Histone Deacetylase Inhibitor Induces the Expression of Select Epithelial Genes in Mouse Utricle Sensory Epithelia-Derived Progenitor Cells

Abstract

Mouse utricle sensory epithelial cell–derived progenitor cells (MUCs), which have hair cell progenitor and mesenchymal features via epithelial-to-mesenchymal transition (EMT) as previously described, provide a potential approach for hair cell regeneration via cell transplantation. In this study, we treated MUCs with trichostatin A (TSA) to determine whether histone deacetylase inhibitor is able to stimulate the expression of epithelial genes in MUCs, an essential step for guiding mesenchymal-like MUCs to become sensory epithelial cells. After 72 h of TSA treatment, MUCs acquired epithelial-like features, which were indicated by increased expression of epithelial markers such as Cdh1, Krt18, and Dsp. Additionally, TSA decreased the expression of mesenchymal markers, including Zeb1, Zeb2, Snai1, and Snai2, and prosensory genes Lfng, Six1, and Dlx5. Moreover, the expression of the hair cell genes Atoh1 and Myo6 was increased in TSA-treated MUCs. We also observed significantly decreased expression of Hdac2 and Hdac3 in TSA-treated MUCs. However, no remarkable change was detected in protein expression using immunofluorescence, indicating that TSA-induced HDAC inhibition may contribute to the initial stage of the mesenchymal-to-epithelial phenotypic change. In the future, more work is needed to induce hair cell regeneration using inner ear tissue–derived progenitors to achieve an entire mesenchymal-to-epithelial transition.

Introduction

Mammalian inner ear hair cells are the primary receptors for the auditory and vestibular system. These hair cells are vulnerable to a number of insults, such as sound overstimulation, aging, genetic disorders, and ototoxic drugs. The mammalian inner ear has a limited ability in to regenerate hair cells (Forge et al., 1993; Roberson and Rubel, 1994; Warchol et al., 1993). Therefore, damage to mammalian hair cells usually results in irreversible and permanent hearing loss. In recent years, regeneration of hair cells has yielded promising results (Groves et al., 2013; Hu and Ulfendahl 2013; Okano and Kelley 2012; Ronaghi et al., 2012; Shi and Edge 2013), and hair cell progenitors have been identified in the adult mammalian utricle and cochlea (Li et al., 2003; Mizutari et al., 2013; White et al., 2006). However, the existence of a specific progenitor population is still controversial (Hu and Ulfendahl 2013). It has been reported that sphere-forming cells exist in the inner ear sensory epithelia (Li et al., 2003; Oshima et al., 2007). Some studies have suggested that supporting cells were the likely source of progenitors (Kelley et al., 1995; Mizutari et al., 2013; White et al., 2006). Recent studies have found that a subpopulation of supporting cells, such as Sox2- and Lgr5-positive supporting cells, have the capacity to become new hair cells (Chai et al., 2012; Shi et al., 2012, 2013).

Recently, cell reprogramming technology has been used in stem cell research. A dedifferentiation program was suggested for inducing differentiated somatic cells to become pluripotent embryonic-like stem cells (Halley-Stott et al., 2013; Takahashi et al., 2007; Takahashi and Yamanaka, 2006; Yu et al., 2007). Such a dedifferentiation program occurs in mammalian pancreatic epithelial cell cultures, in which pancreatic cells dedifferentiate into mesenchymal-like cells that are able to re-enter the cell cycle, probably via epithelial-to-mesenchymal transition (EMT) (Gallo et al., 2007; Gershengorn et al., 2004; Joglekar and Hardikar, 2010). In the auditory system, we have generated mouse utricle sensory epithelial cell–derived progenitor cells (MUCs) via EMT (Zhang and Hu, 2012). We found that adult mouse utricle sensory epithelial cells underwent EMT to become MUCs that were able to propagate in adherent cultures, express mesenchymal markers, and lose the expression of epithelial markers. Remarkably, these mesenchymal-like MUCs expressed prosensory cell markers, which indicated that MUCs were prosensory-like hair cell progenitors that may be able to differentiate into new hair cells. Prosensory cells are hair cell progenitors because they differentiate into hair and supporting cells during development (Kelley, 2006). In suspension cultures, we found that MUCs formed spherical structures and started to express epithelial markers and bundle-like structures, which indicates that MUCs may possess the potential to become epithelial hair-like cells. However, the yield of epithelial bundle–bearing cells is limited in MUC suspension cultures and/or hanging drops. Furthermore, floating spheres in the suspension culture may drift with fluids and are difficult to handle in vitro and/or in a site-specific implantation, which may restrict its future in vivo application in animal research. In this study, we maintained mesenchymal-like MUCs in adherent cultures to test new methods that were able to guide MUCs to become a number of epithelial cells expressing epithelial genes such as Cdh1. However, how adherent-cultured MUCs can be stimulated to become epithelial cells is still an open question.

Previous studies suggest that epigenetic modifications are involved in the regulation of gene expression (Cheung and Rando, 2013; Liu et al., 2008; Papp and Plath, 2013). Histone modification, which is accomplished by the balance of histone acetyltransferases (HATs) and histone deacetylases (HDACs), is one of the common epigenetic regulation mechanisms involved in the control of the accessibility of the transcriptional machinery to promoter and enhancer elements in the genome. It has been reported that HDAC inhibitors have multiple biologic effects consequent to alternations in histone acetylation (Chikamatsu et al., 2013). Several HDAC inhibitors have been used not only as anticancer drugs but also as a transdifferentiation reagent. For example, an HDAC inhibitor could induce the expression of differentiation marker proteins, morphological change, and functional cytodifferentiation in both human endometrial stromal and epithelial cells (Uchida et al., 2006). In addition, trichostatin A (TSA) preserved E-cadherin gene expression via inhibition of EMT in hepatocytes (Kaimori et al., 2010) and renal proximal tubular epithelial cells (Yoshikawa et al., 2007). However, it is unclear whether a HDAC inhibitor is able to stimulate mesenchymal-like progenitors to restore the expression of epithelial features. In this study, we treated prosensory-like MUCs with TSA to determine whether a HDAC inhibitor is able to induce the expression of epithelial marker Cdh1 in MUCs.

