Oxidative stress and ROS metabolism via down-regulation of sirtuin 3 expression in Cmah-null mice affect hearing loss

Cmah disruption result in histological abnormalities in vestibular otoconial epithelia and the cochlear sensory epithelium, and neuronal cell loss of inner ears

Cmah-null mouse, like the previous observations, showed age-related hearing loss [5, 30]. To determine the lack of CMAH protein expression, we performed immunohistochemistry (IHC) in inner ear from WT- and Cmah-null mice. As shown in Fig. 1A, the Neu5Gc, which is a final product by hydroxylation of CMP-Neu5Ac to CMP-Neu5Gc by CMP-Neu5Ac hydro-xylase, was highly expressed in the WT, whereas the Cmah-null mice were completely deficient in inner ears. In histological analysis, 12 month- but not 5 week-old (Supplementary Fig. 1). Cmah-null mice had histological abnormalities in the vestibular otoconial epithelia and the cochlear sensory epithelium of inner ears (Fig. 1B). The apical surface of the vestibular otoconial epithelia contains unusual deposits of apparently a cellular material among the stereociliary bundles (Fig. 1B upper). Further, we observed degeneration of outer hair cell throughout the cochlea, with collapse of the outer organ of corti in the basal (high-frequency) turn of 12 month-old mice (Fig. 1B; bottom), whereas both of the cochlear and vestibular sensory epithelia of WT-mice appeared normal, even in older mice. Next, we performed the TUNEL assay to investigate the apoptotic cell death. The result shows significantly increased numbers of apoptotic cells in supporting cells of cochlear tissues from Cmah-null mice, compared to control (Fig. 1C).

In order to examine neuronal cell loss in Cmah-null mouse, we investigate the several neuronal cell-specific gene expressions in cochlear tissues. In this study, we examined glial fibrillary acidic protein (GFAP) expression for a neural injury marker or astrocyte marker, neurofilament (NF) expression for an axonal injury marker, vimentin (VIM) expression for a neural precursor cells marker, and enolase (NSE) expression for neuronal specific marker. In IHC, cochlear tissues of Cmah-null mice showed significant reductions in expressions of several neuronal cell markers such as GFAP, NSE, NF, and VIM in with respect to control mice (Fig. 1D). Reverse transcriptase quantitative real-time polymerase chain reaction (RT-qPCR) analysis also showed significantly down-regulated gene expression for same neuronal cell markers (Gfap, Nse, Nf, and Vim) compared to control (Fig. 1E), indicate that RT-qPCR results mimic IHC results.

Differential gene expression profile in cochlea of Cmah-null mice

Expression profiles in cochlear tissues of WT- and Cmah-null mice (n=3, respectively) were compared using an Illumina MouseRef-8 v2 Expression BeadChip, which is composed of 25,697 probe sets for mouse mRNA. To analyze differential gene expression in WT- and Cmah-null mice, we focused on the gene expressions, which showed ≥1.5 folds changes with a significant p-value (p < 0.05). In this study, we performed scatter plot analysis in order to estimate global gene expression change. As shown in Fig. 2A, mRNA expression was normalized by the scatter plots: of 25,697 genes, 631 genes were up-regulated, whereas 729 genes were down-regulated in Cmah-null mice-derived cochlear tissues compared to control mice-derived cochlear tissues (Fig. 2B). Additional File 1 and 2 showed complete lists of these genes.

Functional analysis of differentially expressed genes in cochlea of Cmah-null mice

To explore their biological functions, GO classification by functional annotation of the DAVID (Database for Annotation, Visualization and Integrated Discovery) tools categorize each genes with significant differential expression in cochlea of Cmah-null mice. Fig. 2C showed the top 10 enriched functional annotations in three categories according to GO classification. These functional annotations consisted of generation of precursor metabolites and energy (69 genes), regulation of transcription from RNA polymerase II promoter (60 genes), biological adhesion (58 genes), regulation of cell death (57 genes), regulation of biosynthetic process (55 genes), regulation of programmed cell death (55 genes), homeostatic process (55 genes), regulation of nitrogen compound metabolic process (50 genes), regulation of transcription (46 genes), regulation of biosynthetic process (44 genes) for biological process; calcium ion binding (77 genes), structural molecule activity (61 genes), RNA binding (59 genes), protein dimerization activity (39 genes), carbohydrate binding (35 genes), cofactor binding (34 genes), transcription activator activity (32 genes), transcription factor binding (30 genes), actin binding (30 genes), coenzyme binding (27 genes) for molecular function; mitochondrion (213 genes), extracellular region (163 genes), organelle membrane (115 genes), mitochondrial part (113 genes), cytoskeleton (100 genes), membrane-enclosed lumen (100 genes), organelle envelope (95 genes), extracellular region part (94 genes), mitochondrial membrane (84 genes), and extracellular space (61 genes) for cellular component.

