University of Southern California Dissertations and Theses

The role of HES/HEY transcriptional repressors in specification and maintenance of cell fate in the mouse organ of Corti

(USC Thesis Other)

The role of HES/HEY transcriptional repressors in specification and maintenance of cell fate in the mouse organ of Corti

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content i

The role of HES/HEY transcriptional
repressors in specification and
maintenance of cell fate in the mouse organ
of Corti

by
Yassan Abdolazimi

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GENETIC, MOLECULAR AND CELLULAR BIOLOGY)

August 2015

Copyright 2015 Yassan Abdolazimi

ii

Dedication
To my parents and Kaveh.
For their patience and unceasing encouragement.

iii

Acknowledgements
First of all, I offer my sincerest gratitude to Dr. Neil Segil, my advisor, for his
support, patience, and enthusiasm during my PhD, and for giving me the room to follow
my ideas. I also want to thank my committee members, Dr. Robert Maxson, Dr. Gage
Crump and Dr. Kris Kobielak for their invaluable advice and guidance.
I also would like to thank all the current and former members of the Segil lab for
their constructive criticism, technical assistance and friendly advice in the past six years.
I am extremely thankful to Dr. Zlatka Stojanova for sharing expertise and
encouragement extended to me. I would like to thank my former fellow graduate student
and current post-doc, Dr. Litao Tao, for his help with the bioinformatic analysis of
sequencing data, and also Dr. Robert Rainey for his help with the editing. I express my
warm thanks to Llamas and Welly Makmura for animal care and assistance with FACS.
A very special thanks goes out to Dr. Qi-Long Ying for the mESC line, Dr. Jane Johnson
for the Atoh1 enhancer-reporter plasmid, Dr. Greta Segil for the Atoh1-mCherry
reporter, Dr. Takahiro Ohyama for numerous plasmids and Dr. Verdon Taylor for the
Hes5 plasmid. I must also acknowledge Flow Cytometry Core Facility, Lora Barsky and
Bernadette Masinsin, Microscopy Core manager Dr. Seth Ruffins at USC Eli and Edyth
Broad Center and transgenic mouse facility at University of California, Irvine.
I am grateful to have had the opportunity to study and research in the stimulating
environment of House Research Institute and University of Southern California.

iv

Table of Contents
Dedication ........................................................................................................................ii
Acknowledgements ......................................................................................................... iii
Table of Contents ............................................................................................................iv
List of Figures ................................................................................................................ viii
List of Tables ...................................................................................................................xi
Introduction ..................................................................................................................... 1
Chapter 1: Background ................................................................................................... 5
Structure of the inner ear and hearing ................................................................... 5
Development of the inner ear ................................................................................ 8
Organ of Corti development .................................................................................. 9
Differentiation of hair cells and supporting cells .................................................... 9
Regulation of Atoh1 ............................................................................................. 11
Notch signaling pathway .................................................................................. 14
HDACs as interaction partners of HES/HEY factors .................................... 19
HDACs Alter Gene Transcription ................................................................. 21
HDAC inhibitors ............................................................................................ 21
FGF pathway ................................................................................................... 23
Wnt pathway .................................................................................................... 26
Six1/Eya1 and Sox2......................................................................................... 28
Molecular biology of Atoh1 regulation by HES/HEY transcriptional repressors 29

v

Chapter 2: The mechanism of HES5-mediated repression of Atoh1 in postnatal
supporting cells ............................................................................................................. 30
Introduction ......................................................................................................... 30
Results ................................................................................................................ 33
Derepression of Atoh1 is sufficient to induce Atoh1 transcription .................... 33
HES5 targets the Atoh1 promoter region for repression .................................. 38
HES5 recruits the co-repressor GRG/TLE to the Atoh1 promoter in a Notch-
dependent fashion ........................................................................................... 43
Notch inhibition induces histone acetylation at the Atoh1 promoter in supporting
cells .................................................................................................................. 49
Atoh1 is upregulated in prosensory progenitors, not just in nascent hair cells . 54
Summary ............................................................................................................. 59
Chapter 3: The mechanism of HEY2-mediated repression of Atoh1 in postnatal
supporting cells ............................................................................................................. 63
Introduction ......................................................................................................... 63
Results ................................................................................................................ 65
HEY2 inhibits Atoh1 through the promoter region ............................................ 65
HEY2 inhibitory effect is not dependent on its C-terminal motif ....................... 70
HEY2-mediated repression of Atoh1 is dependent on HDAC activity .............. 72
Salermide and TSA stimulate Atoh1 expression in neonatal pillar cells ........... 77
Salermide and TSA are not capable of blocking HES5-mediated repression of
Atoh1 ............................................................................................................... 80
The failure of Salermide or TSA alone to induce transdifferentiation of pillar
cells is not changed by blocking Notch or FGF signaling ................................. 82
HDAC expression analysis in the organ of Corti .............................................. 84
HEY2 and SIRT2 interact in vitro ..................................................................... 87
vi

RNA-Seq. analysis of Salermide and TSA-treated p75
+
supporting cell .......... 88
Discussion ........................................................................................................... 89
Chapter 4: Atoh1 autoregulation ................................................................................... 93
Introduction ......................................................................................................... 93
Results ................................................................................................................ 94
ATOH1 binds the E-box site in the enhancer ................................................... 94
The Atoh1 enhancer acts as an auto-regulatory element in response to ectopic
expression of Atoh1 ......................................................................................... 96
Mutation of the E-box reduces the induction of enhancer activity in the GER .. 97
Atoh1 enhancer is not active in the organ of Corti in the Atoh1 knock out ..... 100
Discussion ......................................................................................................... 101
Chapter 5: Conclusions ............................................................................................... 104
The same promoter element in Atoh1 is inhibited by either HEY2 or HES5 to
maintain the fate of neighboring supporting cells ..................................................... 104
Derepression of the Atoh1 locus is sufficient to stimulate Atoh1 induction ........ 105
HEY2 and HES5 rely on different cofactors to inhibit Atoh1 .............................. 107
Atoh1 silencing in supporting cells is reinforced by “activator insufficiency” ...... 108
Atoh1 is upregulated in prosensory progenitors immediately prior to repression in
nascent supporting cells .......................................................................................... 109
Ongoing challenges to regeneration of hair cells .............................................. 112
Chapter 6: Materials and Methods .............................................................................. 114
Experimental animals and generation of transgenic mouse line ....................... 114
DNA extraction and PCR genotyping: ............................................................... 114
Cell culture ........................................................................................................ 115
Cochlear explant culture .................................................................................... 116
Immunohistochemistry and imaging .................................................................. 116
vii

Small molecule inhibitors ................................................................................... 117
FACS-purification .............................................................................................. 117
Cochlear cells ................................................................................................ 117
Neural progenitor cells (NPCs) ...................................................................... 118
OC1 cells ....................................................................................................... 118
RNA extraction and real time quantitative PCR (qPCR) .................................... 119
Cloning plasmids and reporter constructs ......................................................... 119
Hes5 expression plasmids ............................................................................. 119
Atoh1 reporter plasmids ................................................................................. 120
Site-directed mutagenesis ................................................................................. 121
Chromatin immunoprecipitation (ChIP)-qPCR ................................................... 121
Reporter assay .................................................................................................. 123
Flow cytometry .................................................................................................. 123
Western blotting ................................................................................................ 124
Proximity ligation assay ..................................................................................... 124
Immunoprecipitation assay ................................................................................ 124
ATAC-seq and peak-calling ............................................................................... 125
Electrophoretic mobility shift assay (EMSA) ...................................................... 126
Electroporation .................................................................................................. 126
Cell counts ........................................................................................................ 126
RNA-seq ............................................................................................................ 127
References .................................................................................................................. 132
Appendix ..................................................................................................................... 145

viii

List of Figures
Figure 1. Schematic illustration of the mammalian inner ear ........................................... 7
Figure 2. Cellular organization of the organ of Corti ........................................................ 7
Figure 3. Development of the mouse inner ear ............................................................... 8
Figure 4. Schematic representation of Atoh1 locus and the sequence of Atoh1 enhancer
based on mouse chromosome 6, mm10 ................................................................ 13
Figure 5. Schematic representation of Notch signaling pathway. Modified from (Kiernan
2013). ..................................................................................................................... 16
Figure 6. Schematic representation of Hes and Hey expression pattern in the cross
section of cochlea .................................................................................................. 18
Figure 7. Classification of HDACs ................................................................................. 20
Figure 8. Inhibitory profile of TSA and SAHA for inhibition of HDAC1-9 ........................ 22
Figure 9. FGF ligands and their receptor specificity ...................................................... 25
Figure 10. Supporting cells transdifferentiate into hair cells in the absence of Notch
signaling ................................................................................................................. 35
Figure 11. HES5 protein stability in vitro ....................................................................... 35
Figure 12. Derepression of Atoh1 in supporting cells is sufficient to drive expression,
and occurs through promoter elements, not through the 3’ autoregulatory enhancer
............................................................................................................................... 37
Figure 13. HES5 represses Atoh1 through the promoter region ................................... 40
Figure 14. Mutational analysis of C-site function. HES5 repressed Atoh1 through the
highly conserved binding sites in the promoter region ........................................... 42
Figure 15. HES5 does not have inhibitory effect on wild type or mutated Atoh1 enhancer
reporter .................................................................................................................. 42
Figure 16. The HES5 WRPW motif is required for the repression of the Atoh1 promoter
and for interaction with GRG/TLE .......................................................................... 45
Figure 17. Sox1-GFP knock-in mouse embryonic stem cell (mESC) line 46C expresses
GFP and Nestin upon neural differentiation ........................................................... 47
ix

Figure 18. GRG/TLE localizes to the Atoh1 locus in a Notch dependent manner in
neural progenitors .................................................................................................. 48
Figure 19. Notch inhibition leads to histone acetylation at Atoh1 promoter in supporting
cells ........................................................................................................................ 51
Figure 20. ATAC-seq in postnatal hair cells and supporting cells ................................. 54
Figure 21. Misexpression of mutant promoter transgene in vivo shows HES/HEY
binding sites in the Atoh1 promoter are required for the proper silencing of Atoh1 in
supporting cells ...................................................................................................... 56
Figure 22. HES/HEY binding sites are required to rerepress Atoh1 in supporting cells
after Notch signaling is first inhibited, allowing Atoh1 levels to rise, and then
restored .................................................................................................................. 57
Figure 23. Expression level of class I, II and IV HDACs in FACS-purified p27-positive
supporting cells (SCs) compared to p27-negative cells (non-SC) in P1 cochlea
measured by qPCR ................................................................................................ 58
Figure 24. Basic helix-loop-helix sequence of Hes5 and Hey2 protein ......................... 64
Figure 25. Hey2 is capable of repressing Atoh1 transcription through both the promoter
and enhancer regions ............................................................................................ 68
Figure 26. HEY2’s inhibitory effect on the Atoh1 enhancer is not due to N-box or C-site-
dependent DNA binding ......................................................................................... 69
Figure 27. HEY2 binds to Atoh1 promoter in vitro ......................................................... 70
Figure 28. HEY2 basic domain, but not the YQPW, motif is required for the repression
of the Atoh1 promoter ............................................................................................ 71
Figure 29. HEY2-mediated repression of Atoh1 is dependent on HDAC activity .......... 75
Figure 30. Salermide and TSA do not induce cell proliferation in the sensory epithelium
of organ of Corti ..................................................................................................... 75
Figure 31. HDAC inhibition by Salermide and TSA leads to transdifferentiation of inner
pillar cells ............................................................................................................... 77
Figure 32. Purification of pillar cells with FACS ............................................................. 78
Figure 33. Real-time quantitative PCR showing the fold difference in expression level of
Atoh1, Hey2 and p75 between DMSO- , Salermide/TSA-, TSA-, and Salermide-
treated p75
+
supporting cells .................................................................................. 79
x

Figure 34. Salermide and TSA cannot induce Atoh1 expression in HES5-expressing
cells in vitro or supporting cells ex vivo .................................................................. 81
Figure 35. Salermide or TSA alone cannot induce transdifferentiation of pillar cells in the
absence of Notch signaling .................................................................................... 84
Figure 36. Expression analysis of Sirts and other HDACs in the sensory epithelium .... 86
Figure 37. SIRT2 interacts with HEY2 in vitro ............................................................... 87
Figure 38. HEY2 can prevent binding of ATOH1 to the enhancer E-box....................... 91
Figure 39. ATOH1/TCF3 heterodimers bind the E-box in the Atoh1 enhancer ............. 95
Figure 40. Atoh1 autoregulates ..................................................................................... 96
Figure 41. The reporter constructs used in electroporation of postnatal cochlear
explants .................................................................................................................. 97
Figure 42. Ectopic expression of Atoh1 induces the enhancer activity in GER if the E-
box is intact ............................................................................................................ 99
Figure 43. The Atoh1 enhancer is not active in the Atoh1
-/-
mouse cochlea at E14.5.. 101
Figure 44. Schematic representation of the Atoh1 promoter sequence and the predicted
transcription factor binding sites ........................................................................... 107
Figure 45. A model of Atoh1 regulation during organ of Corti development ................ 111

xi

List of Tables
Table 1. Fold difference in expression level of hair cell and supporting cell markers
between DMSO- and Salermide/TSA-treated p75
+
supporting cells assessed by
RNA-seq ................................................................................................................ 88
Table 2. Primers used in real time quantitative PCR. .................................................. 127
Table 3. Primers used in cloning. ................................................................................ 128
Table 4. Primers used in site-directed mutagenesis. ................................................... 129
Table 5. Primers used in ChIP-qPCR. Numbers refer to the 5’ end of the amplicons. 130
Table 6. Oligonucleotides used in electrophoretic mobility shift assay. ...................... 131

1

Introduction
The organ of Corti, the auditory sensory epithelium in mammals, is a mosaic
composed of the auditory hair cells and the surrounding supporting cells, both of which
originate from the same group of postmitotic prosensory progenitors. This mosaic is
achieved by upregulation of Atoh1 in nascent hair cells and suppression of Atoh1 in
neighboring supporting cells. Atoh1, a mammalian basic helix-loop helix activator and
homolog of Drosophila proneural gene atonal (Helms and Johnson 1998), is the earliest
known marker for hair cell differentiation (Bermingham et al. 1999; Chen et al. 2002)
and the only factor that is both necessary and sufficient for the differentiation of hair
cells. Atoh1 knockout mice do not form any hair cells (Bermingham et al. 1999).
Moreover, it has been demonstrated that the ectopic expression of Atoh1 in non-
sensory regions within the cochlea, such as the greater epithelial ridge (GER) (Zheng
and Gao 2000) or lateral epithelial ridge (LER) (Yang et al. 2013) is sufficient to induce
the formation of ectopic hair cells in postnatal rat inner ears. Repression of Atoh1 in
supporting cells is achieved through the Notch signaling-mediated lateral inhibition.
Differentiating hair cells express Notch ligands such as Delta1 and Jagged2 (Morrison
et al. 1999) that bind to Notch receptors on neighboring cells; as a result, the Notch
intracellular domain (NICD) is cleaved and activates the transcription of Hes/Hey family
of bHLH transcriptional repressors (reviewed in (Pierfelice et al. 2011; Kiernan 2013)),
which in return suppress Atoh1 expression (Zheng et al. 2000; Zine et al. 2001; Tateya
et al. 2011). Due to this Notch-mediated lateral inhibition, the neighbor of a developing
hair cell is inhibited from differentiating as a hair cell and thus acquires a supporting cell

2

fate. Consistent with this example of lateral inhibition, deletion of Hes/Hey genes results
in the formation of supernumerary hair cells in the mouse cochlea (Zine et al. 2001;
Hayashi et al. 2008a; Li et al. 2008). Similarly, disrupting Notch signaling by
pharmacological inhibitors of ɣ-secretase activity, such as DAPT during embryonic
development and postnatally leads to upregulation of Atoh1 and direct
transdifferentiation of supporting cells to a hair cell-like state (Yamamoto et al. 2006;
Doetzlhofer et al. 2009; Mizutari et al. 2013).
HES and HEY transcriptional repressors are related to the Hairy and Enhancer-of-
split type of proteins in Drosophila and are the primary targets of the Notch signaling
pathway ((Fischer and Gessler 2007)). HES/HEY factors repress transcription both
passively and actively. In passive repression, they form nonfunctional heterodimers with
bHLH activators and in this way sequester the activators (Liu et al. 2006; Kageyama et
al. 2007). In active repression, HES/HEY factors bind to consensus DNA site called N-
box (CACNAG) and the class C site (CACGNG) (Akazawa et al. 1992; Liu et al. 2006;
Grogan et al. 2008; Zheng et al. 2011; Heisig et al. 2012). Once bound to target DNA,
HES/HEY factors recruit further corepressors including GRG/TLE cofactors and
different histone deacetylases (HDACs) to inhibit their target genes (Fisher et al. 1996;
Grbavec et al. 1998; Chen et al. 1999; Yu et al. 2001; Winkler et al. 2010).
The Atoh1 promoter and enhancer both contain predicted HES/HEY binding sites
that suggest Atoh1 could be a direct target of repression by these factors. Previous
studies revealed a 1.7 Kb autoregulatory enhancer of the Atoh1 gene which lies 3.4 Kb
3’ of the coding region. In transgenic studies, this enhancer is sufficient to recapitulate
the expression of Atoh1 in the cochlea and other Atoh1 expression domains (Helms et

3

al. 2000; Lumpkin et al. 2003). The Atoh1 promoter region immediately upstream of the
transcription start site is not as well characterized as the enhancer. It has been shown
that the 15 kb sequence upstream of the Atoh1 transcription start site alone was not
sufficient to support the expression of a LacZ transgene in the neural tube (Helms et al.
2000).
Despite many studies indicating that Notch signaling through HES and HEY
repressors is required for maintaining a supporting-cell identity by inhibiting Atoh1
expression, the molecular mechanisms on which this inhibition is based have not been
determined. In this study, we investigated the role of the HES5 and HEY2 in the
suppression of Atoh1, and thus the maintenance of supporting cell fate. We present
evidence that HES5/HEY2 directly repress Atoh1 expression through the conserved
binding sites in the Atoh1 promoter region. We also demonstrate that the transcriptional
corepressor GRG/TLE, and HDAC activity contribute to the repression of the Atoh1
promoter and that inhibition of Notch signaling induces epigenetic changes, in particular
an increase in H3K9ac histone mark that facilitates Atoh1 expression. In response to
the loss of HES/HEY-mediated repression in supporting cells and without any need for
de novo protein synthesis, Atoh1 is derepressed. We propose that derepression of
Atoh1 is sufficient to stimulate transdifferentiation to a hair cell-like phenotype.
The first chapter in this dissertation provides some background information
regarding the development and differentiation of hair cells and supporting cells and also
describes the different signaling pathways/factors that have been shown to influence the
expression of Atoh1 in the cochlea. Chapters 2 and 3 will dissect the molecular
mechanism of Atoh1 repression by HES/HEY transcriptional repressors. Chapter 4

4

examines the role of the Atoh1 enhancer in Atoh1 positive autoregulation. The
significance of this project is discussed in chapter 5. Materials and methods are
described in chapter 6.

5

Chapter 1: Background
This chapter reviews some of the important factors and the molecular pathways that
regulate the development of the mouse organ of Corti and are related to the work done
in this dissertation.
Structure of the inner ear and hearing
The inner ear, a fluid-filled structure is divided into two parts; the vestibular and
auditory systems. The vestibular system consists of three semicircular canals, utricle,
and saccule which are involved in the sense of balance and orientation. The auditory
system responsible for the perception of sound is the cochlea, a snail-shaped duct or
tube that coils around a central canal called modiolus through which cochlear nerve
fibers and blood vessels pass. A cross section shows that the cochlea is divided into
three compartments: scala vestibuli, scala media and scala tympani. The middle
compartment is scala media, also known as the cochlear duct that is filled with
endolymph, an extracellular fluid with high K
+
and low Na
+
concentrations. The
uppermost compartment (the scala vestibuli) and the lowermost compartment (scala
tympani) are filled with perilymph, a typical extracellular fluid. The cochlear duct is
separated from the scala tympani and scala vestibuli by the basilar membrane and
Reissner’s membrane, respectively (Figure 1A). The cochlear duct contains organ of
Corti, the auditory sensory epithelium of the inner ear which sits on the top of the basilar
membrane (Durrant and Lovrinic 1995; Kierszenbaum 2007) (Figure 1B).
The organ of Corti, which extends the length of the cochlear duct, contains the
sensory hair cells and nonsensory supporting cells, both of which originate from the

6

same group of prosensory progenitor (Kelley 2006). Hair cells are the mechanosensory
cells that transduce the auditory stimuli and generate the electrochemical response that
otic neurons will transmit to the brain. Loss of hair cells leads to hearing loss (Brigande
and Heller 2009). In mammals the hair cells are arranged in three rows of outer hair cell
(OHCs) and one row of inner hair cells (IHCs). Inner and outer hair cells are separated
from each other by the tunnel of Corti. The apical surface of the hair cells contains actin-
rich filaments known as steriocilia that are in contact with a structure called tectorial
membrane (Figure 2). Sound waves induce vibration of the middle ear that is
transmitted to endolymph in the cochlear duct resulting in the displacement of the basal
membrane and consequently the movement of hair cells. As hair cells move, the
steriocilia in contact with the tectorial membrane are bent which triggers the opening of
mechanoelectrical transduction (MET) channels. An influx of K
+
into the hair cells
through the MET channels leads to depolarization and release of neurotransmitter at the
basal end of the hair cell onto the auditory afferent nerve fibers (Purves D 2001; Dror
and Avraham 2010). IHCs are the primary transducers of auditory stimuli to neural
signals; 95% of afferent auditory nerve fibers synapse on inner hair cells. On the other
hand, there are efferent axons that mostly synapse on the outer hair cells. OHCs mainly
function as cochlear amplifiers to fine-tune the “sensitivity” and “frequency selectivity” of
the cochlea (Purves D 2001; Brigande and Heller 2009; Dror and Avraham 2010).
Supporting cells that surround the hair cells to form an alternating mosaic are not
sensory but are required for hair cell function and provide physiological support for hair
cells (Driver and Kelley 2009) (Figure 2). These highly differentiated and specialized
cells are given different names based on their position and morphology. Deiters’ cells,

7

for example, surround the outer hair cells and pillar cells are located between the inner
and outer hair cells and create the tunnel of Corti (Kelley 2006).

Figure 1. Schematic illustration of the mammalian inner ear. (A) Schematic of the
ear, showing the outer, middle and inner ear. The inner ear is composed of cochlea and
vestibule. (B) Schematic of the across section of the cochlear duct showing the different
compartment, tectorial membrane (TM) and the organ of Corti in the middle. Modified
from (Dror and Avraham 2010).

Figure 2. Cellular organization of the organ of Corti. Inner hair cell (IHC), outer hair
cells (OHC) and different type of supporting cells are shown. Hair cells are innervated
by cochlear nerve. Modified from (Wan et al. 2013).

8

Development of the inner ear
The development of the inner ear begins at embryonic day 8.5 (E8.5) with the formation
of otic palcode: the bilateral thickening of the surface ectoderm near the hindbrain. By
E9.5, the developing placode invaginates to form the otocyst or otic vesicle (Haddon
and Lewis, 1996; Sanchez-Calderon et al., 2007). At this stage, a population of
neuroblasts from the ventral region of the otocyst delaminate and migrate to form the
developing VIII cranial ganglion (Driver and Kelley 2009), after which the otocyst
gradually undergoes extensive morphological changes and patterning including
extension and coiling, resulting in the formation of the entire inner ear composed of the
dorsal vestibular and ventral auditory regions by P0 (Kelley 2006). The developmental
changes of mouse inner ear are shown in Figure 3.

Figure 3. Development of the mouse inner ear. Modified from (Kelley 2006).

9

Organ of Corti development
Beginning around E12, specific regions in the cochlear duct become specified to
develop as prosensory patches (domains) that will eventually give rise to all of the cells
(hair cells and supporting cells) within the organ of Corti (Kelley 2007). At this time, the
prosensory cells are still dividing. At about E12.5, prosensory cells in the apical region
start to express p27
Kip1
, a cyclin-dependent kinase inhibitor (CKI), and exit cell cycle.
The expression of p27
Kip1
and the cell cycle exit that follows, happen in an apical to
basal pattern, so that by E14.5 the gradient of p27
kip1
expression reaches the basal
region. The wave of p27
Kip1
expression forms a zone of non-proliferating cells (ZNPC)
along the length of the cochlear duct whose borders represent the prosensory domain
or the future organ of Corti. Deletion of p27
Kip1
gene leads to prolonged cell division in
prosensory domain, formation of supernumerary hair cells and supporting cells and also
severe hearing loss (Chen and Segil 1999; Chen et al. 2002; Lee et al. 2006).
Differentiation of hair cells and supporting cells
Once the postmitotic prosensory domain is established, the cellular differentiation
begins. Hair cells are the first distinguishable differentiated cells, with supporting cells
developing with a delay (Kelley 2006). Hair cell differentiation starts near the base of the
cochlea and extends toward the apex. In addition to the basal to apical gradient, there is
a medial to lateral gradient; inner hair cells develop first followed by outer hair cells. The
earliest known gene expressed in the prosensory domain associated with the
differentiation of hair cells is the transcription factor Atoh1, which is first detected near
the cochlear base and extends toward the apex in the next three to four days. As a

10

result of the uncoupling of cell cycle exit and differentiation, the last cells to exit the cell
cycle in the mid-basal region of the cochlea are some of the first to differentiate into hair
cells (Chen et al. 2002)
Although generally accepted that Atoh1 expression starts in the post-mitotic prosensory
cells between E13.0 and E14.5 in the mid-basal region of the cochlea, there is some
discrepancy on the spatiotemporal pattern of Atoh1 expression depending on the
method of detection (RNA or protein). In situ hybridization and Atoh1/LacZ knock-in line
show Atoh1 mRNA expression as early as E13.0 at the mid-basal in a broad domain of
prosensory cells, and the expression was also observed in cells other than those that
will become hair cells (Bermingham et al. 1999; Lanford et al. 2000; Woods et al. 2004).
On the other hand, Atoh1 immunohistochemistry and Atoh1enhancer/GFP transgenic
line suggest that the expression of Atoh1 protein starts at E14.5 in the mid-basal region
only in nascent hair cells (Chen et al. 2002; Lumpkin et al. 2003). In any case, Atoh1
expression becomes restricted to cells that will develop as hair cells and Atoh1 is
sufficient and necessary for their development. Hair cells do not form in the absence of
Atoh1 (Bermingham et al. 1999). Misexpression of Atoh1 in non-sensory regions within
the organ of Corti, such as the greater epithelial ridge (GER), is sufficient to induce the
formation of hair cells (Zheng and Gao 2000).
The mechanisms that control the gradient of cell cycle exit and differentiation are not
fully understood. It remains unknown what factors or signals regulate the expression of
p27
kip1
or Atoh1 in the cochlea. The basal to apical gradient of differentiation does not
require an intact organ of Corti since the sequential differentiation happens in cochlear
epithelium that have been cut into pieces and kept in organ culture (Montcouquiol and

11

Kelley 2003). Although the differentiation of hair cells and supporting cells is completed
before the mouse is born, maturation of the organ of Corti is not complete until the onset
of hearing between P10 and P14 (Pujol and Hilding 1973; Lim and Anniko 1985)

The differentiation of supporting cell during development is largely dependent on
Notch and FGF signaling (see below). In the mammalian organ of Corti there are at
least five different types of supporting cells, characterized based on their morphology,
location and the expression of molecular markers. From the lateral to medial edge of the
organ these cell are Hensen’s cell, Deiters’ cells, pillar cells, inner phalangeal cells and
border cells (Figure 2). The many functions of supporting cells include: (1) maintenance
of hemeostasis of hair cells by recycling K
+
ions through transporters and gap junctions,
(2) ejection of hair cells and phagocytosis of debris after hair cell injury, (3) providing
structural support for the sensory epithelium through their microtuble-rich cytoskeleton,
and (4) production of extracellular matrix (reviewed in (Monzack and Cunningham 2013;
Wan et al. 2013).
Regulation of Atoh1
Although Atoh1 plays a critical role in the differentiation of hair cells, little is known about
the factors or signaling pathways that regulate this gene or the DNA elements in the
Atoh1 gene that are involved in its regulation in the mouse organ of Corti.
The mouse Atoh1 gene containes a 1053 bp exon that codes for a 351 amino acid-
long protein (37 kDa). Atoh1 is a mammalian bHLH with 70% amino acid sequence
identity to the bHLH domain of the Drosophila proneural gene atonal. Outside of the
bHLH domain, there is little similarity (reviewed in (Mulvaney and Dabdoub 2012)).

12

The bHLH domain confers DNA binding and protein binding specificity. The degree
of conservation between Atoh1 and atonal allows full rescue of Atoh1 null mutant mice
by atonal (Wang et al. 2002). Transient transgenic mouse assay revealed a 1.7 Kb
enhancer of Atoh1 gene 3.4 Kb 3’ of coding region (Helms et al. 2000) (Figure 4). This
enhancer sequence, composed of two elements A (561 bp) and B (544 bp) separated
by 300 bp, was used to generate an Atoh1enhancer/GFP transgenic mouse line, in
which the expression of GFP transgene matches the expression of endogenous Atoh1
in the cochlea as demonstrated by immunohistochemistry with ATOH1 antibody (Chen
et al. 2002; Lumpkin et al. 2003). This enhancer sequence is sufficient to recapitulate
the expression of Atoh1 in other expression domains including the developing neural
tube and cerebellum (Helms et al. 2000). Within the Atoh1 enhancer, a class A, E-box
site (CAGCTG) was identified. The class A, E-boxes are the binding site for bHLH
transcriptional activators like ATOH1 (Akazawa et al. 1995). Mutation of the E-box leads
to the dramatic decrease in Atoh1 expression in the neural tube, suggesting the E-box
is required for the activity of Atoh1 enhancer (Helms et al. 2000). Similar requirement for
the activity of the enhancer in the chick neural tube is also reported (Ebert et al. 2003).
Further experiments revealed that ATOH1 binds to the enhancer E-box and ATOH1
itself is required for the expression of Atoh1 in the neural tube, as the expression of
Atoh1enhancer/lacZ transgene is not induced in the Atoh1 knock-out mouse neural tube
(Helms et al. 2000).

13

Figure 4. Schematic
representation of
Atoh1 locus and
the sequence of
Atoh1 enhancer
based on mouse
chromosome 6,
mm10. Transcription
factor binding sites
that have been
experimentally
verified are shown in
color. The predicted
but not verified sites
are shown in black.
Modified from
(Jarman and Groves
2013).