Materials and Methods

MUC culture and TSA treatment

Primary cultures of MUCs have been generated using pure utricle sensory epithelia collected from 3-month-old Swiss Webster mouse in our previous study (Zhang and Hu, 2012). MUCs used in this study were obtained from frozen stock at passages 33–35. MUCs were transferred from the frozen vial into a 15-mL conical tube containing prewarmed Dulbecco's modified Eagle medium (DMEM)/F12 (Invitrogen), followed by centrifugation and resuspension. MUCs were maintained in a 60-mm culture dish containing DMEM/F12-GlutaMAX and 10% fetal bovine serum (FBS; Invitrogen) at 37°C in a 5% CO₂ incubator. MUCs were cultured for a few passages, and passages 37–42 of MUCs were used in this study. When MUCs reached approximately 70–80% confluence, TSA dissolved in dimethyl sulfoxide (DMSO; Fisher) was added to culture medium at 0.1, 0.3, 0.5, or 1 μM. In control cultures, vehicle (DMSO, diluted to 1:10⁴) was added to the MUCs. MUCs were observed daily using phase-contrast microscopy and imaged with a digital camera. Half of the culture medium was changed at 1–2 days, and the cultures were maintained for 72 h.

Viability and proliferation assay

The viability of MUCs was assessed with calcein and propidium iodide (PI) (both from Invitrogen). After treatment with 0.1, 0.3, 0.5, or 1 μM TSA for 72 h, MUCs were rinsed with phosphate-buffered saline (PBS) and incubated in the culture medium containing 2 μM calcein and 0.3 μg/mL PI for 30 min at 37°C in a 5% CO₂ incubator. MUC samples were photographed using epifluorescence microscopy (n=5). Analysis of variance (ANOVA) was performed, and p<0.05 was determined as statistical significance in this study.

To evaluate cell proliferation, a bromodeoxyuridine (BrdU) incorporation assay was used as previously described (Zhang and Hu, 2012). Briefly, BrdU (3 μg/mL; Sigma) was added to the MUC cultures for 24 h; these cultures had already been treated with TSA for 48 h. At the end of the culture period, MUC samples were fixed in 4% paraformaldehyde (Sigma). The samples were denatured using 1 M HCl and preincubated in blocking solution containing 5% donkey serum (Millipore) and 0.2% Triton X-100 (Sigma) in PBS. MUC samples were incubated in antibodies specific for BrdU (1:100; BD Biosciences) overnight at 4°C, followed by DyLight conjugated secondary antibodies incubation at room temperature for 2 h. A universal nucleus marker, 4′,6-diamidino-2-phenylindole, (DAPI; Invitrogen), was used to label all of the nuclei in the samples. MUC samples were observed using epifluorescence microscopy or confocal microscopy (both from Leica). The percentage of BrdU-incorporated cells=(BrdU-positive cells)/(DAPI-positive cells). ANOVA was performed, and p<0.05 was determined as statistical significance.

RNA extraction, reverse transcription PCR, and real-time quantitative RT-PCR

At the end of the culture period, MUCs in the TSA and control groups were harvested, and the total RNA was extracted using RNeasy Mini Kit (Qiagen), followed by cDNA synthesis using a QuantiTect Reverse Transcription Kit (Qiagen) according to manufacturer's protocols. Reverse transcription (RT)-PCR was performed using a thermal cycler (Eppendorf ), and a Bio-Rad CFX system was used for quantitative PCR using SsoAdvanced™ SYBR^® Green Supermix (Bio-Rad). Quantitative PCR was applied when RT-PCR indicated obvious gene expression changes. Both RT-PCR and quantitative PCR used the same pairs of primers (Table 1) which met the following criteria: (1) The melting point in the melting curve analysis was 80–90°C and (2) the efficiency values of primers were 90–110%. The mean of quantification cycle (Cq) was calculated with Bio-Rad CFX software in a regression mode set. All cDNA samples were tested by quantitative PCR at a final concentration of 20 μg/mL. The relative expression values of mRNA were delta/delta values normalized by the internal control gene Gapdh. Cq values >40 were considered to be no expression of mRNA in genes. A two-tailed Student's t-test was used to compare the expression of genes in TSA and control groups. In this study, p<0.05 was considered statistically significant.

Table 1.

Primers for RT-PCR and Quantitative PCR Analyses

Gene	Forward: 5′→3′	Reverse: 5′→3′	Product length (bp)
Gapdh	GGCCGCATCTTCTTGTGCAGT	TGCAAATGGCAGCCCTGGTGA	111
Cdh1	ATTCAAAGTGGCGACAGACGGC	ACCTGGGTACACGCTGGGAAACAT	223
Krt8	CAAGGTGGAACTAGAGTCCCG	CTCGTACTGGGCACGAACTTC	187
Krt18	ACTCCGCAAGGTGGTAGATGA	TCCACTTCCACAGTCAATCCA	162
Tjp1	AGATGTTTATGCGGACGGTGGCGT	AGCGTCACTGTGTGCTGTTCCCAT	171
Dsp	AGCCCTTTACAAGGCCATCAGCGT	TGTTCCACTGAACCAGCGTCCACA	198
Zeb1	ACTGCAAGAAACGGTTTTCCC	GGCGAGGAACACTGAGATGT	127
Zeb2	CCACGCAGTGAGCATCGAA	CAGGTGGCAGGTCATTTTCTT	131
Snai1	ATGCACATCCGAAGCCACACG	TGGAGCAAGGACATGCGGGAGAA	245
Snai2	CATCCTTGGGGCGTGTAAGTC	GCCCAGAGAACGTAGAATAGGTC	186
Fn1	GTGACACTTATGAGCGCCCTA	CCACTTGTCGCCAATCTTGTA	137
Vim	AAGCCGAAAGCACCCTGCAGTCAT	AGGTCAGGCTTGGAAACGTCCACA	202
Hes1	AGCACAGAAAGTCATCAAAGCC	ATGTCTGCCTTCTCTAGCTTGG	142
Six1	AAGGAAAGGGAGAACACCGAA	TGGAGCAGAAGGACCGAGTTC	152
Dlx5	CACCACCCGTCTCAGGAATC	GCTTTGCCATAAGAAGCAGAGG	125
Jag1	CCTCGGGTCAGTTTGAGCTG	CCTTGAGGCACACTTTGAAGTA	150
Cdkn1b	GCGGTGCCTTTAATTGGGTC	TTCGGGGAACCGTCTGAAAC	197
Isl1	ATGATGGTGGTTTACAGGCTAAC	TCGATGCTACTTCACTGCCAG	174
Lfng	GCCGTCAAGACCACCAGAAAG	GGTCATACTCCACAGCCATCTT	208
Sox2	GCGGAGTGGAAACTTTTGTCC	CGGGAAGCGTGTACTTATCCTT	157
Bmp4	CATGAGGGATCTTTACCGGCTC	TCTCCAGATGTTCTTCGTGATGG	140
Myo7a	GCACTTCATCATCGGCAACGGCAT	GCTGCTCTTGGATGGGTTGTGTGT	100
Myo6	GGCATCGTCCCAAGAGATTTTC	CCACAATGTCAAAGTTCGGTACA	150
Atoh1	CCCGTCCTTCAACAACGACAAG	AGGTGATGGTGGTCATTTTTGC	156
Hdac1	TGCTGTGAACTACCCACTGC	TGAAGCAACCTAACCGGTCC	158
Hdac2	AGGAGGTCGTAGGAATGTTGC	AGTTGTTCTGACTTGGCTCCTT	168
Hdac3	GCATTCGAGGACATGGGGAA	GGTCCAGATACTGGCGTGAG	255