Next, we examined differentially expressed genes that are enriched in pathways by KEGG pathway analysis. Table 1 show the 11 most significantly enriched pathways. Of note, most of them are closely associated with oxidative damage and metabolism, supporting the pathogenesis hypothesis of age-related dysfunction. These results are consistent with prior knowledge of AHL [5, 11, 29]. In addition, ingenuity analysis identified more putative diseases and disorders in cochlear tissues of Cmah-null mice. These are as follows: neurological disease, psychological disorders, skeletal and muscular disorders, infectious disease, and hereditary disorder (Table 2). Among them, neurological disease consisted of movement disorders (237 genes, 1.65E-13), disorder of basal ganglia (178 genes, 7.88E-13), neuromuscular disease (190 genes, 5.69E-11), neurological signs (151 genes, 6.19E-10), dyskinesia (143 genes, 1.34E-09), Huntington's disease (135 genes, 1,44E-09), progressive motor neuropathy ( 97 genes, 7.78E-06), damage of brain (31 genes, 1.25E-04), seizures (77 genes, 2.22E-04), hypomyelination (11 genes, 4.76E-04), damage of cerebral cortex (11 genes, 8.94E-04) , dysmyelination of sciatic nerve (4 genes, 9.96E-04), and seizure disorder (84 genes, 1.09E-03) (Fig. 3A). Furthermore, we determined the canonical pathways by IPA tool using genes differentially expressed in cochlear tissues of Cmah-null mice. Of them, the pathway classified aging-related functional groups as follow: glutathion-mediated detoxification, protein ubiquitination pathway, phagosome maturation, mitochondrial dysfunction, glutathione redox reaction I, acetyl-coA biosynthesis, oxidative phosphorylation, and TCA cycle for oxidative stress and mitochondrial dysfunction (Group I) and Wnt/β-catenin signaling (Group II) (Fig. 3B). Taken together, these results provide further information for evaluation of the potential effects and altered functions caused by Cmah gene disruption.

Alteration of genes involved in oxidative damage and ROS metabolism

To clarify differential mRNAs expression of oxidative stress and antioxidant defense-related genes, we examined the expression of genes related to oxidative damage and ROS metabolism pathway in the cochlear tissues from WT- and Cmah-null mice, using a pathway-focused PCR array. RT-qPCR analysis revealed a significant down-regulation in most of gene expressions in Cmah-null mice with AHL, except one gene, Apoe (Table 3). Down-regulated genes consisted of 3 specific subsets for oxidative damage and dysregulated ROS metabolism as follows: 1) glutathion peroxidases (Gpxs), peroxiredoxins (Tpxs), peroxidases, and antioxidant-related genes, 2) superoxide dismutases (Sod), genes involved in superoxide metabolism, genes involved in ROS metabolism, and oxidative damage responsive genes for ROS metabolism, and 3) oxygen transporters. Therefore, these results suggested specific subsets of oxidative stress genes that appear to play an important role in AHL in the Cmah-null mice.

Molecular mechanisms of sirtuins and oxidative stress regulation, and Wnt signaling that involved in Cmah-null mice with hearing loss

To address the underlying mechanisms of AHL in Cmah-null mouse, we analyzed the expression of their genes that involved sirtuins, oxidative stress regulation, Wnt signaling, mitochondrial dysfunction identified by pathway analysis on microarray and a pathway-focused PCR array. As shown in Fig. 4A, Sirt3 protein expression was significantly reduced in cochlear tissues of Cmah-null mouse, compared to their control. Also, RT-qPCR analysis showed that Sirt3, 4, and 5 gene expressions is significantly reduced in Cmah-null mice than in controls (Fig. 4B). To determine the expression of AHL related genes in the cochlear tissues of Cmah-null mouse, we checked the expression profiles of these genes on the microarray analysis (Fig. 4C). The result shows that the data sets obtained from the microarray analysis accurately matched with RT-qPCR analysis in Cmah-null mouse (Fig. 4C and D). Like that, expressions of Hif1α and Foxo3a mRNAs as transcription factor regulated by Sirt3 as well as Foxo1 and MnSod mRNAs expressions as downstream regulator of Sirt3 was decreased in Cmah-null mice (Fig. 4D).