Moreover, in the Atoh1 null mutants carrying the Atoh1enhacner/GFP transgene, GFP
is not expressed in the hair cells (Raft et al. 2007). These lines of evidence led to the
conclusion that Atoh1 autoregulates and that this autoregulation is necessary to
maintain Atoh1 expression. However, it is not known whether this enhancer sequence

14

or, other yet to be identified regulatory sequences, are required for the initial expression
of Atoh1.
Several signaling pathways that regulate Atoh1 have been identified, such as
Notch, fibroblast growth factor (FGF), Wnt, bone morphogenetic protein (BMP), and
sonic hedgehog (SHH) signaling pathways, although it is not clear whether these
pathways directly or indirectly regulate Atoh1 expression. Several factors have also
been identified that can affect Atoh1 expression. For the scope of this thesis, the roles
of Notch, FGF and Wnt signaling pathways and a few factors in the development of the
organ of Corti are discussed below.
Notch signaling pathway
Notch, an ancient signaling pathway conserved across all metazoans, was first
discovered in Drosophila where partial loss of function in the Notch locus cause
Notches (and hence the name) in the wing margin (Mohr 1919). Notch gene was
subsequently, identified and sequenced, which suggested a transmembrane protein
involved in intracellular interaction as its molecular function (Wharton et al. 1985; Kidd
et al. 1986). The intracellular domain of Notch with multiple domains is involved in
signaling. The extracellular domain of Notch includes 36 epidermal growth factor (EGE)-
like repeats involved in binding to ligand. The Notch pathway includes Notch receptors
and Notch ligands, and because Notch ligands and receptors are both transmembrane
proteins, Notch mediates signaling between adjacent or neighboring cells.
In mammals there are four Notch receptors (Notch1-4) and five ligands (Jagged
(Jag) 1-2, Delta-like (Dll) 1, 3 and 4. The binding of ligand to the Notch receptor triggers
two proteolytic cleavage events in the Notch receptor. The first cleavage (S2), which is

15

mediated by ADAM family of metalloproteases, releases the Notch extracellular domain
(NECD). NECD bound to the ligand is endocytosed and degraded by the ligand-
expressing cell. However, the second cleavage (S3) mediated by ɣ-secretase complex
(containing Presenilin), releases the Notch intracellular domain (NICD), the active form
of Notch. NICD, a coactivator, then can translocate to the nucleus and interact with the
DNA-binding factor RBP-J (recombination signal sequence-binding protein Jĸ, also
called as CSL or CBF1) and activate transcription of Notch target genes (Figure 5). In
the absence of NICD, RBP-J represses target genes by recruiting corepressors. The
binding of NICD and other coactivators such as MAML, switches the repressor to
activator (Borggrefe and Oswald 2009; Pierfelice et al. 2011).
The best characterized Notch targets are Hes and Hey of ehich there are six Hes
genes (Hes1, 2,3,5,6 and 7) and three Hey genes (Hey1, 2, and L) in the mouse
genome. Except for Hes2, Hes3 and Hes6, all the Hes and Hey genes are downstream
effectors of canonical Notch signaling pathway (Jarriault et al. 1995; Nishimura et al.
1998; Ohtsuka et al. 1999). Hes and Hey genes encode bHLH (basic helix-loop-helix)
transcriptional repressors. In addition to the bHLH domain that is involved in DNA
binding and dimerization of these factors, HES and HEY proteins contain an orange
domain and a C-terminal domain. The Orange domain includes two α-helices and
serves as an additional interface for the selection of bHLH heterodimer partners
(Dawson et al. 1995; Taelman et al. 2004). The C-terminal domain in HES proteins
consists of the tetrapeptide Trp-Arg-Pro-Trp or WRPW that interacts with corepressors
such as GRG/TLE (Grbavec and Stifani 1996 ; Grbavec et al. 1998). Instead of WRPW
tetrapeptide, the HEY proteins contain an YRPW or YQPW peptide, and are not able to

16

interact with GRG/TLE cofactors, instead they interact with N-CoR and mSin3A
corepressors which recruits HDAC1 (Iso et al. 2001) along with Sirt1, a member of
another histone deacetylase family (Takata and Ishikawa 2003).

Figure 5. Schematic
representation of Notch
signaling pathway.
Modified from (Kiernan
2013).

HES and HEY factors repress transcription both actively and passively. In active
repression, HES factors bind to consensus DNA sites called N-box (CACNAG) and the
class C, E-box site (CACGNG). They also bind to class B, E-box sites to some degree,
but not class A (Iso et al. 2003). HEY factors preferred binding sites are class B and C,
E-boxes (Iso et al. 2001) although their binding to N box has also been reported
(Grogan et al. 2008). HES and HEY factors can form heterodimers. Such heterodimers
are able to bind to their target sequences with a higher affinity and repress transcription
more efficiently than homodimers (Iso et al. 2001). In passive repression, HES and
HEY factors form nonfunctional heterodimers with bHLH activators that bind to E-box
sites and in this way sequester the activators (Liu et al. 2006; Kageyama et al. 2007).
Notch signaling is involved in the development of many tissues and organs. In the
inner ear, it is necessary for the formation of the sensory organs and differentiation of
hair cells and supporting cells. In the organ of Corti, Atoh1-expressing hair cells are

17

surrounded and separated by supporting cells; hair cells are not in direct contact with
one another. This mosaic of alternating hair cell-supporting cell is achieved by Notch
signaling though a process known as lateral inhibition. A simple description of lateral
inhibition is that through a negative feedback loop, a differentiating cell prevents its
neighboring cells from acquiring the same cell fate.
In the cochlea, the expression of Notch ligands Dll1 and Jag2 starts in the single
row of inner hair cells at around E14.5 and gradually expands to the outer hair cells
(Lanford et al. 1999; Morrison et al. 1999). These Notch ligands bind to Notch1
receptors, initially expressed throughout the prosensory domain (Lanford et al. 1999),
on neighboring cells. As a result, NICD is cleaved that activates the transcription of Hes
and Hey repressors (reviewed in (Pierfelice et al. 2011; Kiernan 2013)), which in turn
suppress Atoh1 expression (Zheng et al. 2000; Zine et al. 2001; Tateya et al. 2011).
Due to this Notch-mediated lateral inhibition, the neighbor of a developing hair cell is
inhibited from differentiating as a hair cell and acquires a supporting cell fate. several
lines of evidence support the inhibition of Atoh1 in supporting cells by Notch signaling
pathway through HES/HEY repressors. First, HES/HEY factors are expressed in the
supporting cells, and in a combinatorial pattern (Zine et al. 2001; Hayashi et al. 2008a;
Tateya et al. 2011) (Figure 6). Second, deletion of any of HES/HEY genes results in the
formation of supernumerary hair cells in the mouse cochlea (Zine et al. 2001; Hayashi
et al. 2008a; Doetzlhofer et al. 2009). Third, inhibition of Notch signaling by DAPT, a
gamma-secretase inhibitor leads to the appearance of extra hair cells by
transdifferentiation of supporting cells (Yamamoto et al. 2006; Doetzlhofer et al. 2009).
Fourth, it has been demonstrated that co-transfection of GER with Atoh1 and Hes1

18

inhibits the formation of extra hair cells (Zheng et al. 2000). How the Notch ligands are
initially expressed only in the developing hair cells is not clear. The Atoh1 promoter and
enhancer both contain candidate HES/HEY binding sites that suggest Atoh1 could be
directly inhibited by these repressors. This dissertation focuses on understanding the
regulation of Atoh1 by Notch-HES/HEY signaling.

Figure 6. Schematic representation of Hes and
Hey expression pattern in the cross section of
cochlea. Inner hair cell (ihc), outer hair cell (ohc),
pillar cells (p), Deiters’ cells (d), Hensen cells (h),
and inner phalangeal cells (i). Modified from
(Doetzlhofer et al. 2009).

In addition to its role in lateral inhibition, Notch signaling also plays an earlier role in
the specification of the prosensory domain known as lateral induction. Evidence for
inductive role of Notch stems from studies on Jag1 knockout. Notch induction is
mediated by JAG1 ligand that is initially expressed throughout the prosensory domain
starting at around E12.5 and becomes restricted to supporting cells by E17.5 (Morrison
et al. 1999). In the Jag1 conditional KO, the development of the prosensory domain in
the cochlea is incomplete; absence of hair cells and supporting cells in the base as well
as reduced rows of hair cells in the rest of cochlea is observed. Moreover, in this KO
prosensory markers Sox2 and p27
kip1
show reduced expression domain that is not due
to cell death or lack of proliferation (Brooker et al. 2006; Kiernan et al. 2006; Pan et al.
2010). These observations led to the conclusion that Jag1-Notch signaling is required
for the specification of the progenitors.

19

What are the effectors of Notch lateral induction? HEY factors are likely candidates.
HEY1 and HEY2 are both expressed in the prosensory domain by E12.5, the time of
prosensory specification (Hayashi et al. 2008a). Another candidate is the transcription
factor SOX2. In mice with loss of function alleles of Sox2, the prosensory domain fails to
establish due to a lack of p27
kip1
expression, and the differentiation of hair cells and
supporting cells is impaired (Kiernan et al. 2005b), resulting in phenotypes similar to
Jag1 KO. Additionally, ectopic activation of Notch in the nonsensory regions of the inner
ear induces SOX2 expression and leads to generation of hair and supporting cells
(Hartman et al. 2010; Pan et al. 2010; Liu et al. 2012b).
HDACs as interaction partners of HES/HEY factors
HES/HEY factors, either directly or indirectly, recruit histone deacetylases (HDACs),
which are the enzymes that catalyze the removal of acetyl groups from lysine residues
in histones is called a histone deacetylase or HDAC. An enzyme with the reverse
activity is called histone acetyltransferase (HAT). The first histone deacetylase, HDAC1,
was isolated and cloned in 1996 (Taunton et al. 1996). HDACs are divided in two
families: the histone deacetylase family (also referred to as classical HDAC family) and
Sir2 regulator (or Sirtuin family). HDCAs are further grouped into separate categories or
classes based on their sequence similarities. Classes I, II and IV HDACs belong to
histone deacetylase family, and class III HDACs belong to Sirtuin family. The numbering
of Class I, II, and IV HDACs are based on their chronological order of discovery. The
different families and classes of HDACs are shown in Figure 7.

20

Figure 7. Classification of HDACs. Modified from (Seto and Yoshida 2014).

The main difference between the two families of HDACs involves their catalytic
mechanism. The classical HDACs require a zinc molecule as a cofactor in their active
site and therefore are inhibited by chelating compounds such as vorinostat and
trichostatin A that bind Zn
2+
. However, these inhibitors are not effective against Sirtuins
since they require NAD+ as a cofactor (Witt et al. 2009).
Multiple studies suggest that each Class I, II, IV HDAC might have distinct histone
substrate specificity. Finding the specific substrate for each HDAC has been difficult
given the functional redundancy of HDACs and their low histone deacetylase activity
when purified (Johnson et al. 2002; Vermeulen et al. 2004; Zhang et al. 2004; Seto and
Yoshida 2014).
Sirts have more obvious histone substrate specificity. For example Sirt1, 2 and 6
can deacetylate H4K16ac and H3K9ac (Vaquero et al. 2004; Vaquero et al. 2006;
Michishita et al. 2008) and Sirt2 and Sirt7 have been shown to be involved in the
deacetylation of H3K18ac (Barber et al. 2012; Eskandarian et al. 2013).

21

HDACs Alter Gene Transcription
Transcriptional activation is generally correlated with histone acetylation, while
repression works through histone deacetylation. Histone acetylation/deacetylation can
alter gene expression by two major mechanisms (reviewed in (Verdone et al. 2005;
Choi and Howe 2009)):
1) Changing the chromatin conformation by changing the structure of
nucleosomes: addition of acetyl group to lysine resides in histone N-terminal
tails neutralizes the histone positive charge and therefore weakens the
interaction with the negatively charged DNA backbone. On the other hand,
histone deacetylation by HDACs increases the positive charges on histones and
can strengthen histone–DNA interaction and repress transcription.
2) Providing a protein binding surface: acetylation of lysine creates a docking site
for the bromodomain that exists in many chromatin modifying factors.
In addition to histones, HDACs can deacetylate many acetylated nonhistone
proteins (Choudhary et al. 2009; Peng et al. 2012). Studies in bacteria lacking histones
also suggest this conserved function of HDACs (Wang et al. 2010).
Acetylation/deacetylation of proteins can affect 1) protein stability, 2) protein-protein
interaction, 3) DNA binding affinity (Dai and Faller 2008; Seto and Yoshida 2014). As a
result, HDACs can indirectly regulate the activity of transcription factors and signaling
pathways.
HDAC inhibitors
Given the important role of HDACs in regulation of many cellular processes, the
dysregulation of these enzymes can lead to numerous diseases and disorders. The best

22

studied role of HDACs is in cancer development. Both upregulation and downregulation
of HDACs have been reported in cancer cells. Any change in HDACs expression/activity
can result in the higher expression/activity of oncogenes or lower expression/activity) of
tumor suppressor genes (Huang et al. 2005; Noh et al. 2011; Shahbazi et al. 2014).
Therefore, development of HDAC inhibitors has been pursued as a therapeutic
approach. One of the first identified HDAC inhibitors is trichostatin A (TSA), which was
initially identified as an antifungal antibiotic isolated from a Streptomyces strain
(Yoshida et al. 1987). The HDAC inhibitory effect of TSA was subsequently discovered
when it was shown that it can increase the acetylation of histones and inhibit HDAC
activity in mammalian cell lines (Yoshida et al. 1990). Another potent HDAC inhibitor
developed in 1998 is Vorinostat or suberoylanilide hydroxamic acid (SAHA) (Richon et
al. 1998), which was the first HDAC inhibitor approved for chemotherapy (Bolden et al.
2006). TSA and SAHA are pan-HDAC inhibitors of class I, II, and IV HDACs, although
not all HDACs show similar sensitivity to these inhibitors (Bradner et al. 2010) (Figure
8).

Figure 8. Inhibitory profile of TSA and SAHA for inhibition of HDAC1-9. Modified
from (Bradner et al. 2010).

23

Inhibitors of Sirtuins have also been developed since misexpression of these
enzymes has been reported in cancer cells (Wang et al. 2008a; Audrito et al. 2011). Of
the inhibitors characterized so far, most inhibit Sirt1 and/or Sirt2. The first identified
Sirtuin inhibitors were sirtinol and splitomicin (Bedalov et al. 2001), followed by
salermide, tenovin, EX-527, and AGK2 (reviewed in (Villalba and Alcain 2012; Seto and
Yoshida 2014)).
Synthesis of Salermide was first reported in 2009 (Lara et al. 2009). Lara and
colleagues showed that Salermide can inhibit both Sirt1 and Sirt2 in vitro and induce
apoptosis specifically in cancer cell lines. The proapoptotic effect of Salermide was
shown to be independent of Sirt2 and p53 but dependent on Sirt1. Somewhat
contradictory results were reported by Peck et al., who showed that inhibition of both
Sirt1 and Sirt2 is required to induce apoptosis in breast cancer cell lines (Peck et al.
2010). They also demonstrated that p53 was critical for Salermide-induced apoptosis.
FGF pathway
Fibroblast growth factors (FGF) are a large family of secreted polypeptides found
through evolution from nematode to human; 18 genes have been identified in mammals.
FGFs can be further grouped into six subfamilies based on sequence similarities and
phylogeny (Figure 9). FGFs function by binding to specific tyrosine kinases receptors,
the FGF receptors (FGFR1-4) that induces receptor dimerization and
transautophosphorylation of the kinase domain. The activated receptors then
phosphorylate adaptor proteins such as FRS2 (FGFR substrate 2), GRB2 (growth factor
receptor-bound 2) and the tyrosine phosphatase SHP2 that eventually transduce the
signals by activating different signaling pathways including the Ras/MAP kinase, PLCɣ

24

and the PI3K-AKT (reviewed in (Thisse and Thisse 2005; Ornitz and Itoh 2015). The
FGFs and FGFR1-3 genes have alternative splicing; for example, FGF8a, FGF8b,
FGF8e, and FGF8f are different splicing variants of FGF8. The different isoforms
possess different ligand-receptor binding affinity (Mueller et al. 2002; Goetz and
Mohammadi 2013). In addition to FGFRs, FGFs have high affinity for heparin. Heparin
(or its in vivo counterpart heparan sulfate) can either facilitate or increase the stability of
FGF-FGFR binding (Ornitz 2000). Sprouty (Spry) is an intracellular negative regulator of
signaling via FGFRs that interacts with GRB2 to inhibit the RAS-MAPK pathway and to
regulate the PI3K-AKT pathway (reviewed in (Kim and Bar-Sagi 2004). The mouse spry
family is composed of four members, Spry1-Spry4, of which spry1 and 2 are expressed
in the mouse organ of Corti (Shim et al. 2005; Mansour et al. 2013)
FGFs are involved in the regulation of many developmental processes including
patterning, migration, differentiation and cell proliferation. In the cochlea, FGF signaling
plays important roles during different developmental phases of the organ of Corti. FGF3
and FGF10 as ligands for FGFR2 are critical for the induction of otic vesicle. In mice
lacking these genes, otic vesicles do not form (Pirvola et al. 2000; Alvarez et al. 2003;
Wright and Mansour 2003). FGF signaling is also required for the specification of the
prosensory domain. In the FGFR1 conditional knock out mice fewer hair and supporting
cells develop (Pirvola et al. 2002). Inhibition of FGF receptors with SU5402 in cochlear
explants triggers a similar effect and expression analysis suggests that FGF20 is the
likely ligand for FGFR1 (Hayashi et al. 2008l). Interestingly, the expression of FGF20 is
reportedly regulated by Notch signaling as inhibition of Notch in the cochlear explants at
E12.5 by DAPT, a ɣ-secretase inhibitor decreases the expression level of FGF20.

25

Moreover, it was shown that FGF20 can rescue the inhibition of Notch when added to
the cochlear explants at E13.5, suggesting that the role of Notch in the specification of
the prosensory domain is at least partially mediated by FGF20 (Munnamalai et al.
2012).

Figure 9. FGF ligands and their
receptor specificity. Modified
from (Ornitz and Itoh 2015).

FGF signaling also plays a role in the development of supporting cells, particularly
pillar cells. In mice with targeted disruption of FGFR3, pillar cells fail to differentiate and
the tunnel of Corti does not form, which leads to hearing loss (Colvin et al. 1996). In situ
analysis of cochlea showed that expression of FGFR3 starts at E15.5-E16 in a band of
cells (along the length of the cochlear duct) that will develop as pillar cells, outer hair
cells and Deiters’ cells (Mueller et al. 2002; Hartman et al. 2007). By P0, expression of
FGFR3 is restricted to pillar cells and at lower levels in Deiters’ cells (Mueller et al.
2002; Jacques et al. 2007). Inhibition of FGF signaling by SU5402 in cochlear cultures
suggests that the continuous activation of FGFR3 is required for the differentiation of
pillar cells. On the other hand, increased activation of FGFR3 by treating the cochlear
explants with FGF2 or FGF17 leads to an increase in the number of developing pillar

26

cells, as evident by the expression of NGFR or p75 (Mueller et al. 2002; Jacques et al.
2007). The ligand for FGFR3 is FGF8 expressed by inner hair cells. Targeted deletion
of FGF8 leads to defects in development of pillar cells (Jacques et al. 2007). Just as
loss of FGF signaling leads to an increase in the number of developing pillar cells,
deletion of Sprouty2, a negative regulator of FGF signaling, causes an increase in the
number of developing outer hair cells and supporting cells (four rows of outer hair cells
and Dieters’ cells instead of three). Moreover, in Spry2 null mutants one extra row of
pillar cells form by transformation of Deiters’ cell fate between P5 and P10. Interestingly,
the cellular transformation is rescued when one copy of FGF8 is deleted from the
Sprouty KO background (Shim et al. 2005). These observations together suggested a
model of FGF signaling gradient; progenitor cells that are closer to the source of FGF8
(inner hair cells) and express FGFR3 at higher levels develop as pillar cells, while those
that have weaker FG8-FGFR3 signaling (farther from the source of FGF8 and lower
FGFR3 expression) differentiate as Deiters’ cells (Shim et al. 2005; Mansour et al.
2013).
A role for FGF signaling in the differentiation of pillar cells in postnatal organ of Corti
has also been reported. In this case, FGF cooperates with Notch signaling to maintain
the expression of Hey2 in pillar cells. As a result, inhibition of both signaling pathways is
required to downregulate Hey2 expression and induce the transdifferentiation of pillar
cells to hair-cell-like cells (Doetzlhofer et al. 2009).
Wnt pathway
Wnts (Wingless-related integration site) are a group of glycoproteins that once
secreted, act on their neighbor cells (paracrine) or the Wnt –secreting cells (autocrine)

27

by binding to the receptor Fzd (Frizzled) and the co-receptor LRP5/6 (low-density
lipoprotein receptor related protein5 or 6). The second messenger of canonical Wnt
signaling is transcriptional co-activator β-catenin. In the absence of active wnt signaling,
β-catenin is targeted for ubiquitination by the Axin complex comprising Axin,
adenomatosis polyposis coli (APC), and glycogen synthase kinase 3 (GSK3), and
subsequent degradation by the proteasome pathway. As a result of this continuous
degradation, β-catenin target genes are kept silenced by TCF/LEF (T cell
factor/lymphoid enhancer factor) factors. When Wnt binds to the Fzd and LRP5/6, the
Axin complex is recruited to the plasma membrane where it binds to LRP5/6. This
prevents degradation of β-catenin by the Axin complex. Accumulation of β-catenin and
its translocation to the nucleus leads to the activation of Wnt target genes with the help
of TCF/LEF (reviewed in (Moon 2005; MacDonald et al. 2009).
Wnt signaling regulates many different developmental processes including cell fate
specification, differentiation, and proliferation (Logan and Nusse 2004). During the
cochlear development, Wnt is involved in (1) proliferation of the prosensory progenitors,
and (2) differentiation of hair cells. Jacques, et al., showed that activation of Wnt
signaling by LiCl increases proliferation in the prosensory cells in the mitotic E12 or
post-mitotic E13.5 cochlear explants, as evident by a significant increase in the number
of Sox2
+
Brdu
+
cells. This group also demonstrated that activation or inhibition of Wnt in
E13.5 cochlear explants increases or decreases the number of developing hair cells,
respectively, which correlated with a decrease or an increase in the expression level of
Atoh1. These results suggested that Wnt signaling is required for the differentiation of
hair cells in the organ of Corti (Jacques et al. 2012). A recent study confirmed this idea

28

by showing that deletion of β-catenin decreases the number of developing hair cells (Shi
et al. 2014).
How Wnt might be regulating the expression of Atoh1 is not clear. It was recently
shown that β-catenin can bind to the Atoh1 enhancer through TCF/LEF and increase
Atoh1 expression in Neuro2a cells, suggesting a mechanism for direct regulation of
Atoh1 by Wnt signaling (Shi et al. 2010).
Six1/Eya1 and Sox2
In addition to the HES/HEY binding sites, Atoh1 enhancer contains potential binding
sites for several other transcription factors, including SIX1 and SOX2 (Figure 4).
Besides the role of SIX1 and its coactivator EYA1 in the development of otic vesicle
(Xu et al. 1999; Zheng et al. 2003), the SIX1/EYA1 complex is also involved in the
differentiation of hair cells, in cooperation with SOX2. Recently, it was reported that
Eya1 and Sox2 are expressed in the prosensory domain before E13.5, while Six1
expression disappears between E12.5 and E13.5 and reappears by E13.5, before
Atoh1 upregulation, in a gradient similar to Atoh1 expression (Ahmed et al. 2012).
Further experiments showed that ectopic co- expression of SIX1 and EYA in GER
(greater epithelial ridge) of cochlear explants is enough to induce the formation of
supernumerary MyoVII
+
hair-cell-like cells. Interestingly, not all the MyoVII
+
cells were
ATOH1 positive, although all were positive for POU4F3, downstream target of Atoh1
(Masuda et al. 2011), which suggests an Atoh1-independent mechanism for hair cell
development. Co-expression of SOX2 with SIX1/EYA1 increases the percentage of
transfected cells that express ATOH1 and POU4F3 after 2 days in vitro (DIV), but
reduces the number of ATOH1
+
cells after 6DIV. These results suggest that SOX2 may

29

block hair cell differentiation by interfering with POU4F3 or other factors that are
required to maintain the expression of Atoh1 (Ahmed et al. 2012), indicating that
downregulation of SOX2 in the developing hair cells is required for further maturation of
hair cells as reported by other groups (Dabdoub et al. 2008). Biochemical assays
showed that SIX1 can directly bind to Atoh1 enhancer B, while SOX2 is able to bind to
enhancer A (Ahmed et al. 2012).
Molecular biology of Atoh1 regulation by HES/HEY transcriptional repressors

Given the key role of Atoh1 in hair cell differentiation during development as well as
regeneration, understanding how its expression is regulated is essential for stimulating
regeneration in mammals. Several lines of evidence suggest that Notch signaling
through the HES and HEY family of transcription factors inhibits Atoh1 in supporting
cells (discussed above). However, much less is known about the molecular mechanism
by which Notch signaling keeps Atoh1 silent in supporting cells or about the events that
lead to supporting cell transdifferentiation when Notch signaling is blocked. In this
dissertation, we investigated the role of the HES/HEY repressors in the suppression of
Atoh1 expression, and thus the maintenance of supporting cell fate.

30

Chapter 2: The mechanism of HES5-mediated repression of
Atoh1 in postnatal supporting cells
Introduction
The organ of Corti, the auditory sensory epithelium in mammals, is a mosaic
composed of auditory hair cells and surrounding supporting cells, both of which
originate from a postmitotic prosensory domain that extends along the length of the
embryonic cochlear duct (Ruben and Sidman 1967; Chen and Segil 1999; Chen et al.
2002). In mammals, the sensory hair cells in the inner ear fail to regenerate, leading to
permanent hearing loss (Chardin and Romand 1995; Brigande and Heller 2009).
However, in non-mammalian vertebrates, hair cell death is followed by the direct
transdifferentiation of the surrounding supporting cells, followed later by a wave of
supporting cell proliferation and additional transdifferentiation (Roberson et al. 2004;
Cafaro et al. 2007). Although this does not occur in mammals, we and others have
discovered that perinatal mouse supporting cells are also capable of direct
transdifferentiation to a hair cell-like state. This can occur either in vitro in purified
supporting cells following cell division (White et al. 2006), or ex vivo following a
blockade of Notch signaling in the perinatal organ of Corti (Yamamoto et al. 2006;
Doetzlhofer et al. 2009). Supporting cell transdifferentiation may provide a future target
for therapeutic manipulation.
Supporting cell transdifferentiation in response to the blockade of Notch signaling
is mediated by the transcriptional upregulation of the bHLH transcription factor Atoh1,
the mammalian homologue of the Drosophila proneural gene Atonal (Helms and

31

Johnson 1998). In mammals, Atoh1 is developmentally responsible for inducing hair cell
differentiation within the cochlear prosensory domain. This domain forms along the
ventral wall of the outgrowing cochlear duct, and in this context, Atoh1 is both
necessary and sufficient for hair cell differentiation (Bermingham et al. 1999; Zheng and
Gao 2000; Chen et al. 2002). As part of the Notch signaling pathway, Atoh1 induction of
Notch ligands in nascent hair cells is necessary for the Notch-mediated lateral inhibition
and differentiation of the neighboring cells as supporting cells (Lanford et al. 2000; Gazit
et al. 2004; Kiernan et al. 2006).
The complex molecular strategy by which Atoh1 is regulated in the prosensory
domain and its role in maintaining the perinatal sensory mosaic of hair cells and
supporting cells is not well understood. Transcriptional regulation of Atoh1 is partly
governed by a 3’ autoregulatory enhancer that is sufficient to drive accurate transgenic
reporter expression in many Atoh1-dependent tissues (Helms et al. 2000; Lumpkin et al.
2003), and which in addition to an E-box, harbors additional transcription factor motifs
capable of influencing Atoh1 expression (Ebert et al. 2003; Shi et al. 2010; Ahmed et al.
2012). ATOH1 autoregulates by binding to an E-box site, and this autoregulation is
required for the enhancer activity in the neural tube and hair cells (Helms et al. 2000;
Raft et al. 2007). However, it is not clear how the enhancer functions during the initial
phase of Atoh1 upregulation in the prosensory domain, or during the transdifferentiation
induced by the relief from Notch-mediated lateral inhibition.
Prior to the onset of Atoh1 expression, all the cells of the cochlear prosensory
domain withdraw from the cell cycle and remain postmitotic for the life of the animal
(Lee et al. 2006). Starting at E14 in the mouse, Atoh1 is first upregulated in the base of

32

the postmitotic prosensory domain, then expands its expression in a basal-to-apical
wave. By E17.5, almost the entire prosensory domain is patterned into a mosaic of hair
cells and supporting cells (Chen et al. 2002). During patterning of the prosensory
domain, nascent hair cells express Notch ligands such as Delta1 (Dll1) and Jagged2
(Jag2) (Lanford et al. 1999; Morrison et al. 1999) that bind to Notch receptors on
neighboring cells. As a result, the Notch intracellular domain (NICD) is cleaved that
activates the transcription of the Hes/Hey family of bHLH transcriptional repressors
(reviewed in (Kiernan 2013)), which in turn suppress Atoh1 expression in the nascent
supporting cells. Due to this Notch-mediated lateral inhibition, a developing hair cell’s
neighbor is inhibited from differentiating as a hair cell and acquires a supporting cell
fate. Consistent with this example of lateral inhibition, deletion of Hes/Hey genes results
in the formation of supernumerary hair cells in the mouse cochlea (Zheng et al. 2000;
Zine et al. 2001; Hayashi et al. 2008a; Li et al. 2008; Tateya et al. 2011).
HES and HEY transcriptional repressors are related to the Hairy and Enhancer-of-
split type of proteins in Drosophila and are the primary targets of the Notch signaling
pathway (reviewed in (Fischer and Gessler 2007)). HES/HEY factors repress
transcription both passively and actively. In passive repression, they sequester bHLH
activators by forming nonfunctional heterodimers (Chin et al. 2000; Liu et al. 2006). In
active or DNA binding–dependent repression, HES/HEY factors bind to consensus DNA
site called N-box (CACNAG) and the class C-site (CACGNG) (Akazawa et al. 1992; Liu
et al. 2006; Grogan et al. 2008; Zheng et al. 2011; Heisig et al. 2012). Once bound to
target DNA, HES factors recruit Groucho-related gene in mouse, transducin-like
enhancer of split in human (GRG/TLE) corepressors through their C-terminal

33

tetrapeptide WRPW (Trp-Arg-Pro-Trp) motif (Fisher et al. 1996; Grbavec et al. 1998).
GRG/TLE factors are able to inhibit basal transcriptional machinery (Yu et al. 2001),
recruit further corepressors including histone deacetylases (HDACs) (Chen et al. 1999;
Winkler et al. 2010), and form closed chromatin structures (Sekiya and Zaret 2007) to
silence target genes.
The Atoh1 promoter and enhancer both contain candidate HES/HEY binding sites
that suggest Atoh1 could be directly inhibited by these repressors. While the importance
of HES/HEY repressors in the maintenance of supporting cell fate and
transdifferentiation has been recognized, the molecular mechanisms of their role in
Atoh1 regulation have not been determined. In this chapter, we show that the Notch
effector HES5 actively and directly suppresses the Atoh1 promoter with the help of
corepressor GRG/TLE and HDACs, and that the loss of Notch-mediated repression
leads to a permissive chromatin configuration by favoring an acetylated state. We
propose a model in which Atoh1 expression and hair cell differentiation are primed
through a mechanism in which loss of HES/HEY repression is sufficient to allow Atoh1
levels to rise above a threshold needed to trigger the autoregulatory enhancer. Our data
also suggest that the initiation of Atoh1 upregulation occurs in prosensory progenitors
before being silenced by lateral inhibition in supporting cells, rather than being
upregulated strictly in selected nascent hair cells.
Results
Derepression of Atoh1 is sufficient to induce Atoh1 transcription
Blocking the Notch signaling pathway with the ɣ-secretase inhibitor DAPT in the
neonatal organ of Corti leads to transdifferentiation of supporting cells into hair cell-like

34

cells without inducing cellular proliferation (Yamamoto et al. 2006; Doetzlhofer et al.
2009). We confirmed the transdifferentiation results in cochlear explants cultured with
and without DAPT for 72 hours. The Atoh1 enhancer/β-globin promoter/GFP transgenic
line, which harbors a transgene expressing GFP under the control of a minimal β-globin
promoter, and the previously defined Atoh1 autoregulatory enhancer (Lumpkin et al.
2003) was used as a reporter for Atoh1 expression and supporting cell-to-hair cell
transdifferentiation. As previously reported (Yamamoto et al. 2006; Doetzlhofer et al.
2009), inhibition of Notch signaling by DAPT increased the number of GFP-positive hair
cell-like cells, and reduced the number of supporting cells labeled with Prox1, an early
marker of supporting cell differentiation, suggesting their direct transdifferentiation
(Figure 10A). Expression levels of Atoh1 mRNA, as well as the Notch effectors Hes5
and Hey1 were compared in FACS-purified supporting cells treated without and with
DAPT using real time quantitative PCR (qPCR). Inhibition of Notch signaling increased
Atoh1 expression greater than 10-fold in supporting cells within 24 hours of treatment.
This increase correlated with a significant decrease in the expression level of Hey1 and
particularly Hes5 (p < 0.005) (Figure 10B).
Among Hes and Hey factors that are expressed in the organ of Corti, Hes5 is the
most sensitive to loss of Notch signaling. In addition, HES5 is expressed in Deiters’
cells, a subpopulation of supporting cells that readily transdifferentiate into hair cells
when Notch signaling is lost (Figure 10A, B and (Doetzlhofer et al. 2009)). We
hypothesized that if HES5 is needed to directly silence constitutive Atoh1 transcriptional
activity, inhibition of protein synthesis would derepress Atoh1 expression as a

35

consequence of HES5 protein turnover in supporting cells. In other systems, HES5
family members, HES1 and HES7, have been shown to have short protein half-lives

Figure 10. Supporting cells transdifferentiate into hair cells in the absence of
Notch signaling. (A) Postnatal day 1 (P1) Atoh1 enhancer/β-globin promoter/GFP
transgenic organ of Corti was cultured for 72h in DMSO (control) or DAPT (ɣ-secretase
inhibitor) and immunostained with anti-Prox1 antibody to label supporting cells (Deiter’s
and pillar cells). Appearance of ectopic GFP
+
hair cell-like cells in DAPT are
accompanied by a loss of Prox1
+
supporting cells (white bracket). (B) The observed
transdifferentiation of supporting cells in (A) correlates with upregulation of Atoh1 and
downregulation of Hes/Hey factor mRNA in FACS-purified supporting cells (p27/GFP
+
)
after 24h treatment with DAPT. n = 4. Values are mean ± SEM. . [*] p < 0.05, [**] p <
0.005. Scale bar 100 µm in A.