Immunofluorescence

At 72 h after TSA or vehicle (DMSO) treatment, MUCs were fixed in 4% paraformaldehyde at room temperature for 10 min. MUCs were preincubated for 30 min in the blocking solution containing 5% donkey serum and 0.2% Triton X-100 in PBS, followed by incubation in primary antibodies for 3 h at room temperature. Secondary antibodies were Dylight-488, -549, or -649 conjugated antibodies (1:500; all from Jackson Immunoresearch). DAPI was used to label all the cells in the sample. Most of the primary antibodies have been used in our previous studies (Hu et al., 2012; Zhang and Hu, 2012), including E-cadherin (1:100; Santa Cruz), ZO-1 (tight junction protein; 1:100; Life Technologies); ZEB1 (1:100; Santa Cruz), N-cadherin (1:100; Millipore), fibronectin (1:100; Millipore), P27^Kip1 (1:100; Thermo Scientific), myosin VIIa (1:200; Developmental Studies Hybridoma Bank), and myosin VI (1:100; Sigma). MUC samples were observed using epifluorescence microscopy, and digital images were captured using a QImaging monochrome cooled CCD camera.

Results

Viability and proliferation of MUCs following TSA treatment

To evaluate the effective and nontoxic concentration of TSA used for prosensory-like MUCs, a dose–response study was performed using viability assays. A calcein viability study revealed viable cells labeled with calcein showing strong fluorescence in the cytoplasm (Fig. 1A). The quantitative study indicated that 97.65±0.70%, 97.86±0.69%, 98.12±1.40%, 90.65±2.05%, and 99.19±0.28% of MUCs were labeled with calcein when treated with 0.1, 0.3, 0.5, and 1 μM TSA and control medium for 72 h, respectively (Fig. 1B). Statistical analysis suggested significant difference in the calcein viability assay (p<0.001; ANOVA). The Tukey post hoc test revealed that there was no significant difference in the viability of MUCs treated with 0.1, 0.3, and 0.5 μM TSA and the control medium, but these cells were significantly different from MUCs treated with 1 μM TSA (p<0.05). In PI staining, the nuclei of dead cells were labeled with PI and fluorescence was shown in the nuclei (Fig. 1A). The cell counting showed 2.36±0.66%, 2.14±1.49%, 1.88±1.56%, 9.35±0.25%, and 1.03±0.25% of PI-positive MUCs in the treatment with 0.1, 0.3, 0.5, and 1 μM TSA and control medium for 72 h, respectively (Fig. 1B), which was statistically significant (p<0.001; ANOVA). A Tukey post hoc test did not indicate significant difference in the PI staining of MUCs treated with 0.1, 0.3, and 0.5 μM TSA and the control medium, but MUCs treated with 1 μM TSA were significantly different from other treatments (p<0.05).

FIG. 1.

Viability and proliferation of MUCs following TSA treatment. (A) Calcein and PI staining assays were used to label viable and dead MUCs (arrows; enlarged in the insert), respectively, when MUCs were treated with control medium (ctrl) and medium containing 0.1–1 μM TSA using epifluorescence microscopy. Scale bar, 100 μm. (B) The percentage of viable MUCs in TSA and control groups. The percentage of viable cells=(calcein positive cells)/(calcein positive cells+PI positive cells). Quantitative study indicated that 97.65±0.70%, 97.86±0.69%, 98.12±1.40%, 90.65±2.05%, and 99.19±0.28% of MUCs were labeled with calcein when treated with 0.1, 0.3, 0.5, and 1 μM TSA and control medium for 72 h. A Tukey post hoc test revealed no significant difference in the viability of MUCs treated with 0.1–0.5μ TSA and control medium. However, the percentage of viable cells were significantly lower in 1 μM TSA than other groups. (*) p<0.05. (C) The percentage of BrdU-incorporated cells in control and TSA groups. After treatment with TSA for 72 h, BrdU incorporation study revealed that 12.62±5.85%, 9.89±6.98%, 9.87±6.39%, and 68.28±9.57% of MUCs incorporated BrdU in 0.1, 0.3, and 0.5 μM TSA and control medium, respectively. The percentage of BrdU-incorporated cells in the control group was significantly higher than the TSA groups. (*) p<0.05; ANOVA, Tukey post hoc test. There was no significant difference in TSA groups (0.1, 0.3, and 0.5 μM). (D) The number of BrdU-positive cells (arrows) was decreased following TSA treatment. BrdU-incorporated MUCs were also labeled by DAPI, a universal nuclei marker. Scale bar, 100 μm.