To observe the regulatory factors in Wnt signaling, which might control by ROS in Cmah-null mouse, we analyzed the expression of Wnt signaling-related genes (Lrp5, Lrp6, Wnt3a, Wnt5a, Gsk3β, and Catenin). The result shows that Wnt signaling-related gene expressions in Cmah-null mouse were significantly decreased, whereas the Wnt receptor gene expression was not altered (Fig. 4D). Next, we examined mitochondrial dysfunction-related gene expression. The expression of these genes (Crb1, Mtfp1, RhoT2, Soc2, and Atp5f1) with a role as regulators and mediators of mitochondrial molecular transport was significantly down-regulated (Fig. 4D). We therefore investigated whether altered genes affect the mitochondria expression in sensory neurons of Cmah-null mouse by immunofluorescence. As shown in Fig. 4E, mitochondrial positive signals were significantly decreased in Cmah-null mice, compared to control mice. Taken together, we proposed a key underlying mechanism for explaining AHL in Cmah-null mice (Fig. 4F), suggested that AHL in Cmah-null mice was caused by down-regulation of Sirt3 mRNA expression, which resulted in histological abnormalities in vestibular otoconial epithelia and the cochlear sensory epithelium, and neuronal cell loss in inner ears due to mitochondrial dysfunction by excessive oxidative stress.

Animal ethics

All animal experiments were approved and performed under the guidelines of the Konkuk University Animal Care and Experimentation Community [IACUC approval number: KU12045]. Cmah knockout mice were kindly provided by RIKEN Center for Developmental Biology (CDB, Japan).

Immunohistochemistry (IHC)/ immunofluorescence

For histology, inner ears were harvested after intracardiac perfusion with 4% paraformaldehyde, postfixed overnight, decalcified in 8% EDTA, and embedded in paraffin. Sections were stained with Hematoxylin QS for light microscopy. For IHC, the cochlear tissues were fixed with 4% paraformaldehyde and then embedded on slides. After deparaffination and rehydration in PBS, sections were blocked in background Sniper solution. After washing, the samples were incubated with specific primary antibodies for Neu5Gc (1:200 dilution; Sialix, San Diego, CA, USA), GFAP, NSE, NF, VIM (prediluted: Leica Biosystems Inc. Buffalo Grove, IL, USA), and Sirt3 (1:200 dilution; Santa Cruz Biotehnology, Inc. Dallas, TX, USA) at 4°C overnight. After the incubation, the samples were washed and incubated with horseradish peroxidase-conjugated secondary antibody. Samples were then stained with ImmPACT^TM DAB peroxidase substrate (Vector Laboratories; CA, USA) to visualize the signal. Samples were also stained with Hematoxylin QS to provide background information for reference. The samples were mounted using VECTORSHIELD HardSet mounting medium (Vector Laboratories; CA, USA) and observed using fluorescence microscopy (Olympus; Japan). For immunofluorescence, the samples were incubated with primary antibodies for mitochondrial marker (1:1000 dilution, Abcam, Cambridge, UK) at 4°C overnight. After incubate with Alexa Fluor 488-labeled goat anti-mouse (1:500), the samples were mounted using VECTORSHIELD HardSet mounting medium with DAPI (Vector Laboratories) and observed using fluorescence microscopy (Olympus; Japan).

Apoptotic cell staining

The terminal deoxynucleotidyl transferase–mediated nick end labeling (TUNEL) method was employed for detection of apoptosis using the in situ cell detection kit, tetramethylrhodamine red (TMR, red fluorescence) from Roche Applied Science (Lewes, East Sussex, UK) according to the manufacturer's instructions. For detection of the characteristic chromatin condensation and nuclear fragmentation associated with apoptosis, the samples were mounted using VECTORSHIELD HardSet mounting medium with DAPI (Vector Laboratories) and observed using fluorescence microscopy (Olympus; Japan).