Figure 11. HES5 protein stability in vitro. (A) 293 cells were transfected with CMV-
FLAG-Hes5 plasmid. After 24h cells were treated with cycloheximide (CHX) alone, or
cycloheximide and proteasome inhibitors, and collected at the indicated time points.
Whole-cell extracts were made according to (Gallagher, 2007) and western blotted with
anti-FLAG antibody. (B) The intensity of the signals relative to actin was measured

36

using Image J (NIH). Values shown in B are average of two independent experiments.
HES5 is degraded with a half-life of 2.5 hours. Proteasome pathway inhibitors MG132
or Lactacystin increased HES5 stability, while the protease inhibitor PMSF had no
effect.
(Hirata et al. 2002; Hirata et al. 2004). A time-course experiment in vitro in 293 cells
transfected with flag-tagged HES5 indicated that its half-life is about 2.5 hours, and that
selective inhibition of the proteasome pathway significantly increases HES5 stability
(Figure 11). To test this hypothesis in laterally-inhibited supporting cells, we examined
the effect of cycloheximide (CHX), an inhibitor of protein synthesis, on Atoh1 expression
in P1 organ of Corti cultures. Supporting cells were FACS-purified after 6 hours of
treatment with CHX, and qPCR analysis indicated more than nine-fold increase in Atoh1
expression (Figure 12A). If HES5 directly inhibits Atoh1 expression and is degraded
with the half-life of 2.5 hours by the proteasome pathway, then we hypothesized that
stabilizing HES5 protein by inhibiting its proteasome degradation would prevent Atoh1
derepression in supporting cells, even following the loss of Notch signaling. Inhibition of
Notch signaling by DAPT for 12 hours caused upregulation of Atoh1 in FACS-purified
supporting cells (Figure 12B). When DAPT and the proteasome inhibitor MG132 were
added to the cochlear cultures together, Atoh1 upregulation was lost (MG132 alone had
no significant effect on Atoh1 expression, data not shown).

37

Figure 12. Derepression of Atoh1 in supporting cells is sufficient to drive
expression, and occurs through promoter elements, not through the 3’
autoregulatory enhancer. The effect of protein synthesis, Notch and proteasome
inhibition in the organ of Corti. All values are mean ± SEM. (A) Inhibition of protein
synthesis with cycloheximide (CHX) for 6h induces Atoh1 expression in FACS-purified
supporting cells. n = 3. (B) Inhibition of protein degradation by MG132 prevents DAPT-
induced upregulation of Atoh1 in FACS-purified supporting cells treated for 12h. n = 3.
(C and D) CHX induces endogenous Atoh1, but not the Atoh1 enhancer/β-globin
promoter/GFP transgene. P1 cochlear organ cultures from Atoh1 enhancer/β-globin
promoter/GFP mice were incubated without and with CHX for 6h, after which organ
cultures were dissociated and FACS-sorted to eliminate hair cells. Atoh1 and GFP
levels were measured in GFP-negative cells (non-hair cells of the cochlear epithelia,
predominantly supporting cells). n = 4. [*] p < 0.05, [**] p < 0.005.

These observations support the hypothesis that the factors permissive for induction
of Atoh1 expression are present constitutively in the supporting cells, but that Atoh1 is
maintained in a silent state in supporting cells by the action of HES/HEY repressive
transcription factors. However, it does not show if these factors act through the
promoter, the autoregulatory enhancer, or other regulatory elements associated with the
Atoh1 locus. To elucidate this, we tested the effect of CHX on sensory epithelial cells,
which have been depleted of hair cells by FACS purification using the Atoh1
enhancer/β-globin promoter/GFP transgenic (these include supporting cells, as well as
cells of the GER, LER and surrounding tissue). In these cells, we compared the

38

expression of Atoh1 mRNA from the endogenous Atoh1 locus and the expression of the
Atoh1 enhancer/β-globin promoter/GFP mRNA. The ability of the Atoh1 enhancer to
faithfully recapitulate Atoh1 expression, and the presence of a potential HES/HEY
binding site overlapping with the previously identified Atoh1 E-box necessary for
autoregulation (Helms et al. 2000) suggested that the enhancer might harbor the
negative regulatory element(s) bound by HES/HEY factors. As expected, inhibition of
protein synthesis by CHX for 6 hours led to the upregulation of the endogenous Atoh1
(Figure 12C). However, to our surprise, the Atoh1 enhancer/β-globin promoter/GFP was
not induced (Figure 12D). Since this transgene contains a complete copy of the
autoregulatory enhancer (1.4 kb construct, (Lumpkin et al. 2003)), Our result suggested
that the enhancer was not involved in conferring Notch-mediated negative regulation on
the Atoh1 locus, and that the enhancer sequence alone is not sufficient to initiate
expression when repression is relieved.
HES5 targets the Atoh1 promoter region for repression
The Atoh1 enhancer/β-globin promoter/GFP transgene, which failed to be induced
in supporting cells upon inhibition of protein synthesis, contains a β-globin minimal
promoter (Lumpkin et al. 2003) that lacks a canonical N-box (CACNAG) or C-site
(CACGNG) seen in other HES target genes (Liu et al. 2006; Grogan et al. 2008). In
contrast, the endogenous Atoh1 promoter contains four sites within 226 bp of the
transcription start site (Figure 13A), suggesting that HES5 might directly regulate Atoh1
expression through the promoter.
To test the functional significance of the HES5 binding sites in the Atoh1 promoter
and enhancer region, we conducted transient transfection assays with two reporter

39

constructs. The first reporter includes the Atoh1 enhancer linked to 226 bp of the Atoh1
promoter (containing the predicted HES5 cognate C-sites) driving the expression of
GFP (Atoh1 enhancer/Atoh1 promoter
wt
/GFP, green) (Figure 13B). In the second
reporter, instead of the Atoh1 promoter, the β-globin basal promoter was inserted to
drive the expression of mCherry (Atoh1 enhancer/β-globin promoter/mCherry, red)
(Figure 13C). These two reporter constructs were tested by transfection into 293 cells
along with a plasmid expressing Hes5, or a control lacking Hes5 expression, and the
percentage of cells expressing GFP and mCherry was determined by flow cytometry.
The reporter with the Atoh1 promoter containing four class C-sites (GFP) was strongly
inhibited by HES5 with the percentage of GFP
+
cells decreasing by more than 60%
(Figure 13B). Hes5 expression had no effect on the control reporter containing the β-
globin minimal promoter (mCherry) (Figure 13C). These results suggest that HES5
represses Atoh1 transcription through its promoter region, and not its identified
autoregulatory enhancer, in spite of the presence of consensus HES/HEY binding sites.
To determine if the predicted HES/HEY binding sites in the Atoh1 promoter region
are required for HES5-mediated repression of Atoh1, we mutated the C-sites in the
Atoh1 reporter construct (Figure 14A). GFP reporter constructs containing either wild
type or mutated promoter sequences were transfected into 293 cells in the absence or
presence of the Hes5 expression vector, and the number of GFP positive cells were
analyzed by flow cytometry as before. In the absence of Hes5 expression, the mutation
of C-sites had no effect on the expression of GFP (Figure 14B). However, mutation of
the C-sites in the Atoh1 promoter increased the percentage of GFP
+
cells in the
presence of HES5, indicative of their requirement for HES5-mediated inhibition. The C-

40

sites had an additive effect: the more sites that were mutated, the greater the loss of the
HES5 inhibitory effect on the Atoh1 promoter (Figure 14B). The percentage of GFP
+

cells inhibited by Hes5 transfection increased from 31.1% (± 0.74) with the wild type
promoter, to 68.2% (± 0.75) in the single C3 mutant, to 74.1% (± 0.59) in the C3/C4
double-mutant, and to 93.2% (± 3.02) when all C-sites were mutated.

Figure 13. HES5 represses Atoh1 through the promoter region. (A) Schematic
diagram of the murine Atoh1 promoter indicating the C-sites and their conservation
among vertebrates. (B and C) Ectopic Hes5 expression requires the Atoh1 promoter
(compare 4B to 4C) to repress expression in heterologous 293 cells. Reporters and
expression constructs were transiently transfected and expression was quantified by
flow cytometry. Reporter in 4B contains the Atoh1 enhancer and 226 bp of Atoh1
promoter (containing the predicted C-sites) driving the expression of GFP. Reporter in
4C contains the Atoh1 enhancer and β-globin basal promoter driving the expression of
mCherry. Reporters were cotransfected with a control (empty vector) or a plasmid
expressing Hes5. The number of cells expressing GFP and mCherry was determined
by flow cytometry after 48h. Values are mean ± SEM, n = 6. [***] p < 1 x 10
-7
.

41

42

Figure 14. Mutational analysis of C-site function. HES5 repressed Atoh1 through
the highly conserved binding sites in the promoter region. Five out of six
nucleotides of each C-site were mutated. (A) Quantification as in 4B and 4C with the
indicated plasmids reported as percentage of cells expressing GFP relative to CMV-
RFP co-transfected control (not shown) set to 100% in the absence of Hes5 expression
(empty vector). Mutation of C-sites in Atoh1 promoter region reduces Hes5 inhibitory
effect on Atoh1 promoter-reporter. Flow cytometry. Values are mean ± SEM, n = 4.
[*] p < 0.0003, [**] p < 2 x 10
-5
, [***] p < 4 x 10
-7
. (B) In the absence of HES5, wild type
and mutated reporters have similar expression. Flow cytometry analysis of 293 cells
transfected with the indicated plasmids for 48h.

Although there is a conserved N-box and a class C-site present in the Atoh1
enhancer, HES5 was not able to inhibit the enhancer-driven reporter (Figure 13C). We
wondered if the reporter assays were not sensitive enough to detect the weak HES5
binding to the Atoh1 enhancer. Transient reporter assays showed that HES5 expression
had no effect on the wild type, or mutated enhancer N-box or C-site (Figure 15).

Figure 15. HES5 does not have inhibitory effect on wild type or mutated Atoh1
enhancer reporter. Quantification of flow cytometry analysis of 293 cells transfected
with the indicated plasmids, reported as percentage of cells expressing mCherry relative
to GFP (not shown) set to 100% in the absence of Hes5 expression. Values are mean ±
SEM, n = 3.

43

HES5 recruits the co-repressor GRG/TLE to the Atoh1 promoter in a Notch-
dependent fashion
HES1 and HES5 proteins have been reported to recruit GRG/TLE corepressors
(GRG/TLE1-4) through their conserved C-terminal WRPW motif (Grbavec and Stifani
1996 ; Grbavec et al. 1998). Therefore, we hypothesized that, in the absence of the
WRPW motif, HES5 would not be able to inhibit Atoh1 expression. To test this, we
deleted the WRPW motif in the Hes5 expression plasmid (Hes5 ΔWRPW) (Figure 16A).
Deletion of this C-terminal motif had no effect on flag-tagged Hes5 expression levels in
transient transfections assays, or on HES5 protein cellular localization (data not shown).
Wild type or ΔWRPW Hes5 plasmid was transfected into 293 cells with the Atoh1
enhancer/Atoh1 promoter
wt
/GFP reporter. As expected, wild type HES5 reduced the
percentage of GFP-expressing cells significantly (30.2% ± 0.8, p < 0.0005). However,
loss of the HES5 ΔWRPW severely reduced the ability of HES5 to inhibit Atoh1 reporter
expression (88% ± 0.89) (Figure 16B). This result suggested that the HES5 WRPW
motif is critical for the repression of Atoh1.
To demonstrate that HES5 physically interacts with GRG/TLE cofactors, we
performed proximity ligation assay (PLA) (Soderberg et al. 2006) in 293 cells
transfected with flag-tagged Hes5 plasmid. We detected GRG/TLE: FLAG-HES5 PLA
signals when wild type FLAG-Hes5 was transfected into the 293 cells (Figure 16C). PLA
signals were drastically reduced when mutant Hes5 (ΔWRPW) was transfected (Figure
16C’), and there were no PLA signals in the absence of FLAG primary antibody (Figure
16C”), suggesting that HES5 and GRG/TLE interact through the WRPW motif of HES5.

44

The result of PLA assay were also confirmed with Coimmunoprecipitation of flag-tagged
HES5 and GRG/TLE in 293 cells (Figure 16E).

45

Figure 16. The HES5 WRPW motif is required for the repression of the Atoh1
promoter and for interaction with GRG/TLE. (A) Schematic representation of mouse
HES5 protein showing the amino acid sequence and different domains. Four amino
acids of the C-terminal WRPW motif were deleted in Hes5 ΔWRPW. (B) Quantification
of flow cytometry analysis of 293 cells transfected with the indicated plasmids reported
as a percentage of cells expressing GFP relative to RFP control (not shown, empty
vector control set to 100%). When Hes5 C-terminal motif WRPW was deleted, HES5
inhibitory effect on Atoh1 reporter was significantly reduced. Values are mean ± SEM, n
= 3. [***] p < 0.0005. (C) Proximity ligation assay showing requirement for the WRPW
motif for interaction with endogenous GRG/TLE. Confocal images of 293 cells
transfected with FLAG-Hes5 wild type or WRPW-deleted expression plasmids (red dots
indicate interaction). DAPI (blue). Scale bar = 20 µm. (D) Quantification of PLA signals
shown in (C). (E) Coimmuniprecipitation assay in 293 cells showing that flag-tagged
HES5 and GRG/TLE interact in vito.

We next wanted to test whether the interaction of HES5 and GRG co-repressor
could be detected at the Atoh1 promoter. Since the paucity of supporting cells in the
organ of Corti makes reliable transcription factor ChIP currently impractical, we chose to
use neural progenitor cells (NPCs) as an in vitro model (Figure 17). NPCs, like
supporting cells, also rely on active Notch signaling that is essential for maintenance of
their progenitor state (Ohtsuka et al. 2001; Hatakeyama et al. 2004), and inhibition of
the Notch signaling pathway by DAPT induces differentiation of embryonic and neural
stem cells (Crawford and Roelink 2007; Borghese et al. 2010), providing an in vitro
model for molecular studies of Notch-mediated gene regulation. Consistent with this,
we observed more than five-fold (p <0.005) upregulation of Atoh1, and a complete loss
of Hes5 mRNA in NPCs following DAPT treatment (Figure 18A). Also, like supporting
cells, NPCs express GRG4 at higher levels compared to the other GRG factors (Figure
18B and C).

46

I
Day 4 of differentiation
Day 7 of differentiation
Relative mRNA

47

Figure 17. Sox1-GFP knock-in mouse embryonic stem cell (mESC) line 46C
expresses GFP and Nestin upon neural differentiation. ESCs (line 46C) were
cultured and differentiated to neural progenitor cells (NPCs) following a monolayer cell
culture protocol (Ying and Smith 2003; Ying et al. 2003). The expression of GFP allows
monitoring of the differentiation procedure and also FACS-purification to obtain a highly
purified population of NPCs. (A-C) Fluorescent and (A’-C’) brightfield images of
differentiation of mES cells. mESCs were cultured without (A and A’) and with
differentiation media for 4 days (B and B’), or 7 days (C and C’). (D-H) ESCs in
differentiation media downregulate Oct4 and upregulate Nestin by day 7 as shown by
immunostaining (D-G), and quantitative real time PCR (H). (I) The fluorescence
activated cell sorting (FACS) analysis of 46C ES cells on day 4 and 7 of differentiation.
As the cell differentiate, the percentage of Sox1-GFP poisitive cell increases. Scale bars
= 50 µm.

48

Figure 18. GRG/TLE localizes to the Atoh1 locus in a Notch dependent manner in
neural progenitors. Neural progenitors (NPCs) were differentiated from ESCs and
FACS-purifed (Sox1-GFP
+
). All Values are mean ± SEM. (A) Atoh1 and Hes5 mRNA
expression levels in 46C ES cells and NPCs that were treated with control (DMSO) or ɣ-
secretase inhibitor (DAPT) for 48h. Similar to supporting cells, inhibition of Notch
signaling by DAPT induces the expression of Atoh1 and represses the expression of
Hes5 in NPCs. n = 4. (B) GRG4 is the predominant family member expressed in
supporting cells and neural progenitors. qPCR analysis of FACS-purified supporting
cells (p27
+
SC), non-supporting cells (p27
-
non SC), 46C ES cells and NPCs. n = 4. (C)
GRG4 is expressed in NPCs and supporting cells. Immunoblotting of GRG4 with 25,000
FACS-purified NPCs, purified p27/GFP
+
supporting cells (SCs) and p27/GFP
-
non-
supporting cells (non SCs) at P1. Histone H2B was used as control.
(D) Schematic of the Atoh1 locus in mouse showing the promoter and the
autoregulatory enhancer regions. Numbers refer to the regions amplified in ChIP-qPCR.

49

The primers used to measure the enrichment after ChIP are shown with arrows (see
Supplemental Material). (E) HES5 localizes to the proximal promoter and not the
enhancer by chromatin immunoprecipitation followed by qPCR of NPCs transfected with
CMV-FLAG-Hes5 plasmid (or CMV-RFP, control). Result is reported as fold enrichment
(Hes5 transfected % input/RFP transfected % input). n = 3. (F) GRG/TLE localizes to
the Atoh1 promoter in a Notch-dependent manner. ChIP-qPCR with anti-pan GRG/TLE
anibody and primers scanning the Atoh1 locus in NPCs treated with control (DMSO) or
ɣ-secretase inhibitor (DAPT) for 24 hours after 7 days of differentiation. n = 3. [*] p <
0.05, [**] p < 0.005.

To test whether HES5 directly interacts with the Atoh1 promoter region, NPCs
were transfected with a flag-tagged Hes5 expression plasmid, chromatin
immunoprecipitation (ChIP) was performed with FLAG antibody and the chromatin was
analyzed by qPCR (Figure 18D). As before, flag-tagged Hes5 transfection was used
because of the lack of a reliable HES5 antibody. HES5 can be seen to physically
interact with the Atoh1 promoter region, but not the enhancer, confirming the reporter
assays (Figure 18E).
As shown, blocking Notch activity in NPCs leads to a rise in the endogenous Atoh1
level and a drop in Hes5 mRNA. ChIP-qPCR in untransfected NPCs with a pan
TLE/GRG antibody showed that endogenous GRG/TLE was also normally enriched at
the Atoh1 promoter region (Figure 18F). In addition, GRG/TLE was lost from the Atoh1
promoter region when Notch signaling was blocked by DAPT (Figure 18F), suggesting
that it is the interaction of HES5 with this co-repressor in NPCs that maintains them in
an undifferentiated state.
Notch inhibition induces histone acetylation at the Atoh1 promoter in supporting
cells
We investigated whether Atoh1 upregulation in response to the blockade of Notch
activity is through the recruitment of HDACs to the Atoh1 locus through the interaction

50

with GRG/TLE. GRG/TLE-mediated repression has been reported to be partially
dependent on HDAC activity (Winkler et al. 2010). In neonatal supporting cells purified
from cochlear cultures treated with the HDAC inhibitor trichostatin A (TSA) for 6 hours,
Atoh1 was upregulated (p <0.05,Figure 19A), suggesting that the repression of Atoh1 in
supporting cells is at least partially dependent on HDAC activity. Similar experiments in
Hes5-transfected 293 cells suggested a similar conclusion (Figure 19B). Furthermore,
inhibition of HAT (histone acetyltransferase) activity by curcumin (Balasubramanyam et
al. 2004) led to a reduction in the level of DAPT-induced Atoh1 mRNA, further indicating
a need for ongoing histone acetylation to induce Atoh1 (Figure 19C). Curcumin, either
on its own or in conjunction with DAPT, had no effect on the expression level of the
Notch effectors Hes5 and Hey1 (data not shown). This prompted us to ask if inhibition of
Notch signaling in supporting cells, which we have shown induces Atoh1 expression,
brings about changes in the acetylation status of histones at the Atoh1 locus. We
performed ChIP-qPCR with FACS-purified DMSO- and DAPT-treated supporting cells
for H3K9ac, a histone modification highly correlated with actively transcribed genes
(Wang et al. 2008b). Inhibition of Notch signaling for 24 hours increased the H3K9ac
level in transdifferentiating supporting cells at the promoter region (Figure 19D). This
result strongly suggests that the upregulation of Atoh1 in supporting cells is
accompanied by epigenetic changes that render the chromatin more permissive at the
Atoh1 locus.

51

Figure 19. Notch inhibition leads to histone acetylation at Atoh1 promoter in
supporting cells. (A) Inhibition of histone deacetylase (HDAC) activity by trichostatin A
(TSA) for 6h induces Atoh1 expression in FACS-purfied supporting cells. Values are
mean ± SEM, n = 3. (B) HES5 cannot repress Atoh1 in the absence of HDAC activity.
Hes5 overexpression in 293 cells in the absence and presence of trichostatin A (TSA)
for 15h. Values are mean ± SEM, n = 3; [**] P< 0.001, Student's t-test. (C) Inhibition of
histone acetyltransferase (HAT) activity by curcumin for 24h blocks DAPT-induced
Atoh1 activation in FACS-purfied supporting cells (Lfng
+
). n = 3. (D) Inhibition of Notch
signaling (DAPT for 24h) increases H3K9 acetylation analyzed by ChIP-qPCR at the
Atoh1 promoter in FACS-purified supporting cells. Values are mean ± SE, n = 3. [*] p <
0.05, [**] p < 0.005

In addition to the recruitment of HDACs and deacetylation of histones, GRG/TLE
cofactors are able to reduce chromatin accessibility by aggregation of nucleosomes
(Sekiya and Zaret 2007). We conducted transposase-accessible chromatin (ATAC-seq)

52

(Buenrostro et al. 2013) on at least two biological replicates of FACS-purified hair cells
(AtohGFP fusion knock-in, (Rose et al. 2009)), supporting cells and transdifferentiating
supporting cells (treated with DAPT for 24h) to identify open chromatin regions at the
Atoh1 locus. We used Lfng/GFP transgenic line to purify supporting cells. In this line,
GFP is expressed in Deiter’s, outer pillar and inner phalangeal cells, so the HES5-
expressing cells are more represented in the sorted supporting cells compared to
p27/GFP. As shown in Figure 20, although hair cells and freshly FACS-purified
supporting cells showed a significant difference in the accessibility of chromatin at
Atoh1 promoter and enhancer (P< 0.05), inhibition of Notch signaling by DAPT in
supporting cells did not increased chromatin accessibility at these regions, although a
trend at the promoter was observed

53

A
C
B
D

54

Figure 20. ATAC-seq in postnatal hair cells and supporting cells. Chromatin
accessibility at Atoh1 promoter and enhancer was assessed by ATAC-seq in FACS-
purified hair cells, supporting cells, 24h DMSO- and DAPT-treated supporting cells at
P1 (average of two replicates is shown). (A and B) Atoh1 promoter and enhancer are
significantly more accessible in hair cells compared to supporting cells. (C and D)
Inhibition of Notch signaling by DAPT for 24h in supporting cells does not result in an
increase in chromatin accessibility at Atoh1 promoter or enhancer region. The predicted
HES/HEY C-sites and transcriptional start site are shown in (A) and (C). Atoh1
enhancer element B is marked in (B) and (D).

Atoh1 is upregulated in prosensory progenitors, not just in nascent hair cells
The mutational analysis of the Atoh1 promoter-reporter in vitro demonstrated the
importance of the class C-sites for Atoh1 repression in 293 cells. To better understand
the mechanism of HES-mediated lateral inhibition in vivo, we generated a double-
transgenic mouse line carrying both the wild type and mutated Atoh1 promoter reporter
constructs. During organ of Corti development, Atoh1 is upregulated in nascent hair
cells in a basal-to-apical wave (Chen et al. 2002). Evidence suggests that lateral
inhibition is responsible for repressing Atoh1 in the prosensory progenitors surrounding
the nascent hair cells, and allowing their differentiation as supporting cells (Kiernan et
al. 2005a). We hypothesized that if Atoh1 was not selectively upregulated in hair cells,
but instead was upregulated in groups of prosensory progenitors immediately in front of
(apical to) the visible wave of hair cell differentiation, then the C-sites in the Atoh1
promoter would be needed to silence Atoh1 expression in what will become nascent
supporting cells. This process, in turn, might be delayed in the mutated, compared to
the wild type reporter construct during the developmental rise in Atoh1 expression. In
our double-transgenic line, the wild type Atoh1enhancer-promoter drives the expression
of tdTomato (Atoh1 enhancer/Atoh1 promoter
wt
/tdTomato), and the fully C-site-mutated
Atoh1enhancer-promoter construct drives the expression of GFP (Atoh1

55

enhancer/Atoh1 promoter
mut
/GFP) (Figure 21A). At E16.5 in the middle region of the
cochlear duct, while the expression of tdTomato was present in inner and outer hair
cells, GFP was strongly expressed in the hair cell region, as well as being misexpressed
in the differentiating supporting cell layer (Figure 21D, arrowheads). In a slightly more
apical and less differentiated position along the cochlear duct, only nascent inner hair
cells expressing the wild type tdTomato transgene are observed, with a commensurate
increase in the number of progenitors expressing GFP from the mutant promoter
(Figure 21B, C), consistent with the well documented earlier differentiation of inner,
relative to outer hair cells (Chen et al. 2002). Together, these results reinforce the
importance of the HES/HEY binding sites in the Atoh1 promoter region during
development for the Notch-mediated lateral inhibition. They also indicate that Atoh1 is
upregulated in prosensory progenitors, not just in nascent hair cells. At P1, however,
both wild type and mutant promoter constructs (tdTomato and GFP, respectively) were
specifically expressed in the hair cells, and not in the surrounding supporting cells
(Figure 21D).This suggests that HES/HEY repression of endogenous Atoh1 mRNA was
sufficient at this time to keep the expression of endogenous ATOH1 protein below the
threshold needed to trigger the Atoh1 enhancer in vivo.

56

Figure 21. Misexpression of mutant promoter transgene in vivo shows HES/HEY
binding sites in the Atoh1 promoter are required for the proper silencing of Atoh1
in supporting cells. (A) Schematics show wild type and mutant constructs used to
generate the double-transgenic mouse line Atoh1 enhancer/Atoh1 promoter
wt
/tdTomato
and Atoh1 enhancer/Atoh1 promoter
mut
/GFP. (B and C) The expression pattern of the
wild type and mutant transgnes at E16.5 and P1 is shown in cross section. (B)
Illustration showing that mutant transgene expression is not silenced in groups of
progenitor cells in the mid-apical region of the cochlear duct at a time when only inner
hair cells are positive for wild type transgene expression. Asterisk (*) indicates debris
causing auto-fluorescence. (C) At E16.5, the GFP from the mutant promoter transgene
is misexpressed in the supporting cell layer (white arrowheads), in addition to the hair
cells, indicating need for active repression for supporting cell silencing, while tdTomato
from the wild type transgene is rapidly silenced by lateral inhibition. At P1, the
expression of both wild type and mutant promoters is limited to hair cells, indicating that

57

low levels of ATOH1 (activator insufficiency) is sufficient to maintain silencing from the
mutant transgene. Scale bar is 50 µm in B and 20 µm in C and D.

To further test the importance of C-site repression in the Atoh1 promoter, we
cultured P1 cochlear explants and treated with DAPT for 18 hours to inhibit Notch
signaling and induce the wild type and mutant reporters in supporting cells, after which
DAPT was washed out from half of the explants to allow re-expression of Notch
Hes/Hey effectors. Explants were collected after an additional 6 hour incubation, and
the expression level of Atoh1, GFP and tdTomato were compared between the two
groups by qPCR. As shown in Figure 22, the expression of tdTomato from the wild type
promoter transgene, and endogenous Atoh1 were reinhibited following DAPT washout
by the restored Notch-mediated lateral inhibition, but the expression of GFP from the
mutated promoter transgene failed to be rerepressed.