After treatment with TSA for 72 h, a BrdU incorporation assay revealed that 12.62±5.85%, 9.89±6.98%, 9.87±6.39%, and 68.28±9.57% of MUCs incorporated BrdU in 0.1, 0.3, and 0.5 μM TSA and control medium groups, respectively (Fig. 1D). The number of surviving cells in the 1 μM TSA group was the lowest in all treatments, so that few cells were detected with BrdU antibody, which was not appropriate for statistical analysis (not shown). Statistical analysis revealed that the percentage of BrdU-incorporated MUCs in the control group was significantly higher than those in 0.1, 0.3, and 0.5 μM TSA groups (p<0.05; ANOVA), whereas there was no significant difference in the 0.1–0.5 μM TSA treatment groups (Fig. 1C). This observation indicated that a decreased number of MUCs entered the S phase in response to TSA treatment. On the basis of cell viability and proliferation results, 0.5 μM TSA seemed to be an effective concentration that caused less toxicity to MUCs.

TSA treatment increased the expression of epithelial markers

When MUCs were treated with 0.1 or 0.3 μM TSA, RT-PCR showed that there was no significant change in the expression of epithelial genes, including Cdh1, Krt8, Krt18, Dsp, and Tjp1 (Fig. 2A). On the contrary, the expression of Cdh1, Krt18, and Dsp was significantly increased when MUCs were exposed to 0.5 μM TSA. These data indicated that TSA treatment may have a dose-dependent effect on the expression of selected epithelial genes of MUCs. In the meantime, we found that the expression of Tjp1 and Krt8 was not changed following 0.1–0.5 μM TSA treatment (Fig. 2A). In quantitative PCR, the corresponding fold changes of Cdh1, Krt18, Krt8, and Tjp1 from TSA to control groups in quantitative PCR were approximately 4.11±0.47, 2.82±0.30, 1.19±0.82, and 0.63±0.17, respectively (Fig. 2B). Statistical analysis revealed that there was a significant difference in the expression of Cdh1 and Krt18 between the TSA and control groups (p<0.05; ANOVA), whereas not with Krt8 and Tjp1. The mean Cq value of Dsp was not detectable in the control MUCs, but it was 31.12±0.45 in TSA-treated MUCs. However, the changes in gene expression did not seem to translate into protein expression changes in the immunofluorescence study. In immunofluorescence, E-cadherin (an epithelial protein encoded by Cdh1) and ZO-1 (a tight-junction protein encoded by Tjp1) were expressed at similar levels in TSA-treated and control groups (Fig. 2C).

FIG. 2.

MUCs expressed epithelial genes following TSA treatment. (A) RT-PCR study. The relative expression of epithelial genes in MUCs was related to TSA doses. The gene expression of Cdh1, Krt18, and Dsp did not show significant changes after treatment with 0.1 or 0.3 μM TSA. However, expression of Cdh1, Krt18, and Dsp was increased following 0.5 μM TSA treatment. Tjp1 and Krt8 were expressed stably in response to different TSA doses. (B) Quantitative PCR analysis. The normalized relative mRNA expression values of Cdh1, Krt8, Krt18, and Tjp1 of 0.5 μM TSA-treated MUCs exhibited 4.11±0.47-, 1.19±0.82-, 2.82±0.30-, and 0.63±0.17-fold changes compared to control medium–treated MUCs. Significant difference was detected in the expression of Cdh1 and Krt18 between TSA- and control medium–treated MUCs. (*) p<0.05; ANOVA. (C) Immunofluorescence study. Similar expression levels of ZO-1 (tight junction protein encoded by Tjp1; arrows, enlarged in the insert) and E-cad (E-cadherin, encoded by Cdh1; arrowheads, enlarged in the insert) were observed in MUCs treated with the control medium and 0.5 μM TSA. DAPI was used as a universal nuclei marker. Scale bar, 100 μm.

TSA treatment decreased the expression of mesenchymal markers

RT-PCR suggested that TSA-treated MUCs exhibited decreased expression of mesenchymal genes, including Zeb1, Zeb2, Snai1, and Snai2 compared to control MUCs (Fig. 3A). The other mesenchymal genes, including Cdh2, Fn1, and Vim, presented similar levels in 0.5 μM TSA-treated and control MUCs. In quantitative PCR analyses of normalized relative mRNA expression between TSA-treated and control MUCs, Zeb1, Zeb2, Snai1, and Snai2 were significantly decreased following TSA treatment, and were quantified as 0.13±0.03, 0.09±0.03, 0.04±0.00, and 0.35±0.05 fold decreases, respectively (Fig. 3B) (p<0.01 in Zeb1 and Snai2; p<0.001 in Zeb2 and Snai1; Student's t-test). In immunofluorescence, MUCs in the control and TSA groups were stained with mesenchymal markers such as ZEB1, fibronectin, and N-cadherin (Fig. 3C). The relative fluorescence intensity of ZEB1 was decreased in MUCs treated with 0.5 μM TSA (Fig. 3C), whereas no obvious change was observed in other tested markers.

FIG. 3.

TSA reduced the expression of mesenchymal markers in MUCs. (A) RT-PCR study. Compared to the control group, MUCs treated with 0.5 μM TSA showed decreased expression of Zeb1, Zeb2, Snai1, and Snai2. Other mesenchymal genes, including Cdh2, Fn1, and Vim, expressed similar levels in TSA-treated and control MUCs. (B) Quantitative PCR analysis. The normalized relative mRNA expression levels of Zeb1, Zeb2, Snai1, and Snai2 were 0.13±0.03-, 0.09±0.03-, 0.04±0.00-, and 0.35±0.05-fold lower than those of the control MUCs, respectively, which was statistically significant between TSA-treated and control MUCs. (*) p<0.01; (**) p<0.001; Student's t-test. (C) Immunofluorescence study. The relative fluorescence intensity of ZEB1 of TSA-treated MUCs was slightly decreased compared to control MUCs (double arrowheads, enlarged in the insert), whereas no obvious change was observed in fibronectin (FN; arrowheads, enlarged in the insert) or N-cadherin (N-cad; arrows, enlarged in the insert). Scale bar, 100 μm.

