional differentially-expressed genes that likely contribute to cochlear cell proliferation.

Weaknesses:

The primary weakness is that the initial genes studied here, galectins 1 and 3 and Myc, are all associated with tumor formation or cancer progression, so targeting these genes raises concerns about tumor formation in the cochlea.