Figure 22. HES/HEY binding sites are required to rerepress Atoh1 in supporting
cells after Notch signaling is first inhibited, allowing Atoh1 levels to rise, and then
restored. Schematic shows experimental time course: P1 cochlear cultures from
double-transgenic mouse line (Figure 21A) were treated with DAPT for 18h after which

58

DAPT was washed out from half of the explants, and the explants were collected after
an additional 6h in cultures. Expression levels of endogenous Atoh1, GFP, and
tdTomato between the two groups using qPCR show that endogenous Atoh1 and wild
type transgene are actively repressed by returning Notch signaling, but the mutant
transgene fails to be rerepressed in the same time period (6h). Values are mean ± SEM,
n = 3. [*] p < 0.05.

Figure 23. Expression level of class
I, II and IV HDACs in FACS-purified
p27-positive supporting cells (SCs)
compared to p27-negative cells
(non-SC) in P1 cochlea measured by
qPCR.

59

Summary
The results presented in this chapter suggest that repression of Atoh1 in supporting
cells is directly mediated by active repression of the Atoh1 promoter through the Notch
effector HES5 as well as lack of ATOH1 (activator insufficiency) to induce the Atoh1
enhancer activity.
Our data shows that the Notch effector HES5 is critical for Atoh1 suppression.
Inhibition of protein synthesis results in the degradation of HES5 and consequent
upregulation of Atoh1 in supporting cells (Figure 11and Figure 12). Induction of Atoh1 in
supporting cells in the absence of de novo protein synthesis is an example of
derepression where loss of the repressor allows the activation of the target gene (Riccio
et al. 2008; Kulic et al. 2015). Consistent with these observations, stabilizing HES5
protein by inhibiting the proteosome pathway prevents DAPT-induced Atoh1 in
supporting cells (Figure 12B). Small numbers of supporting cells in the organ of Corti
and lack of specific HES5 antibody made us assess the degradation rate of FLAG-
HES5 protein in 293T cells. There are no previous reports on HES5 protein turnover in
other cell types, but HES1, also a target of proteasomal degradation, has been shown
to have a similar half-life in cell types such as fibroblasts and embryonic stem cells
(Hirata et al. 2002; Kobayashi et al. 2009).
Our results show that HES5 represses Atoh1 expression by directly binding to the
promoter region of Atoh1 and that this binding is dependent on four conserved C sites
located within the proximal 226 bp of the transcription start site (Figure 13). Mutation in
all four of the C sites abolishes the inhibitory effect of HES5 as shown in reporter
assays (Figure 14). Although there is a conserved C site and an N box in Atoh1

60

enhancer, we did not detect HES5 binding to the enhancer region (Figure 18D), and
mutations of these sites had no effect on the expression of Atoh1enhancer reporter
(Figure 15) indicating that the Atoh1 enhancer is insufficient to mediate Atoh1
suppression through Notch lateral inhibition.
Our results further suggest that HES5–dependent repression of Atoh1 promoter is
mediated by the recruitment of GRG/TLE corepressors and HDACs. We show that the
HES5 WRPW motif is needed for the repression of Atoh1 (Figure 16B), HES5 and
GRG/TLE physically interact through WRPW motif (Figure 16C) and that GRG/TLE,
likely GRG/TLE4, localizes to the TSS of Atoh1 in a Notch-dependent manner (Figure
18E). We also show that inhibition of HDAC activity in FACS-purified supporting cells
lead to up-regulation of Atoh1 (Figure 19A). In addition, 24 hours after Notch inhibition,
we observed a modest but significant increase in histone acetylation, specifically
H3K9ac in the promoter region, consistent with the increase in expression of Atoh1
(Figure 19D). This is not surprising as Notch signaling has been shown to prevent the
recruitment of the histone acetyltransferase p300 and the acetylation of histones
(Krishnamoorthy et al. 2015), and transcriptional repression by HES1, HEY1 and HEY2
proteins has been associated with histone deacetylase activity (Iso et al. 2001; Takata
and Ishikawa 2003; Weber et al. 2015). The HDAC inhibitor that we used in this study
(trichostatin A or TSA) is a pan HDAC inhibitor that blocks the activity of HDACs in class
I and II (see chapter 1, (Seto and Yoshida 2014)). Further studies are required to
identify which HDAC or HDACs are specifically involved in Atoh1 repression in
supporting cells. The expression analysis of these enzymes suggests that multiple

61

HDACs are expressed in supporting cells (Figure 23), perhaps reflecting the
heterogeneity in supporting cell population.
I also examined chromatin accessibility of Atoh1 locus using ATAC-seq in hair cells,
supporting cells and supporting cells treated with DMSO or DAPT for 24h. In spite of the
observation that chromatin at the Atoh1 locus (promoter and enhancer) in hair cells is
significantly more accessible than in supporting cells (Figure 20), we observed little
change in chromatin after 24h of Notch inhibition suggesting that immediate transition to
a highly open chromatin state is not required for the induction of Atoh1 expression but
the chromatin is responsive to activation. An alternative explanation is that supporting
cell transdifferentiation is incomplete at 24 hours post-DAPT treatment; longer inhibition
of Notch signaling will lead to a detectable increase in chromatin accessibility at Atoh1
locus in supporting cells. This notion is more likely considering that supporting cells in
the basal regions of the cochlea are less sensitive to DAPT compared to apical regions
as observed by the upregulation of Atoh1 enhancer/β-globin promoter/GFP first in the
apical region after DAPT treatment (Doetzlhofer et al. 2009).
It is unclear what mechanisms are responsible for inducing Atoh1 expression
developmentally. Whether or not upregulation is due to a new positively acting event or
a reduction in repression mediated by the existing factors such as HEY repressors, is
unknown. Regardless, we show that similar to postnatal supporting cells, in the absence
of direct repression due to mutation of HES/HEY binding sites in the Atoh1 promoter,
Atoh1 has aberrant expression in nascent supporting cells (Figure 21). This suggests
that Atoh1 levels transiently rise in the progenitors of both hair cell and supporting cell,
and that active repression through the Atoh1 promoter is required to inhibit this process

62

of transcriptional activation in nascent supporting cells. Our data suggest that the
reported low level of Atoh1 transcription in the prosensory domain at E13.5 (Lanford et
al. 2000; Woods et al. 2004) prior to the obvious upregulation of Atoh1 in the nascent
hair cells (Chen et al. 2002; Lumpkin et al. 2003), does not rise to the threshold needed
to trigger autoregulation (activator insufficiency, discussed more in chapter 4 and
conclusion). A recent study that used inducible Atoh1
Cre
to label Atoh1-expressing cells
showed that in E13 cochlea induced for one day in vitro, 30% of the cells from the
Atoh1-expressing cells develop as supporting cells (Driver et al. 2013). Our data not
only confirms this observation, but also provides a mechanism for the subsequent
repression of Atoh1 in supporting cell.

63

Chapter 3: The mechanism of HEY2-mediated repression of
Atoh1 in postnatal supporting cells
Introduction
HEY factors, HEY1, HEY2 and HEYL (also known as Hrt1/2/3, Hesr1/2/3,
Herp2/1/3, or Chf2/1/3) are direct targets of the canonical Notch signaling pathway,
although their regulation by other signaling pathways have also been reported (Maier
and Gessler 2000; Nakagawa et al. 2000; Iso et al. 2003; Hayashi et al. 2008a;
Doetzlhofer et al. 2009; Benito-Gonzalez and Doetzlhofer 2014).
HEY factors are involved in many cellular processes including angiogenesis,
somitogenesis, myogenesis, and gliogenesis (reviewed in (Iso et al. 2003)). In the
cochlea, Hey1 and Hey2 are expressed in the prosensory domain by E12.5 (Hayashi et
al. 2008a), but are downregulated as they differentiate. By P1, Hey2 expression is
limited to pillar cells, while Hey1 is expressed in Deiters’ and Hensen cells (Hayashi et
al. 2008a; Doetzlhofer et al. 2009) . HeyL expression is not detected until E16.5, and by
P1 its expression restricts to GER and Deiters’ cells (Hayashi et al. 2008a). The function
of HEY factors in prosensory cells is not clear. They have been suggested as effectors
of Notch lateral induction, required for the specification of prosensory domain (Hayashi
et al. 2008a). However, it has been shown that neither the specification of prosensory
domain, nor the Hey2 expression is affected in RBPj knock-out mice (Basch et al.
2011), suggesting the involvement of other signaling pathways regulating Hey
expression. In fact, a recent study showed that Hey1 and Hey2 are controlled by
Hedgehog signaling in prosensory cells and that Hedgehog is critical to keep

64

prosensory cells in an undifferentiated state by preventing the upregulation of Atoh1
(Benito-Gonzalez and Doetzlhofer 2014). In the postnatal organ of Corti, Hey2
expression in pillar cells is regulated by both the FGF and Notch signaling pathways;
inhibition of Notch and FGF leads to downregulation of HEY2, upregulation of Atoh1 in
pillar cells and transdifferentiation of these cells to hair cell-like cells (Doetzlhofer et al.
2009). Involvement of other signaling pathways in the regulation of HEY factors in the
postnatal cochlea has not been reported.
Structurally, the main difference between HES and HEY proteins is a proline
residue in the basic region of HES members that is replaced by glycine in HEY factors,
suggesting that HEY2 and HES5 could have different DNA binding properties (Figure
24).

Figure 24. Basic helix-loop-helix sequence of Hes5 and Hey2 protein. Black areas
indicate conserved residues.

Nevertheless, HEY factors, similar to HES factors, are repressors that can inhibit
transcription both actively and passively (Chin et al. 2000; Nakagawa et al. 2000; Iso et
al. 2001; Sun et al. 2001; Heisig et al. 2012) and HEY-mediated repression of target
genes is partially dependent on HDAC activity . It has been shown that HEY1 recruits
NcoR and Sin3a, both of which are part of the Sin3 corepressor complex that recruits
histone deacetylase1 (HDAC1) (Iso et al. 2001; Gould et al. 2009). HEY2 has been
shown to interact with Sirt1 (Takata and Ishikawa 2003) and HDAC2 that correlates with
a reduction in H3K27 acetylation (Weber et al. 2015). The cofactors assisting HEY2 to
inhibit Atoh1 in pillar cells are not known.

65

Although the inhibitory effect of HEY2 on Atoh1 has been shown (Hayashi et al.
2008a; Doetzlhofer et al. 2009; Benito-Gonzalez and Doetzlhofer 2014), the molecular
mechanisms on which this inhibition is based have not been determined. In the current
chapter, the mechanism of HEY2-mediated repression of Atoh1, as well as the role of
histone deacetylases in this process was investigated. We provide evidence that HEY2
repression of the Atoh1 gene involves direct binding of this repressor to the conserved
binding site in the Atoh1 promoter region, and that HDAC activity contributes to the
repression of Atoh1. We show that pharmacological inhibition of NAD
+
-dependent
HDACs by Salermide and NAD
+
-independent HDACs by TSA (discussed in chapter 1)
in organotypic cultures lead to downregulation of supporting cell markers and up-
regulation of Atoh1 in pillar cells without inducing proliferation, suggesting their direct
transdifferentiation. These results demonstrate the importance of histone deacetylase
activity for maintaining the pillar cell fate and suggest that a combination of NAD
+
-
dependent and -independent histone deacetylases may work through HEY2-mediated
repression of Atoh1.
Results
HEY2 inhibits Atoh1 through the promoter region
HEY proteins are bHLH transcriptional repressors that bind to the proximal promoter
regions of their target genes, although DNA-dependent repression is not their only
mode of repression (Heisig et al. 2012; Tu et al. 2012). The preferred binding sites for
HEY factors are class B and C, E boxes (CACGTG/CACGCG) (Iso et al. 2001; Heisig et
al. 2012).

66

The Atoh1 promoter and enhancer both contain conserved HEY binding sites
(Figure 25A) suggesting that Atoh1 expression might be directly regulated by HEY2. To
test whether HEY2 inhibits Atoh1 expression through the promoter and/or enhancer
region of Atoh1, in vitro reporter assays were performed with a reporter construct
containing both Atoh1 enhancer sequence and 226 bp of Atoh1 promoter sequence
driving the expression of GFP (Atoh1 enhancer/Atoh1 promoter/GFP, similar to Figure
13B). This reporter construct was transfected into 293 cells without and with a Hey2
expression plasmid and the percentage of GFP positive cells were determined by flow
cytometry. An RFP expression plasmid was also co-transfected in all assays as
endogenous control (not shown). Expression of Hey2 significantly reduced the
percentage of GFP positive cells (55.12% ± 1.26 relative to controls), suggesting
HEY2’s inhibitory effect on the reporter (Figure 25C). To find out whether HEY2’s effect
on the reporter construct was due to the Hey binding sites in the promoter region, the C-
sites were mutated and the mutated reporter was tested again in reporter assays. In the
absence of Hey2 expression, the mutation of C-sites had no significant effect on the
expression of GFP (Figure 14B). Mutation of the C3 site decreased the repressive
effect of Hey2 co-transfection by 18%, as seen by an increase in the percentage of
GFP-reporter positive cells (72.99% ± 0.33 relative to the wild type control), indicative of
its requirement for HEY2-mediated inhibition. Mutation of additional C sites only slightly
increased the percentage of GFP
+
cells (77.78% ± 1.58), suggesting that (1) the C3-site
is the most critical for Atoh1 repression by HEY2 and (2) part of HEY2-mediated
repression of Atoh1 is likely independent of the C-sites in the promoter region.

67

To test whether HEY2 promoter-independent repression of Atoh1 by HEY2 could
occur through the enhancer, reporter assays were performed with a construct that
contained the Atoh1 enhancer, and β-globin basal promoter (Atoh1 enhancer/β-globin
promoter/mCherry, similar to Figure 13C). The β-globin promoter lacks HEY2 preferred
binding sites. When Hey2 was co-transfected with this reporter construct, the
percentage of mCherry
+
cells decreased to 82% (± 1.84) of controls (Figure 26B).
However, mutations of the N-box or C-site in the enhancer region did not change the
percentage of mCherry
+
cells; HEY2 was still able to inhibit the enhancer-driven reporter
(~ 20%). This result indicates that HEY2 is a weak inhibitor of the enhancer, but this
inhibition is independent of the HEY binding sites in the enhancer sequence.
To confirm that HEY2 inhibits Atoh1 by directly interacting with the Atoh1 promoter
region, chromatin immunoprecipitation (ChIP) assays in mouse OC1 (organ of Corti1)
cells was performed, since the paucity of supporting cells in the organ of Corti makes
reliable transcription factor ChIP difficult. To perform the HEY2-ChIP assay, OC1 cells
were transfected with a FLAG-Hey2 expression plasmid, and chromatin was
precipitated with a FLAG antibody. FLAG-HEY2 bound DNA was analyzed with PCR
primers encompassing the Atoh1 gene promoter, exon (coding region), and the
enhancer A and B (Figure 27). Assessment by ChIP-qPCR revealed that HEY2 binds
to Atoh1 promoter region, but not the enhancer. This result confirms the reporter assays
performed with the Atoh1 enhancer/Atoh1-promoter/GFP construct and also shows that
HEY2’s inhibition of the enhancer does not rely on HEY2-DNA binding.

68

Figure 25. Hey2 is capable of repressing Atoh1 transcription through both the
promoter and enhancer regions. (A) Schematic presentation of Atoh1 locus in mouse
based on mm9, showing the promoter region and enhancer element (enhancer A and
enhancer B) 3’ to the coding region. Hes/Hey predicted binding sites N box and class C
site are shown. Atoh1 enhancer contains one copy of the class C site and one copy of N
box. The N box overlaps with the E-box which is the binding site for bHLH
transcriptional activators like Atoh1 itself. Atoh1 promoter contains four copies of the
class C site. Numbers refer to the regions amplified in ChIP-qPCR. The site of primers
used to measure the enrichment after ChIP is shown with arrows. (B) The schematic of
the Atoh1engacner-promoter reporter construct used in transient transfection assays. In
this reporter Atoh1 enhancer and 226 bp of Atoh1 promoter (containing the four
A
B
C

69

predicted C sites) drive the expression of GFP. Five out of six nucleotides of each C site
were mutated. (C) Quantification of flow cytometry analysis in 293 cells transfected with
the indicated plasmids. Hey2 repression of Atoh1 promoter was dependent on the
conserved C3 site. Values are mean ± SEM, n = 3; [***] P < 0.0005, [**] P < 0.005,
Student's t-test.

Figure 26. HEY2’s inhibitory effect on the Atoh1 enhancer is not due to N-box or
C-site-dependent DNA binding. (A) The schematic of the Atoh1 enhancer reporter
construct used in transient transfection assays. In this reporter beta globin basal
promoter was inserted to drive the expression of mCherry. (B) Quantification of flow
cytometry analysis in 293 cells transfected with the indicated plasmids. The C-site and
N-box were mutated. The wild type and mutated enhancer reporter constructs were
transfected into 293 cells with an empty vector or a plasmid expressing Hey2, and the
number of cells expressing mCherry was determined by flow cytometry after 48h. Hey2
weak inhibitory effect on Atoh1 enhancer was not dependent on N-box or C-site.
Values are mean ± SEM, n = 3.

A
B

70

Figure 27. HEY2 binds to Atoh1
promoter in vitro. ChIP-qPCR
result in OC1 cells that were
transfected with CMV-FLAG-Hey2
plasmid or CMV-GFP plasmid
(control) for 48 hours. Chromatin
immunoprecipitation was performed
with anti-FLAG antibody. The result
is reported as fold enrichment (Hey2
transfected % input/GFP transfected
% input). Shown are the values as
mean ± SEM for three independent
replicates. Hey2 is significantly
enriched with p < 0.05 (Student's t-
test) at the Atoh1 promoter region
when compared to the coding
region.

HEY2 inhibitory effect is not dependent on its C-terminal motif
Previous reports have shown that the basic domain and not the C-terminal
tetrapeptide motif of HEY1 and HEY2 proteins is required for their repression activity
(Nakagawa et al. 2000; Iso et al. 2001). In order to determine which domain of HEY2 is
required for repression of Atoh1, mutant versions of HEY2, ∆basic and ∆YQPW were
made in which most of the basic domain and all of the C-terminal domain is deleted,
respectively (Figure 28A). The deletion of the basic domain and YQPW motif had no
effect on the expression level of HEY2 as evident by immunoblotting (Figure 28B). The
Hey2 wild type and mutant expression plasmids were transfected into 293 cells with the
Atoh1 enhancer-Atoh1 promoter reporter (Figure 25A) and the number of GFP
+
cells
were determined by flow cytometry as before. As shown in Figure 28C, the wild type
HEY2 was able to inhibit the Atoh1 reporter, but the ∆basic version of HEY2 had no
inhibitory effect; the percentage of GFP
+
cells was similar to the percentage of cells in
the absence of HEY2. On the other hand, deletion of the YQPW motif had no effect on

71

HEY2’s ability to inhibit the reporter. This result suggests that the Hey2 basic domain is
critical for its repressive activity on the Atoh1 promoter.

Figure 28. HEY2 basic domain, but not the YQPW, motif is required for the
repression of the Atoh1 promoter. (A) Schematic representation of mouse HEY2
protein showing the different domains that is FLAG-tagged. Four amino acids of the C-
terminal YQPW motif and 9 amino acids of the basic domain were deleted in ΔYQPW
and Δbasic, respectively and. (B) Immunoblotting with anti-FLAG and histone H2B
(control) antibody from FLAG-Hey2 WT, ∆basic or ∆YQPW transfected 293 cells.
Deletion in the basic or YQPW domain does not affect Hey2 expression. (C)
Quantification of flow cytometry analysis of 293 cells transfected with the indicated
plasmids reported as percentage of cells expressing GFP relative to RFP (not shown)
A
B
C

72

set to 100% in the absence of Hey2 expression. When Hey2 basic domain was deleted,
HEY2 inhibitory effect on Atoh1 reporter was significantly reduced hence the
percentage of GFP positive cells increased. Values are mean ± SEM, n = 3. [***] p <
0.0005.
HEY2-mediated repression of Atoh1 is dependent on HDAC activity
The known function of HEY proteins in recruiting HDACs (Iso et al. 2001; Takata
and Ishikawa 2003; Weber et al. 2015) led us to hypothesize that HDAC activity was
required in the HEY2-mediated repression of Atoh1. This hypothesis was initially tested
in 293 cells using trichostatin A (TSA), a pan HDAC inhibitor of class I and II HDACs,
and Salermide, an inhibitor of Sirt1 and Sirt2, members of class III HDACs (discussed in
chapter 1). 293 cells were treated with DMSO, TSA, Salermide, or Salermide and TSA
together (referred to as Salermide/TSA), in the absence or presence of the Hey2
expression plasmid. Overexpression of HEY2 in 293 cells downregulates Atoh1
expression, but if HEY2 is expressed in the presence of Salermide/TSA, Atoh1
downregulation is lost (Figure 29A). In the presence of TSA alone or Salermide alone,
HEY2 is still able to downregulate Atoh1 expression, suggesting that HEY2-mediated
repression of Atoh1 is dependent on both classes of HDACs.
This result encouraged us to test the HDAC inhibitors in cochlear explants. Atoh1
enhancer/β-globin promoter/GFP transgenic line was used in which the GFP driven by
Atoh1 enhancer labels hair cells. Cochlear cultures were established at postnatal day 1
(P1) and treated with TSA and/or Salermide. When explants were treated for 72h with
Salermide/TSA, extra GFP
+
cells appeared between the inner and outer hair cells,
where normally pillar cells are located. Single treatment with TSA or Salermide had no
significant effect (Figure 29B-E). Quantification of Atoh1/GFP positive cells in the
treated explants showed a significant increase (Figure 29F). To determine how

73

differentiated the extra Atoh1/GFP
+
cells are, the expression of Myosin VI (Myo VI),
downstream target of Atoh1 was examined. As shown in Figure 29G, Atoh1/GFP
positive cells are also Myo VI positive.
Given that Salermide and TSA did not induce cell proliferation (Figure 30), we
speculated that the observed supernumerary GFP
+
cells are transdifferentiating pillar
cells. To test this, the expression of p75, a marker of inner pillar cells (Shim et al. 2005;
Mansour et al. 2013) was examined. As shown in Figure 31A, p75 staining appears as a
band between the inner and outer hair cells, but when Salermide/TSA are added to
organ cultures for 72h, p75 staining is drastically reduced. TSA or Salermide alone did
not have significant effects on p75 expression. In order to confirm that Salermide and
TSA specifically affect the differentiation status of inner pillar cells, I also examined the
expression of CD44, a marker of outer pillar cells (Mansour et al. 2013). In all of the
treated cochlear explants CD44 staining was detected (Figure 31B), suggesting that the
extra GFP
+
cells that were observed (Figure 29B-E), were indeed inner pillar cells.

74

A
B
D
C
E
TSA
DMSO Salermide
Salermide
and TSA
F
G

75

Figure 29. HEY2-mediated repression of Atoh1 is dependent on HDAC activity. (A)
HEY2 ability to downregulate Atoh1 expression in 293 cells is lost when HDAC activity
is blocked by Salermide and TSA. Salermide or TSA alone does not have any effect.
Values are mean ± SEM, n = 3; [*] p < 0.03, [**] p < 0.0005, Student's t-test. (B-E)
Treatment of Atoh1/GFP cochlear explants with Salermide and TSA together (C) leads
to appearance of supernumerary GFP positive cells between the outer and inner hair
cells. (F) Quantification of Atoh1/GFP positive cells in P1 cochlear explants treated with
Salermide and TSA for 72h. A minimum of three cochlear cultures were analyzed for
each condition (mean number of cells per 100 µm ± SEM). (G) The digital cross section
of Atoh1/GFP cochlear cultures treated for 72h with Salermide and TSA. The extra GFP
positive cells that appear in the presence of Salermide and TSA (C) are Myo VI positive
(red). Scale bar = 20 µm. (n.s.) not significant.

Figure 30. Salermide and TSA do not induce cell proliferation in the sensory
epithelium of organ of Corti. No Edu (Red) incorporation in cochlear cultures treated
for 60h. Nuclei are counterstained with DAPI (blue). The layer below basilar membrane
is shown as positive control. Scale bar = 25 µm.

76

A
B
C
D
C'

77

Figure 31. HDAC inhibition by Salermide and TSA leads to transdifferentiation of
inner pillar cells. (A-C’) P1 Atoh1/GFP transgenic cochlear explants were stained with
p75 (A) CD44 (B) or Prox1 (C) antibody (red) after being treated with Salermide and
TSA for 72h. p75 and Prox1 but not CD44 expression is repressed by Salermide and
TSA together suggesting the extra Atoh1/GFP
+
cells are transdifferentiating inner pillar
cells. Some TSA-treated Prox1
+
cells are also GFP
+
(C’). (D) Quantification of Prox1
+

cells/100 µm as in (C). The number of Prox1 positive supporting cells is reduced by
Salermide and TSA together. Scale bar is 50 µm in (A) and (C) and 20 µm in (B).

If Salermide and TSA are able to induce transdifferentiation of inner pillar cells to
hair cell-like cells, then a decrease in the number of supporting cells is expected. To test
this, expression of Prox1, a marker of Deiters’ and pillar cells (Kirjavainen et al. 2008)
was examined. As shown in Figure 31C, when cochlear cultures were treated with
Salermide/TSA, the number of Prox1
+
cells in the pillar cell region decreased, but not
when treated with TSA or Salermide alone, although with TSA alone, a few Atoh1/GFP
+

cells between the inner and outer rows of hair cells could be detected that were also
Prox1
+
(Figure 31C’). Quantification of Prox1
+
cells showed a significant decrease and
suggested about one row of supporting cells is missing when treated with both inhibitors
(Figure 31D).
Salermide and TSA stimulate Atoh1 expression in neonatal pillar cells
To confirm that Salermide and TSA can induce the expression of the endogenous
Atoh1 gene and not only the Atoh1 enhancer reporter, pillar cells were FACS-purified
using the p75 antibody staining (Figure 32) to analyze gene expression.

78

Figure 32. Purification of pillar cells with FACS. (A) FACS plot of P1 wild type
cochlear cells stained with anti-p75

antibody. The gate shows the sorted p75
+

supporting cells. (B) Verification of supporting cell purity by qPCR showing the p75-fold
difference between the p75
-
and p75
+
sorted cells. Values are mean ± SEM, n = 4;
[*] p < 0.05.

Organ cultures were treated with Salermide and TSA alone or together and 24h
later organs were dissociated and pillar cells FACS-purified using an antibody to p75. In
p75
+
supporting cells, when both Salermide and TSA were added to cochlear cultures,
Atoh1 was upregulated (1.7 fold, p < 0.05). Atoh1 was also upregulated with TSA (1.48
fold, p < 0.05) or Salermide (1.33 fold) alone, but to a lesser degree (Figure 33). On the
other hand, Hey2 expression is not affected by Salermide and TSA together, and is a bit
increased with TSA alone (1.33 fold, p < 0.05). This suggests that upregulation of Atoh1
in response to Salermide and TSA is not due to downregulation of Hey2, and loss of
HEY2-mediated repression. p75 expression, consistent with antibody staining (Figure
31A), is significantly downregulated (4.76 fold, p < 0.005).

A B

79

Figure 33. Real-time quantitative PCR showing the fold difference in expression
level of Atoh1, Hey2 and p75 between DMSO- , Salermide/TSA-, TSA-, and
Salermide-treated p75
+
supporting cells. Values are mean ± SEM, n = 4; [*] p < 0.05,
[**] p < 0.005, Student's t-test.

80

Salermide and TSA are not capable of blocking HES5-mediated repression of
Atoh1
The ability of Salermide/TSA to block the repression of Atoh1 in pillar cells, but not
Deiters’ cells (Figure 29B-E and Figure 31C), suggested that only HEY2-mediated
repression of Atoh1 is sensitive to the double treatment with these inhibitors. To test
this, 293 cells were treated with DMSO, TSA, Salermide or Salermide/TSA together in
the absence or presence of the HES5 expression plasmid (similar to Figure 29A).
Overexpression of HES5 in 293 cells downregulates Atoh1 expression (0.41 fold, p <
0.01) (similar to HEY2), but if HES5 is expressed in the presence TSA, Atoh1
downregulation is reduced (0.79 fold) (Figure 34A, yellow bars). In the presence of
Salermide, HES5 is still able to downregulate Atoh1 expression (0.41 fold, p < 0.01),
and addition of Salermide/TSA did not have an additive effect (0.87 fold) (Figure 34,
purple bars vs yellow bars), suggesting that HES5-mediated repression of Atoh1 is only
dependent on TSA-sensitive or classical HDACs, as was also shown in the previous
chapter (Figure 19). To confirm this result in supporting cells, the expression level of
Atoh1, Hes5 and Hey2 was examined in supporting cells FACS-purified from p27/GFP
transgenic cochlear cultures that had been treated with the inhibitors for 15h and 40h.
As shown in Figure 34B, Salermide/TSA cannot upregulate Atoh1 expression in HES5-
expressing supporting cells, which might explain the failure of Deiters’ cell
transdifferentiation in the presence of Salermide and TSA (Figure 31).

81

Figure 34. Salermide and TSA cannot induce Atoh1 expression in HES5-
expressing cells in vitro or supporting cells ex vivo. (A) Hes5 ability to
downregulate Atoh1 expression in 293 cells is partially inhibited by TSA. Salermide
does not have any significant effect. Values are mean ± SEM, n = 4; [*] p < 0.05, [**] p <
0.002, Student's t-test. (n.s.) not significant. (B) The effect of TSA and Salermide on the
expression level of Atoh1, Hes5 and Hey2 in supporting cells FACS-purified from
p27/GFP transgenic cochlear cultures after 15h and 40h of treatment. The purified
supporting cells mostly comprise of HES5-expressing Deiters’ cells. Values are mean ±
SEM, n = 3.
A
B

82

The failure of Salermide or TSA alone to induce transdifferentiation of pillar cells
is not changed by blocking Notch or FGF signaling
Hey2 expression is regulated by both Notch and FGF signaling in the postnatal
cochlea (Doetzlhofer et al. 2009). Pillar cells do not transdifferentiate into hair cell-like
cells when only Notch singling or FGF signaling is blocked. However, in the absence of
both Notch and FGF signaling, they do transdifferentiate (Figure 35A), similar to the
effect of Salermide and TSA. This led us to hypothesize that Salermide and TSA can
affect these signaling pathways upstream of Hey2 (i.e. Salermide inhibits FGF, and TSA
inhibits Notch, or vice versa). To test this possibility, the Atoh1/GFP transgenic cochlear
explants were treated with DAPT or SU5402, inhibitors of Notch and FGF signaling,
respectively, in combination with TSA or Salermide. As shown in Figure 35B, addition of
Salermide or TSA alone with Notch inhibitor DAPT or FGF inhibitor SU5402, did not
affect pillar cell fate as evidenced by the unchanged Prox1 staining observed. Also, I
examined the expression of other known Notch and FGF effectors that are expressed in
the supporting cells. QPCR analysis of FACS-purified p75
+
supporting cells showed that
there is not a global effect of these inhibitors on these two pathways. For example,
Hes5, another effector of Notch signaling, is upregulated by Salermide and TSA. Of
FGF effectors, Etv4 is down-regulated, while Etv5 does not change or is upregulated.
FGFR3, expressed by pillar cells and Dieters’ cells, is not significantly altered. This
result suggests that the upregulation of Atoh1 in inner pillar cells in response to
Salermide and TSA is not due to changes in the abundance of Notch and FGF signaling
pathway elements.