TSA treatment affects the expression of hair cell genes, hair cell progenitor genes, and HDACs

In our RT-PCR study, TSA treatment led to downregulation of prosensory genes, including Lfng, Six1, and Dlx5 (Fig. 4A). In quantitative PCR, the mean Cq value of Lfng was 29.25±0.35 in the control MUCs, but the expression of Lfng was decreased and could not be detected in MUCs treated with 0.5 μM TSA. The fold changes of normalized relative mRNA expression of Six1 and Dlx5 between TSA and control MUCs were approximately 0.19±0.01 and 0.04±0.00, respectively (p<0.001; Student's t-test; Fig. 4A). The expression of other progenitor genes, such as Hes1, Isl1, Jag1, Cdkn1b, Bmp4, and Sox2, did not demonstrate significant changes in RT-PCR between the TSA and control groups (Fig. 4A); therefore, quantitative PCR was not performed. Immunofluorescence study showed that the prosensory protein P27^kip1 was expressed consistently in TSA-treated and control MUCs (Fig. 4C).

FIG. 4.

TSA treatment affected the expression of hair cell and its progenitor markers. (A) The expression of the prosensory genes Lfng, Six1, and Dlx5 was decreased in TSA-treated MUCs in RT-PCR. No significant change was detected in the expression of other progenitor genes, including Hes1, Isl1, Jag1, Cdkn1b, Bmp4, and Sox2 between TSA and control groups. The normalized relative mRNA expression of Six1and Dlx5 in the TSA group was 0.19±0.01- and 0.04±0.00-fold lower than that of the control group, respectively. (*) p<0.001; Student's t-test. (B) Upregulated expression of hair cell genes Atoh1 and Myo6 was detected after treatment with 0.5 μM TSA for 72 h, but Myo7a expression was slightly increased in RT-PCR. The normalized relative expression levels of Atoh1 and Myo6 in the TSA group were 2.31±0.36- and 2.10±0.05-fold higher than that in the control group, which was statistically significant (p<0.05; Student's t-test). (C) Immunofluorescence. P27^kip1 (arrowhead, enlarged in the insert) expressed constantly in TSA-treated and control MUCs, whereas myosin VI (MyoVI) staining was slightly increased in the TSA group. Scale bar, 100 μm.

In the study of hair cell genes, RT-PCR indicated that 0.5 μM TSA-treated MUCs increased the expression of Atoh1, Myo6, and Myo7a. In quantitative PCR, normalized relative expression levels of Atoh1 and Myo6 in the TSA group were 2.10±0.05- and 2.31±0.36-fold higher than those in the control group, which was statistically significant (p<0.05 in Myo6, p<0.01 in Atoh1; Student's t-test; Fig. 4B). No obvious change was observed in the expression of Myo7a in quantitative PCR. MUCs in the TSA and control groups were labeled by antibodies specific for myosin VI with slightly increased expression in the TSA group (Fig. 4D), whereas the myosin VIIa immunostaining was very weak in both groups (not shown).

Additionally, we studied the expression of HDACs following TSA treatment. We found that after treatment with 0.5 μM TSA for 72 h, Hdac1 expression did not change significantly in RT-PCR. In contrast, quantitative PCR study indicated a significant decrease in Hdac2 and Hdac3 expression following TSA treatment, which was 0.28±0.06- and 0.16±0.06-fold lower than that of controls, respectively (p<0.05; Student's t-test) (Fig. 5).

FIG. 5.

RT-PCR analysis showed that the expressions of Hdac2 and Hdac3 in MUCs were slightly decreased in TSA-treated MUCs, whereas Hdac1 expression did not change obviously. The normalized relative mRNA expression values of Hdac2 and Hdac3 in TSA-treated MUCs were approximately 0.28±0.06-, 0.16±0.06-fold lower than those of control MUCs. (*) p<0.05; Student's t-test.

Discussion

In this study, we treated mesenchymal-like MUCs with TSA to determine whether a HDAC inhibitor could induce the expression of epithelial markers in MUCs maintained in adherent cultures. We found that TSA treatment stimulated MUCs to upregulate the expression of epithelial markers and downregulate the expression of mesenchymal markers. Additionally, MUCs decreased the expression level of prosensory genes and increased the expression of hair cell genes, including Atoh1 and Myo6.

TSA is a potent class I and II HDAC inhibitor that has been applied to alter gene expression by interfering with the removal of acetyl groups from lysine in the histone tail. TSA is a cell-toxic chemical that is able to affect the viability and proliferation of a variety of cell types. First, cell viability following TSA treatment varies in different cell lines. For example, retinoblastoma cells could hardly endure 0.2 μM TSA (Karasawa and Okisaka, 2004), whereas mutant human lymphocytes were able to survive through 48 μM TSA (Smith et al., 2014). Second, TSA can arrest the cell cycle in a variety of cell types (Kong et al., 2014; Sikandar et al., 2011; Smith et al., 2014). For instance, in cochlear and vestibular tissue explant cultures, TSA showed the ability to decrease supporting cell proliferation (Slattery et al., 2009). In our study, MUCs presented a dose–response to TSA treatment. The number of viable cells was not significantly changed when MUCs were exposed to 0.1–0.5 μM TSA, but a higher dose of TSA (1 μM) remarkably increased the proportion of dead cells in MUCs. Although MUCs survived well in lower concentrations of TSA from 0.1 to 0.5 μM, TSA at this dosage range significantly suppressed cell proliferation compared to control MUCs, and this was indicated in the BrdU incorporation assay. Our study suggests that TSA led to an inhibitory effect on MUCs proliferation at lower concentrations (≤0.5 μM) and caused cells apoptosis at relatively higher concentrations (≥1 μM).

In addition to its effect on cell viability and proliferation, TSA has been implicated in affecting the differentiation of stem/progenitor cells (Karantzali et al., 2008; Lim et al., 2013). In general, stem/progenitor cells usually downregulate the expression of stem cell genes during differentiation into a specific cell type. However, TSA-induced cell differentiation showed variable stem cell gene expression in different stem/progenitor cell lines. Take Sox2 as an example. Sox2 is a stem cell marker involved in maintenance and self-renewal of stem cells (Takahashi et al., 2007; Wegner and Stolt, 2005). In the auditory system, Sox2 was detected in both prosensory cells (Neves et al., 2012) and supporting cells (Liu et al., 2012). Sox2 was upregulated in oligodendrocyte progenitor cells (Foti et al., 2013) and prostate cancer cells (Kong et al., 2012) after TSA treatment, whereas its expression was slightly reduced in embryonic stem cell differentiation (Karantzali et al., 2008). In our study, TSA did not significantly affect the expression of Sox2 in MUCs. The discrepancy of these observations suggests that the role of TSA in the expression of progenitor genes may depend on the cell line. Additionally, we found in this study that TSA decreased the expression of Six1 and Dlx5, indicating MUCs might have been partially directed to cell differentiation fate.