Microarray analysis of gene expression

WT- and Cmah-null mice with same age (12 months) and genetic background (C57BL/6J) were used (n=3 per each group) for microarray analysis. Total RNA was extracted and purified from the cochlear tissues of WT- and Cmah-null mice using Trizol and RNeasy columns (Qiagen; Valencia, CA, USA) according to the manufacturer's protocol. The RNA quality was verified using an Agilent Bioanalyzer 2100 (Agilent Technologies; Palo Alto, CA, USA) using the RNA 6000 Pico Assay. Generation of double-stranded cDNA, preparation and labeling of cRNA, hybridization to Mouse Ref-8 v2.0 Expression BeadChip (Illumina, Inc.; San Diego, CA, USA), washing, and scanning were all performed according to the standard Illumina protocol. Arrays were scanned using the Illumina Bead Array Reader Confocal Scanner. Microarray data have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE70659 (Gene expression profile in the cochlear tissue of Cmah-null mouse).

In silico data analysis

Differentially expressed genes (DEG) were subjected to intensive investigation to identify their biological functions. GO analysis was performed as described using DAVID (http://david.abcc.ncifcrf.gov) to study the biological function of the regulated genes [48]. The differentially expressed genes were compared to all of the genes on the microarrays and a P value computed using a Fisher's exact test. KEGG is a collection of online databases dealing with genomes, enzymatic pathways, and biological chemicals [49]. The KEGG PATHWAY Database records networks of molecular interactions in the cell that includes organism-specific network maps (http://www.genome.jp/kegg/). In addition, canonical pathways and functions analyses were created by IPA (Ingenuity Systems, Mountain View, CA, USA), which assists with microarray data interpretation via grouping DEG into known functions, pathways and gene interaction networks, based primarily on human and rodent studies. Significance was set at a P value of 0.05.

Pathway-focused PCR array

The Mouse Oxidative Stress and Antioxidant Defense RT² Profiler™ PCR array (SABiosciences Corp., Valencia, CA) was used to measure the alteration of gene expression in the cochlear tissues from WT- and Cmah-null mice. The RT² Profiler™ PCR Array (SABiosciences Corp.) is a commercially available set of optimized RT-qPCR primer assays on a 96-well plate. Total RNA was then treated with DNase I, reverse-transcribed using an RT² First Strand Kit (Qiagen), and brought to a final volume of 120 μL. cDNA from individual samples was used as a template for the PCR array, according to the array instructions, using SYBR green on an ABI ViiA^TM 7 system (Applied Biosystems). Data were analyzed using SABiosciences RT2 Profiler PCR Data Analysis software, available at http://pcrdataanalysis.sabiosciences.com/pcr/arrayanalysis.php, and were considered significant at >4-fold change. Relative quantitation for each gene was determined by normalization to 5 housekeeping genes (Actb, B2m, Gapdh, Gusb, and Hsp90ab1), comparing the WT and Cmah-null groups by using the 2-ΔΔCt method.

Quantitative real-time polymerase chain reaction (RT-qPCR)

The total RNA obtained from the cochlea of WT- and Cmah-null mice was reverse-transcribed with the QuantiTect Reverse Transcription Kit (Qiagen; Valencia, CA, USA) as recommended by the manufacturer. To evaluate gene expression, RT-qPCR was performed on an ABI ViiA^TM 7 system (Applied Biosystems; Foster City, CA, USA) with SYBR Green as the fluorescence detection method (Bio-Rad; Hercules, CA, USA). The mouse Gadph gene was used as an internal control to normalize the RT-qPCR efficiency and for quantification of gene expression (Supplementary Table 1 and 2). The RT-qPCR was performed in triplicate for each sample.

Statistical analysis

All experimental data was presented as means ± standard deviation (SD). Each experiment was performed at least three times. For statistical analysis (Fig. 1 and 4), one-way analysis of variance (ANOVA) was performed to determine whether there were differences among the groups, and Fisher's post-test was performed to determine significance between pairs of groups. In all experiments, *P < 0.05, **P < 0.01, and ***P < 0.001 were considered significant.