83

B
A

84

Figure 35. Salermide or TSA alone cannot induce transdifferentiation of pillar
cells in the absence of Notch signaling. (A) Prox1
+
supporting cells in the pillar cell
region transdifferentiate only when both FGF and Notch signaling pathways are
blocked. P1 Atoh1/GFP transgenic cochlear explants were cultured for 72 h in the
presence or absence of DAPT (Notch inhibitor) and SU5402 (FGF inhibitor) and stained
with Prox1 antibody (red). White and yellow brackets show Deiters’ cells and pillar cells,
respectively. (B) P1 Atoh1/GFP transgenic cochlear explants were stained with Prox1
antibody (red) after being treated with DAPT, Salermide,TSA and Su5402 for 72h. (C)
qPCR result showing the expression of Notch and FGF signaling pathway components
in FACS-purified p75
+
supporting cells treated with DMSO, Salermide, TSA or both
Salermide and TSA for 24h. Values are mean ± SEM, n = 4; [*] p < 0.05, [**] p < 0.005,
Student's t-test. Scale bar = 50 µm.

HDAC expression analysis in the organ of Corti
Salermide is an inhibitor of both SIRT1 and SIRT2 (see chapter 1). In order to
determine which HDAC is the potential target of this inhibitor in inner pillar cells, the
expression level of Sirt1 and Sirt2 was analyzed in different subpopulations of
supporting cells and hair cells from the organ of Corti. QPCR analysis showed Sirt2 is
expressed at higher levels in hair cells, and pillar cells when compared to the whole
organ of Corti. Sirt1 did not show any relative enrichment (Figure 36A). Staining using
anti-SIRT2 antibody confirmed that SIRT2 is expressed in hair cells, as well as inner
C

85

pillar cells (Figure 36B), the cells that transdifferentiate to hair cell-like cells by TSA plus
Salermide treatment. This suggests that SIRT2 is probably the target of inhibition by
Salermide the expression level of classical HDACs (1-11) that could potentially be
inhibited by TSA was also examined. QPCR analysis suggested that HDAC1, HDAC4,
HDAC7 and HDAC10 are expressed at higher levels in p75
+
pillar cells, relative to other
cells in the organ of Corti (Figure 36C).

86

Figure 36. Expression analysis of Sirts and other HDACs in the sensory
epithelium. (A) Real-time quantitative PCR showing the relative expression level of
Sirt1 and Sirt2 in hair cell (HC), p75
+
supporting cell (p75 SC) and Lfng+ supporting
cells (Lfng SC) compared to the whole organ of Corti (OC) at P1. (B) Cochlear cross
section at P1 stained with anti Myo VII (green) and Sirt2 (red) antibody showing Sirt2
expression in inner hair cell (IHC), outer hair cells (OHC), inner pillar cell (IPC) and
phalangeal cells (PhC). (C) QPCR showing the expression level of HDAC1 to HDAC11
in p75
+
supporting cells compared to the whole organ of Corti at P1. Only HDAC1, 4, 7,
and 10 showed significant differential expression. Values are mean ± SEM, n = 3; [*] p <
0.005, Student's t-test. Scale bar is 10 µm in (B).
A
B
C

87

HEY2 and SIRT2 interact in vitro
If SIRT2 is the corepressor assisting HEY2 to repress Atoh1 expression, then it is
expected that HEY2 and SIRT2 interact. To test this hypothesis, proximity ligation assay
(PLA) (Soderberg et al. 2006) was performed. As a control to test the specificity of the
PLA assay, we compared the Hey2-Δbasic construct that does not have any repressive
activity on Atoh1 expression (Figure 28C). When PLA was performed with FLAG and
SIRT2 antibody in 293 cells that were transfected with wild type FLAG-Hey2 expression
plasmid, PLA signals were detected largely over the nuclei, indicating HEY2-SIRT2
interaction. There was little or no interaction in the cells receiving the Hey2-Δbasic
mutant expression plasmid, or if the FLAG antibody was omitted (technical control
testing the specificity of the assay). The number of HEY2:SIRT2 PLA signal deposits
were drastically reduced when the Hey2-Δbasic construct was transfected instead of the
wild type. The absence of PLA-detected interaction suggests that in the absence of
HEY2 DNA binding, SIRT2 and HEY2 do not interact.

Figure 37. SIRT2 interacts with HEY2 in vitro. Confocal images of 293 cells
transfected with WT or ∆basic FLAGHey2 expression plasmid and labled with
Sirt2:FLAG PLA (red) and DAPI (blue). Scale bar is 20 µm.

88

RNA-Seq. analysis of Salermide and TSA-treated p75
+
supporting cell
To get a better understanding of the gene expression changes in pillar cells in response
to these inhibitors, FACS-purified pillar cells using the p75 antibody from organ cultures
that had been treated with DMSO or Salermide/TSA together were analyzed using RNA
sequencing (RNA-seq). The Salermide/TSA-induced differential gene expression was
analyzed using the quantification software in Partekflow.
Changes in expression level of hair cell (HC) and supporting cell (SC) markers in FACS-
purified p75
+
SC in response to Salermide and TSA treatment for 18h (compared to DMSO
treatment, 2 replicates pooled).
Gene
symbol
Fold
change
replicate 1
Fold
change
replicate 2
Fold
change
replicate 3
average P value Marker of
Atoh1 2.64 1.13 2.5 2.1 ↑ 0.585 HC
Myo6* 2.2 1.74 1.54 1.82 ↑ 0.037 HC
Myo7a 0.99 1.11 0.79 0.96 ↑ 0.580 HC
Hes6* 1.5 1.47 1.88 1.61 ↑ 0.042 HC
Prestin* 5.6 3.17 3.52 4.10 ↑ 0.0497 HC
Hey2 1.3 0.96 1.72 1.46 ↑ 0.270 PC
p75* 0.22 0.18 0.23 0.21 ↓ 0.0004 PC
Etv4 0.64 0.54 0.34 0.51 ↓ 0.119 PC
FGFR3 1.88 1.23 3.89 2.33 ↑ 0.237 SC
HeyL* 0.36 0.38 0.25 0.33 ↓ 0.002 SC
Hes5 2.84 1.05 3.54 2.48 ↑ 0.179 SC
Sox2* 0.66 0.49 0.65 0.6 ↓ 0.009 SC

Table 1. Fold difference in expression level of hair cell and supporting cell
markers between DMSO- and Salermide/TSA-treated p75
+
supporting cells
assessed by RNA-seq. Hair cell markers are upregulated while supporting cell markers

89

are downregulated. HC hair cell, SC supporting cell, PC pillar cell. [*] p < 0.05,
Student's t-test.

As shown in Table 1, the hair cell markers such as Atoh1, Myo6, and Prestin are
upregulated in response to Salermide and TSA. On the other hand, supporting cell
markers such as Sox2, Etv4 and p75 are downregulated. This is consistent with the
observed transdifferentiation of inner pillar cells to hair cell-like cells (Figure 29) and the
gene expression analysis by qPCR (Figure 35).
Discussion
The results presented in this chapter indicate that HEY2- and HES5-mediated
repression of Atoh1 (discussed in chapter 2) have similarities and differences. The
similarities are:
 HEY2 binds to the Atoh1 promoter region.
 HEY2 represses Atoh1 transcription through the promoter region.
 HEY2 inhibitory effect is dependent on the C3 site in the Atoh1 promoter.
The differences are:
 Not all the C-sites in Atoh1 promoter are required for HEY2 repression.
 HEY2 has weak inhibitory effect on the Atoh1 enhancer activity that is
independent of the conserved C-site or N-box present in the enhancer.
 HEY2 repressive activity is not dependent on its C-terminal motif.
 HEY2-mediated repression of Atoh1 is dependent on TSA-sensitive and
insensitive HDACs.

90

HEY2 binding to the C-site in the Atoh1 promoter region is in agreement with the
previous reports. A recent study using ChIP-seq showed that HEY factors bind mostly to
the proximal promoter region of their target genes and identified CACGTG, a class B-
site and CACGCG, a class C-site as their preferential binding motifs (Heisig et al. 2012).
Binding to these sites was also reported by other groups (Nakagawa et al. 2000; Iso et
al. 2001). Interestingly, not all the bound regions by HEY factors, including Hey2’s
promoter itself, have the identified motifs, suggesting that HEY’s sequence-specific
binding might be affected by binding partners (Nakagawa et al. 2000; Heisig et al.
2012).
The HEY2 inhibitory effect on the Atoh1 enhancer could be explained by passive,
DNA-independent, mechanisms of repression. It has been reported that HEY2 binds the
aryl-hydrocarbon receptor nuclear translocator (ARNT) and prevents ARN-dependent
transcription of the VEGF promoter (Chin et al. 2000). It has also been shown that
HEY1 can inhibit MyoD-dependent transcription of the myogenin promoter by
preventing the formation of the MyoD-E47 heterodimer (Sun et al. 2001). The results of
EMSA with Atoh1 enhancer probes suggests that HEY2 might interfere with
autoregulation (discussed in chapter 4) by preventing ATOH1 binding to DNA and
therefore inhibit the enhancer activity (Figure 38).
HEY proteins recruit different HDACs to repress target genes. HEY1 can recruit
HDAC1 indirectly through the NcoR and Sin3a complex (Iso et al. 2001; Gould et al.
2009) and HEY2 has been shown to interact with Sirt1 (Takata and Ishikawa 2003) and
HDAC2 (Weber et al. 2015). Our results further suggest that unlike HES5, HEY2-
mediated repression of Atoh1 occurs by recruitment of different classes of HDACs;

91

inhibition of NAD
+
-independent HDACs by TSA, and the NAD
+
-dependent HDAC SIRT2
by Salermide, leads to upregulation of Atoh1 and transdifferentiation of inner pillar cells
(Figure 29 and Figure 31).

Figure 38. HEY2 can prevent binding of ATOH1 to the enhancer E-box.
Electrophoretic mobility shift assay with IRDye800-labeled 30 bp-long oligonucleotide
containing the enhancer E-box and reticulocyte lysate alone (lane 1 and 11), ATOH1
(lane 2-5), TCF3 (a potential heterodimerzation partner of ATOH1, lane 6), ATOH1 plus
TCF3 (lanes 7-10), HEY2 (lane 12), HEY2 plus ATOH1 (lane 13) or HEY2, ATOH1 and
TCF3 (lane 14) is shown. The specificity of ATOH1 and ATOH1/TCF3 dimer binding to
the probe was determined by adding 50 fold molar excess of unlabeled wild type
oligonucleotide (lane 3 and 8), unlabeled E-box mutant oligonucleotide (lane 4 and 9),
or unlabeled N-box mutant oligonucleotide (lane 5 and 10). The arrows show ATOH1-
or ATOH1/TCF3-DNA complex. The two gels shown were run at the same time in the
same electrophoresis apparatus.

How does inhibition of HDACs lead to upregulation of Atoh1 in inner pillar cells? In
addition to deacetylation of histones and chromatin-associated repression of Atoh1
transcription (see chapter 1), SIRT2 could affect HEY2-mediated repression of Atoh1 by
other mechanisms such as alteration of subcellular localization and/or protein stability of
HEY2.

92

Shuttling of transcription factors between the nucleus and cytoplasm is a
mechanism of transcriptional regulation. The acetylation of transcription factors can
affect the accessibility of nuclear localizations signals to nuclear import and export
proteins (Dai and Faller 2008). Several studies suggest that SIRT1 and SIRT2 can
regulate the localization of transcription factors. SIRT2, for example is able to
deacetylate FOXO1 and FOXO3a and promote their nuclear localization (Jing et al.
2007; Wang et al. 2007), so it is conceivable that Atoh1 upregulation in response to
Salermide/TSA may be due to a decrease in the availability of HEY2 in the nucleus.
A question for future study will be whether inhibition of SIRT2 leads to upregulation
of Atoh1 by decreasing the stability of HEY2 protein or other factors, yet to be identified,
that maintain the pillar cell fate by Atoh1 inhibition. Acetylation of transcription factors
can promote their degradation by ubiquitination (Caron et al. 2005) and it has been
reported that SIRT2 can stabilize MYC oncoproteins by reducing the proteasome-
mediated degradation of MYCs (Liu et al. 2013).
Altogether, the data presented in this chapter suggest that HES and HEY
repressors, despite having similar domains and motifs, rely on distinct repression
mechanisms.

93

Chapter 4: Atoh1 autoregulation
Introduction
In the previous chapters, HES5- and HEY2-mediated repression of Atoh1 was
discussed. One main question that remains to be addressed is how Atoh1 is up-
regulated in the absence of HES/HEY repression. The experimental results presented in
this chapter suggest that derepression of Atoh1 promoter sites (chapters 2 and 3)
eventually leads to activation of the Atoh1 autoregulatory enhancer that allows
sustained and high expression levels of Atoh1. The Atoh1 enhancer sequence was
identified by Jane Johnson’s laboratory more than fifteen years ago (Helms et al. 2000).
In an extensive study, Helms, et al., used LacZ reporter constructs with different length
sequences upstream and downstream of Atoh1 coding region in transient transgenic
mouse assays to more precisely determine the sequences regulating the expression of
Atoh1. Analysis of transgenic embryos at E10.5 showed that a 1.7 Kb enhancer
sequence, 3.4 Kb downstream of the Atoh1gene, was enough to direct specific
expression of Atoh1 in the neural tube. Further analysis of a stable transgenic line
carrying the LacZ sequence under the control of the1.7 Kb enhancer, showed Atoh1-
specific expression of LacZ in the developing inner ear and cerebellum (Helms et al.
2000). These observations led to the conclusion that the 1.7 Kb enhancer is sufficient to
recapitulate Atoh1 specific expression in different expression domains of Atoh1.
Comparison of the enhancer sequence between mouse and human showed two highly
conserved regions, termed enhancer A and enhancer B. Within enhancer B, a
conserved E-box site (CAGCTG) was identified that when mutated, drastically reduced

94

the LacZ transgene expression in the neural tube, suggesting the E-box is required for
the activity of the Atoh1 enhancer. A similar requirement for the activity of the enhancer
in the chick neural tube was also reported (Ebert et al. 2003). The class A, E-boxes are
the binding site for bHLH transcriptional activators like ATOH1 (Akazawa et al. 1995). In
Vitro assays showed that ATOH1 itself binds to the enhancer, and that the binding is
dependent on an intact E-box (Helms et al. 2000). Further experiments revealed
ATOH1 itself is required for activity of the enhancer in the neural tube as the expression
of Atoh1 enhancer/lacZ transgene could not be detected in the Atoh1-null background
(Atoh1
-/-
, Atoh1/LacZ
-/+
) neural tube (Helms et al. 2000). These lines of evidence led to
the conclusion that in the nervous system ATOH1 autoregulates through the enhancer
E-box, and that ATOH1 itself is necessary to maintain the expression of Atoh1. Whether
the activity of the Atoh1 enhancer in the organ of Corti requires similar autoregulation or
an intact E-box, is not clear which is the subject of study in this chapter.
Results
ATOH1 binds the E-box site in the enhancer
First, I tested whether ATOH 1 can bind to the E-box present in the enhancer using
the electrophoretic mobility shift assay (EMSA). ATOH1 and a potential
heterodimerzation partner TCF3 (from the E family of proteins also known as E47 or
E2A) (Akazawa et al. 1995) were in vitro transcribed and translated using a rabbit
reticulocyte lysate (Figure 39). The EMSA with IRDye-labeled oligonucleotide identical
to the E-box, and its flanking sequence in the enhancer showed that ATOH1/TCF3
heterodimers are able to bind this sequence (lane 8). ATOH1 alone also binds the E
box-containing oligonucleotide, but the binding is weaker (lane 2). TCF3 alone did not

95

show binding (lane 6). The specificity of ATOH1 binding was tested by competition with
wild type unlabeled oligonucleotide in 12.5- to 100-fold molar excess, which inhibited
ATOH1 and ATOH1/TCF3 binding to the labeled probe (lanes 3 and 9-11). To examine
the requirement for an intact E-box site for ATOH1 binding, competitor oligonucleotides
with two point mutations in the E-box sequence, in 12.5- to 100-fold molar excess was
added, and was not able to compete (lanes 4 and 12-14). However, the probe with point
mutations in the overlapping N-box had no effect on ATOH1 or ATOH1/TCF3 binding as
it was able to compete with the wild type probe (lanes 15-17). This result indicates that
Atoh1 binding to its ehancer is dependent on the E-box, as was reported previously
(Helms et al. 2000).

Figure 39. ATOH1/TCF3 heterodimers bind the E-box in the Atoh1 enhancer.
Electrophoretic mobility shift assay with reticulocyte lysate alone (lane 1 and 7), with
ATOH1 (lane 2-5), with TCF3 (lane 6) or with ATOH1 plus TCF3 (lanes 8-17) is shown.
IRDye800-labeled oligonucleotide containing the enhancer E-box and its flanking
sequences were mixed with in vitro transcribed and translated ATOH1 and/or TCF3,
incubated at 37°C for 30 minutes and run on 6% polyacrylamide gel. After the run was
completed, the gels were scanned using the LI-COR Odyssey scanner.

96

The Atoh1 enhancer acts as an auto-regulatory element in response to ectopic
expression of Atoh1
If ATOH1 binds to its enhancer sequence, we expected that Atoh1 would be able to
induce the enhancer-mediated expression. To test this, we took advantage of the Atoh1
enhancer/β-globin promoter/GFP transgenic line in which GFP is driven by the
enhancer and expressed in the hair cells of the inner ear, similar to the expression of
endogenous Atoh1 as demonstrated by immunohistochemistry with Atoh1 antibody
(Chen et al. 2002; Lumpkin et al. 2003).
A CMV-RFP expression plasmid with or with an Atoh1 expression plasmid was
electroporated into GER (greater epithelial ridge) of Atoh1 enhancer/β-globin
promoter/GFP transgenic mouse at P1. Over-expression of Atoh1 in this area, which
normally does not express Atoh1, induced the enhancer activity in the transgene and
resulted in the appearance of GFP-positive cells in the GER (Figure 40). This suggests
the existence of an Atoh1-positive, autoregulatory feedback loop in the organ of Corti.

Figure 40. Atoh1
autoregulates. Atoh1
overexpression in
GER of Atoh1
enhancer/β-globin
promoter/GFP
transgenic line
induces the enhancer
activity and leads to
expression of GFP,
suggesting that Atoh1
autoregulates.

97

Mutation of the E-box reduces the induction of enhancer activity in the GER
In order to test that Atoh1 autoregulation (Figure 40) is dependent on the enhancer
E-box, we made use of two reporter constructs that were introduced in the previous
chapters: Atoh1 enhancer/β-globin promoter/mCherry and Atoh1 enhancer/β-globin
promoter/GFP (Figure 41). These two reporters were electroporated into the wild type
cochlear explants at P1, with or without an Atoh1 expression plasmid (CMV-Atoh1
similar to Figure 40) and examined after 2 days in culture. In the absence of Atoh1
expression, the reporters are not expressed in the cochlear cultures (Figure 42A-C). In
the presence of Atoh1 expression, GFP
+
and mCherry
+
cells appear (Figure 42A’-C”).
This result is consistent with Figure 40, and confirms that ATOH1 can induce its own
enhancer activity.

Figure 41. The reporter constructs used in electroporation of postnatal cochlear
explants. (A) The schematic of Atoh1 enhancer/β-globin promoter/GFP reporter, Atoh1
enhancer/β-globin promoter/mCherry reporter (both wild type) and Atoh1 expression
plasmid that were used in Figure 42A-C”. (B) The schematic of wild type Atoh1
enhancer/β-globin promoter/GFP reporter, Atoh1 enhancer/β-globin promoter/mCherry
reporter with mutated E-box, and Atoh1 expression plasmid that were used in Figure
42Error! Reference source not found.D-F”.

A B

98

99

Figure 42. Ectopic expression of Atoh1 induces the enhancer activity in GER if
the E-box is intact. (A-C”) Wild type Atoh1 enhancer/β-globin promoter/GFP and
Atoh1 enhancer/β-globin promoter/mCherry reporters were electroporated into cochlear
explants at P1 without (A-C) and with (A’-C’) Atoh1 expression plasmid.
Overexpression of Atoh1 induces the enhancer activity in both reporters and leads to
appearance of GFP
+
and mCherry
+
cells. (A”-C”) higher magnification of (A’-C’). Wild
type Atoh1 enhancer/β-globin promoter/GFP and E-box mutated Atoh1 enhancer/β-
globin promoter/mCherry reporters were electroporated into cochlear explants at P1
without (D-F) and with (D’-F’) Atoh1 expression plasmid. Overexpression of Atoh1 only
induces the enhancer activity in the reporter with wild type E-box (GFP) and leads to
appearance of GFP
+
cells. The reporter with mutated E-box (mCherry) is not induced by
ectopic ATOH1. (D”-E”) higher magnification of (D’-E’). (G) quantification of GFP
positive cells that co-express mCherry as shown in A’-C’ and D’-F’. Values are
percentage of mean ± SEM, n = 3; [*] p < 0.005, Student's t-test. Scale bar is 100 µm.

In contrast, Atoh1 ectopic expression failed to induce the enhancer activity when the
E-box in Atoh1 enhancer/β-globin promoter/mCherry reporter was mutated (Figure 42D-
G), and as a result over-expression of Atoh1 only led to the appearance of GFP
+
cells.
This result shows that not only Atoh1 can autoregulate through the enhancer sequence
in the non-sensory epithelium of the cochlea, but that this autoregulation is dependent
on ATOH1 binding to the enhancer E-box.
G
*

100

Atoh1 enhancer is not active in the organ of Corti in the Atoh1 knock out
To address the requirement for ATOH1 itself in the activity of Atoh1 enhancer
during the development, the expression of the Atoh1 enhancer/β-globin promoter/GFP
transgene in the Atoh1 mutant background was examined. Mice hemizygous for the
Atoh1 enhancer/β-globin promoter/GFP transgene and heterozygous for the Atoh1
mutant allele were crossed with Atoh1 heterozygous mutant mice. Embryos from this
cross were examined by fluorescent microscopy for GFP expression in the cochlea at
E14.5, when expression of Atoh1 in the inner ear has started. Animals hemizygous for
the Atoh1 enhancer/β-globin promoter/GFP transgene and homozygous for the Atoh1
mutant allele (Atoh1
-/-
, Atoh1GFP
+/-
) were compared to Atoh1
+/+
, Atoh1GFP
+/-

littermates. In the Atoh1 knock-out mouse organ of Corti, the Atoh1 enhancer failed to
induce the expression of GFP transgene, suggesting that Atoh1 enhancer is not active
in the absence of Atoh1 expression, and indicating that an autoregulatory loop is
required for the Atoh1 enhancer activity (Figure 43).

101

Figure 43. The Atoh1 enhancer is not active in the Atoh1
-/-
mouse cochlea at
E14.5. The expression of Atoh1 enhancer/β-globin promoter/GFP transgene in the
Atoh1 wild type (A) and Atoh1 mutant background (B) in a whole mount of mouse
cochlea. The expression of GFP (bracket) is not located in the Atoh1-deficient organ of
Corti, but rather it appears to be misexpressed in some other cell types (black
arrowheads) at embryonic stage 14.5.

Discussion
The results presented in this chapter confirm the previous reports in mouse and
chicken neural tube suggesting the autoregulatory function of ATOH1 through a
conserved E-box motif in the enhancer element B. The autoregulation of Atoh1 through
an E-box motif is not surprising since atonal, its Drosophila homolog, also autoregulates
in multiple expression domains (Sun et al. 1998), and the binding of other bHLH

102

activators to E-box motifs in their target genes has been shown in numerous studies.
For example, the bHLH myogenic factor MyoD and the proneural gene Mash1 (Ascl1)
bind to the E-box in the enhancer sequence of the MCK (muscle creatine kinase) gene
and activate its expression (Lassar et al. 1989; Johnson et al. 1992). Interestingly,
MyoD also autoregulates. In a study by Thayer, et al., it was shown that overexpression
of MyoD in 10T1/2 cells increases the expression of endogenous MyoD1 and converts
them to myogenic cells. This effect was not observed when MyoD1 was overexpressed
in fibroblast and adipoblast cell lines (Thayer et al. 1989). The binding of bHLH
activators to E-boxes have even been reported in plants. In Arabidopsis, AtbHLH112 is
a salt-induced transcriptional activator that binds to the E-box motifs of stress-
responsive genes to increase the stress tolerance (Liu et al. 2015).
Similar binding site preference by bHLH factors raises the question as how these
factors activate specific sets of target genes. One possibility is that the interaction of
bHLH factors with other cell-type specific (co)factors provides the specificity needed for
gene activation (Groves et al. 2013). This idea is supported by the observation that the
bHLH factor Neurog1, needed for the differentiation of sensory neurons in the inner ear,
is not able to substitute for Atoh1. In mice in which the Atoh1 coding region has been
replaced with Neurog1, the organ of Corti is severely abnormal, (although the
phenotype is not as severe as in Atoh1-null mice), indicating that Neurog1 can only
partially rescue hair cell differentiation, possibly by activating a subset of Atoh1 target
genes (Jahan et al. 2012).
Other than Atoh1 itself, other direct targets of Atoh1 in hair cells are not known. The
small number of hair cells present in the cochlea makes the application of techniques

103

such as transcription factor chromatin immunoprecipitation (ChIP) unfeasible. However,
a recent study used ChIP and deep sequencing to identify the Atoh1 targetome in
granule cell precursors (GCPs) in the cerebellum that express Atoh1 and constitute the
majority of cells in this organ (Klisch et al. 2011). The analysis of Atoh1 ChIP-seq data
in GCPs identified a consensus 10-nucleotide long Atoh1 binding motif called Atoh1 E-
box associated motif (AtEAM) that is present in the regulatory regions of more than half
of Atoh1 target genes in cerebellum. Whether Atoh1 targets in hair cell contain such a
motif is not clear. Many of the genes bound by Atoh1 in GCPs are involved in the
regulation of proliferation and cell migration that probably are not expressed in the post-
mitotic hair cells, so a difference in the Atoh1 targetome between the hair cell and GCPs
is expected.
Other known activators of the Atoh1 enhancer are the factor SIX1 and its binding
partner EYA1, and TCF/LEF downstream of the Wnt pathway. Given that these factors
can induce Atoh1 by binding to the Atoh1 enhancer, as has been reported (Ahmed et al.
2012; Jacques et al. 2012) and that they are expressed in the prosensory domain
before hair cell differentiation (discussed in chapter 1), it is interesting that the Atoh1
enhancer is not active in the Atoh1-null background (Figure 43 and (Helms et al. 2000;
Raft et al. 2007)). One possible explanation is that the SIX1/EYA1 complex or TCF/LEF
require other factor(s) to activate Atoh1 enhancer that are not present in the progenitors
of hair cells.

104

Chapter 5: Conclusions
Although mammalian supporting cells have similarities to those in birds, such as
sharing a common progenitor cell or being in close contact with hair cells (Stone and
Cotanche 2007), they show very limited spontaneous regenerative capacity through
direct transdifferentiation or mitotic regeneration (Cox et al. 2014). One likely
mechanism to limit the plasticity of supporting cells is by the Notch signaling pathway.
This is evident in the neonatal cochlea by the robust increase in hair cell numbers
through transdifferentiation of supporting cells when Notch signaling is blocked
(Yamamoto et al. 2006; Doetzlhofer et al. 2009). A few reports also suggest that
modulation of Notch signaling after damage in mature cochlea can stimulate hair cell
regeneration by inducing the expression of Atoh1 (Hori et al. 2007; Korrapati et al. 2013;
Mizutari et al. 2013). This thesis project helps to explain several important aspects of
Notch-dependent regulation of Atoh1 that underlie the differentiation of sensory
epithelia, as well as the potential of supporting cells to undergo direct
transdifferentiation. These results are discussed below and summarized in a model of
Atoh1 regulation in Figure 45.
The same promoter element in Atoh1 is inhibited by either HEY2 or
HES5 to maintain the fate of neighboring supporting cells
Our results show that HES5 and HEY2 repress Atoh1 expression by directly binding
to the promoter region of Atoh1 and that this binding is dependent on highly conserved
C-sites located within the proximal 226 bp of the transcription start site. One main
difference between HES5 and HEY2 in this regard is that mutation in all four of the C-

105

sites is required to abolish the inhibitory effect of ectopically expressed HES5 on
promoter function (Figure 14), suggesting a synergistic role; however, HEY2’s
repressive activity, which is not as complete as Hes5’s, is only affected by mutation of
the C3 site (Figure 25C). A Hes/Hey repressor sequence present in regulatory elements
of Atonal homologues throughout evolution indicates that this repressor function is
extremely ancient (Rebeiz et al. 2005). There are fewer Hes/Hey motifs associated with
proneural genes in non-vertebrate lineages, just as there are fewer homologues of
Hes/Hey genes (Rebeiz et al. 2005), suggesting that duplication of the binding sites in
the Atoh1 promoter may have paralleled the expansion of the Hes/Hey family of genes,
leading to specialization rather than redundancy in mammals. Although further
experiments are required to test the DNA-binding activity of other HES/HEY factors
expressed in supporting cells, the combinatorial pattern of Hes/Hey expression in the
organ of Corti (Doetzlhofer et al. 2009), and the different patterning phenotypes of the
Hes/Hey mutant mice suggest a role for these sites in fine-tuning the expression of
Atoh1 and therefore, sculpting organ of Corti structure.
Derepression of the Atoh1 locus is sufficient to stimulate Atoh1
induction
Blocking Notch activity leads to the rapid upregulation of Atoh1 in most supporting
cell subtypes in P1 mouse organ of Corti, correlated with the loss of the Notch effectors
HES/HEY, and leading to direct transdifferentiation into hair cells (Figure 10 and
(Yamamoto et al. 2006; Doetzlhofer et al. 2009)). Interestingly, we show that blocking
de novo protein synthesis with cycloheximide is sufficient to elicit a stimulatory response

106

similar to that of blocking Notch activity with DAPT, suggesting that rapid degradation of
HES/HEY factors and the subsequent derepression of Atoh1 is responsible for initiating
its transcriptional stimulation (Figure 11 and Figure 12). Similar to the loss of HES5- and
CHX-mediated derepression of Atoh1, HES1 protein has previously been shown to
have a short half-life (Hirata et al. 2002; Kobayashi et al. 2009), and its loss of
expression leads to selective derepression and transcriptional stimulation of target
genes (Riccio et al. 2008). The short half-life of HES1 in neural progenitors is critical for
its autonomous oscillatory expression, regulated by a negative feedback loop that
maintains the stem cell state of neural progenitors (Hirata et al. 2002; Shimojo et al.
2008).
An interesting question that remains to be addressed is if in addition to the
derepression of the Atoh1 promoter in transdifferentiating supporting cells, there are
specific transcription factor(s) required for upregulation of Atoh1. Stimulation of Atoh1
transcription after blocking protein synthesis indicates that these factors, if critical, are
already present in supporting cells. Similar to the process of hair cell differentiation in
the murine cochlea, the Drosophila sensory organ precursor cell (SOP) develops within
a proneural clusters (PNCs) through increased expression levels of proneural genes
such as Achaete (Ac) and Scute (Sc) and Notch-mediated lateral inhibition (Barad et al.
2011). A recent study has shown that SOP development can happen in the absence of
the proneural activity if the function of Extramacrochaetae (Emc), the only member of
the Id proteins in Drosophila, is abolished. In this case, the activity of Daughterless (Da)
, the ubiquitously expressed bHLH factor ortholog of Tcf3, seems to be enough for the
development of SOPs (Troost et al. 2015). So, it is conceivable that to induce

107

transdifferentiation of supporting cells, all that is needed is the help of general
transcriptional machinery in the absence of active Notch signaling.
A similar transcriptional derepression does not occur from the enhancer/β-globin
promoter/GFP transgene (Figure 12D), indicating that activation is mediated through the
226 bp Atoh1 promoter and not through the enhancer. Binding motifs for several
transcription factors are present in the promoter and may serve as activators in this
context (Factor X in Figure 45).