EMT and mesenchymal-to-epithelial transition (MET) play crucial roles in development, organogenesis, and tumor metastasis. Previous studies demonstrated that TSA inhibited transforming growth factor-β (TGFβ)-induced EMT in liver (Kaimori et al., 2010; Lei et al., 2010) and heart cells (Liu et al., 2008), which played a role in preventing hepatic and cardiac fibrogenesis (Niki et al., 1999). In the current study, we studied the expression of epithelial and mesenchymal genes in MUCs to evaluate TSA's effect on MET. We found that TSA was able to promote the expression of the epithelial genes Cdh1 and Krt18 and decreased the expression of the mesenchymal genes Zeb1, Zeb2, Snai1, and Snai2 at the gene level. Although it has been described in many cell lines that the inhibition of E-cadherin was dependent on the transcription factors Zeb1 and Snai1 (Kakihana et al., 2009; Peinado et al., 2004), the detailed regulation mechanism is still unclear. We speculate that the role of TSA in MET may involve two aspects. First, TSA may serves as an HDAC inhibitor to reverse HDAC-related Cdh1 silencing. It was suggested that the expression of Cdh1 was upregulated via the activation of histone acetylation in its promoter area under the effect of TSA (von Burstin et al., 2009). A study of pancreatic cancer cells found that TSA could restore the expression of Cdh1 via interfering with the SNAIL/HDAC1/HDAC2 repressor complex. In line with this observation, we found in our study that TSA reduced the expression of Zeb1, Zeb2, Snai1, Snai2, and HDAC2, but increased the expression of the epithelial gene Cdh1 in MUCs. Second, TSA may direct MET via upregulation of the prosensory gene Six1, a homeodomain transcription factor involved in organ development (Zheng et al., 2003; Zou et al., 2004) and tumorigenesis (Wu et al., 2014). A study in breast cancer cells found that Six1 was able to induce EMT mediated with TGFβ activation (Micalizzi et al., 2010). In our study, TSA suppressed the expression of Six1 concomitant with increased expression of Cdh1. It is possible that TSA-induced Six1 downregulation may play a role in inhibiting Six1-related EMT, which subsequently leads to the upregulation of epithelial genes such as Cdh1. Overall, our data suggest that TSA may cause MET of MUCs, including upregulation of the expression of epithelial genes and downregulation of the expression of mesenchymal genes.

In addition to our selected gene expression study, we investigated the protein expression of MUCs following TSA treatment using immunofluorescence, including epithelial, mesenchymal, prosensory, and hair cell proteins. We found that no significant change was observed in most of the proteins studied using immunofluorescence, indicating that TSA may contribute to the initial gene expression changes during MET, which have not translated into a protein level. A solid conclusion needs more efforts on a comprehensive understanding of both gene and protein levels to elucidate the underlying mechanisms. In addition, it is noted that TSA was applied to MUCs for 72 h in the current study. It is possible that a longer TSA treatment period may be required to translate gene expression changes into the protein level. Therefore, further work may be necessary to determine whether longer TSA treatment, such as 5–10 days, may have effects on protein expression and cell morphology of MUCs. Finally, it is interesting to understand whether TSA is needed to maintain gene and protein expression changes, and this should be an independent study to be explored in our future research.

In summary, our present study investigated a novel epigenetic approach to induce stem/progenitor cells to change their cell fate. We focused on the effect of the HDAC inhibitor TSA on the gene expression levels of epithelial, mesenchymal, and prosensory markers in prosensory-like MUCs. Our data showed that TSA induced significant changes in the expression of a group of selected genes, which indicates that TSA may have stimulated the initial stages of mesenchymal-to-epithelial transition at the gene expression level. Our results suggested the potential of MUCs to differentiate toward an epithelial cell fate using epigenetic methods, which may provide evidence to further understand the molecular pathways that mediate a complete restoration of epithelial and hair cell features. Finally, additional study will be needed to clarify how to guide MUCs to become specialized and functional inner ear hair cells in the future.

Footnotes

Acknowledgments

The authors thank Neelkumar Patel for his technical support, and Fei Nie, Yang Zhou, and Xiaoyang Li for valuable comments to the manuscript. Antibodies were obtained from Developmental Studies Hybridoma Bank. The study is supported by the National Institute on Deafness and Other Communication Disorders, National Institutes of Health (NIDCD/NIH grant no. R03DC011597) and the Carls Foundation.

Author Disclosure Statement

The authors declare that there are no conflicts of interest.

References

Chai

, Kuo

, Wang

, Liaw

E.J.

, Xia

, Jan

T.A.

, Liu

, Taketo

M.M.

, Oghalai

J.S.

, Nusse

, Zuo

, and Cheng

A.G.

(2012). Wnt signaling induces proliferation of sensory precursors in the postnatal mouse cochlea. Proc. Natl. Acad. Sci. USA, 109, 8167–8172.

Cheung

T.H.

, and Rando

T.A.

(2013). Molecular regulation of stem cell quiescence. Nat. Rev. Mol. Cell Biol., 14, 329–340.

Chikamatsu

, Ishii

, Murata

, Sakakura

, Shino

, Toyoda

, Takahashi

, and Masuyama

(2013). Alteration of cancer stem cell-like phenotype by histone deacetylase inhibitors in squamous cell carcinoma of the head and neck. Cancer Sci., 104, 1468–1475.

Foti

S.B.

, Chou

, Moll

A.D.

, and Roskams

A.J.

(2013). HDAC inhibitors dysregulate neural stem cell activity in the postnatal mouse brain. Int. J. Dev. Neurosci., 31, 434–447.

Forge

, Li

, Corwin

J.T.

, and Nevill

(1993). Ultrastructural evidence for hair cell regeneration in the mammalian inner ear. Science, 259, 1616–1619.

