Fig 2: Hair cell and support cell subtypes and dynamics during development.(a) Highlights the position of support cells and hair cells in UMAP space relative to the whole dataset (i.e. E16, E18, P3, and P7 combined). (b) Cluster analysis in support cells and hair cells. (c and d) show expression of markers and Notch pathway genes, respectively, relative to cell cluster, pseudotime and developmental time. Additional color bars indicate whether markers are known or novel in specificity, implicated in vestibular diseases and dysfunction or in the Notch pathway. (e) Pseudotime of the support cell–hair cells trajectory in cristae. Note that the trajectory between support cells and hair cells is continuous. (f) The relative contributions of support cell and hair cell subtypes to the composition of the support cell and hair cell clusters at the indicated developmental stages. Note that the contribution of the Id1HI support cell subtype decreases as that of Srxn1HI support cell subtype and hair cells increases. (g) ‘Blendplots’ show the overlap of marker specificity in UMAP. (h) shows differential expression of the Notch target Hes1 and the Hsp70 component Hspa1a across developmental stages in the support cell clusters. (i) Id1 (magenta) localization in a subset of Sox2+ (green) support cells in P0 B6 crista relative to phalloidin (cyan). (j) Atoh1 (green), Myo7a (blue) and Sox2 (green) IF in optical sections of P0 B6 mouse crista. Arrowheads indicate Atoh1+/Sox2+ nuclei in the support cell layer. (k) Hair cell counts made manually in whole B6 cristae at P0 (n = 15), P14-21 (n = 7), 3 months (n = 30), and 6 months (n = 6). Note the addition of ~500 hair cells. Asterisks indicate p<1×10-7. (l) The Id1+ percentage of Sox2+ support cells at P0 (n = 6) and P22 (n = 5). (m) The number Atoh1+ cells in the Sox2+ support cell layer per crista at P0 (n = 5) and P7 (n = 5). Scale bars = 50 µm.

Fig 3: (A) HSPA1A and (B) HSPA1B protein expression in DBMSC, DPMSC and pMSC, immediately at commencement of heat stress (0H) and during recovery at 37 °C at 1 h (1H), 6 h (6H), and 24 h (24H). More HSPA1A (A) in control samples as compared to HSPA1B (B). Protein expression is reported to be statistically significant ** at p-values < 0.05.

Fig 4: HSPA1B Expression Dominates HSPA1A Expression. (GE: Gene expression. PE: Protein expression) The time courses plotted here are from merged gene and protein expression Log2 data for the three PDSCs. For each cell type, the results are at least from two individual experiments. (A,B) Smoothened scatterplot of the averaged Log2 fold change in expression with respect to expression in control cells. (C,D) Smoothened scatterplot of the averaged fold change in expression with respect to expression at 0H. (E) Ratio of HSAP1B expression to HSPA1A expression. (F) Heat-map representation of the HSPA1B/HSPA1A ratios. HSPA1B dominance is higher in protein expression and is very well observed even 24 h post-stress exposure.

Fig 5: Interaction of Hsp70 with the 20 S proteasome is increased during recovery from oxidative stress. (A) HT22 cells were treated with 0.5 mM hydrogen peroxide for 0.5 hr. After recovery, cells were lysed using PBS. Lysates were used for immunoprecipitation of Hsp70 as described in methods. Immunoprecipitates (IP) were used for immunoblot against Hsp70, proteasomal subunit a4 and 20S core subunits. The columns display the ratios of co-immunoprecipitated proteasomal subunits per immunoprecipitated Hsp70 and are means ± SD, n = 3–4*P < 0.05 vs. control. (B) Hsp70 immunoprecipitation was performed from brain homogenates of three 18-month old 129/SV mice. The immunoprecipitates (IP) were used for immunoblots against Hsp70 and the proteasomal subunit a4. In panels C) and D) HT22 cells were transfected with HSPA1A-AC-GFP expression vector 24 hr after seeding. Further 24 hr later cells were incubated with 0.5 mM hydrogen peroxide for 0.5 hr. (C) At the indicated time points after treatment, cells were lysed using PBS, lysates were applied to a 10–40% sucrose gradient. Density gradient centrifugation was performed as described in methods. Fractions were analyzed for proteasomal activity and fractions with a high proteasomal activity were pooled and used for immunoblot against Hsp70 and 20S core subunits. The columns are the ratios between the amount of Hsp70 and the content of proteasomal subunits. The columns are means ± SD, n = 4–5, *P < 0.05 vs. control. Panel D) demonstrates the results of FRAP experiments 2 hr after hydrogen peroxide treatment. To inhibit the proteasome, experiments were performed in the presence of 2 µM LC. The data presented show the recovery time t1/2 after photobleaching in the presence of LC with the t1/2 of the non-LC-treated cells set as 100%. The insert shows the absolute recovery time t1/2 of all samples in seconds. The columns are the means ± SD, n = 4–5. *P < 0.05 vs. hydrogen peroxide control. Representative fluorescent images of one illustrative FRAP experiment are shown.