Figure 44. Schematic representation of the Atoh1 promoter sequence and the
predicted transcription factor binding sites. TFC3,TCF4 and LEF-1 downstream of
the Wnt signaling pathway, and SMAD downstream of BMP and TGF signaling pathway
have binding sites in the Atoh1 proximal promoter region. The transcription start site is
shown with an arrow. The TRANSFAC+ PROTEOME tool in Biobase Digital Databases
was used to predict the binding sites.
HEY2 and HES5 rely on different cofactors to inhibit Atoh1
How does HES5 or HEY2 binding to the Atoh1 promoter bring about the silencing of
transcription? Notch signaling has been shown to prevent the recruitment of the histone
acetyltransferase p300 and the acetylation of histones (Krishnamoorthy et al. 2015),
and transcriptional repression by HES/HEY proteins has been linked to interaction with
HDACs (Iso et al. 2001; Takata and Ishikawa 2003; Weber et al. 2015). Our results

108

indicate that HES5-dependent repression of the Atoh1 promoter is mediated by the
recruitment of GRG/TLE corepressors, and the presence of NAD
+
-independent HDAC
activity in a Notch-dependent manner (Figure 18 and Figure 19). HEY2 repressive
activity on the other hand, relies on both NAD
+
-dependent (SIRT) and -independent
histone deacetylases (classical HDACs) to inhibit Atoh1 in pillar cells. The observation
that epigenetic repression of Atoh1 is context dependent, such that neighboring
supporting cells (pillar vs. Deiters’ cells) rely on different repressive mechanisms,
suggest that any regenerative approach based on a single epigenetic modulator of
Atoh1 expression may not lead to the effective or robust induction of Atoh1 needed for
hair cell formation.
Atoh1 silencing in supporting cells is reinforced by “activator
insufficiency”
The active repression by HES/HEY factors through interaction with the Atoh1
promoter raises the question of why the Atoh1 enhancer transgene containing only the
β-globin minimal promoter that lacks HES/HEY binding motifs is silent in supporting
cells. We hypothesize that this silencing is an example of “activator insufficiency”
(Barolo and Posakony 2002), in which active repression maintains ATOH1 protein
below a threshold needed to stimulate the autoregulatory enhancer activity (discussed
in chapter 4). In this context, the Atoh1 enhancer acts in a switch-like manner, so that
once the threshold is crossed, Atoh1 is robustly stimulated. Activator insufficiency is
demonstrated by the failure of the Atoh1 enhancer alone (Atoh1 enhancer/β-globin
promoter/GFP) to initiate transcriptional upregulation when protein synthesis is blocked,

109

while the endogenous Atoh1 gene is upregulated (Figure 12). The autoregulatory
enhancer, however, might be important to maintain the expression of Atoh1 at high
levels, since in Atoh1 knock-out mice carrying the Atoh1 enhancer/β-globin
promoter/GFP transgene, GFP is not expressed (Figure 43 and (Raft et al. 2007)).
Atoh1 is upregulated in prosensory progenitors immediately prior to
repression in nascent supporting cells
Our results using the mutant Atoh1 promoter transgenic mouse, indicate that the
HES/HEY motifs are needed to repress Atoh1 expression in prosensory progenitors
destined to become supporting cells immediately after Atoh1 upregulation. This is
consistent with recent evidence from a number of labs that suggests that rather than
being selectively upregulated in nascent hair cells, Atoh1 upregulation occurs in
progenitors of both hair cells and supporting cells. First, loss of Notch inhibition through
the conditional mutation of Rbpj leads to the broader expression of Atoh1 within the
prosensory domain (Basch et al. 2011; Yamamoto et al. 2011). Second, lineage tracing
experiments indicated that significant numbers of supporting cells, derived from Atoh1-
cre
pr
expressing prosensory cells (Driver et al. 2013). Third, an Atoh1-GFP knock-in
construct showed low levels of GFP in multiple rows of cells at the leading edge of the
wave of hair cell differentiation, likely nascent supporting cells, prior to complete Atoh1
silencing (Cai et al. 2013). The transgene data support these observations by showing
that the mutant promoter allows expression in both nascent hair cells and nascent
supporting cells due to ineffective Notch-mediated repression during the early Atoh1

110

induction, and prior to silencing through active repression of the endogenous Atoh1
gene and consequent insufficiency of activator (Figure 21).
Our data also suggest that the reported low level of Atoh1 transcription in the
prosensory domain prior to E13.5 (Lanford et al. 2000; Woods et al. 2004), before the
frank upregulation of Atoh1 in the nascent hair cells (Chen et al. 2002; Lumpkin et al.
2003), does not rise to the threshold needed to trigger enhancer autoregulation
(activator insufficiency, Figure 45A, B). Of course, this begs the question of what is
driving the rise in Atoh1 levels between E13.5 and E14.5 in the base of the prosensory
domain. This could occur either through a different, specific derepression signal, such
as the downregulation of HEY factors as recently suggested (Benito-Gonzalez and
Doetzlhofer 2014), or by the stimulation of unknown transcriptional inductive event(s)
that occur at this time, acting through other motifs in the 226 bp promoter (Factor X,
Figure 45).
Together these data indicate that Atoh1 levels transiently rise in a basal-to-apical
gradient within the prosensory domain in the progenitors of both hair cells and
supporting cells, and that active repression through the Atoh1 promoter is required to
inhibit this process of transcriptional activation in nascent supporting cells. Our results
argue for a system by which the HES/HEY proteins, expressed in a complex
combinatorial pattern within the organ of Corti, interact with multiple binding motifs in the
Atoh1 promoter and recruit different cofactors to sculpt the precise hair cell and
supporting cell mosaic required for function.

111

Figure 45. A model of Atoh1 regulation during organ of Corti development. (A)
Between E13.5 and 14.5, Atoh1 is upregulated in the cochlear prosensory domain that
will become the organ of Corti. We hypothesize that prior to upregulation, Atoh1 levels
are kept low through direct repression by HEY transcription factors and activator
insufficiency. TF-X represents a hypothetical basal transcriptional activity needed to
prime Atoh1 autoregulatory function. (B) Initial upregulation of Atoh1 could occur either
through promoter-mediated active stimulation and/or derepression across the
prosensory domain. (C) Once the Atoh1 autoregulatory threshold is achieved in
selected nascent hair cells, Notch-mediated active repression is triggered in the
surrounding prosensory cells to stimulate the silencing of Atoh1 through HES/HEY
interaction, leading to the onset of supporting cell differentiation, and the patterning of
the cellular mosaic of the organ of Corti. At P1, cell fate is maintained by continuing
Notch-mediated repression through recruitment of Grg/HDACs. Loss of Notch activity
leads to derepression of the Atoh1 promoter, increased Atoh1 expression through
autoregulation, and transdifferentiation of supporting cells to a hair cell-like state.

112

Ongoing challenges to regeneration of hair cells
While the inhibition of Notch signaling in the embryonic and neonatal cochlear
cultures leads to robust formation of supernumerary hair cells, the cochlea loses
competence to respond to the blockage of Notch signaling with age (unpublished data,
Segil Lab and (Hori et al. 2007)). The reason for this limited responsiveness is not
clearly understood. A recent study suggests that it might be due to the gradual
downregulation of Notch signaling pathway in the organ of Corti (Maass et al. 2015). If
that is the case, it would be interesting to explore the possible role of other signaling
pathways, such as FGF and SHH (sonic hedgehog), in the maintenance of hair and
supporting cell fate in the more mature cochlea and examine the effect of their inhibition
in hopes of inducing hair cell regeneration.
Although many studies have focused on the regulation and expression of Atoh1 in
the cochlea as a therapeutic approach for hair cell regeneration, recent studies suggest
that there is an age-dependent limit on the ability of Atoh1 to induce hair cells.
Expression of Atoh1 in supporting cells or non-sensory cochlear epithelium of mice can
lead to formation of new hair cells; however, this potential declines rapidly and fewer
and fewer hair cells are induced as the mice age (Kelly et al. 2012; Liu et al. 2012a).
One possible explanation for the failure of mature cochlea in mammals to respond to
the activation of Atoh1, is that Atoh1 requires other (co)factors, such as E12 or E47, to
activate its target genes which are downregulated in the ear with age. Therefore,
expression of Atoh1 alone in mature supporting cells will not be enough to induce the
Atoh1 targets and promote hair cell differentiation. This suggests that a complete

113

understanding of Atoh1 function in the context of inner ear is essential for any Atoh1-
based strategy to restore auditory function.

114

Chapter 6: Materials and Methods
Experimental animals and generation of transgenic mouse line
The experimental procedures were approved by the Animal Care and Use
Committees of University of Southern California and the House Research Institute.
P27/GFP BAC transgenic (White et al. 2006), Atoh1 enhancer/ -globin promoter/GFP
transgenic (Lumpkin et al. 2003), and Atoh1-GFP fusion knock-in (Rose et al. 2009) and
Lfng/GFP BAC transgenic generated by GENSAT project (Korrapati et al. 2013) mice
were described previously. For generating Atoh1-reporter double transgenic line, Atoh1
enhancer/Atoh1 promoter
wt
/tdTomato and Atoh1 enhancer/Atoh1 promoter
mut
/GFP
plasmids were linearized, and founder animals were generated by pronuclear
coinjection (UCI Transgenic mouse facility; FVB background). Positive founders were
identified by PCR genotyping using tail DNA (see Supplemental Material). Postnatal
transgenic mice were identified by direct observation of GFP or tdTomato fluorescence
in the cerebellum. For timed mating, animals were put together in the evening and the
next morning designated as day 0.5 (E0.5).
DNA extraction and PCR genotyping:
Tail clips were digested in 150 μL of Zandy buffer and Proteinase K at 65⁰C for 4
hours after which 200 μL of 3M Potassium Acetate and 200 μL of phenol-chloroform-
isoamyl alcohol was added and samples were centrifuged at 13,200 rpm for 20 min at
4⁰C. The aqueous layer from each sample was transferred to clean new tube and DNA
was precipitated by 1 ml ethanol (Koptec) 100% and centrifugation at 13,200 rpm for 20

115

min at 4⁰C. The DNA pellet was washed once with 70% ethanol, dried and resolved in
30 μL dH2O. The following primes were used for genotyping; GFP-F:
GTGAAGTTCGAGGGCGACAC, GFP-R: CGGACTGGGTGCTCAGGTAG, tdTomato-F:
GTGACCGTGACCCAGGACTC, tdTomato-R: TGACGGCCATGTTGTTGTCCTC.
Cell culture
293 cells were cultured in DMEM (Invitrogen) supplemented with 10% fetal bovine
serum (GIBCO), 2 mM L-glutamine, and 10 mM HEPES maintained in a 5% CO2
incubator at 37°C. Mouse embryonic stem cell line 46C, a Sox1-GFP knock-in line, was
a generous gift from Qi-Long Ying (University of Southern California, USA). Embryonic
stem cells were cultured and differentiated to neural progenitor cells (NPCs) as reported
(Ying and Smith 2003). In brief, 46C ES cells were cultured on gelatin-coated tissue
culture plates in GMEM (Invitrogen) supplemented with 15% Embryonic Stem (ES)
Screened FBS (HyClone), 1 mM sodium pyruvate (Invitrogen), non-essential amino acid
(Invitrogen), 0.1% 2-Mercaptoethanol (Invitrogen) and 10 ng/ml leukemia inhibitory
factor (LIF) (Millipore). For neural differentiation, 46C cells were cultured on gelatin-
coated 6-well plates at a density of 1×104/cm2 (105 cells per well). Culture medium was
DMEM/F12 combined 1:1 with Neurobasal medium (Invitrogen) supplemented with 1%
B27 (Invitrogen), 0.5% N2 (Invitrogen), 25 µg/ml bovine serum albumin (Sigma).
Medium was changed every other day for six days. On day six the medium was
changed to DMEM/F12 combined 10:1 with Neurobasal medium supplemented with
0.1% B27, 0.5% N2 and 20 ng/ml FGF and EGF (R&D).

116

Cochlear explant culture
Cochlea of postnatal day 1 (P1) pups were dissected in PBS. Reissner’s membrane
and basal hook segment were removed. Cochlear explants were cultured on SPI black
membrane in DMEM/F12 with 1% N2 supplement and 100U/ml Penicillin. All cultures
were incubated in a 5% CO2/ 5% O2 humidified incubator (Forma Scientific).
Immunohistochemistry and imaging
Cochlear explants, 293 cells and NPCs were fixed in 4% paraformaldehyde/PBS for
20 minutes at room temperature and kept in PBS at 4 °C until the staining was
performed. E16.5 whole heads or P1 inner ears were fixed in 4%
paraformaldehyde/PBS for 24h at 4 °C, placed in 30% sucrose/PBS for 48h at 4 °C,
incubated in O.C.T (Tissue Tek) at room temperature for 30 minutes, and frozen in
liquid nitrogen. 14 micron sections were cut on a Leica cryrostat and mounted on
superfrost plus slides (Fisher Scientific). The following primary antibodies were used:
anti-Nestin (Developmental Studies Hybridoma Bank), anti-FLAG (Sigma F7425), anti-
Myo VI (Proteus BioSciences), anti-Prox1 (Novus Biologicals), anti-p75 (Millipore
AB1554) and anti-Sirt2 (Santa Cruz sc-20966). Species specific Alexa 594 or Alexa 633
(life technologies) secondary antibodies were used. Mounting Medium with DAPI
(VECTASHIELD) was used for nuclear staining and mounting of the samples. Confocal
images were taken on a scanning confocal microscope (Zeiss; LSM 780). Fluorescent
imaging was done with a Zeiss Axio observerA1.

117

Small molecule inhibitors
The inhibitors/chemicals used in this study and their final concentrations were as
follows: 10 µM ɣ-secretase inhibitor DAPT (Calbiochem, EMD Millipore); 100 µM
proteasome inhibitor MG132 (sigma); 100 µg/ml protein synthesis inhibitor
cycloheximide (Sigma); 1mM protease inhibitor PMSF (Sigma); 10 µM proteasome
inhibitor lactacystin; 30 µM HAT inhibitor curcumin (Enzo life sciences); 40 µM (organ
explants) and 100 µM (293 cells) Sirt1/Sirt2 inhibitor Salermide (Sigma); and 200 nM
HDAC inhibitor trichostatin A (Sigma).
FACS-purification
All cells were sorted on BD FACSAria II cell sorter with lasers at 488- and 561-nm
wavelength and 100um nozzle.
Cochlear cells
The preparation of samples for FACS was done as described previously with minor
modifications. For each sample, 8-12 neonatal cochleas were incubated in 500 µl PBS
containing 0.1% trypsin (Invitrogen) and 1 mg/ml collagenase (Worthington) for 8
minutes at 37 °C, and digestion stopped by addition DMEM/F12, 5% FBS (Invitrogen).
The tissue was centrifuged at 4°C for 10 minutes at 1000 rpm. The supernatant was
removed, the tissue was triturated for 2 minutes in 300 to 500 µl fresh DMEM/F12, 5%
FBS and passed through a 40 um cell strainer (BD). For RNA extraction, the cells were
sorted directly into RNA lysis buffer (Zymo research) and kept at -80 °C until the RNA
isolation was performed. For ATAC-seq, cells were sorted into DMEM/F12, 10% FBS.

118

For p75 purification, after the trituration, single cells were incubated with p75
antibody (1:500 dilution in DMEM/F12, 5% FBS) for 35 to 45 minutes on ice. After which
1ml DMEM/F12, 5% FBS was added and cells were centrifuged at 1500 rmp for 10 min
at 4°C. Next, the supernatant was removed, cells were resuspened in DMEM/F12, 5%
FBS containing 1:500 dilution Alexa 594 or Alexa 633 and kept on ice for 30-40 min.
After the secondary antibody incubation, cells were washed with 1 ml of DMEM/F12, 5%
FBS as above, and resuspended in fresh DMEM/F12, 5% FBS before passing through
a 40 um cell strainer.
Neural progenitor cells (NPCs)
NPCs transfected with pCS2-CMV-FLAGHes5 or pCS2-CMV-H2BmRFP (control)
were collected 48h post transfection and RFP positive cells were FACS-purified.
Untransfected NPCs were used to determine the RFP positive and RFP negative gates.
Each sample was 2 x 10
5
RFP positive cells sorted into DMEM, 2% FBS containing 1
mM PMSF (Sigma) and protease inhibitor cocktail (1:100 of stock) (Calbiochem).
OC1 cells
OC1 cells transfected with pCL-CMV-FLAGHes5/SV40-eGFP or pCL-SV40-eGFP
(control) were collected 48h post transfection and GFP positive cells were FACS-
purified. Untransfected OC1 cells were used to determine the GFP positive and GFP
negative gates. Each sample was 2 x 10
5
GFP positive cells sorted into DMEM, 2%
FBS containing 1 mM PMSF and protease inhibitor cocktail.

119

RNA extraction and real time quantitative PCR (qPCR)
Total RNA was isolated using Zymo Research Quick-RNA MicroPrep kit, with
DNAse I digestion (Qiagen). RNA from supporting cells (3500-10,000 cells), NPCs,
(5000-10,000 cells) and non-hair cells (100,000 cells) was reverse transcribed by
Quanta qScript cDNA Synthesis Kit. Real time quantitative PCR (qPCR) was performed
with Power SYBR master mix (Applied Biosystems) on 7500 or ViiA 7 real time PCR
system with fast 96-well block (Applied Biosystems). Relative quantification of gene
expression was analyzed by the comparative ΔΔCt method (Livak and Schmittgen
2001) with Gapdh gene as the endogenous reference. Two-tailed Student’s t-test was
used to determine the statistical significance. P-values < 0.05 were considered to be
statistically significant. The primers used in qPCR are listed in Table 2.
Cloning plasmids and reporter constructs
Hes5 expression plasmids
A Hes5 plasmid including 1.6kb of Hes5 promoter and 1.4kb Hes5 gene (including
the 5’-untranslated region, three exons, two intron, 3’-untranslated region and
polyadenylation signal) was generously provided by Verdon Taylor (University of Basel,
Switzerland). This plasmid had a unique BamHI restriction site engineered after the
translation start site (ATG). 3X FLAG oligonucleotides with BamHI sticky ends
(BamHIFLAG-F: 5’-
GATCCGACTATAAAGACGACGACGACAAAGCGGCGGATTACAAGGAT
GATGATGATAAGGCTGCAGATTATAAGGACGATGACGATAAG-3’ and BamHIFLAG-
R: 5’-GATCCCTTATCGTCATCGTCCTTATA ATCTGCAGCCTTATCATCATC

120

ATCCTTGTA ATC CGCCGCTTTGTCGTCGTCGTCTTTATAGTC-3’) were annealed
and inserted downstream of the translation start site to make a FLAGHes5 fusion
construct. To make the pCS2-CMV-FLAG-Hes5 expression vector, pCS2-CMV-
H2BmRFP plasmid (generous gift from Takahiro Ohyama, University of Southern
California) was digested by StuI and KpnI to remove the histone H2B and mRFP
sequence. The FLAGHes5 region was amplified (Hes5StuI-F: 5’-
TATGAATTCAGGCCTGCGCCAGTCCGG-3’ and Hes5KpnI-R: 5’-
CACGGCGGTACCTGGAGGGCTTGATAT-3’) and cloned into the StuI and KpnI sites
downstream of the CMV sequence. The FLAGHes5 gene was also placed downstream
of the CMV sequence in a pCL plasmid that included the SV40-eGFP sequence to
make a dual expression vector of pCL-CMV-FLAGHes5/SV40-eGFP.
Atoh1 reporter plasmids
Atoh1 reporter plasmids
The Atoh1 enhancer/β-globin promoter/GFP plasmid, in which the mouse Atoh1
enhancer and the human β-globin basal promoter drive the expression of GFP, was a
generous gift from Jane Johnson, University of Texas (Lumpkin et al. 2003). The
mCherry reporter (Figure 2) was constructed by cloning a histone H2B-mCherry fusion
into the Atoh1 enhancer/β-globin promoter/GFP plasmid to replace the GFP coding
region. Atoh1 enhancer/Atoh1 promoter
wt
/GFP reporter was constructed using a 226 bp
fragment upstream of the Atoh1 transcription start site, and 177 bp of Atoh1 5’-UTR to
replace the β-globin promoter sequence in Atoh1 enhancer/β-globin promoter/GFP. For
the Atoh1 enhancer/Atoh1 promoter
wt
/tdTomato reporter, tdTomato sequence was PCR
amplified from pFUGW-tdtomato (Addgene plasmid 22478) and inserted in the Atoh1

121

enhancer/Atoh1 promoter
wt
/GFP reporter to replace the GFP sequence. Expression of
the reporters carrying the Atoh1 autoregulatory enhancer element was observed in 293
cells, likely as a result of low level Atoh1 expression as previously reported (Neves et al.
2012).
Hey2 expression plasmid was generously provided by Manfred Gessler (University
of Wuerzburg, Germany). The cloning primers are listed in table Table 3.
Site-directed mutagenesis
Deletion or point mutations in the Atoh1 reporter plasmids, Hes5 and Hey2
expression plasmids were introduced by using a Finnzyme Phusion site-directed
mutagenesis kit following the manufacturer’s instructions. The primers used for
mutagenesis are listed in Table 4.
Chromatin immunoprecipitation (ChIP)-qPCR
For HES5 ChIP, NPCs in each well of a 6-well culture dish were transfected with 15
µg of pCS2-CMV-FLAG-Hes5 or pCS2-CMV-H2BmRFP (control) using Lipofectamine
LTX (Invitrogen). NPCs were FACS-purified into DMEM, 2% FBS containing 1 mM
PMSF (Sigma) and protease inhibitor cocktail (Calbiochem). ChIP from NPCs was
performed as described (Dahl and Collas 2008), using 2x 10
5
cells (FLAG-Hes5); 1x 10
6

cells (pan-GRG/TLE). For H3K9ac ChIP, supporting cells were FACS-purified from
p27/GFP transgenic mouse line (~25,000 supporting cell per sample). FACS-purified
NPCs were fixed with 1% formaldehyde for 10 minutes and quenched by 125 mM
Glycine for 5 minutes at room temperature. Cross-linked cells were centrifuged at 470g
for 10 minutes at 4°C, washed twice with ice-cold PBS, lysed with 50 µl lysis buffer (50

122

mM Tris–HCl, pH 8.0, 10 mM EDTA, 1% (wt/vol) SDS, PMSF and protease inhibitor
cocktail for 8 minutes on ice. RIPA buffer (10 mM Tris–HCl, pH 7.5, 140 mM NaCl, 1
mM EDTA, 0.5 mM EGTA, 1% (vol/vol) Triton X-100, 0.1% (wt/vol) SDS, 0.1% (wt/vol)
Na-deoxycholate, protease inhibitor cocktail and PMSF) was added before sonication.
Chromatin was sonicated to an average size of 200-300bp with High Intensity Ultrasonic
Processor (50 Watt Model) on ice for 4 X 30s, with 30s intervals. After sonication,
samples were centrifuged at 12000g for 10 minutes at 4°C and 1% of the supernatant
chromatin was set aside as input. The remaining chromatin was added to antibody-
Dynabeads protein G complexes (4 µg FLAG antibody (Sigma F1804) or GRG/TLE
antibody (Santa Cruz sc-13373 X) and 25 µl protein G beads (Invitrogen) were
preincubated on a rotator for 2h at 4°C). The tubes containing the chromatin-antibody-
protein G and the input tubes were placed on a rotor for 16h at 4°C. The beads were
then captured in magnetic rack, washed with RIPA and TE. The washed beads were
reverse cross-linked with elution buffer (20 mM Tris–HCl, pH 7.5, 5 mM EDTA, 50 mM
NaCl, 1% (wt/vol) SDS, 50 µg/ml proteinase K) at 1300 rpm for 4h at 68°C. ChIP and
input DNA was purified by phenol–chloroform–isoamyalcohol extraction and ethanol
precipitation and dissolved in TE. qPCR reactions were done in triplicate. H3K9ac ChIP
with about 25,000 supporting cells (from p27/GFP mouse line) per experiment was done
as above with minor differences. FACS-purified supporting cells were cross-linked for 8
minutes and sonicated for 8 X 30s, with 30s intervals. 2.4 µg H3k9ac antibody (Active
Motif 39137) and 10 µl Dynabeads Protein A (Invitrogen) per samples were used. ChIP-
qPCR primers are listed in Table 5.

123

Reporter assay
2 x 10
5
293 cells were seeded in each well of 24-well plate culture dish the night
before transfection. Cells in each well were transfected with 150 ng of Atoh1 reporter
plasmid (GFP or mCherry) and 600 ng of pCS2-CMV-Hes5 plasmid or empty plasmid
by Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. 48 hours
after transfection cells were collected and the number of GFP and mCherry positive
cells was determined by flow cytometry. For the mutational analysis of the C cites in the
Atoh1 promoter region, 293 cells in 24-well plates were transfected with 150 ng of wild
type or mutant GFP-expressing Atoh1 reporter plasmid and 600 ng of pCS-CMV-Hes5
plasmid or empty plasmid. 150 ng of RFP plasmid was also cotransfected as an internal
control for transfection efficiency. 48h post -transfection the number of GFP and RFP
positive cells was determined by flow cytometry analysis.
Flow cytometry
48h post-transfection 293 cells were collected and suspended in 250 µl DMEM/10%
FBS, passed through a 40 µm cell strainer and analyzed using BD FACSAria II
cytometer with lasers at 488- and 561-nm wavelengths and 100um nozzle. Cells were
initially gated (P1) using forward scatter (FSC-A) and side scatter (SSC-A). Two
sequential gates were used to exclude the cellular debris and clumps (P2: SSC-H vs.
SSC-W, P3: FSC-H vs. FSC-W). Untransfected and single-transfected (GFP, mCherry
or RFP) cells were used to determine the gates for the positive and negative
populations as well as to set up the compensation. For each sample about 1x105 cells

124

in gate P3 were analyzed. BD FACSDiva software was used to operate the system and
analyze the cell populations.
Western blotting
Proteins extracted from transfected 293 cells were separated under reducing
conditions on 12% NuPage Novex Bis-Tris gels (Invitrogen) and transferred to PVDF
membranes (Millipore) using the XCell II Blot Module (life technologies). Membranes
were blocked using Odyssey blocking buffer (LI-COR Biosciences) and probed with the
following antibodies: anti-FLAG (Sigma F7425); anti-actin (Sigma A3853); anti-TLE4
(Novus NB100-92363); anti-histone H2B (Millipore 07-371); IRDye 800CW Goat anti-
Rabbit IgG; and IRDye 680RD Goat anti-Mouse IgG (LI-COR).
Proximity ligation assay
To visualize GRG/TLE:FLAG-HES5 interaction, proximity ligation assay (Soderberg
et al. 2006) was performed using the Duolink kit (Sigma-Aldrich) in transfected 293 cells
according to manufacturer’s manual with the following antibodies: rabbit anti-FLAG
(Sigma F7425); goat anti-pan GRG/TLE (Santa Cruz sc-13373 X).
Immunoprecipitation assay
293 cells were suspended in RIPA buffer (10 mM Tris–HCl, pH 7.5, 140 mM NaCl, 1
mM EDTA, 0.5 mM EGTA, 1% (vol/vol) Triton X-100, 0.1% (w/vol) SDS, 0.1% (wt/vol)
with protease inhibitor cocktail. Cell extracts were incubated with Dynabeads protein G
and anti-FLAG antibody (Sigma F1804) at 4 °C overnight. Dynabeads with pulled-down
protein complexes were washed with RIPA buffer three times and proteins were

125

released in 1X laemmli buffer (60 mM Tris-Cl, 2% SDS, 10% glycerol, 5% 2-
mercaptoehtanol and 0.002% bromophenol blue) at 100 °C for 5 min. Input and
immunoprecipitated samples were subjected to western blotting.
ATAC-seq and peak-calling
To profile for open chromatin, we performed “Assay for Transposase Accessible
Chromatin” (ATAC-seq) according to a previously described protocol (Buenrostro et al.
2013) with minor modifications. Briefly, 15,000 to 20,000 hair cells (Atoh1GFP fusion
knock-in) or supporting cells (Lfng/GFP BAC transgenic) were purified by FACS per
experimental sample. After sort cells were spun at 500g for 5 min at 4°C, washed with
cold PBS once, and lysed in 50 µl cold lysis buffer (10 mM Tris-HCl, pH 7.4, 10 mM
NaCl, 3 mM MgCl2, 0.1% IGEPAL CA-630). Immediately after lysis, nuclei were spun
at 500g for 10 min at 4 °C, supernatant was discarded, nuclear pellet was resuspended
in the transposase reaction mix (TD buffer and Tn5 transposase, Illumina, #FC-121-
1030) and incubated at 37°C for 30 min. Immediately following transposition, the
transposed DNA was purified with Qiagen MinElute Kit, eluted in TE and stored at -80
°C until library preparation. To amplify transposed DNA fragments and make libraries,
NEBNext High-Fidelity 2x PCR Master Mix (New England Labs #M0541) and custom
Nextera PCR primers were used with the following PCR conditions: 72°C for 5 min; 98
°C for 30 s; and thermocycling at 98°C for 10 s, 63°C for 30 s and 72°C for 1 min. After
5 cycles, the PCR was hold at 4 °C and a qPCR side reaction was run with 5 µl of the
PCR reaction to determine the additional number of cycles needed for the remaining 45
µl PCR reaction. All libraries were amplified for 11-13 cycles and purified using AMPure
XP beads after which they were run on Bioanalyzer (Agilent Technologies) to determine

126

library sizes. Libraries where sequenced on a NextSeq 500 for an average of 40 million
reads per sample. Reads were trimmed based on quality score and then aligned to
mouse genome assembly mm9 by BWA in PartekFlow. After removing PCR duplicates
and the reads aligned to mitochondrial DNA, peaks were called by peak caller MACS
(Zhang et al. 2008), and tag numbers in each peak region for individual replicates were
counted by HOMER software.
Electrophoretic mobility shift assay (EMSA)
EMSA with invitro transcribed and translated proteins was performed as reported
before (Helms et al. 2000), except that IRdye labled oligonucleotides (probes) were
used. The probes are listed in table Table 6. To prepare the probes, 20 µM forward and
reverse oligonucleotides were mixed in TE and heated at 100°C for 7 min in PCR
machine after which the PCR machine was turned off to allow the probes anneal slowly
over night. 0.5 µM double strand oligonucleotides were used in each EMSA reaction.
Electroporation
Electroporation of postnatal cochlear explants were done as previously reported
(Doetzlhofer et al. 2009). The following plasmids were used: pCBA-CMV-Atoh1, pCS2-
CMV-H2BmRFP, Atoh1 enhancer/β-globin promoter/GFP, and Atoh1 enhancer/β-globin
promoter/mCherry.
Cell counts
Image J (NIH) was used to measure the length of the cochlea and count the number
of cells. Numbers of cells were reported as an average number of cells per 100 µm

127

length of the cochlea. Hair cells were identified by GFP (Atoh1 enhancer/β-globin
promoter/GFP transgenic line) and supporting cells were identified by Prox1 antibody
staining.
RNA-seq
RNA was extracted with NucleoSpin RNA XS kit (Clontech) and libraries were made
with Low Input RNA Kit (Clontech) from 6,000 to 10,000 FACS-purified p75
+
supporting
cells. Five samples were sequenced by Illumina NextSeq500 High Output Kit for 40 bp
paired end reads (40bp reads). The sequencing data were trimmed and aligned against
mouse assembly mm9 using Partek Flow. Differential gene expression was done by the
imbedded GSA in Partex Flow.