Gallo

, Gambelli

, Gava

, Sasdelli

, Tellone

, Masini

, Marchetti

, Dotta

, and Sorrentino

(2007). Generation and expansion of multipotent mesenchymal progenitor cells from cultured human pancreatic islets. Cell Death Differ., 14, 1860–1871.

Gershengorn

M.C.

, Hardikar

A.A.

, Wei

, Geras-Raaka

, Marcus-Samuels

, and Raaka

B.M.

(2004). Epithelial-to-mesenchymal transition generates proliferative human islet precursor cells. Science, 306, 2261–2264.

Groves

A.K.

, Zhang

K.D.

, and Fekete

D.M.

(2013). The genetics of hair cell development and regeneration. Annu. Rev. Neurosci., 36, 361–381.

Halley-Stott

R.P.

, Pasque

, and Gurdon

J.B.

(2013). Nuclear reprogramming. Development, 140, 2468–2471.

10.

, and Ulfendahl

(2013). The potential of stem cells for the restoration of auditory function in humans. Regen. Med., 8, 309–318.

11.

, Luo

, Zhang

, Lu

, Dong

, Monsell

, and Jiang

(2012). Generation of human inner ear prosensory-like cells via epithelial-to-mesenchymal transition. Regen. Med., 7, 663–673.

12.

Joglekar

M.V.

, and Hardikar

A.A.

(2010). Epithelial-to-mesenchymal transition in pancreatic islet beta cells. Cell Cycle, 9, 4077–4079.

13.

Kaimori

, Potter

J.J.

, Choti

, Ding

, Mezey

, and Koteish

A.A.

(2010). Histone deacetylase inhibition suppresses the transforming growth factor beta1-induced epithelial-to-mesenchymal transition in hepatocytes. Hepatology, 52, 1033–1045.

14.

Kakihana

, Ohira

, Chan

, Webster

R.B.

, Kato

, Drabkin

H.A.

, and Gemmill

R.M.

(2009). Induction of E-cadherin in lung cancer and interaction with growth suppression by histone deacetylase inhibition. J. Thorac. Oncol., 4, 1455–1465.

15.

Karantzali

, Schulz

, Hummel

, Hubner

, Hatzopoulos

, and Kretsovali

(2008). Histone deacetylase inhibition accelerates the early events of stem cell differentiation: Transcriptomic and epigenetic analysis. Genome Biol., 9, R65.

16.

Karasawa

, and Okisaka

(2004). Inhibition of histone deacetylation by butyrate induces morphological changes in Y79 retinoblastoma cells. Jpn. J. Ophthalmol., 48, 542–551.

17.

Kelley

M.W.

(2006). Regulation of cell fate in the sensory epithelia of the inner ear. Nat. Rev. Neurosci., 7, 837–849.

18.

Kelley

M.W.

, Talreja

D.R.

, and Corwin

J.T.

(1995). Replacement of hair cells after laser microbeam irradiation in cultured organs of corti from embryonic and neonatal mice. J. Neurosci., 15, 3013–3026.

19.

Kong

, Ahmad

, Bao

, Li

, Banerjee

, and Sarkar

F.H.

(2012). Histone deacetylase inhibitors induce epithelial-to-mesenchymal transition in prostate cancer cells. PLoS One, 7, e45045.

20.

Kong

, Ji

, Xie

, Li

, Mao

, Wang

, Liu

, and Liu

(2014). Telomere elongation facilitated by trichostatin a in cloned embryos and pigs by somatic cell nuclear transfer. Stem Cell Rev. Rep., 10, 399–407.

21.

Lei

, Zhang

, Pan

, Hu

, Wang

, Yuan

, Shu

, and Song

(2010). Histone deacetylase 1 is required for transforming growth factor-beta1-induced epithelial-mesenchymal transition. Int. J. Biochem. Cell Biol., 42, 1489–1497.

22.

, Liu

, and Heller

(2003). Pluripotent stem cells from the adult mouse inner ear. Nat. Med., 9, 1293–1299.

23.

Lim

S.Y.

, Sivakumaran

, Crombie

D.E.

, Dusting

G.J.

, Pebay

, and Dilley

R.J.

(2013). Trichostatin A enhances differentiation of human induced pluripotent stem cells to cardiogenic cells for cardiac tissue engineering. Stem Cells Transl. Med., 2, 715–725.

24.

Liu

, Levin

M.D.

, Petrenko

N.B.

, Lu

M.M.

, Wang

, Yuan

L.J.

, Stout

A.L.

, Epstein

J.A.

, and Patel

V.V.

(2008). Histone-deacetylase inhibition reverses atrial arrhythmia inducibility and fibrosis in cardiac hypertrophy independent of angiotensin. J. Mol. Cell. Cardiol., 45, 715–723.

25.

Liu

, Walters

B.J.

, Owen

, Brimble

M.A.

, Steigelman

K.A.

, Zhang

, Mellado Lagarde

M.M.

, Valentine

M.B.

, Yu

, Cox

B.C.

, and Zuo

(2012). Regulation of p27Kip1 by Sox2 maintains quiescence of inner pillar cells in the murine auditory sensory epithelium. J. Neurosci., 32, 10530–10540.

26.

Micalizzi

D.S.

, Wang

C.A.

, Farabaugh

S.M.

, Schiemann

W.P.

, and Ford

H.L.

(2010). Homeoprotein Six1 increases TGF-beta type I receptor and converts TGF-beta signaling from suppressive to supportive for tumor growth. Cancer Res., 70, 10371–10380.

27.

Mizutari

, Fujioka

, Hosoya

, Bramhall

, Okano

H.J.

, Okano

, and Edge

A.S.

(2013). Notch inhibition induces cochlear hair cell regeneration and recovery of hearing after acoustic trauma. Neuron, 77, 58–69.

28.

Neves

, Vachkov

, and Giraldez

(2012). Sox2 regulation of hair cell development: incoherence makes sense. Hear. Res., 297, 20–29.

29.

Niki

, Rombouts

, De Bleser

, De Smet

, Rogiers

, Schuppan

, Yoshida

, Gabbiani

, and Geerts

(1999). A histone deacetylase inhibitor, trichostatin A, suppresses myofibroblastic differentiation of rat hepatic stellate cells in primary culture. Hepatology, 29, 858–867.

30.

Okano

, and Kelley

M.W.

(2012). Stem cell therapy for the inner ear: Recent advances and future directions. Trends Amplif., 16, 4–18.

31.

Oshima

, Grimm

C.M.

, Corrales

C.E.

, Senn

, Martinez

, Geleoc

G.S.

, Edge

, Holt

J.R.

, and Heller

(2007). Differential distribution of stem cells in the auditory and vestibular organs of the inner ear. J. Assoc. Res. Otolaryngol., 8, 18–31.

32.

Papp

, and Plath

(2013). Epigenetics of reprogramming to induced pluripotency. Cell, 152, 1324–1343.

33.

Peinado

, Ballestar

, Esteller

, and Cano

(2004). Snail mediates E-cadherin repression by the recruitment of the Sin3A/histone deacetylase 1 (HDAC1)/HDAC2 complex. Mol. Cell. Biol., 24, 306–319.

34.

Roberson

D.W.

, and Rubel

E.W.

(1994). Cell division in the gerbil cochlea after acoustic trauma. Am. J. Otol., 15, 28–34.

35.

Ronaghi

, Nasr

, and Heller

(2012). Concise review: Inner ear stem cell—an oxymoron, but why?. Stem Cells, 30, 69–74.

36.

Shi

, and Edge

A.S.

(2013). Prospects for replacement of auditory neurons by stem cells. Hear. Res., 297, 106–112.

37.

Shi

, Kempfle

J.S.

, and Edge

A.S.

(2012). Wnt-responsive Lgr5-expressing stem cells are hair cell progenitors in the cochlea. J. Neurosci., 32, 9639–9648.

38.

Shi

, Hu

, and Edge

A.S.

(2013). Generation of hair cells in neonatal mice by beta-catenin overexpression in Lgr5-positive cochlear progenitors. Proc. Natl. Acad. Sci. USA, 110, 13851–13856.

39.

Sikandar

, Dizon

, Shen

, Li

, Besterman

, and Lipkin

S.M.

(2011). The class I HDAC inhibitor MGCD0103 induces cell cycle arrest and apoptosis in colon cancer initiating cells by upregulating Dickkopf-1 and non-canonical Wnt signaling. Oncotarget, 1, 596–605.

40.

Slattery

E.L.

, Speck

J.D.

, and Warchol

M.E.

(2009). Epigenetic influences on sensory regeneration: Histone deacetylases regulate supporting cell proliferation in the avian utricle. J. Assoc. Res. Otolaryngol., 10, 341–353.

41.

Smith

, Fox

, Mejia

, Ruangpradit

, Saberi

, Kim

, Choi

, Oh

, Wang

, Choi

, Li

, Hendrickson

E.A.

, Takeda

, Muller

, and Myung

(2014). Histone deacetylase inhibitors selectively target homology dependent DNA repair defective cells and elevate non-homologous endjoining activity. PLoS One, 9, e87203.

42.

Takahashi

, and Yamanaka

(2006). Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell, 126, 663–676.

43.

Takahashi

, Tanabe

, Ohnuki

, Narita

, Ichisaka

, Tomoda

, and Yamanaka

(2007). Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 131, 861–872.

44.

Uchida

, Maruyama

, Nagashima

, Ono

, Masuda

, Arase

, Sugiura

, Onouchi

, Kajitani

, Asada

, and Yoshimura

(2006). Human endometrial cytodifferentiation by histone deacetylase inhibitors. Hum. Cell., 19, 38–42.

45.

von Burstin

, Eser

, Paul

M.C.

, Seidler

, Brandl

, Messer

, von Werder

, Schmidt

, Mages

, Pagel

, Schnieke

, Schmid

R.M.

, Schneider

, and Saur

(2009). E-cadherin regulates metastasis of pancreatic cancer in vivo and is suppressed by a SNAIL/HDAC1/HDAC2 repressor complex. Gastroenterology, 137, 361–371.

46.

Warchol

M.E.

, Lambert

P.R.

, Goldstein

B.J.

, Forge

, and Corwin

J.T.

(1993). Regenerative proliferation in inner ear sensory epithelia from adult guinea pigs and humans. Science, 259, 1619–1622.

47.

Wegner

, and Stolt

C.C.

(2005). From stem cells to neurons and glia: A Soxist's view of neural development. Trends Neurosci., 28, 583–588.

48.

White

P.M.

, Doetzlhofer

, Lee

Y.S.

, Groves

A.K.

, and Segil

(2006). Mammalian cochlear supporting cells can divide and trans-differentiate into hair cells. Nature, 441, 984–987.

49.

, Ren

, Li

, Yu

, Chen

, Huang

, and Liu

(2014). Six1: A critical transcription factor in tumorigenesis. Int J Cancer, 2014. Article first published online: 11 FEB 2014 DOI:10.1002/ijc.28755.

50.

Yoshikawa

, Hishikawa

, Marumo

, and Fujita

(2007). Inhibition of histone deacetylase activity suppresses epithelial-to-mesenchymal transition induced by TGF-beta1 in human renal epithelial cells. J. Am. Soc. Nephrol., 18, 58–65.

51.

, Vodyanik

M.A.

, Smuga-Otto

, Antosiewicz-Bourget

, Frane

J.L

, Tian

, Nle

, Jonsdottir

G.A.

, Ruotti

, Stewart

, Slukvin

, and Thomson

J.A.

(2007). Induced pluripotent stem cell lines derived from human somatic cells. Science, 318, 1917–1920.

52.

Zhang

, and Hu

(2012). Sensory epithelial cells acquire features of prosensory cells via epithelial to mesenchymal transition. Stem Cells Dev., 21, 1812–1821.

53.

Zheng

, Huang

, Wei

Z.B.

, Silvius

, Tang

, and Xu

P.X.

(2003). The role of Six1 in mammalian auditory system development. Development, 130, 3989–4000.

54.

Zou

, Silvius

, Fritzsch

, and Xu

P.X.

(2004). Eya1 and Six1 are essential for early steps of sensory neurogenesis in mammalian cranial placodes. Development, 131, 5561–5572.