Table 2. Primers used in real time quantitative PCR.
Gene Forward (5’ 3’) Reverse (5’ 3’)
Atoh1 GAGTGGGCTGAGGTAAAAGAGT GGTCGGTGCTATCCAGGAG
Hes5 GCACCAGCCCAACTCCAA GGCGAAGGCTTTGCTGTGT
GFP CTGCTGCCCGACAACCA TGTGATCGCGCTTCTCGTT
tdTomato CCTCCTCCGAGGACAACAAC GCCCTCGATCTCGAACTCG
HeyL GCGCAGAGGGATCATAGAGA TCGCAATTCAGAAAGGCTACT
Gapdh TGTGTCCGTCGTGGATCTGA CCTGCTTCACCACCTTCTTGA
Hey2 AAGCGCCCTTGTGAGGAAA TCGCTCCCCACGTCGAT
Hey1 CACTGCAGGAGGGAAAGGTTAT CCCCAAACTCCGATAGTCCAT
Oct4 ACATCGCCAATCAGCTTGG AGAACCATACTCGAACCACATCC
Nestin GCTGGAACAGAGATTGGAAGG CCAGGATCTGAGCGATCTGAC
Sox2 CTGTTTTTTCATCCCAATTGCA GGAGATCTGGCGGAGAATA
GRG/TLE
1
GACAGCCTAAGAGGCACAGAT GGTCCTCGTTAGACACATCCA
GRG/TLE
2
TGAGGACCAACCGTCAGAG GCTGGACTGTCTGTGAGGT

128

GRG/TLE
3
TGGATGTCTCTAATGAGGACCC TTCAGACCACGGGCTTTGTC
GRG/TLE
4
ATTGCAGCTCGCTATGACAGT GAGGAGTCGTGTCTTGTCCAG
hAtoh1 TTGTCCGAGCTGCTACAAACG GAGAAGCGAGTCCGGCAAC
hGAPDH CTGGGCTACACTGAGCACC AAGTGGTCGTTGAGGGCAATG
P75 GAGGCACCGCTGACAACCT CCACAAGGCCCACAACCA
Etv4 TGTGGCAGTTTCTGGTGGC GAGCCTGGCAACCTCTTCAG
Etv5 GGCGGGTATTTCTCCAGCAG TGCTTGACTTTGCCTTCCAGTC
FGFR3 CTGGTGACCGAGGACAATGTG AGGTAGCCGGCCATTTGTG
Sirt1 CTGTTGGTTCCAGTACTGCAGAC GGTATTGATTACCCTCAAGCCGC
Sirt2 CAAGAAGGCTTACAGGGACGTG CTTCCAGTTCCTTCTTCCATCCG
HDAC1 ATTCCTGCGTTCTATTCGCCCAGA TTAGCAGTTCCAGGATGGCCAG
HDAC2 GGAGGCTACACAATCCGGAATG TCTGGAGTGTTCTGGTTTGTCATG
HDAC3 ACCAAGAGCCTTAATGCCTTCAAC
GCAGCTCCAGGATACCAATTACTAT
G
HDAC4 CAGCAGAGGCTGAATGTGAGC CTGGGAAGAAGTTCCCATCGTCA
HDAC5 GAGCAAACACTGGAGCTGTGTACAG CTCCATGGGCTCCTCTGCT
HDAC6 TGCCACCATGAAGCCTCTGA GTTCTGGTGGGCAGCGTTCT
HDAC7 ATGGGGGATCCTGAGTACCTG ACCCTCTAAGGCCAACACCAC
HDAC8 ACCGAATCCAGCAAATCCTCAAC CCCTGCAGTCACAAATTCCACA
HDAC9 CAGCAGCAGATCCACATGAAC TCCACACAAGCACTGCTGTCAC
HDAC10 GCCACAGCAGCAACATTGGATG CTGTGGCAGTGGAGTGGAGAAG
HDAC11 GGTGGAGAGGAATGTCAGGAGGT ACCACGCGGAAAACCACTTCATC

Table 3. Primers used in cloning.
Primer Sequence (5’ 3’)
dtomato-F1 TGTCCGCTAAATTCTGGCCG
dtomato-F2 ACAGGATCCCGCCACCATG
dtomato-R1 CGAATTCTTACTTGTACAGCTCGTCC
dtomato-R2 GCCATACGGGAAGCAATAGCATGATAC
dtomato EcoRV-F CGACACGATATCATGGTGAGCAAGGG
dtomato SacI-R CGTACGAGCTCTTACTTGTACAGCTCGT

129

Atoh1enhancer-
SacI mut-F
GCGAATTGATAATCCACCGCG
Atoh1enhancer-
SacI-R
CCTATAGTGAGTCGTATTGCAATTCACTGG
Atoh1enhancer-
SacI-F
GTGGTCGACAGATCTCAATGAAGTTTG
Atoh1enhancer-
SacI mut-R
CGCGGTGGATTATCAATTCGC
Atoh1enhancer-
EcoRV GFP mut-F
ACGTCATGGATATCAAGGGCGAG
Atoh1enhancer-
EcoRV GFP-R
CTTCTACCTTTCTCTTCTTTTTTGGAGCCA
Atoh1enhancer-
EcoRV GFP-F
GAGCTGTTCACCGGGGTGGT
Atoh1enhancer-
EcoRV GFP mut-R
CTCGCCCTTGATATCCATGACGT
Atoh1 promoter F1 CTTTCTGTCAGCATTGGAAGGCCCG
Atoh1 promoter R1 GCGGGACATCGCACTGCAAT
Atoh1 C box EcoRI-
F1
TGCGAATTCCCAGAGCCAGAGCT
Atoh1 C box
BamHI-F1
TATGGATCCCCAGAGCCAGAGCT
Atoh1 C box NcoI-
R1
TATCCATGGCGCACTGCAATGGC
Atoh1 site C EcoRI-
F1
TGCGAATTCGTCAGCATTGGAAGGC
Atoh1 site C
BamHI-F1
CATGGATCCGTCAGCATTGGAAGGC
Hes5 F-XhoI GTATGATCTCGAGCAGAAGCCCAGAC
Hes5 R- XbaI CAGTGCTCTAGAGCAGAGAGTCTACAG

Table 4. Primers used in site-directed mutagenesis.
Primer Sequence (5’ 3’) Modification
Atoh1 enhancer
E box mut-F
ACGCGCTGTataCTGGTGAGCG
5’
phosphorylation
Atoh1 enhancer
E box-R
GCTCCGCTCCAGACGCTC
5’
phosphorylation
Atoh1 enhancer
N box mut-F
GTCAGCTGGCAAGCGCACTC
5’
phosphorylation
Atoh1 enhancer
N box-R
AGCGCGTGCTCCGCTCC
5’
phosphorylation
Atoh1 enhancer
class C site mut-
GGAGCGGAGataGCGCTGTCA
5’
phosphorylation

130

F
Atoh1 enhancer
class C site-R
AGACGCTCCGCACCGGG
5’
phosphorylation
Atoh1 promoter
C1mut-R
CACACCAGGTAGTTATCAACGAAGCTTCTC
5’
phosphorylation
Atoh1 promoter
C1-F
CGATCTCCGAGTGAGAGGGGGA
5’
phosphorylation
Atoh1 promoter
C2mut-R
CCCTTTAAAAAGCGGATAACTCCAGTGTATCT
CC
5’
phosphorylation
Atoh1 promoter
C2-F
GCGCAGCGCCTTCAGCAA
5’
phosphorylation
Atoh1 promoter
C3mut-F
CTGGGCAGAATAACTACTGGCGCA
5’
phosphorylation
Atoh1 promoter
C3-R
TCAAAGAAGGGGGCGGGCAG
5’
phosphorylation
Atoh1 promoter
C4mut-F
GCTCCTCGATAACTCCTGCCCG
5’
phosphorylation
Atoh1 promoter
C4-R
CAATCAAAGAGTTTTTACACCCAGTGAAAAG
5’
phosphorylation
Hes5 WRPW
deletion-F
TGACCCAGCGGCCGACC
5’
phosphorylation
Hes5 WRPW
deletion-R
GAGGCCGCAGGCGGGTT
5’
phosphorylation
Hes5 basic
deletion-F
CGGGACCGCATCAACAGCA
5’
phosphorylation
Hes5 basic
deletion-R
CAGctgcgggaggagaggc
5’
phosphorylation
Hey2-YQPW
deletion-F
GGGACAGAAGTTGGAGCCTTT
5’
phosphorylation
Hey2-YQPW
deletion-R
AGGTTTATTGTTTGTCCCAGTGCT
5’
phosphorylation
Hey2-basic
deletion-F
CGGGATCGAATAAATAACAGTTTATCTG
5’
phosphorylation
Hey2-basic
deletion-R
CTTTCTTGCCATAATCTGAGAGGTAG
5’
phosphorylation

Table 5. Primers used in ChIP-qPCR. Numbers refer to the 5’ end of the
amplicons.
Primer Forward (5’ 3’) Reverse (5’ 3’)
Amplic
on size
(bp)
Atoh1-
promoter
CCCTCACTCAGGTCGCCT
G
CGTGCGAGGAGCCAATCA 205

131

Table 6. Oligonucleotides used in electrophoretic mobility shift assay.
Primer Sequence (5’ 3’) Modification
Atoh1 F-WT ACGCGCTGTCAGCTGGTGAGCGCACTC -
Atoh1 R-WT GAGTGCGCTCACCAGCTGACAGCGCGT -
Atoh1 F-WT ACGCGCTGTCAGCTGGTGAGCGCACTC 5' IRDye 800
Atoh1 F-Ebox
Mut
ACGCGCTGTATACTGGTGAGCGCACTC -
Atoh1 R-Ebox
Mut
GAGTGCGCTCACCAGTATACAGCGCGT -
Atoh1 F-Nbox
Mut
ACGCGCTGTCAGCTGGCAAGCGCACTC -
Atoh1 R-Nbox
Mut
GAGTGCGCTTGCCAGCTGACAGCGCG -

(-354)
Atoh1-TSS
(-87)
GGGGAGCCGGGGGAGAT
ACAC
ACCAGGTCGCGTGCAACGAAG 93
Atoh1-exon
(+1156)
ACATCTCCCAGATCCCAC
AG
GGGCATTTGGTTGTCTCAGT 119
Atoh1
enhancer-B
(+5346)
AGAGCGGCTGACAATAGA
GG
GTGCGCTCACCAGCTGAC 93
Atoh1
enhancer-A
(+4264)
CACACCCCATTAACAAGC
TG
GTCTGGCATATGGGGAATGA 112

132

References
Ahmed M, Wong EY, Sun J, Xu J, Wang F, Xu PX. 2012. Eya1-Six1 interaction is
sufficient to induce hair cell fate in the cochlea by activating Atoh1 expression in
cooperation with Sox2. Developmental cell 22: 377-390.
Akazawa C, Ishibashi M, Shimizu C, Nakanishi S, Kageyama R. 1995. A mammalian
helix-loop-helix factor structurally related to the product of Drosophila proneural
gene atonal is a positive transcriptional regulator expressed in the developing
nervous system. J Biol Chem 270: 8730-8738.
Akazawa C, Sasai Y, Nakanishi S, Kageyama R. 1992. Molecular characterization of a
rat negative regulator with a basic helix-loop-helix structure predominantly
expressed in the developing nervous system. J Biol Chem 267: 21879-21885.
Alvarez Y, Alonso MT, Vendrell V, Zelarayan LC, Chamero P, Theil T, Bosl MR, Kato S,
Maconochie M, Riethmacher D et al. 2003. Requirements for FGF3 and FGF10
during inner ear formation. Development 130: 6329-6338.
Audrito V, Vaisitti T, Rossi D, Gottardi D, D'Arena G, Laurenti L, Gaidano G, Malavasi F,
Deaglio S. 2011. Nicotinamide blocks proliferation and induces apoptosis of
chronic lymphocytic leukemia cells through activation of the p53/miR-34a/SIRT1
tumor suppressor network. Cancer research 71: 4473-4483.
Balasubramanyam K, Varier RA, Altaf M, Swaminathan V, Siddappa NB, Ranga U,
Kundu TK. 2004. Curcumin, a novel p300/CREB-binding protein-specific inhibitor
of acetyltransferase, represses the acetylation of histone/nonhistone proteins and
histone acetyltransferase-dependent chromatin transcription. J Biol Chem 279:
51163-51171.
Barad O, Hornstein E, Barkai N. 2011. Robust selection of sensory organ precursors by
the Notch-Delta pathway. Current opinion in cell biology 23: 663-667.
Barber MF, Michishita-Kioi E, Xi Y, Tasselli L, Kioi M, Moqtaderi Z, Tennen RI, Paredes
S, Young NL, Chen K et al. 2012. SIRT7 links H3K18 deacetylation to
maintenance of oncogenic transformation. Nature 487: 114-118.
Barolo S, Posakony JW. 2002. Three habits of highly effective signaling pathways:
principles of transcriptional control by developmental cell signaling. Genes &
development 16: 1167-1181.
Basch ML, Ohyama T, Segil N, Groves AK. 2011. Canonical Notch signaling is not
necessary for prosensory induction in the mouse cochlea: insights from a
conditional mutant of RBPjkappa. The Journal of neuroscience : the official
journal of the Society for Neuroscience 31: 8046-8058.
Bedalov A, Gatbonton T, Irvine WP, Gottschling DE, Simon JA. 2001. Identification of a
small molecule inhibitor of Sir2p. Proceedings of the National Academy of
Sciences of the United States of America 98: 15113-15118.
Benito-Gonzalez A, Doetzlhofer A. 2014. Hey1 and Hey2 control the spatial and
temporal pattern of mammalian auditory hair cell differentiation downstream of
Hedgehog signaling. The Journal of neuroscience : the official journal of the
Society for Neuroscience 34: 12865-12876.

133

Bermingham NA, Hassan BA, Price SD, Vollrath MA, Ben-Arie N, Eatock RA, Bellen HJ,
Lysakowski A, Zoghbi HY. 1999. Math1: an essential gene for the generation of
inner ear hair cells. Science 284: 1837-1841.
Bolden JE, Peart MJ, Johnstone RW. 2006. Anticancer activities of histone deacetylase
inhibitors. Nature reviews Drug discovery 5: 769-784.
Borggrefe T, Oswald F. 2009. The Notch signaling pathway: transcriptional regulation at
Notch target genes. Cellular and molecular life sciences : CMLS 66: 1631-1646.
Borghese L, Dolezalova D, Opitz T, Haupt S, Leinhaas A, Steinfarz B, Koch P,
Edenhofer F, Hampl A, Brustle O. 2010. Inhibition of notch signaling in human
embryonic stem cell-derived neural stem cells delays G1/S phase transition and
accelerates neuronal differentiation in vitro and in vivo. Stem Cells 28: 955-964.
Bradner JE, West N, Grachan ML, Greenberg EF, Haggarty SJ, Warnow T, Mazitschek
R. 2010. Chemical phylogenetics of histone deacetylases. Nature chemical
biology 6: 238-243.
Brigande JV, Heller S. 2009. Quo vadis, hair cell regeneration? Nature neuroscience
12: 679-685.
Brooker R, Hozumi K, Lewis J. 2006. Notch ligands with contrasting functions: Jagged1
and Delta1 in the mouse inner ear. Development 133: 1277-1286.
Buenrostro JD, Giresi PG, Zaba LC, Chang HY, Greenleaf WJ. 2013. Transposition of
native chromatin for fast and sensitive epigenomic profiling of open chromatin,
DNA-binding proteins and nucleosome position. Nature methods 10: 1213-1218.
Cafaro J, Lee GS, Stone JS. 2007. Atoh1 expression defines activated progenitors and
differentiating hair cells during avian hair cell regeneration. Developmental
dynamics : an official publication of the American Association of Anatomists 236:
156-170.
Cai T, Seymour ML, Zhang H, Pereira FA, Groves AK. 2013. Conditional deletion of
Atoh1 reveals distinct critical periods for survival and function of hair cells in the
organ of Corti. The Journal of neuroscience : the official journal of the Society for
Neuroscience 33: 10110-10122.
Caron C, Boyault C, Khochbin S. 2005. Regulatory cross-talk between lysine acetylation
and ubiquitination: role in the control of protein stability. BioEssays : news and
reviews in molecular, cellular and developmental biology 27: 408-415.
Chardin S, Romand R. 1995. Regeneration and mammalian auditory hair cells. Science
267: 707-711.
Chen G, Fernandez J, Mische S, Courey AJ. 1999. A functional interaction between the
histone deacetylase Rpd3 and the corepressor groucho in Drosophila
development. Genes & development 13: 2218-2230.
Chen P, Johnson JE, Zoghbi HY, Segil N. 2002. The role of Math1 in inner ear
development: Uncoupling the establishment of the sensory primordium from hair
cell fate determination. Development 129: 2495-2505.
Chen P, Segil N. 1999. p27(Kip1) links cell proliferation to morphogenesis in the
developing organ of Corti. Development 126: 1581-1590.
Chin MT, Maemura K, Fukumoto S, Jain MK, Layne MD, Watanabe M, Hsieh CM, Lee
ME. 2000. Cardiovascular basic helix loop helix factor 1, a novel transcriptional
repressor expressed preferentially in the developing and adult cardiovascular
system. J Biol Chem 275: 6381-6387.

134

Choi JK, Howe LJ. 2009. Histone acetylation: truth of consequences? Biochemistry and
cell biology = Biochimie et biologie cellulaire 87: 139-150.
Choudhary C, Kumar C, Gnad F, Nielsen ML, Rehman M, Walther TC, Olsen JV, Mann
M. 2009. Lysine acetylation targets protein complexes and co-regulates major
cellular functions. Science 325: 834-840.
Colvin JS, Bohne BA, Harding GW, McEwen DG, Ornitz DM. 1996. Skeletal overgrowth
and deafness in mice lacking fibroblast growth factor receptor 3. Nature genetics
12: 390-397.
Cox BC, Chai R, Lenoir A, Liu Z, Zhang L, Nguyen DH, Chalasani K, Steigelman KA,
Fang J, Rubel EW et al. 2014. Spontaneous hair cell regeneration in the neonatal
mouse cochlea in vivo. Development 141: 816-829.
Crawford TQ, Roelink H. 2007. The notch response inhibitor DAPT enhances neuronal
differentiation in embryonic stem cell-derived embryoid bodies independently of
sonic hedgehog signaling. Developmental dynamics : an official publication of the
American Association of Anatomists 236: 886-892.
Dabdoub A, Puligilla C, Jones JM, Fritzsch B, Cheah KS, Pevny LH, Kelley MW. 2008.
Sox2 signaling in prosensory domain specification and subsequent hair cell
differentiation in the developing cochlea. Proceedings of the National Academy of
Sciences of the United States of America 105: 18396-18401.
Dahl JA, Collas P. 2008. A rapid micro chromatin immunoprecipitation assay
(microChIP). Nat Protoc 3: 1032-1045.
Dai Y, Faller DV. 2008. Transcription Regulation by Class III Histone Deacetylases
(HDACs)-Sirtuins. Translational oncogenomics 3: 53-65.
Dawson SR, Turner DL, Weintraub H, Parkhurst SM. 1995. Specificity for the
hairy/enhancer of split basic helix-loop-helix (bHLH) proteins maps outside the
bHLH domain and suggests two separable modes of transcriptional repression.
Molecular and cellular biology 15: 6923-6931.
Doetzlhofer A, Basch ML, Ohyama T, Gessler M, Groves AK, Segil N. 2009. Hey2
regulation by FGF provides a Notch-independent mechanism for maintaining
pillar cell fate in the organ of Corti. Developmental cell 16: 58-69.
Driver EC, Kelley MW. 2009. Specification of cell fate in the mammalian cochlea. Birth
Defects Res C Embryo Today 87: 212-221.
Driver EC, Sillers L, Coate TM, Rose MF, Kelley MW. 2013. The Atoh1-lineage gives
rise to hair cells and supporting cells within the mammalian cochlea.
Developmental biology 376: 86-98.
Dror AA, Avraham KB. 2010. Hearing impairment: a panoply of genes and functions.
Neuron 68: 293-308.
Durrant JD, Lovrinic JH. 1995. Bases of Hearing Science Williams & Wilkins.
Ebert PJ, Timmer JR, Nakada Y, Helms AW, Parab PB, Liu Y, Hunsaker TL, Johnson
JE. 2003. Zic1 represses Math1 expression via interactions with the Math1
enhancer and modulation of Math1 autoregulation. Development 130: 1949-
1959.
Eskandarian HA, Impens F, Nahori MA, Soubigou G, Coppee JY, Cossart P, Hamon
MA. 2013. A role for SIRT2-dependent histone H3K18 deacetylation in bacterial
infection. Science 341: 1238858.

135

Fischer A, Gessler M. 2007. Delta-Notch--and then? Protein interactions and proposed
modes of repression by Hes and Hey bHLH factors. Nucleic Acids Res 35: 4583-
4596.
Fisher AL, Ohsako S, Caudy M. 1996. The WRPW motif of the hairy-related basic helix-
loop-helix repressor proteins acts as a 4-amino-acid transcription repression and
protein-protein interaction domain. Molecular and cellular biology 16: 2670-2677.
Gazit R, Krizhanovsky V, Ben-Arie N. 2004. Math1 controls cerebellar granule cell
differentiation by regulating multiple components of the Notch signaling pathway.
Development 131: 903-913.
Goetz R, Mohammadi M. 2013. Exploring mechanisms of FGF signalling through the
lens of structural biology. Nature reviews Molecular cell biology 14: 166-180.
Gould F, Harrison SM, Hewitt EW, Whitehouse A. 2009. Kaposi's sarcoma-associated
herpesvirus RTA promotes degradation of the Hey1 repressor protein through
the ubiquitin proteasome pathway. Journal of virology 83: 6727-6738.
Grbavec D, Lo R, Liu Y, Stifani S. 1998. Transducin-like Enhancer of split 2, a
mammalian homologue of Drosophila Groucho, acts as a transcriptional
repressor, interacts with Hairy/Enhancer of split proteins, and is expressed during
neuronal development. Eur J Biochem 258: 339-349.
Grbavec D, Stifani S. 1996 Molecular interaction between TLE1 and the carboxyl-
terminal domain of HES-1 containing the WRPW motif. Biochem Biophys Res
Commun 223: 701-705.
Grogan SP, Olee T, Hiraoka K, Lotz MK. 2008. Repression of chondrogenesis through
binding of notch signaling proteins HES-1 and HEY-1 to N-box domains in the
COL2A1 enhancer site. Arthritis Rheum 58: 2754-2763.
Groves AK, Zhang KD, Fekete DM. 2013. The genetics of hair cell development and
regeneration. Annual review of neuroscience 36: 361-381.
Hartman BH, Hayashi T, Nelson BR, Bermingham-McDonogh O, Reh TA. 2007. Dll3 is
expressed in developing hair cells in the mammalian cochlea. Developmental
dynamics : an official publication of the American Association of Anatomists 236:
2875-2883.
Hartman BH, Reh TA, Bermingham-McDonogh O. 2010. Notch signaling specifies
prosensory domains via lateral induction in the developing mammalian inner ear.
Proceedings of the National Academy of Sciences of the United States of
America 107: 15792-15797.
Hatakeyama J, Bessho Y, Katoh K, Ookawara S, Fujioka M, Guillemot F, Kageyama R.
2004. Hes genes regulate size, shape and histogenesis of the nervous system by
control of the timing of neural stem cell differentiation. Development 131: 5539-
5550.
Hayashi T, Kokubo H, Hartman BH, Ray CA, Reh TA, Bermingham-McDonogh O.
2008a. Hesr1 and Hesr2 may act as early effectors of Notch signaling in the
developing cochlea. Developmental biology 316: 87-99.
Hayashi T, Ray CA, Bermingham-McDonogh O. 2008l. Fgf20 is required for sensory
epithelial specification in the developing cochlea. The Journal of neuroscience :
the official journal of the Society for Neuroscience 28: 5991-5999.
Heisig J, Weber D, Englberger E, Winkler A, Kneitz S, Sung WK, Wolf E, Eilers M, Wei
CL, Gessler M. 2012. Target gene analysis by microarrays and chromatin

136

immunoprecipitation identifies HEY proteins as highly redundant bHLH
repressors. PLoS genetics 8: e1002728.
Helms AW, Abney AL, Ben-Arie N, Zoghbi HY, Johnson JE. 2000. Autoregulation and
multiple enhancers control Math1 expression in the developing nervous system.
Development 127: 1185-1196.
Helms AW, Johnson JE. 1998. Progenitors of dorsal commissural interneurons are
defined by MATH1 expression. Development 125: 919-928.
Hirata H, Bessho Y, Kokubu H, Masamizu Y, Yamada S, Lewis J, Kageyama R. 2004.
Instability of Hes7 protein is crucial for the somite segmentation clock. Nature
genetics 36: 750-754.
Hirata H, Yoshiura S, Ohtsuka T, Bessho Y, Harada T, Yoshikawa K, Kageyama R.
2002. Oscillatory expression of the bHLH factor Hes1 regulated by a negative
feedback loop. Science 298: 840-843.
Hori R, Nakagawa T, Sakamoto T, Matsuoka Y, Takebayashi S, Ito J. 2007.
Pharmacological inhibition of Notch signaling in the mature guinea pig cochlea.
Neuroreport 18: 1911-1914.
Huang BH, Laban M, Leung CH, Lee L, Lee CK, Salto-Tellez M, Raju GC, Hooi SC.
2005. Inhibition of histone deacetylase 2 increases apoptosis and p21Cip1/WAF1
expression, independent of histone deacetylase 1. Cell death and differentiation
12: 395-404.
Iso T, Kedes L, Hamamori Y. 2003. HES and HERP families: multiple effectors of the
Notch signaling pathway. J Cell Physiol 194: 237-255.
Iso T, Sartorelli V, Poizat C, Iezzi S, Wu HY, Chung G, Kedes L, Hamamori Y. 2001.
HERP, a novel heterodimer partner of HES/E(spl) in Notch signaling. Molecular
and cellular biology 21: 6080-6089.
Jacques BE, Montcouquiol ME, Layman EM, Lewandoski M, Kelley MW. 2007. Fgf8
induces pillar cell fate and regulates cellular patterning in the mammalian
cochlea. Development 134: 3021-3029.
Jacques BE, Puligilla C, Weichert RM, Ferrer-Vaquer A, Hadjantonakis AK, Kelley MW,
Dabdoub A. 2012. A dual function for canonical Wnt/beta-catenin signaling in the
developing mammalian cochlea. Development 139: 4395-4404.
Jahan I, Pan N, Kersigo J, Calisto LE, Morris KA, Kopecky B, Duncan JS, Beisel KW,
Fritzsch B. 2012. Expression of Neurog1 instead of Atoh1 can partially rescue
organ of Corti cell survival. PloS one 7: e30853.
Jarman AP, Groves AK. 2013. The role of Atonal transcription factors in the
development of mechanosensitive cells. Seminars in cell & developmental
biology 24: 438-447.
Jarriault S, Brou C, Logeat F, Schroeter EH, Kopan R, Israel A. 1995. Signalling
downstream of activated mammalian Notch. Nature 377: 355-358.
Jing E, Gesta S, Kahn CR. 2007. SIRT2 regulates adipocyte differentiation through
FoxO1 acetylation/deacetylation. Cell metabolism 6: 105-114.
Johnson CA, White DA, Lavender JS, O'Neill LP, Turner BM. 2002. Human class I
histone deacetylase complexes show enhanced catalytic activity in the presence
of ATP and co-immunoprecipitate with the ATP-dependent chaperone protein
Hsp70. J Biol Chem 277: 9590-9597.

137

Johnson JE, Birren SJ, Saito T, Anderson DJ. 1992. DNA binding and transcriptional
regulatory activity of mammalian achaete-scute homologous (MASH) proteins
revealed by interaction with a muscle-specific enhancer. Proceedings of the
National Academy of Sciences of the United States of America 89: 3596-3600.
Kageyama R, Ohtsuka T, Kobayashi T. 2007. The Hes gene family: repressors and
oscillators that orchestrate embryogenesis. Development 134: 1243-1251.
Kelley MW. 2006. Regulation of cell fate in the sensory epithelia of the inner ear. Nat
Rev Neurosci 7: 837-849.
-. 2007. Cellular commitment and differentiation in the organ of Corti. The International
journal of developmental biology 51: 571-583.
Kelly MC, Chang Q, Pan A, Lin X, Chen P. 2012. Atoh1 directs the formation of sensory
mosaics and induces cell proliferation in the postnatal mammalian cochlea in
vivo. The Journal of neuroscience : the official journal of the Society for
Neuroscience 32: 6699-6710.
Kidd S, Kelley MR, Young MW. 1986. Sequence of the notch locus of Drosophila
melanogaster: relationship of the encoded protein to mammalian clotting and
growth factors. Molecular and cellular biology 6: 3094-3108.
Kiernan AE. 2013. Notch signaling during cell fate determination in the inner ear.
Seminars in cell & developmental biology 24: 470-479.
Kiernan AE, Cordes R, Kopan R, Gossler A, Gridley T. 2005a. The Notch ligands DLL1
and JAG2 act synergistically to regulate hair cell development in the mammalian
inner ear. Development 132: 4353-4362.
Kiernan AE, Pelling AL, Leung KK, Tang AS, Bell DM, Tease C, Lovell-Badge R, Steel
KP, Cheah KS. 2005b. Sox2 is required for sensory organ development in the
mammalian inner ear. Nature 434: 1031-1035.
Kiernan AE, Xu J, Gridley T. 2006. The Notch ligand JAG1 is required for sensory
progenitor development in the mammalian inner ear. PLoS genetics 2: e4.
Kierszenbaum AL. 2007. Histology and Cell Biology : an Introduction to Pathology.
Mosby/Elsevier.
Kim HJ, Bar-Sagi D. 2004. Modulation of signalling by Sprouty: a developing story.
Nature reviews Molecular cell biology 5: 441-450.
Kirjavainen A, Sulg M, Heyd F, Alitalo K, Yla-Herttuala S, Moroy T, Petrova TV, Pirvola
U. 2008. Prox1 interacts with Atoh1 and Gfi1, and regulates cellular
differentiation in the inner ear sensory epithelia. Developmental biology 322: 33-
45.
Klisch TJ, Xi Y, Flora A, Wang L, Li W, Zoghbi HY. 2011. In vivo Atoh1 targetome
reveals how a proneural transcription factor regulates cerebellar development.
Proceedings of the National Academy of Sciences of the United States of
America 108: 3288-3293.
Kobayashi T, Mizuno H, Imayoshi I, Furusawa C, Shirahige K, Kageyama R. 2009. The
cyclic gene Hes1 contributes to diverse differentiation responses of embryonic
stem cells. Genes & development 23: 1870-1875.
Korrapati S, Roux I, Glowatzki E, Doetzlhofer A. 2013. Notch signaling limits supporting
cell plasticity in the hair cell-damaged early postnatal murine cochlea. PloS one
8: e73276.

138

Krishnamoorthy V, Carr T, de Pooter RF, Akinola EO, Gounari F, Kee BL. 2015.
Repression of ccr9 transcription in mouse T lymphocyte progenitors by the notch
signaling pathway. Journal of immunology 194: 3191-3200.
Kulic I, Robertson G, Chang L, Baker JH, Lockwood WW, Mok W, Fuller M, Fournier M,
Wong N, Chou V et al. 2015. Loss of the Notch effector RBPJ promotes
tumorigenesis. The Journal of experimental medicine 212: 37-52.
Lanford PJ, Lan Y, Jiang R, Lindsell C, Weinmaster G, Gridley T, Kelley MW. 1999.
Notch signalling pathway mediates hair cell development in mammalian cochlea.
Nature genetics 21: 289-292.
Lanford PJ, Shailam R, Norton CR, Gridley T, Kelley MW. 2000. Expression of Math1
and HES5 in the cochleae of wildtype and Jag2 mutant mice. Journal of the
Association for Research in Otolaryngology : JARO 1: 161-171.
Lara E, Mai A, Calvanese V, Altucci L, Lopez-Nieva P, Martinez-Chantar ML, Varela-
Rey M, Rotili D, Nebbioso A, Ropero S et al. 2009. Salermide, a Sirtuin inhibitor
with a strong cancer-specific proapoptotic effect. Oncogene 28: 781-791.
Lassar AB, Buskin JN, Lockshon D, Davis RL, Apone S, Hauschka SD, Weintraub H.
1989. MyoD is a sequence-specific DNA binding protein requiring a region of
myc homology to bind to the muscle creatine kinase enhancer. Cell 58: 823-831.
Lee YS, Liu F, Segil N. 2006. A morphogenetic wave of p27Kip1 transcription directs
cell cycle exit during organ of Corti development. Development 133: 2817-2826.
Li S, Mark S, Radde-Gallwitz K, Schlisner R, Chin MT, Chen P. 2008. Hey2 functions in
parallel with Hes1 and Hes5 for mammalian auditory sensory organ
development. BMC Dev Biol 8: 20.
Lim DJ, Anniko M. 1985. Developmental morphology of the mouse inner ear. A
scanning electron microscopic observation. Acta oto-laryngologica
Supplementum 422: 1-69.
Liu A, Li J, Marin-Husstege M, Kageyama R, Fan Y, Gelinas C, Casaccia-Bonnefil P.
2006. A molecular insight of Hes5-dependent inhibition of myelin gene
expression: old partners and new players. EMBO J 25: 4833-4842.
Liu PY, Xu N, Malyukova A, Scarlett CJ, Sun YT, Zhang XD, Ling D, Su SP, Nelson C,
Chang DK et al. 2013. The histone deacetylase SIRT2 stabilizes Myc
oncoproteins. Cell death and differentiation 20: 503-514.
Liu Y, Ji X, Nie X, Qu M, Zheng L, Tan Z, Zhao H, Huo L, Liu S, Zhang B et al. 2015.
Arabidopsis AtbHLH112 regulates the expression of genes involved in abiotic
stress tolerance by binding to their E-box and GCG-box motifs. The New
phytologist.
Liu Z, Dearman JA, Cox BC, Walters BJ, Zhang L, Ayrault O, Zindy F, Gan L, Roussel
MF, Zuo J. 2012a. Age-dependent in vivo conversion of mouse cochlear pillar
and Deiters' cells to immature hair cells by Atoh1 ectopic expression. The Journal
of neuroscience : the official journal of the Society for Neuroscience 32: 6600-
6610.
Liu Z, Owen T, Fang J, Srinivasan RS, Zuo J. 2012b. In vivo Notch reactivation in
differentiating cochlear hair cells induces Sox2 and Prox1 expression but does
not disrupt hair cell maturation. Developmental dynamics : an official publication
of the American Association of Anatomists 241: 684-696.

139

Livak KJ, Schmittgen TD. 2001. Analysis of relative gene expression data using real-
time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25: 402-
408.
Logan CY, Nusse R. 2004. The Wnt signaling pathway in development and disease.
Annual review of cell and developmental biology 20: 781-810.
Lumpkin EA, Collisson T, Parab P, Omer-Abdalla A, Haeberle H, Chen P, Doetzlhofer
A, White P, Groves A, Segil N et al. 2003. Math1-driven GFP expression in the
developing nervous system of transgenic mice. Gene Expr Patterns 3: 389-395.
Maass JC, Gu R, Basch ML, Waldhaus J, Lopez EM, Xia A, Oghalai JS, Heller S,
Groves AK. 2015. Changes in the regulation of the Notch signaling pathway are
temporally correlated with regenerative failure in the mouse cochlea. Frontiers in
cellular neuroscience 9: 110.
MacDonald BT, Tamai K, He X. 2009. Wnt/beta-catenin signaling: components,
mechanisms, and diseases. Developmental cell 17: 9-26.
Maier MM, Gessler M. 2000. Comparative analysis of the human and mouse Hey1
promoter: Hey genes are new Notch target genes. Biochem Biophys Res
Commun 275: 652-660.
Mansour SL, Li C, Urness LD. 2013. Genetic rescue of Muenke syndrome model
hearing loss reveals prolonged FGF-dependent plasticity in cochlear supporting
cell fates. Genes & development 27: 2320-2331.
Masuda M, Dulon D, Pak K, Mullen LM, Li Y, Erkman L, Ryan AF. 2011. Regulation of
POU4F3 gene expression in hair cells by 5' DNA in mice. Neuroscience 197: 48-
64.
Michishita E, McCord RA, Berber E, Kioi M, Padilla-Nash H, Damian M, Cheung P,
Kusumoto R, Kawahara TL, Barrett JC et al. 2008. SIRT6 is a histone H3 lysine 9
deacetylase that modulates telomeric chromatin. Nature 452: 492-496.
Mizutari K, Fujioka M, Hosoya M, Bramhall N, Okano HJ, Okano H, Edge AS. 2013.
Notch inhibition induces cochlear hair cell regeneration and recovery of hearing
after acoustic trauma. Neuron 77: 58-69.
Mohr OL. 1919. Character Changes Caused by Mutation of an Entire Region of a
Chromosome in Drosophila. Genetics 4: 275-282.
Montcouquiol M, Kelley MW. 2003. Planar and vertical signals control cellular
differentiation and patterning. The Journal of neuroscience : the official journal of
the Society for Neuroscience 23: 9469-9478.
Monzack EL, Cunningham LL. 2013. Lead roles for supporting actors: critical functions
of inner ear supporting cells. Hearing research 303: 20-29.
Moon RT. 2005. Wnt/beta-catenin pathway. Science's STKE : signal transduction
knowledge environment 2005: cm1.
Morrison A, Hodgetts C, Gossler A, Hrabe de Angelis M, Lewis J. 1999. Expression of
Delta1 and Serrate1 (Jagged1) in the mouse inner ear. Mechanisms of
development 84: 169-172.
Mueller KL, Jacques BE, Kelley MW. 2002. Fibroblast growth factor signaling regulates
pillar cell development in the organ of corti. The Journal of neuroscience : the
official journal of the Society for Neuroscience 22: 9368-9377.
Mulvaney J, Dabdoub A. 2012. Atoh1, an essential transcription factor in neurogenesis
and intestinal and inner ear development: function, regulation, and context

140

dependency. Journal of the Association for Research in Otolaryngology : JARO
13: 281-293.
Munnamalai V, Hayashi T, Bermingham-McDonogh O. 2012. Notch prosensory effects
in the Mammalian cochlea are partially mediated by Fgf20. The Journal of
neuroscience : the official journal of the Society for Neuroscience 32: 12876-
12884.
Nakagawa O, McFadden DG, Nakagawa M, Yanagisawa H, Hu T, Srivastava D, Olson
EN. 2000. Members of the HRT family of basic helix-loop-helix proteins act as
transcriptional repressors downstream of Notch signaling. Proceedings of the
National Academy of Sciences of the United States of America 97: 13655-13660.
Neves J, Uchikawa M, Bigas A, Giraldez F. 2012. The prosensory function of Sox2 in
the chicken inner ear relies on the direct regulation of Atoh1. PloS one 7:
e30871.
Nishimura M, Isaka F, Ishibashi M, Tomita K, Tsuda H, Nakanishi S, Kageyama R.
1998. Structure, chromosomal locus, and promoter of mouse Hes2 gene, a
homologue of Drosophila hairy and Enhancer of split. Genomics 49: 69-75.
Noh JH, Jung KH, Kim JK, Eun JW, Bae HJ, Xie HJ, Chang YG, Kim MG, Park WS, Lee
JY et al. 2011. Aberrant regulation of HDAC2 mediates proliferation of
hepatocellular carcinoma cells by deregulating expression of G1/S cell cycle
proteins. PloS one 6: e28103.
Ohtsuka T, Ishibashi M, Gradwohl G, Nakanishi S, Guillemot F, Kageyama R. 1999.
Hes1 and Hes5 as notch effectors in mammalian neuronal differentiation. EMBO
J 18: 2196-2207.
Ohtsuka T, Sakamoto M, Guillemot F, Kageyama R. 2001. Roles of the basic helix-loop-
helix genes Hes1 and Hes5 in expansion of neural stem cells of the developing
brain. J Biol Chem 276: 30467-30474.
Ornitz DM. 2000. FGFs, heparan sulfate and FGFRs: complex interactions essential for
development. BioEssays : news and reviews in molecular, cellular and
developmental biology 22: 108-112.
Ornitz DM, Itoh N. 2015. The Fibroblast Growth Factor signaling pathway. Wiley
interdisciplinary reviews Developmental biology 4: 215-266.
Pan W, Jin Y, Stanger B, Kiernan AE. 2010. Notch signaling is required for the
generation of hair cells and supporting cells in the mammalian inner ear.
Proceedings of the National Academy of Sciences of the United States of
America 107: 15798-15803.
Peck B, Chen CY, Ho KK, Di Fruscia P, Myatt SS, Coombes RC, Fuchter MJ, Hsiao
CD, Lam EW. 2010. SIRT inhibitors induce cell death and p53 acetylation
through targeting both SIRT1 and SIRT2. Molecular cancer therapeutics 9: 844-
855.
Peng L, Ling H, Yuan Z, Fang B, Bloom G, Fukasawa K, Koomen J, Chen J, Lane WS,
Seto E. 2012. SIRT1 negatively regulates the activities, functions, and protein
levels of hMOF and TIP60. Molecular and cellular biology 32: 2823-2836.
Pierfelice T, Alberi L, Gaiano N. 2011. Notch in the vertebrate nervous system: an old
dog with new tricks. Neuron 69: 840-855.
Pirvola U, Spencer-Dene B, Xing-Qun L, Kettunen P, Thesleff I, Fritzsch B, Dickson C,
Ylikoski J. 2000. FGF/FGFR-2(IIIb) signaling is essential for inner ear

141

morphogenesis. The Journal of neuroscience : the official journal of the Society
for Neuroscience 20: 6125-6134.
Pirvola U, Ylikoski J, Trokovic R, Hebert JM, McConnell SK, Partanen J. 2002. FGFR1
is required for the development of the auditory sensory epithelium. Neuron 35:
671-680.
Pujol R, Hilding D. 1973. Anatomy and physiology of the onset of auditory function. Acta
oto-laryngologica 76: 1-10.
Purves D AG, Fitzpatrick D, et al. 2001. Hair Cells and the Mechanoelectrical
Transduction of Sound Waves. in Neuroscience. Sinauer Associates, Sunderland
(MA).
Raft S, Koundakjian EJ, Quinones H, Jayasena CS, Goodrich LV, Johnson JE, Segil N,
Groves AK. 2007. Cross-regulation of Ngn1 and Math1 coordinates the
production of neurons and sensory hair cells during inner ear development.
Development 134: 4405-4415.
Rebeiz M, Stone T, Posakony JW. 2005. An ancient transcriptional regulatory linkage.
Developmental biology 281: 299-308.
Riccio O, van Gijn ME, Bezdek AC, Pellegrinet L, van Es JH, Zimber-Strobl U, Strobl LJ,
Honjo T, Clevers H, Radtke F. 2008. Loss of intestinal crypt progenitor cells
owing to inactivation of both Notch1 and Notch2 is accompanied by derepression
of CDK inhibitors p27Kip1 and p57Kip2. EMBO reports 9: 377-383.
Richon VM, Emiliani S, Verdin E, Webb Y, Breslow R, Rifkind RA, Marks PA. 1998. A
class of hybrid polar inducers of transformed cell differentiation inhibits histone
deacetylases. Proceedings of the National Academy of Sciences of the United
States of America 95: 3003-3007.
Roberson DW, Alosi JA, Cotanche DA. 2004. Direct transdifferentiation gives rise to the
earliest new hair cells in regenerating avian auditory epithelium. J Neurosci Res
78: 461-471.
Rose MF, Ren J, Ahmad KA, Chao HT, Klisch TJ, Flora A, Greer JJ, Zoghbi HY. 2009.
Math1 is essential for the development of hindbrain neurons critical for perinatal
breathing. Neuron 64: 341-354.
Ruben RJ, Sidman RL. 1967. Serial section radioautography of the inner ear.
Histological technique. Archives of otolaryngology 86: 32-37.
Sekiya T, Zaret KS. 2007. Repression by Groucho/TLE/Grg proteins: genomic site
recruitment generates compacted chromatin in vitro and impairs activator binding
in vivo. Molecular cell 28: 291-303.
Seto E, Yoshida M. 2014. Erasers of histone acetylation: the histone deacetylase
enzymes. Cold Spring Harbor perspectives in biology 6: a018713.
Shahbazi J, Scarlett CJ, Norris MD, Liu B, Haber M, Tee AE, Carrier A, Biankin AV,
London WB, Marshall GM et al. 2014. Histone deacetylase 2 and N-Myc reduce
p53 protein phosphorylation at serine 46 by repressing gene transcription of
tumor protein 53-induced nuclear protein 1. Oncotarget 5: 4257-4268.
Shi F, Cheng YF, Wang XL, Edge AS. 2010. Beta-catenin up-regulates Atoh1
expression in neural progenitor cells by interaction with an Atoh1 3' enhancer. J
Biol Chem 285: 392-400.

142

Shi F, Hu L, Jacques BE, Mulvaney JF, Dabdoub A, Edge AS. 2014. beta-Catenin is
required for hair-cell differentiation in the cochlea. The Journal of neuroscience :
the official journal of the Society for Neuroscience 34: 6470-6479.
Shim K, Minowada G, Coling DE, Martin GR. 2005. Sprouty2, a mouse deafness gene,
regulates cell fate decisions in the auditory sensory epithelium by antagonizing
FGF signaling. Developmental cell 8: 553-564.
Shimojo H, Ohtsuka T, Kageyama R. 2008. Oscillations in notch signaling regulate
maintenance of neural progenitors. Neuron 58: 52-64.
Soderberg O, Gullberg M, Jarvius M, Ridderstrale K, Leuchowius KJ, Jarvius J, Wester
K, Hydbring P, Bahram F, Larsson LG et al. 2006. Direct observation of individual
endogenous protein complexes in situ by proximity ligation. Nature methods 3:
995-1000.
Stone JS, Cotanche DA. 2007. Hair cell regeneration in the avian auditory epithelium.
The International journal of developmental biology 51: 633-647.
Sun J, Kamei CN, Layne MD, Jain MK, Liao JK, Lee ME, Chin MT. 2001. Regulation of
myogenic terminal differentiation by the hairy-related transcription factor CHF2. J
Biol Chem 276: 18591-18596.
Sun Y, Jan LY, Jan YN. 1998. Transcriptional regulation of atonal during development
of the Drosophila peripheral nervous system. Development 125: 3731-3740.
Taelman V, Van Wayenbergh R, Solter M, Pichon B, Pieler T, Christophe D, Bellefroid
EJ. 2004. Sequences downstream of the bHLH domain of the Xenopus hairy-
related transcription factor-1 act as an extended dimerization domain that
contributes to the selection of the partners. Developmental biology 276: 47-63.
Takata T, Ishikawa F. 2003. Human Sir2-related protein SIRT1 associates with the
bHLH repressors HES1 and HEY2 and is involved in HES1- and HEY2-mediated
transcriptional repression. Biochem Biophys Res Commun 301: 250-257.
Tateya T, Imayoshi I, Tateya I, Ito J, Kageyama R. 2011. Cooperative functions of
Hes/Hey genes in auditory hair cell and supporting cell development.
Developmental biology 352: 329-340.
Taunton J, Hassig CA, Schreiber SL. 1996. A mammalian histone deacetylase related
to the yeast transcriptional regulator Rpd3p. Science 272: 408-411.
Thayer MJ, Tapscott SJ, Davis RL, Wright WE, Lassar AB, Weintraub H. 1989. Positive
autoregulation of the myogenic determination gene MyoD1. Cell 58: 241-248.
Thisse B, Thisse C. 2005. Functions and regulations of fibroblast growth factor signaling
during embryonic development. Developmental biology 287: 390-402.
Troost T, Schneider M, Klein T. 2015. A re-examination of the selection of the sensory
organ precursor of the bristle sensilla of Drosophila melanogaster. PLoS genetics
11: e1004911.
Tu X, Chen J, Lim J, Karner CM, Lee SY, Heisig J, Wiese C, Surendran K, Kopan R,
Gessler M et al. 2012. Physiological notch signaling maintains bone homeostasis
via RBPjk and Hey upstream of NFATc1. PLoS genetics 8: e1002577.
Vaquero A, Scher M, Lee D, Erdjument-Bromage H, Tempst P, Reinberg D. 2004.
Human SirT1 interacts with histone H1 and promotes formation of facultative
heterochromatin. Molecular cell 16: 93-105.

143

Vaquero A, Scher MB, Lee DH, Sutton A, Cheng HL, Alt FW, Serrano L, Sternglanz R,
Reinberg D. 2006. SirT2 is a histone deacetylase with preference for histone H4
Lys 16 during mitosis. Genes & development 20: 1256-1261.
Verdone L, Caserta M, Di Mauro E. 2005. Role of histone acetylation in the control of
gene expression. Biochemistry and cell biology = Biochimie et biologie cellulaire
83: 344-353.
Vermeulen M, Carrozza MJ, Lasonder E, Workman JL, Logie C, Stunnenberg HG.
2004. In vitro targeting reveals intrinsic histone tail specificity of the Sin3/histone
deacetylase and N-CoR/SMRT corepressor complexes. Molecular and cellular
biology 24: 2364-2372.
Villalba JM, Alcain FJ. 2012. Sirtuin activators and inhibitors. BioFactors 38: 349-359.
Wan G, Corfas G, Stone JS. 2013. Inner ear supporting cells: rethinking the silent
majority. Seminars in cell & developmental biology 24: 448-459.
Wang F, Nguyen M, Qin FX, Tong Q. 2007. SIRT2 deacetylates FOXO3a in response
to oxidative stress and caloric restriction. Aging cell 6: 505-514.
Wang Q, Zhang Y, Yang C, Xiong H, Lin Y, Yao J, Li H, Xie L, Zhao W, Yao Y et al.
2010. Acetylation of metabolic enzymes coordinates carbon source utilization
and metabolic flux. Science 327: 1004-1007.
Wang RH, Sengupta K, Li C, Kim HS, Cao L, Xiao C, Kim S, Xu X, Zheng Y, Chilton B
et al. 2008a. Impaired DNA damage response, genome instability, and
tumorigenesis in SIRT1 mutant mice. Cancer cell 14: 312-323.
Wang VY, Hassan BA, Bellen HJ, Zoghbi HY. 2002. Drosophila atonal fully rescues the
phenotype of Math1 null mice: new functions evolve in new cellular contexts.
Curr Biol 12: 1611-1616.
Wang Z, Zang C, Rosenfeld JA, Schones DE, Barski A, Cuddapah S, Cui K, Roh TY,
Peng W, Zhang MQ et al. 2008b. Combinatorial patterns of histone acetylations
and methylations in the human genome. Nature genetics 40: 897-903.
Weber D, Heisig J, Kneitz S, Wolf E, Eilers M, Gessler M. 2015. Mechanisms of
epigenetic and cell-type specific regulation of Hey target genes in ES cells and
cardiomyocytes. Journal of molecular and cellular cardiology 79: 79-88.
Wharton KA, Johansen KM, Xu T, Artavanis-Tsakonas S. 1985. Nucleotide sequence
from the neurogenic locus notch implies a gene product that shares homology
with proteins containing EGF-like repeats. Cell 43: 567-581.
White PM, Doetzlhofer A, Lee YS, Groves AK, Segil N. 2006. Mammalian cochlear
supporting cells can divide and trans-differentiate into hair cells. Nature 441: 984-
987.
Winkler CJ, Ponce A, Courey AJ. 2010. Groucho-mediated repression may result from a
histone deacetylase-dependent increase in nucleosome density. PloS one 5:
e10166.
Witt O, Deubzer HE, Milde T, Oehme I. 2009. HDAC family: What are the cancer
relevant targets? Cancer letters 277: 8-21.
Woods C, Montcouquiol M, Kelley MW. 2004. Math1 regulates development of the
sensory epithelium in the mammalian cochlea. Nature neuroscience 7: 1310-
1318.
Wright TJ, Mansour SL. 2003. Fgf3 and Fgf10 are required for mouse otic placode
induction. Development 130: 3379-3390.

144

Xu PX, Adams J, Peters H, Brown MC, Heaney S, Maas R. 1999. Eya1-deficient mice
lack ears and kidneys and show abnormal apoptosis of organ primordia. Nature
genetics 23: 113-117.
Yamamoto N, Chang W, Kelley MW. 2011. Rbpj regulates development of prosensory
cells in the mammalian inner ear. Developmental biology 353: 367-379.
Yamamoto N, Tanigaki K, Tsuji M, Yabe D, Ito J, Honjo T. 2006. Inhibition of
Notch/RBP-J signaling induces hair cell formation in neonate mouse cochleas. J
Mol Med 84: 37-45.
Yang J, Cong N, Han Z, Huang Y, Chi F. 2013. Ectopic hair cell-like cell induction by
Math1 mainly involves direct transdifferentiation in neonatal mammalian cochlea.
Neuroscience letters 549: 7-11.
Ying QL, Smith AG. 2003. Defined conditions for neural commitment and differentiation.
Methods Enzymol 365: 327-341.
Ying QL, Stavridis M, Griffiths D, Li M, Smith A. 2003. Conversion of embryonic stem
cells into neuroectodermal precursors in adherent monoculture. Nat Biotechnol
21: 183-186.
Yoshida M, Kijima M, Akita M, Beppu T. 1990. Potent and specific inhibition of
mammalian histone deacetylase both in vivo and in vitro by trichostatin A. J Biol
Chem 265: 17174-17179.
Yoshida M, Nomura S, Beppu T. 1987. Effects of trichostatins on differentiation of
murine erythroleukemia cells. Cancer research 47: 3688-3691.
Yu X, Li P, Roeder RG, Wang Z. 2001. Inhibition of androgen receptor-mediated
transcription by amino-terminal enhancer of split. Molecular and cellular biology
21: 4614-4625.
Zhang X, Wharton W, Yuan Z, Tsai SC, Olashaw N, Seto E. 2004. Activation of the
growth-differentiation factor 11 gene by the histone deacetylase (HDAC) inhibitor
trichostatin A and repression by HDAC3. Molecular and cellular biology 24: 5106-
5118.
Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C,
Myers RM, Brown M, Li W et al. 2008. Model-based analysis of ChIP-Seq
(MACS). Genome biology 9: R137.
Zheng JL, Gao WQ. 2000. Overexpression of Math1 induces robust production of extra
hair cells in postnatal rat inner ears. Nature neuroscience 3: 580-586.
Zheng JL, Shou J, Guillemot F, Kageyama R, Gao WQ. 2000. Hes1 is a negative
regulator of inner ear hair cell differentiation. Development 127: 4551-4560.
Zheng W, Huang L, Wei ZB, Silvius D, Tang B, Xu PX. 2003. The role of Six1 in
mammalian auditory system development. Development 130: 3989-4000.
Zheng X, Tsuchiya K, Okamoto R, Iwasaki M, Kano Y, Sakamoto N, Nakamura T,
Watanabe M. 2011. Suppression of hath1 gene expression directly regulated by
hes1 via notch signaling is associated with goblet cell depletion in ulcerative
colitis. Inflammatory bowel diseases 17: 2251-2260.
Zine A, Aubert A, Qiu J, Therianos S, Guillemot F, Kageyama R, de Ribaupierre F.
2001. Hes1 and Hes5 activities are required for the normal development of the
hair cells in the mammalian inner ear. The Journal of neuroscience : the official
journal of the Society for Neuroscience 21: 4712-4720.

145

Appendix

Cross-sections through postnatal organ of Corti and their schematics illustrating
the expression pattern of different transgenic mouse lines used in this study.

Abstract (if available)

Abstract The organ of Corti is an orderly mosaic of sensory hair cells and supporting cells that relies on the precise regulation of the bHLH transcription factor ATOH1 and its relation to Notch-dependent patterning to establish its intricate structure and function. However, the detailed molecular mechanisms responsible for Atoh1 regulation and the subsequent maintenance of cell fate within the mosaic are poorly understood. Using a transgenic mouse harboring a mutant Atoh1 promoter-reporter, we show that, rather than upregulating in selected hair cells, Atoh1 is first induced in prosensory progenitors, then silenced in nascent supporting cells through newly characterized HES/HEY binding motifs in the Atoh1 promoter. We also show that the Notch effector HES5 interacts with the co-repressor GRG/TLE at these sites, correlating with Notch-dependent silencing of Atoh1 by histone deacetylation. Our data also indicate that relief from repression is sufficient to drive promoter-dependent Atoh1 expression during perinatal supporting cell transdifferentiation, suggesting a mechanism for priming Atoh1 expression and eliciting a switch-like function of its autoregulatory enhancer. Finally, our results demonstrate the importance of histone deacetylase activity for maintaining supporting cell fate and suggest that a combination of both NAD⁺-dependent and -independent histone deacetylases mediate repression of Atoh1 in HEY2-expressing supporting cells.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Transcriptional and epigenetic mechanisms underlying sensory hair cell differentiation and regeneration

PDF

The organ of Corti sensory progenitor cell competence is controlled by SoxC transcription factors

PDF

DNA methylation in the mouse cochlea promotes maturation of supporting cells and contributes to the failure of hair cell regeneration

PDF

The epigenetic landscapes underlying differentiation and plasticity in the developing organ of Corti

PDF

The calcium-sensing receptor in the specification of normal and malignant hematopoietic cell localization in the bone marrow microenvironment

PDF

The role of Prkci in stem cell maintenance and cell polarity using a 3-D culture system

PDF

Role of beta-catenin in mouse epiblast stem cell, embryonic stem cell self-renewal and differentiation

PDF

The molecular mechanism regulating the timing of cell cycle exit during development of the mouse organ of Corti

PDF

Development and evolution of skeletal joints: lessons learned from studying zebrafish

PDF

The role of neuregulin receptors in cell differentiation and the response to inflammatory cytokines in the intestinal epithelium

PDF

Jagged-notch signaling: patterning the vertebrate face

PDF

Utilizing zebrafish and mouse models to uncover the underlying genetics of human craniofacial anomalies

PDF

Investigation of the molecular mechanisms of ototoxicity

PDF

Jagged1 functions downstream of Twist1 in the specification of the coronal suture and the formation of a boundary between osteogenic and non-osteogenic cells

PDF

Characterization of new stem/progenitor cells in skin appendages

PDF

Identification and characterization of adult stem cells in the oral cavity

PDF

Transcriptional regulation of IFN-γ and PlGF in response to Epo and VEGF in erythroid cells

PDF

Molecular basis of mouse epiblast stem cell and human embryonic stem cell self‐renewal

PDF

The mechanism by which extracellular Hsp90α promotes cell migration: implications in wound healing and cancer

PDF

Pleotropic potential of Stat3 in determining self-renewal, apoptosis, and differentiation in mouse embryonic stem cells

Asset Metadata

Creator Abdolazimi, Yassan (author)

Core Title The role of HES/HEY transcriptional repressors in specification and maintenance of cell fate in the mouse organ of Corti

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Genetic, Molecular and Cellular Biology

Publication Date 07/24/2015

Defense Date 06/11/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Atoh1,cochlear development,hair cell,Hes5,Hey2,OAI-PMH Harvest,organ of Corti,Regeneration,Sirt2,supporting cell,transdifferentiation

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Segil, Neil (committee chair), Crump, Gage D. (committee member), Kobielak, Krzysztof (committee member), Maxson, Robert E., Jr. (committee member)

Creator Email abdolazi@usc.edu,yassan.ab@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-605829

Unique identifier UC11300197

Identifier etd-Abdolazimi-3696.pdf (filename),usctheses-c3-605829 (legacy record id)

Legacy Identifier etd-Abdolazimi-3696.pdf

Dmrecord 605829

Document Type Dissertation

Format application/pdf (imt)

Rights Abdolazimi, Yassan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA