University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

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

by

Talon Trecek

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
(DEVELOPMENT, STEM CELLS AND REGENERATIVE MEDICINE)

May 2023

ii

Dedication
To my parents Nina Matthews, Chris Weiss, and Brad Trecek.
For their patience, unconditional support, and belief in in me.

In memory and dedication to Dr. Neil Segil and Dr. Unaiza Hayat.
I remain guided by their outstanding character.

iii

Acknowledgements
My journey through this PhD would be unimaginable without the friends and
mentors I had along the way, but none more so than my mentor Dr. Neil Segil. Many
mentors claim an open-door policy, but Dr. Neil Segil lived it. Neil had an unparalleled
ability to make every idea you had feel interesting and worthy of discussion. This
fostered an incredible atmosphere that made our three-hour weekly lab meeting, and
our often two hour weekly personal meetings feel far too short.
I would also like to thank the current and former members of the Segil lab. The
culture of friendship in our lab is unparalleled; it is the result of Neil’s nature and the
people he brought together. Our daily morning coffees and lunches put hardships in
perspective. To Dr. Litao Tao, thank you for serving as what I can only describe as my
second PhD mentor. Your optimistic, calm, and level-headed nature brought balance to
the entire lab. Everything seems possible and worth pursuing with both you and Neil. To
Juan Llamas, absolutely none of this work would have been possible without your skill,
support, and time. To both Juan and Welly Makmura, thank you for your help with
mouse work. The experience I developed in the lab for my career would not have been
possible without your help.
One of the reasons I joined Neil’s lab was because he could understand my
weaknesses and the challenges I would face. I thank my committee’s patience and
understanding while working with me in Neil’s place. Especially Dr. Andrew Groves &
Dr. Andy McMahon, who have taken on an incredible task of making sure the members
of the Segil lab find their way through all this.
iv

Table of Contents
Dedication ....................................................................................................................... ii
Acknowledgements ......................................................................................................... iii
List of Figures .................................................................................................................. v
Abstract .......................................................................................................................... vii
Chapter 1: Background ................................................................................................... 1
Hearing Loss and Lack of Hair Cell Regeneration ....................................................... 1
Structure of the inner ear ............................................................................................. 2
Inner ear development ................................................................................................. 4
Development of the organ of Corti ............................................................................... 6
Differentiation and trans-differentiation of sensory cells .............................................. 9
Epigenetics of plasticity and capacity ........................................................................ 10
Epigenetics of the inner ear ....................................................................................... 13
Chapter 2: Expression dynamics during early differentiation ......................................... 17
Introduction ................................................................................................................ 17
Transcriptional profiling of the differentiating prosensory domain .............................. 17
Single cell RNAseq profiling of differentiating progenitors ......................................... 22
Co-expression of cell-fate determinants coincides with initiation of differentiation at
the base. .................................................................................................................... 25
Summary Conclusion ................................................................................................. 26
Chapter 3: Epigenetic dynamics during prosensory differentiation ................................ 28
Introduction ................................................................................................................ 28
Hair cell enhancer network is primed in supporting cells but not progenitors ............ 28
Supporting cells fail to activate primed hair cell enhancers ....................................... 30
Progenitor enhancers are decommissioned in hair cells but not supporting cells ...... 31
Transdifferentiation associated enhancer predominately arise during or after
differentiation ............................................................................................................. 33
Common primed enhancer between hair cells and supporting cells enrich for TFs
associated with cell fate potential .............................................................................. 34
Summary & Conclusion ............................................................................................. 39
Chapter 4: A role for Prc1 and Prc2 in cell-fate plasticity .............................................. 40
Atoh1 is epigenetically de-repressed in early supporting cells. .................................. 40
H2AK119ub in maintain in absence of H3K27me3 .................................................... 41
Loss of H3K27me3 at Atoh1 occurs during onset of differentiation ........................... 42
Summary and Conclusion .......................................................................................... 47
Chapter 5: Discussion & Future Directions .................................................................... 48
Factor X ..................................................................................................................... 49
Prc1 bookmarking for H3K27me3? ............................................................................ 50
References .................................................................................................................... 52
Appendix I: Materials and Methods ............................................................................... 64

v

List of Figures
Chapter 1 :
1.1. Structures of the inner ear………………………………………………………..........4
1.2. Development of inner ear……………………………………………….......…………5
1.3. Cochlear duct extension and patterning…………………………..……….....………6
1.4. Waves of cell-cycle exit and differentiation…………………………………..…....…8
1.5. Transdifferentiation of supporting cells………………………………………......….10
1.6. Histones and enhancer states………………………………………………....…….12
1.7. Pioneer factors…………………………………………………………………..……..12
1.8. Enhancer decommissioning in supporting cells………………………………...….14
1.9. The Atoh1-Pou4f3 feedforward circuit……………………………....………………15
1.10. Experimental design for NGS assays……………………………………….....……16
1.11. CUT&RUN profiling of histone marks……………………………………….....……16

Chapter 2 :
2.1. Hair cells and supporting cells share common transcriptomic signature…......…19
2.2. Low Atoh1-fGFP+ population enriches for supporting cell markers………......…21
2.3. Sox2-CreER lineage tracing of prosensory domain………………………............22
2.4. Annotation of single-cell RNA-seq clusters..........................................................24
2.5. Regional expression markers..............................................................................25
2.6. Broad co-expression of competing cell-fate factors.............................................27

Chapter 3 :
3.1. Supporting cells acquire a hair cell like primed epigenetic landscape.................29
3.2. Enhancer activation dynamics.............................................................................32
vi

3.3. Supporting cells skew towards upregulation........................................................33
3.4. Motif signatures of activators and pioneers.........................................................36
3.5. Common de novo motifs......................................................................................37
3.6. Expression of NFI factors at E14.5......................................................................38

Chapter 4 :
4.1. Repressive chromatin dynamics..........................................................................44
4.2. Regional marker expression in single-cell Paired-Tag.........................................45
4.3. E14.5-E15.5 supporting cell de-repress Atoh1 locus...........................................46

Chapter 5 :
5.1. Discussion & Future Directions............................................................................51
vii

Abstract
Embryonic and perinatal supporting cells of the inner ear harbor a latent capacity
for regeneration of sensory hair cells, but this capacity is lost after birth. To understand
the epigenetic basis for how and when this plasticity is established, we mapped the
epigenetic and transcriptional changes occurring as cochlear progenitors differentiate
into hair cells and supporting cells. We purified both differentiating cell types and found
that initially, each cell type epigenetically de-represses regulators of both hair cells and
supporting cells by removing inhibitory H3K27me3 histone marks. This de-repression
occurs with co-expression of both hair cell and supporting cell fate factors in both cell
types. Both cell types acquire a common primed (H3K4me1) enhancer landscape that is
associated with hair cell fate and the transdifferentiation potential of supporting cells.
While supporting cells prime the hair cell enhancer landscape, they fail to epigenetically
activate (H3K27ac) these enhancers. Additionally, while early supporting cells remove
H3K27me3 from the Atoh1 locus, they maintain the repressive Prc1 mark H2K119ub1.
In contrast to hair cells, supporting cell maintain progenitor enhancers in either an active
or primed state. Understanding how early supporting cells prime the hair cell epigenetic
program without activation and its relation to transdifferentiation stands to provide a
novel regenerative approach to restoring sensory hair cells.
1

Chapter 1: Background
Hearing Loss and Lack of Hair Cell Regeneration
The loss of sensory hair cells in the mammalian organ of Corti is the primary
cause of sensorineural hearing loss, which do not regenerate (Corwin and Cotanche
1988; Cruickshanks et al. 2003; Brigande and Heller 2009). Around 17% of adults in the
U.S. experience hearing loss (Brigande and Heller 2009). Sensorineural hearing loss
involves various factors, including genetic mutations, aging, noise exposure, and drugs
that have ototoxic side effects (Cunningham and Tucci 2017). Half of all hearing loss is
caused by genetic mutations, with over 100 genes linked to non-syndromic hearing loss
(Morton and Nance 2006; Van Camp 2022). Acquired hearing loss, such as age-related
hearing loss and noise-induced hearing loss, is caused by mechanical stress on the
cochlea. Hearing aids and cochlear implants remain the primary treatment options for
hearing loss. However, hearing aids only benefit those with mild to moderate hearing
loss and rely on the remaining sensory hair cells and intact auditory circuitry. For
individuals with severe and total sensorineural hearing loss, cochlear implants are the
only rehabilitation option, but they do not restore normal hearing. In addition, they have
much lower success in adults compared to children, as patients must essentially learn a
new language due to how different speech sounds.
In contrast to mammals, birds can regenerate functional hair cells through
supporting cell re-entry into the cell cycle and/or direct transdifferentiation of the
supporting cell into a new hair cell (Ryals and Rubel 1988; Corwin and Cotanche 1988).
Although hair cell regeneration does not occur naturally in mature mammals, under
experimental conditions perinatal supporting cells have a transient capacity to
2

transdifferentiate into hair cells in response to Atoh1 or loss of Notch-mediated lateral
inhibition (Kelly et al. 2012; Z. Liu et al. 2012; Mizutari et al. 2013; Maass et al. 2015).
However, this capacity is lost by postnatal day 6 (White et al. 2006; Takebayashi et al.
2007; Doetzlhofer et al. 2009; Z. Liu et al. 2012; Cox et al. 2014; Bramhall et al. 2014).
Trying to understand how this regenerative capacity is lost and ways to restore it has
been a major focus for the regenerative field.
Structure of the inner ear
The inner ear is responsible for sense of balance and hearing, which it achieves
through two distinct components: the dorsally positioned vestibular system and the
ventral auditory system (Fig. 1.1A). The vestibular system is responsible for sense of
balance and orientation. It is comprised of three semicircular canals, the utricle, and the
saccule. The auditory system is responsible for the perception of sound and is made up
of the cochlea, a snail-shaped tube that wraps around the spiral ganglion, that it is
innervated by the interior of the cochlea is divided into three fluid filled chambers: the
scala vestibuli, scala media, and scala tympani (Fig. 1.1B). The Reissner’s membrane
separates the scala vestibuli and scala media chambers. The scala media and scala
tympani are separated by the hearing organ, the organ of Corti. The scala media, also
known as the cochlear duct, is filled with endolymph, a fluid with high K+ and low Na+
concentrations. The scala vestibuli and scala tympani, are filled with perilymph, with
normal concentrations of K+ and Na+. These concentration differences between the
fluid filled chambers are responsible for maintaining the endolymphatic potential of the
scala media (Abraham and Kierszenbaum 2007; Durrant and Lovrinic 1995) (Fig. 1.1B).
3

The organ of Corti is composed of sensory hair cells and their non-sensory supporting
cells that underlie them (Fig. 1.1C). When sound waves travel through the auditory
canal, to the tympanic membrane, the ossicles of the middle ear transfer these
vibrations to the fluid filled space of the inner ear. The wave then travels through the
scala tympani, displacing the basement membrane that the organ of Corti sits on. Due
to the variable stiffness of the basement membrane, specific sound frequencies
resonate at specific point along the length of the cochlea. This displacement is amplified
by the outer hair cells, which contract and expand with the displacement of the
membrane. This amplified forced leads to a shearing force towards inner hair cells
causing deflection of stereocilia bundles found on the luminal surface of hair cells. This
exposes mechanotransduction channels, causing an influx of K+ from the scala media,
leading to depolarization and release of neurotransmitters at the synapse where
innervating neurons from the spiral ganglion will transmit the signal through the VIII
4

cranial ganglion (Purves et al. 2001; Dror and Avraham 2010; Brigande and Heller
2009).

Figure 1.1: Structures of the inner ear A) Schematic of the ear showing the external ear, middle ear,
and inner ear. The inner ear contains the vestibule organs and the cochlea. B) The cochlear duct includes
three canals (or ‘scala’): the scala vestibuli, the scala media and the scala tympani. The scala media is
comprised of three walls: Reissner’s membrane (RM), the stria vascularis and spiral ligament (SV), and
the cochlear floor, which contains the sensory organ of Corti (OC) flanked by, the inner sulcus (IS) and
outer sulcus (OS). C) The organ of Corti contains two types of hair cells (inner hair cells and outer hair
cells) and several different types of unique supporting cell types, including inner phalangeal cells adjacent
to the inner hair cells, pillar cells separating the inner and outer hair cells, Dieters’ cells interdigitated
among outer hair cells, and Hensen’s cells lateral to the organ of Corti. Hair cells have characteristic
stereocilia bundles on their luminal surface, and the stereocilia of the outer hair cells are in contact with
the tectorial membrane (TM) within the scala media. Neurons of the spiral ganglion (SG) synapse with
hair cells and project centrally to the cochlear nucleus. Panel A adapted from (Dror and Avraham 2010).
Panel B adapted from (Elizabeth Carroll Driver and Kelley 2020)

Inner ear development
The inner ear arises from a thickening
epithelium near the hindbrain known as the
otic placode between E8-8.5 (Le Douarin
1984; Morsli et al. 1998). By E9.5 the otic
placode invaginates to form the otic cup/pit,
which then pinches off from the ectoderm to
form the otic vesicle, also known as the
B
C
5

otocyst (Haddon and Lewis 1996; Fritzsch et al. 2002; Kelley 2006; Abelló and Alsina
2007; Sánchez-Calderón et al. 2007). Neuroblast from the ventral region of the otocyst
delaminate and migrate to form the VIII cranial ganglion: Comprised of both the
cochlear nerve and vestibular nerve, which innervate sensory hair cells throughout inner
ear from the spiral ganglion (Rubel and Fritzsch 2002; Elizabeth C. Driver and Kelley
2009; Satoh and Fekete 2005). The otocyst then undergoes rapid morphological
changes, which results in the formation of the two major sensory systems of the inner
ear (Morsli et al. 1998; Kelley 2006). 1) the dorsal vestibular structures comprising of
the semi-circular canals, utricle, and saccule, which are responsible for sense of
balance and orientation and the 2) ventral auditory structure comprised of the cochlea,
which detects sound (Fig 1.2A).

Figure 1.2 Development of inner ear: Schematic of early stages of inner ear development from E9.5
otocysts to the development of pro-sensory patches and their supporting structures. Adapted from (Kelley
2006).
6

Development of the organ of Corti
At E11, the ventral portion of the inner ear, the cochlear duct, begins undergoing
extension (Fig. 1.3A) (Elizabeth Carroll Driver and Kelley 2020). By E12.5 the cochlear
duct can be divided into five distinct regions: 1) Stria vascularis 2) Future Reissner’s
membrane 3) Kolliker’s organ 4) Prosensory domain and 5) Future outer sulcus (Fig
1.3A). For orientation: the medial side is situated closest to the spiral ganglion, referred
to as the neural side. Kolliker’s organ and Reissner’s membrane are located on the
neural side. The lateral side is the furthest from the neural side, referred to as abneural.
Future outer sulcus and stria vascularis are located on the abneural side. Stria
vascularis and future Reissner’s membrane are considered the roof with the other
regions composing the floor.
Figure 1.3: Cochlear duct extension and patterning Schematic depicting morphological changes
occurring with cochlear duct extension between E11 to E14. On the right, a diagram depicting the
regional identities of the cochlear duct epithelium. Reissner’s membrane (RM, light blue) and stria
vascularis (SV, beige) are consider the roof of the cochlear duct. On the medial side, the inner sulcus
(IS), also referred to as Kolliker’s organs (KO, pink), resides closet to the spiral ganglion (SG, blue). The
future outer sulcus (OS, orange) is found on the lateral side, also called the lower epithelial ridge (LER).
The organ of Corti (OC, green), also called the prosensory (PS) domain, is located between KO and OS.
Adapted from (Elizabeth Carroll Driver and Kelley 2020).

7

Beginning at ~E12.5, prosensory progenitors at the apex exit cell-cycle, marked by
expression of the cell-cycle inhibitor p27Kip1, also known as Cdkn1b, cell-cycle exit
then continues within the prosensory domain traveling from apex-to-base (Fig 1.4A &
B). Expression of p27Kip1 is necessary for cell-cycle exit of prosensory progenitors (P.
Chen and Segil 1999). P27Kip1 expression continues to spread in a wave throughout
the entirety of the prosensory domain from apex to base between E12.5-E14.5 (Fig
1.4B). The prosensory domain and their differentiated progeny will remain permanently
post-mitotic (Lee, Liu, & Segil, 2006).
At around E12, Sox2-expressing progenitors in the prosensory domain are
specified to become either hair cells or supporting cells (Kelley 2007). The mechanisms
driving this specification are unknown, but they are dependent on surrounding
mesenchyme at E12.5. At E12.5, the presence of the surrounding mesenchyme is
necessary for eventual hair cell differentiation. At E13.5, all progenitors will differentiate
into hair cells or supporting cells with or without mesenchyme (Montcouquiol and Kelley
2003). The dynamics of the differentiation wave also appear intrinsic to the epithelium,
as even when cochlea’s are cut in pieces and cultured apart, the differentiation wave is
not affected (Montcouquiol and Kelley 2003).
At E14.5, some post-mitotic prosensory progenitors at the base begin expressing
the proneural transcription factor atonal homolog 1, Atoh1. Atoh1 is the earliest marker
for hair cells and is both necessary and sufficient for hair cell differentiation
(Bermingham et al. 1999). Atoh1 expression in post-mitotic prosensory progenitors
spreads from the base to the apex between E14.5 to E17.5 (Fig 1.4C). Inner hair cells
are the first to express Atoh1, with outer hair cell Atoh1 expression lagging. The delay of
8

Atoh1 expression in outer hair cells is dependent on sonic hedgehog signaling from the
spiral ganglion (Bok et al. 2013). The earliest marker of supporting cell differentiation is
Lfng expression on the neural side boundary with Kolliker’s organ. The first Lfng
expressing cells in the prosensory domain become either inner hair cells or inner
phalangeal cells. Inner hair cells rapidly downregulate Lfng, leading to expression only
being seen at leading edge of differentiation wave (Basch et al. 2016). On the abneural
side, differentiating supporting cells are first marked by Prox1 and Lfng expression. Like
outer hair cells, abneural supporting cell marker expression is delayed relative to the
more neural positioned supporting cells. By E17.5 both hair cells and supporting cells
are both morphologically and molecularly distinct from base to apex.

Figure 1.4: Waves of cell-cycle
exit and differentiation (A)
Dissected out cochlea stained with
BrDu (red), showing a region of
non-proliferation at the apex at
E12.5, followed by expansion of that
region from apex to base by E14.5.
(B) Staining of cell-cycle inhibitor
p27Kip1 (green). Expression occurs
in spreading wave corresponding to
the zone of non-proliferation in (A).
Adapted from (Y.-S. Lee, Liu, and
Segil 2006).

9

Differentiation and trans-differentiation of sensory cells
Prosensory progenitors differentiate into both hair cells and supporting cells
simultaneously through Notch-mediated lateral inhibition. It’s important to note that
Notch only regulates selection of cell fate and not specification of cell fate or initiation of
differentiation. Genetically perturbing Notch signaling does not prevent or alter the onset
of differentiation, only cell-fate choice, and size of prosensory domain (Martín L. Basch
et al. 2011). Prior to Atoh1 up-regulation, prosensory progenitors are bipotent in their
ability to activate both hair cell and supporting cell programs, however, Notch-mediated
lateral inhibition rapidly switches on to ensure mutually exclusive activation of a single
cell-fate (Lanford et al. 1999). When Atoh1 is expressed in early differentiating hair
cells, it upregulates Notch ligands Jag2 and Dll1 that then activate Notch signaling in
neighboring cells. Hey/Hey factors are upregulated by cleaved intracellular Notch
receptor (NICD) and Rbpj, which then directly bind to and repress the Atoh1 promoter
(Fig. 1.5A) (Abdolazimi, Stojanova, and Segil 2016; Zine et al. 2001; Zheng et al. 2000).
Without Atoh1, these cells become supporting cells. Active Notch signaling is necessary
for maintaining supporting cell fate up until post-natal day 6 (P6). After P6, supporting
cells are no longer capable of transdifferentiation into hair cells through Notch inhibition
or Atoh1 overexpression (Fig. 1.5B) (Z. Liu et al. 2012; Kelly et al. 2012; Maass et al.
2015). Interestingly, In non-mammalian species, supporting cells never lose their
capacity to transdifferentiate into hair cells through Notch signaling, allowing
regeneration of hair cells throughout their lifespan (Roberson, Alosi, and Cotanche
2004; Duncan et al. 2006; Hj and Y 1996).
10

Figure 1.5: Transdifferentiation of supporting cells: (A) Diagram depicting a Atoh1 expressing hair cell
(orange) upregulating Notch ligands that go on to activate Notch signaling in adjacent supporting cells
(green) leading to Hes & Hey expression. Hes & Hey factors then bind to the Atoh1 promotor and repress
its transcription. (B) Top-down view of P1 cochlear explants with a control organ on left and Notch
inhibitor (DAPT) treated organ on right. Hair cells show with Atoh1-GFP (Green) and supporting cells with
Prox1 (red). The control has all four rows of hair cells and underlying supporting cells. In the Notch
inhibited organ on the right, there is an increase of Atoh1 positive cells at the expense of the Prox1
expressing cells. Panel B adapted from (Abdolazimi, Stojanova, and Segil 2016).

Epigenetics of plasticity and capacity
While adult mammals lack the regenerative capacity seen in non-mammalian
species, during embryonic and neonatal periods, many tissues harbor some capacity for
regeneration. This plasticity has been linked to the developmental sequences of
induction, specification, and restriction that remodel a cell’s epigenetic landscape as
they progress toward their terminal fate within their lineage (Hemberger, Dean, and Reik
2009). As a result of these processes, cells with a common developmental history retain
a greater capacity for initiating and activating each other’s cell-fate programs. This is
especially true for the few examples of physiological transdifferentiation in the liver,
pancreas, and possibly in the kidney (Gk, L, and Wc 2005; Y et al. 2009; K et al. 2013;
Thorel et al. 2010; Habener and Stanojevic 2012; Assmus et al. 2020). It has been
proposed that loss of the epigenetic memory of these developmental events during
maturation is a major mechanism in the loss of regenerative capacity in adult mammals
11

(Ong and Corces 2012; Yang and Kang 2019; Kang et al. 2016; Nicetto and Zaret 2019;
Goldman and Poss 2020; VandenBosch et al. 2020).
The epigenetic status of histone 3 lysine 27 (H3K27) and histone 3 lysine 4
(H3K4) is highly predictive of the regulatory activity of enhancers and promoters.
Acetylated H3K27 (H3K27ac) with monomethylated H3K4 is highly predictive of active
elements such as promoters and enhancers (Fig. 1.6A) (Creyghton et al. 2010; Zentner,
Tesar, and Scacheri 2011). H3K4me1 alone is considered a primed enhancer mark (Fig
1.6B). Primed enhancers are those that have some of the features an active enhancer
has, but without active marks. H3K4me1 is a priming mark as it is dispensable for
maintenance of active enhancers, but necessary for enhancer activation through de
novo acetylation of H3K27 during cell-fate transition (C. Wang et al. 2016; Bleckwehl et
al. 2021; K. Lee et al. 2019). During development, H3K4me1 status is predictive of
lineage competency and enhancers that will become active in downstream fate
decisions (A. Wang et al. 2015). Thus, priming of enhancer networks are critical for
developmental transitions that require activation of new enhancer networks. The
establishment of primed enhancers is mediated by pioneer factors. Pioneer factors are
transcription factors capable of recognizing and recruiting regulatory complexes to
inaccessible/unmarked chromatin to either prime or activate (Fig 1.7). This special class
of factors are often involved in establishing competency through de novo H3K4me1
priming, sometime referred to as pre-patterning.
12

Figure 1.6 Histones and enhancer states: (A) Schematic representation of the active and primed
enhancer features. Enhancers are associated with incorporation of hypermobile nucleosomes containing
H3.3/H2A.Z histone variants, which compete for DNA binding with TFs. TFs in turn recruit coactivator
proteins that can modify and remodel nucleosomes. H3K4me1 and H3K27ac are the predominant histone
modifications deposited at nucleosomes flanking active enhancer elements. (B) Prior to activation,
enhancers can exist in a primed state, characterized by the presence of H3K4me1. Other features that
have been associated with enhancer priming are presence of pioneer TFs, hypermobile H3.3/H2A.Z
nucleosomes, DNA 5mC hypomethylation, and hydroxylation (5hmC). (C) Example of de novo enhancer
activation. This type of activation results in direct and simultaneous H3K27 acetylation and H3K4
monomethylation. Panel (A) and (B) adapted from (Calo and Wysocka 2013). Panel (C) adapted from (L.-
H. Wang et al. 2021).

Figure 1.7 Pioneer factors: Diagram of pioneer factor targeting chromatin and remodeling nucleosomes
to allow other regulators to bind and impart their activity. This can include activators (green) or repressors
(red). Adapted from (Zaret and Mango 2016).
13

Epigenetics of the inner ear
Our understanding of epigenetic regulation in the inner ear is poor due to the
sparse numbers of cells and the difficulty of obtaining them. However, in recent years
new techniques have enabled profiling the epigenetic states of rare cells such as those
of the inner ear. In particular, genome-wide histone mark profiling with Cleavage Under
Targets and Release Using Nuclease (CUT&RUN), assay for transposable-accessible
chromatin (ATAC-seq), Cleavage Under Targets and Tagmentation (CUT&TAG) and
multi-omic profiling with Paired-Tag (Fig 1.10 & 1.11). These techniques have led to the
discovery of an epigenetic mechanism behind supporting cell transdifferentiation
potential and a common pioneer factor for mechnotranducing sensory cells.
Supporting cell transdifferentiation potential is in part due to the decommissioning of
hair cell enhancers in supporting cells by loss of the enhancer the priming mark
H3K4me1 (Fig. 1.8) (Tao et al. 2021). Postnatal day 1 (P1) supporting cells are capable
of transdifferentiating into hair cells upon Notch inhibition. At P6 this capacity is lost. In
this study, P1 supporting cell were found to keep a number of enhancers that are active
only in hair cells, in a primed state marked by H3K4me1. By P6, H3K4me1 priming at
these hair cell enhancers in supporting cells is lost. Culturing P1 cochlear explants with
an inhibitor that prevent removal of methyl from H3K4 by Lsd1, results in increased
transdifferentiation at P6, effectively expanding the window of transdifferentiation
potential (Tao et al. 2021). When, how, and what establishes H3K4me1 at hair cell
enhancers remains unknown. A factor that can prime the hair cell program, rather than
activate, holds immense potential as a regenerative approach to restoring hair cells, as
it would restore the capacity to respond and regenerate to an appropriate context.
14

Figure 1.8 Enhancer decommissioning
in supporting cells: Transdifferentiation
capable P1 supporting cells maintain hair
cells enhancers in a primed state, marked
by H3K4me1. By P6, H3K4me1 at these
sites is lost. Inhibition of H3K4me1
decommissioning by Lsd1 results in
increased transdifferentiation potential at
P6. Adapted from (Tao et al. 2021).

Pou4f3 acts as a pioneer factor for both sensory hair cells and Merkel cells of
touch dome receptors (H. V. Yu et al. 2021). In both sensory type cells, Atoh1
upregulates Pou4f3 through a pre-established enhancer. Through Pou4f3’s pioneering
activity, opening new enhancers enables Atoh1 binding and Atoh1-mediated activation
of new transcriptional targets (Fig. 1.9A & B) (H. V. Yu et al. 2021). What is responsible
for pre-establishing enhancers for Atoh1 to act on isn’t known. In the context of
supporting cell’s that have primed hair cells enhancers, it suggests more pioneer factors
are necessary for Atoh1’s function and potentially supporting cell transdifferentiation
potential.

15

Figure 1.9: The Atoh1-Pou4f3 feedforward circuit (A) Pou4f3 is a direct target of Atoh1. Atoh1 is up
regulated in the inner ear sensory progenitors. Atoh1 binds to pre-established enhancers upstream of the
Pou4f3 gene to stimulate the expression of Pou4f3. (B) Pou4f3 has pioneer factor activity. POU4F3 can
recognize its cognate motifs in nucleosome occupied DNA and is required for Atoh1 binding in closed
chromatin. The binding of Pou4f3 and Atoh1 does not result in the opening or the activation of the
regulatory elements in the closed chromatin. Adapted from (H. V. Yu et al. 2021).

With the recent findings that supporting cells prime hair cell enhancers and that
Atoh1 acts on pre-established enhancers, there may be a common factor driving
priming of the hair cell program throughout the sensory domain. We hypothesized that a
permissive epigenetic landscape for hair cell fate is established at, or after, the onset of
differentiation throughout the prosensory domain. We first characterized the
transcriptomic changes occurring early in differentiation by bulk RNA-seq of FACS-
purified progenitors, supporting cells, and hair cells (Fig 1.10). We found that supporting
cells express a number of hair cell associated genes at low levels. To confirm this, we
collected single-cell RNAseq of the E14.5 sensory epithelium and found that genes
associated with both hair cell and supporting cell fates are co-expressed at the base
where differentiation is starting. To characterize the epigenetic landscapes of the
differentiating sensory epithelium, we profiled RNA-seq along with epigenetic marks for
repressed (H3K27me3 & H2AK119ub), primed (H3K4me1), and active chromatin
16

(H3K27ac) with CUT&RUN on FACS-purified cell types before (E13.5), at (E14.5-
E15.5), and just following (E17.5) the onset of differentiation (Fig 1.10 & 1.11). These
assays are the groundwork for this thesis, a schematic of the workflow is provided in
(Fig. 1.10 & 1.11). For all bulk data, the same sorting strategy and same reporter lines
were used for all assays, as shown in (Fig. 1.10).
Figure 1.10 Experimental design for NGS assays: (A)
Purification (FACS) and genomic analysis of hair cells,
supporting cells, and sensory progenitors from mouse cochlea.
From top: inner ear (paint fill) mouse head; the spiral cochlea
expressing GFP; depiction of organ of Corti cross-section;
Graphic of cells labeled by each reporter line. From left to right:
Atoh1 fusion GFP (Atoh1-fGFP) labeling hair cells; Lfng GFP
labeling supporting cells; Cdkn1b-GFP labeling E13.5
progenitors. 5-10,000 cells sorted and used for next generation
sequencing (NGS) assays for histone profiling, transcription
factor profiling, expression, with bulk and single-cell. Adapted
from (Tao et al. 2021).

Figure 1.11: CUT&RUN profiling of histone
marks Cells are bound to magnetic concanavalin
beads to allow easy washing between steps. The
nuclease, MNase, is fused to Protein-A to target
antibodies for an epitope of interest. Once
bound, Ca++ is added, leading to digestion and
release of DNA fragments where epitope of
interest was bound. Sequenced fragments
represent where the cutting took place, allowing
inference of antibody target genomic location.
CUT&TAG is similar but uses Tn5 to ligate
adaptors instead of just cutting. Adapted from
(Skene and Henikoff 2017).
17

Chapter 2: Expression dynamics during early differentiation
Introduction
The basis for early supporting cell capacity for transdifferentiation and what
establishes it is poorly understood. While it may seem intuitive to think that
transdifferentiation potential is ‘carry over’ from a progenitor program, in the cochlea
progenitors don’t gain competency for hair cell differentiation until after E12.5
(Montcouquiol and Kelley 2003). Additionally, it’s not clear if supporting cells are a
distinct cell type from E13.5 progenitors, as they express similar marker genes such as
Sox2 and Cdkn1b/p27Kip1. Interestingly, early Cre lineage tracing studies of Atoh1
show many supporting cells had expressed Atoh1 early in differentiation (Elizabeth
Carroll Driver et al. 2013). Recently, a more sensitive lineage tracing strategy using a
homozygous Atoh1-HA-P2A-Cre mouse, showed the entire E14.5 prosensory domain
initiates Atoh1 expression (S. Li et al. 2022). Others have also reported co-expression
of supporting cell markers Prox1 and Lfng with Atoh1 between E14.5-E15.5 (Basch et
al. 2016; Bermingham-McDonogh et al. 2006). If supporting cells come from progenitors
that initiated the hair cell program, then perhaps this could explain their capacity for
transdifferentiation.

Transcriptional profiling of the differentiating prosensory domain
To understand the differentiation and expression dynamics of hair cells and
supporting cells, we first characterized the transcriptional changes during their
differentiation from cochlear progenitors. We performed bulk RNA-seq from FACS-
purified E13.5 Cdkn1b-GFP+ sensory progenitors, E17.5 Lfng-GFP+ supporting cells,
and E17.5 Atoh1-GFP+ hair cells (Fig. 1.8) and conducted a principal component
18

analysis to assess the transcriptional relationships between the three cell types (Fig.
2.1A). Strikingly, E17.5 hair cells and E17.5 supporting cells clustered together, away
from E13.5 PG on PC1 with 83% variance (Fig. 2.1A). Performing a differential
expression analysis comparing E17.5 hair cells to E13.5 sensory progenitors, and E17.5
supporting cells to E13.5 sensory progenitors, more than half of the genes enriched in
differentiating cells (~56%) were upregulated in both hair cells and supporting cells, in
agreement with the similarity observed in PCA (Fig. 2.1B). A gene ontology (GO)
analysis on this set of common up-regulated genes revealed enrichment for sensory-
and hair cell-related GO terms (Fig. 2.1C). This common set included transcription
factors associated with both hair cells and supporting cell fates, such as Atoh1 and
Hes5 respectively (Fig. 2.1D). While many upregulated genes are shared between hair
cells and supporting cells, the level of expression change is cell type specific (Fig. 2.1C
& Fig. 2.2C).

19

Figure 2.1: Hair cells and supporting cells share
common transcriptomic signature (A) Principal
component analysis showing clustering of embryonic
day 17 (E17.5) hair cells (HC, in green) and E17.5
supporting cells (SC, in orange) together, away from
their common E13.5 progenitor (PG, in blue). (B) Venn
diagram of results from differential expression analysis
showing a large overlap of common up-regulated
genes (purple) in hair cells and supporting cells
compared to their common progenitor. (C) Table of
log2 fold changes of hair cell and supporting cell
associated genes that are up regulated in both hair
cells and supporting cells. (D) GO analysis of genes
upregulated in only hair cells, both hair cells and supporting cells, and only supporting cells. (E) Bar chart
average log2 FPKMs of hair cell and supporting cell marker genes from RNA-seq of negative, low, and
high GFP populations at E15.5. Atoh1-Gfp cells are enriched for supporting cell markers and hair cell
markers.

20

The segregation of hair cells and supporting cell types from a common progenitor
is driven in part by Notch-mediated lateral inhibition whereby nascent hair cells signal to
their neighbors to prevent them from adopting a hair cell fate. Since the Atoh1
transcription factor begins to be up-regulated in the progenitors of hair cells and
supporting cells before finally restricting to hair cells, we hypothesized that at early
stages of hair cell and supporting cell differentiation, Atoh1-GFP+ cells would have
higher expression of Notch genes such as Hes5 and Lfng compared to those at later
stages (Lo et al. 2017; Tj et al. 2011). To test this, we performed RNA-seq on FACS-
purified GFP+ E15.5 Atoh1-GFP sensory epithelium to perform a differential analysis
comparing E15.5 hair cells with E17.5 hair cells (Fig. 2.2C). As expected, E15.5 hair
cells had higher levels of supporting cell marker gene expression (Fig. 2.1G). To see if a
similar downregulation of hair cells genes was occurring during supporting cell
maturation, we compared our E17.5 Lfng-GFP+ RNA-seq data to P1 Lfng-GFP+ RNA-
seq data previously published by our lab (Tao et al. 2021). As expected, hair cell-
related genes were downregulated by P1 (Fig. 2.2C).
Interestingly, the GFP+ population from the Atoh1-GFP fusion mice are more
continuous with the GFP negative population at E15.5, while at later stages it becomes
more discrete (data not shown). We sought to characterize this low GFP+ population at
E15. To do this, we performed RNA-seq on FACS-purified GFP negative, low, and high
cells from E15.5 Atoh1-GFP sensory epithelium. Consistent with our hypothesis, the low
Atoh1-GFP population had the highest expression of supporting cells markers
compared to both the negative Atoh1-GFP and high Atoh1-GFP populations (Fig. 2.2G).
The negative-GFP population lacked both supporting cell markers and hair cells marks
21

(Fig. 2.1G & Fig. 2.2A), and instead enriched for cochlear mesenchymal markers such
as Pou3f4 and non-sensory epithelial roof marker Oc90 (Fig. 2.2B) (Phippard et al.
1998; Hartman et al. 2015). This suggests that the low Atoh1-GFP population may
represent not just differentiating hair cells, but the entire sensory epithelium.

Figure 2.2: Low Atoh1-fGFP+ population
enriched for supporting cell markers (A) Bar
chart of average log2 FPKMs of sensory domain
associated genes from RNA-seq of negative (gray),
low (orange), and high Atoh1-fGfp (green)
populations. High and low Atoh1-fGfp populations
show high expression on sensory related genes
compared to negative Gfp population. (B) Bar chart
of average log2 FPKMs of Non-sensory domain
associated genes. The negative Gfp population is
highly enriched for otic mesenchyme marker
Pou3f4 and epithelial roof domain marker Oc90.
(C) Table of log2 fold changes of hair cell and
supporting cell associated genes from comparisons
of E17 versus E15 hair cells (HC) and P1 versus
E17.5 supporting cells (SC). The later stage E17.5
hair cells downregulate supporting genes Hes5 and
Lfng compared to E15.5 hair cells (HC). The later
stage P1 supporting cells downregulate hair cell
genes Atoh1 and Dll1 compared to E17.5
supporting cells.

22

Single cell RNAseq profiling of differentiating progenitors
Our bulk RNA-seq experiments suggested differentiating hair cells and
supporting cells upregulate genes for both cell-fate programs at early stages before
resolving these programs into the two cell types over a period of days. However,
because these experiments were performed from bulk FACS sorted populations, the
expression profiles don’t reveal the dynamics of expression within the sensory
epithelium. To test whether both transcriptional programs for hair cells and supporting
cells are initiated simultaneously in the same cells at the onset of differentiation, we
performed single-cell RNA-seq on FACS-purified cells from cochlea’s of E14 Sox2-
CreER mice crossed to ROSA-Ai9 mice that express a tdTomato-reporter after
recombination. We lineage-labeled the Sox2-expressing prosensory progenitors with
tamoxifen at E13.5 (Fig. 2.3A). This labels the entire prosensory domain that gives rise
to hair cells and supporting cells, as well as part of Kolliker’s organ (Fig 2.3B) (Gu et al.
2016).
Figure 2.3: Sox2-CreER lineage tracing of prosensory domain (A) On the left is a summary graphic of
cells labeled by Sox2CreER following tamoxifen treatment at different stages. (B) Sox2-CreER mice were
crossed with a flox-stop-flox Td tomato reporter line to label Sox2 expressing cells at E13.5, which labels
the prosensory domain and Kolliker’s organ. Sox2-CreER TdT-fl/fl mice were tamoxifen treated at E13 to
label prosensory cells. At E14.5 Td tomato positive cells were FACS-purified and used for single-cell
RNA-seq with 10x. Panel A adapted from (Gu et al. 2016).

23

After quality control filtering, a total of 5,507 Sox2 lineage-traced single cells
remained for subsequent analysis. To better preserve spatial relationships, PAGA was
used for projection and clusters were identified by Louvain clustering (Fig. 2.4A). PAGA
has been shown to better preserve topological information compared to common
projection methods (Wolf et al. 2019). Regional identities were assigned based on
known apex-base and medial-lateral markers (Fig 2.4C, 2.5A&B). Expression of
regional markers on projection plots followed expected spatial relationships (Fig. 2.5A &
2D). For example, the apical marker Fst follows an opposing gradient of expression with
base markers Inhba and Lgr5 (Prajapati-DiNubila et al. 2019; Chai et al. 2011). Nr2f2
and Hmga2 are highly enriched at apex and low/absent in base at E14.5-E15.5
(Golden, Benito-Gonzalez, and Doetzlhofer 2015; Tang et al. 2005). Hair cell and
supporting cell differentiation markers appear within and/or around only the Inhba
expressing domain (Fig. 2.4A&B). Kolliker’s organ cells on the neural side are
separated from Fgfr3 expressing cells on the abneural side by the Tbx2 expressing
inner hair cells (Fig. 2.4B & 2.5B). Together, our data demonstrates retention of spatial
relationships between different regions of the cochlea, allowing pseudo-spatial analysis
on our single-cell RNA-seq datasets.
24

’

Figure 2.4: Annotation of single-cell RNA-seq clusters (A) UMAP of all clusters, unfiltered. On the
right are the annotated clusters based on marker gene expression in B-D. (B) Gene expression violins for
cell-type markers in each cluster (C) Regional marker gene expression violins. Apex-base and medial-
lateral markers show mutually exclusive expression. (D) Cell-cycle gene expression violin plots. Kolliker’s
organ clusters are only cell-cycling population.

25

A.

B.

Figure 2.5: Regional expression markers (A) UMAP plots with overlayed expression of apex marker
genes above and base marker genes below. (B) UMAP plots with overlayed expression of medial marker
genes above and lateral marker genes below. In both (A) and (B), there is clear mutually exclusive
expression gradients of opposing axis marker genes.

Co-expression of cell-fate determinants coincides with initiation of differentiation at the
base.
To focus on the sensory epithelium, Kolliker’s organ clusters were removed, and the
projection plot for the remaining clusters was recalculated (Fig. 2.6A). Atoh1, Prox1 and Hes5
expression co-occur in a gradient starting high at the inferred base and absent at the inferred
apex (Fig. 2.6B). In addition, Lfng expression co-occurs with Atoh1, but only on the neural side
with Fgf8 expressing inner hair cell cluster, consistent with previous observations of Lfng
expression at this stage (Basch et al., 2016). Other hair cell markers that are believed to
be downstream of Atoh1, such as Pou4f3 and Gfi1, showed more restricted expression
26

in the Fgf8 and Insm1 expressing hair cell cluster (Fig. 2.4B & 2.6B). Both Pou4f3 and
Gfi1 are downstream of Atoh1, so their restricted expression in the inferred base region
within a Atoh1 expressing cluster is consistent with their inferred temporal state.
Together this confirms the co-expression of competing hair cell fate and supporting cell
fate determinants. Additionally, it supports that the expression of hair cell fate-
determinants in supporting cells begins at the onset of differentiation with broad
initiation at the base. This is consistent with lineage tracing and expression studies of
Atoh1 and Prox1 which find co-expression between hair cells and supporting cells
between E14-E16 (Bermingham-McDonogh et al. 2006; Elizabeth Carroll Driver et al.
2013; Z. Liu et al. 2012; S. Li et al. 2022).
Summary Conclusion
This expression data suggests that there is an Atoh1-independent mechanism for
initiating the hair cell transcriptional program in all cells from the prosensory domain.
The factor that initiates Atoh1 expression has yet to be identified. The observed
supporting cell expression of Atoh1 by bulk RNA-seq and single-cell RNA-seq along
with lineage tracing studies supporting Atoh1 is induced in supporting cells, suggests
that the factor that initiates Atoh1 expression is also active in supporting cells. We next
asked, if supporting cells come from a progenitor that initiated hair cell fate, how far did
they progress towards hair cell fate? Did they remodel their epigenetic landscape
towards hair cell fate? If this is the case, it would mean the factor(s) that induces Atoh1
expression may be capable of imparting the transdifferentiation potential seen in
supporting cells.

27

A.

B.

Figure 2.6: Broad co-expression of competing cell-fate factors (A) Clusters assigned region labels by
maker gene expression, with the cochlear base region highlighted in red. OS: Future outer sulcus; PG:
Prosensory progenitors; SC: Supporting cells; IHC: Inner hair cells; OHC: Outer hair cells. (B)
Dimensional reduction plots with gene expression overlayed for hair cell-fate associated factors (Atoh1,
Pou4f3, and Gfi) and supporting cell-fate associated factors (Hes5, Prox1, and Lfng). Co-expression of
Atoh1, Lfng, Hes5, and Prox1 occurs in an overlapping gradient from base to apex. The more mature hair
cell markers Pou4f3 and Gfi show restricted expression in highest Atoh1 expressing clusters.

28

Chapter 3: Epigenetic dynamics during prosensory differentiation
Introduction
Given the transcriptional initiation of hair cell-related genes in supporting cells,
we wondered if supporting cells also acquire and retained a hair cell like epigenetic
landscape. To test this, we performed CUT&RUN for H3K4me1, H3K27ac, and
H3K27me3 on FACS-purified E17.5 Lfng-GFP+ supporting cells, E17.5 Atoh1-GFP+
hair cells, and E13.5 Cdkn1b-GFP+ PGs (Fig. 1.8) (Skene and Henikoff 2017). These
epigenetic marks correspond to primed, active, and repressed chromatin, respectively.
To assess the overall relationship between the three cell types we performed a PCA for
each histone mark.
Hair cell enhancer network is primed in supporting cells but not progenitors
For the priming mark, H3K4me1, hair cells and supporting cells overwhelming
cluster together, away from PG, on PC1 with 97% of variance explained (Fig. 3.1A). To
identify the H3K4me1 regions driving this similarity between SC and hair cells, we
performed a differential analysis between the three cell types with ChromStar (Taudt et
al. 2016). We identified 11,844 de novo H3K4me1 shared between hair cells and
supporting cells, which are absent in progenitors (Fig. 3.1B). To assess what functional
role these elements may play, we performed an ontology enrichment analysis with
GREAT (McLean et al., 2010). The top enriched terms were sensory related and hair
cell terms. This included genes like Atoh1 and Dll1, both markers of hair cell fate (Fig.
3.1C). An IGV snapshot with signal tracks for H3K4me1 and ATAC shows two of the
common de novo H3K4me1 peaks at two putative hair cell enhancers upstream of Dll1
(Fig. 3.1D). Interestingly, while there appears to be little difference between supporting
29

cells and hair cells at these common de novo H3K4me1 regions, not all become
accessible in supporting cells as shown by ATAC-seq previously collected by our lab
(data not shown) (H. V. Yu et al. 2021).
Figure 3.1: Supporting cells acquire a hair cell like primed epigenetic landscape (A) Principal
component analysis (PCA) on H3K4me1 regions from FACS-purified E17.5 hair cells (HC), E17.5
supporting cells (SC), and E13.5 progenitors (PG). Hair cells and supporting cells cluster together, away
from their common E13.5 progenitor. (B) Heatmap of differential H3K4me1 modified regions identified by
ChromstaR. (C) Gene ontology analysis performed with GREAT on genes associated with de novo
H3K4me1 peaks identified in (B). Genes associated with these common de novo peaks are highly
enriched for hair cell-related terms. (D) IGV snapshot of the Dll1 locus with H3K4me1 (purple) signal
tracks. Three upstream putative enhancers with common de novo H3K4me1 regions shared between hair
cells and supporting cells are shown highlighted in blue.

30

Supporting cells fail to activate primed hair cell enhancers
While H3K4me1 is a priming mark, it does not indicate whether an enhancer is
active. To identify active elements, we next looked at H3K27ac in the three cell types at
the same stages. A PC analysis of H3K27ac shows PG and supporting cells clustering
together away from hair cells on the major PC 1, with 84% of variance (Fig. 3.2A). We
then performed a differential analysis to identify active regions in each cell type (see
methods) (Fig. 3.2B). We identified 3,434 hair cell specific enhancers and 2,678
supporting cell-specific enhancers (Fig. 3.2C). We next sought to determine if these cell
type specific enhancers were already primed by H3K4me1 in progenitors or if they were
de novo. Of the 3,434 hair cell specific active enhancers, 645 were primed by H3K4me1
in progenitors, while 2,309 were not previously primed by H3K4me1 (Fig. 3.2D). Of the
supporting cell-specific enhancers, 1,359 had H3K4me1 in progenitors while 463 were
de novo (Fig. 3.2E). While the hair cell enhancer network is predominantly de novo,
much of the supporting cell enhancer network is already primed by E13.5 (Fig. 3.2E). As
can be seen in the H3K4me1 heatmaps of hair cell-specific enhancers, supporting cells
also prime hair cell enhancers relative to progenitors, but they fail to acquire the active
H3K27ac mark (Fig. 3.2C & 3.2D). A putative enhancer upstream of Dll1 shows an
example of supporting cells acquiring H3K4me1 but devoid of H3K27ac while hair cells
acquire both (Fig. 3.2F). This region is part of the common de novo H3K4me1 sites
previously identified (Fig. 3.1D). This suggest that the supporting cell enhancer network
is already established while the hair cell enhancer program is being primed in both
supporting cells and hair cells, but only activated in hair cells.

31

Progenitor enhancers are decommissioned in hair cells but not supporting cells
While supporting cells prime many of the same regions as hair cells, they share
most of their active regions marked by H3K27ac with PGs (Fig. 3.2A & B). Of 8,863
supporting cell enhancers, ~32% are supporting cells-specific, ~51% are shared with
progenitors. Of the 5,856 hair cell enhancers, ~59% are hair cell-specific, and ~13% are
shared with progenitors. Hair cells de-acetylate greater that 85%, 5,006, of enhancers
active in progenitors, while supporting cells only de-acetylate 20% of progenitor
enhancers (Fig. 3.2B). With supporting cells primarily activating new enhancers and hair
cells both activating new enhancers while shutting down progenitor enhancers, we
checked to see if this was reflected by changes in gene expression. A differential
expression analysis of E17.5 supporting cells compared to E13.5 progenitors reveal far
more upregulated genes than downregulated, 2,028 up and 555 down (Fig 3.3B). A
comparison of E17.5 hair cells to E13.5 progenitors reveals 2,566 up and 1,549 down
(Fig 3.3A). Thus, the expression changes seen in supporting cells are consistent with
supporting cell maintenance of the progenitor associated enhancers. In contrast to hair
cells where there is both transcriptional and epigenetic silencing of the progenitor
program. With many of supporting cell specific enhancers already primed in progenitors,
and the de novo priming of hair cell enhancers, it would suggest supporting cells are
more likely uncommitted progenitors that have been specified for hair cell fate; while
hair cells show signs of determination with removal of progenitor associated epigenetic
marks.
32

Figure 3.2: Enhancer activation dynamics (A) Principal component analysis (PCA) on distal H3K27ac
regions from FACS-purified E17.5 hair cells (HC), E17.5 supporting cells (SC), and E13.5 progenitors
(PG). E13.5 progenitors and E17.5 supporting cells cluster together away from hair cells on the primary
PC1 (84% variance). (B) Heatmap of distal H3K27ac regions across the three cell types, showing shared
and uniquely H3K27ac modified regions. (C) Hair cell and supporting cell specific enhancers subset from
(B). (D) H3K4me1 signal from E13.5 PG, E17.5 SC, and E17.5 HC shown on heatmap for hair cell
specific enhancer regions identified in (C). A majority (2,039) of hair cell enhancers are de novo. (E) For
supporting cell specific enhancers, a majority (1,359) were already primed by H3K4me1 in E13.5
progenitors. (F) IGV snapshot of the Dll1 locus with H3K4me1 (purple) from figure 3d with H3K27ac
(green) signal tracks included. Upstream putative enhancer shown that was de novo H3K4me1 modified
in both hair cells and supporting cells, but only H3K27 acetylated in hair cells.

33

Figure 3.3: Supporting cells skew towards upregulation (A & B) MA Plots for (A) hair cells versus
E13.5 progenitors (PG) and (E) supporting cells versus E13.5 PG. (A) Hair cells upregulate 2,566 genes
(red) and downregulate 1,549 genes (blue). (B) Supporting cell predominately upregulated genes with
2,028 upregulated compared to only 555 downregulated. Not significant (NS) threshold set at p-value
>0.05 & log fold change <0.5. Up & down regulation significance at p-value <= 0.05 at log fold change
>=0.5.

Transdifferentiation associated enhancer predominately arise during or after
differentiation
Recently our lab identified primed hair cell enhancers in P1 supporting cells that
are decommissioned by removal of H3K4me1 by P6, leading to a loss of
transdifferentiation potential by P6 (Tao et al. 2021). We wanted to know if these
transdifferentiation associated enhancers were the same enhancers we identified as
common de novo H3K4me1 regions in both supporting cells and hair cells. A majority of
the common de novo H3K4me1 overlapped with the transdifferentiation associated
enhancers (Tao et al. 2021). This suggest that the program for establishing the hair cell
enhancer network is not established at the E13.5 progenitor stage but rather sometime
during differentiation around E14.5.
34

Common primed enhancer between hair cells and supporting cells enrich for TFs
associated with cell fate potential
To better understand what regulators may be driving the de novo priming of the
hair cell enhancer network in supporting cells, we performed a motif analysis using AME
(McLeay and Bailey 2010). By analyzing H3K27ac and H3K4me1 patterns at different
stages, we can associate the activating activity of a regulator versus pioneering/priming
activity. We first wanted to know if we can recapitulate the previous findings of a Atoh1-
Pou4f3 feedforward circuit from a motif analysis on regions with different patterns of
priming and activation (H. V. Yu et al. 2021). In this circuit, Atoh1 was found to drive
Pou4f3 expression, which in turn primes new enhancers that Atoh1 could then target
(Fig. 3.4A). In this circuit, the initial elements Atoh1 targets and activates are already
primed before expression of Pou4f3 (Fig. 3.4A). To see if our analysis could recapitulate
these previous findings, we took enhancers that were primed in E13 progenitors, but
only activated by H3K27ac in hair cells and compared to enhancers that remained
inactivated in supporting cells (Fig 3.4B). Motif results from AME were compared with
MotifStack to group merge similar motifs (Fig 3.5) (Ou et al. 2023). All TFs merged
under a motif are listed as associated genes. Only expressed associated transcription
factors were kept (see methods). Consistent with direct Atoh1 regulation, Ebox motifs,
binding sites for Atoh1, were the highest enriched motifs (Fig. 3.4C). One of these ebox
containing elements includes a Pou4f3 enhancer, where Atoh1 was found to bind to and
activate (H. V. Yu et al. 2021).
Next, we wanted to see if we could identify Pou4f3 as the hair cell specific
priming factor with this analysis. To identify hair cell priming factors, hair cell specific
35

H3K4me1 regions were compared to common de novo H3K4me1 regions shared
between hair cells and supporting cells (Fig. 3.4D & F). As expected, A/T rich POU
motifs were enriched in the top hits, including Pou4f3 (Fig. 3.4E). Together this gave us
confidence we can associate potential epigenetic mechanisms with motifs and their
regulatory factors by comparing the changes in primed and active statuses at various
stages between cell types.
To identify the transcription factors that may be responsible for de novo priming
of H3K4me1 regions shared between hair cells and supporting cells, common de novo
H3K4me1 regions were compared to hair cell specific H3K4me1 regions using AME
(Fig. 3.4F & 3.4G). The top enriched motifs were Ebf1, Nfi, SoxE (Sox9 & 10), Sox2,
SoxC (Sox4, Sox11), and Prdm16 (Fig. 3.4G).
36

37

Figure 3.4: Motif signatures of activators and pioneers (A) Diagram of the previously described
Atoh1-Pou4f3 feed-forward circuit in hair cells; whereby Atoh1 activates already primed enhancers,
leading to Pou4f3 driven de novo priming and activation of new enhancers. (B) Diagram of pre-
established H3K4me1 regions at E13.5, which acquire H3K27ac only in hair cells by E17.5. Shown below
are enhancers that remain unchanged between E13.5 progenitors and E17.5 hair cells and supporting
cells. (C) Motif enrichment results table of the pre-established hair cell specific H3K27ac regions, as
shown B, compared to regions that remain unchanged. Shown below is a table of the top motifs enriched
in the pre-established hair cell enhancers and their associated transcription factors. The Ebox motif is the
top enriched motif, with Atoh1 being the most significant match. (D) Diagram of the 5,460 hair cell specific
de novo H3K4me1 enhancers, with hair cell specific Pou4f3 being shown as the pioneer factor. (E) Motif
enrichment results of hair cell specific de novo H3K4me1 enhancers as show in (D), compared to hair cell
and supporting cell common de novo H3K4me1 enhancers as shown in (F). The top enriched motifs are
A/T rich, with the top motif associated with Lmx1a and Pou4f3. (F) Diagram of the 11,738 common de
novo enhancers shared between hair cells and supporting cells. The pioneer factor responsible for this
common set is unknown. (G) Motif enrichment results of hair cell and supporting cell common de novo
H3K4me1 enhancers as shown in (F) compared to hair cell specific de novo H3K4me1 enhancers as
show in (D). The top motifs include known pioneer factors Ebf1, Sox2, and Sox4.

Figure 3.5: Common de novo
motifs Phylogenomic tree of
motifs found to be enriched in
common de novo H3K4me1
enhancers shared between hair
cells and supporting cells.
(PRD16, Prdm16) (COE1, Ebf1)

Ebf1, Sox2, and Sox11 are well documented pioneer factors, capable of
epigenetically potentiating the activity of other TFs (Dodonova et al. 2020; Zhou et al.
2022; Boller et al. 2016; Treiber et al. 2010). Prdm16 regulates adipogenesis, where it
binds enhancers and recruits mediator complex to drive switching of enhancer targeting,
driving cell-fate change (Harms et al. 2015). More recently Prdm16 has been found to
38

be critical for formation and maintenance of Kolliker’s organ where is it expressed from
E13.5 (Ebeid et al. 2022). Sox2 and SoxC are expressed in the sensory epithelium, with
Sox2 levels increasing greatly between E12 to E14. SoxC and Sox2 have both been
shown to play critical roles in differentiation of the sensory epithelium, including a role
for SoxC in specifying sensory fate.
We next examined the expression pattern of these TFs in our E14.5 single-cell
RNA-seq dataset. Ebf1 was expressed in KO and the sensory epithelium, while Prdm16
was expressed exclusively in KO (Fig. 2.5B). Expression of NFI factors were more
dynamic. Nfia was expressed highest at the base of the sensory epithelium, and broadly
in KO (Fig. 3.6). Nfib was expressed broadly through both KO and sensory epithelium
(Fig. 3.6). Nfix is expressed in a small group of cells at the base (Fig. 3.6). Nfic
expression is specific to presumptive base clusters (Fig. 3.6). Sox2 expression was
mostly restricted to the sensory epithelium, with diving cells in KO also displaying high
expression. Sox9, Sox10, Sox4 and Sox11 were highly expressed in both KO and the
sensory epithelium.
Figure 3.6 Expression of NFI
factors at E14.5: Expression of Nfia
and Nfix shows enrichment in the
presumptive base clusters, including
Kolliker’s organ. Expression of Nfib
shows enrichment in presumptive
base of sensory epithelium. Nfic is
sparsely expressed across all cells.

39

Summary & Conclusion
Here, we have found that the priming by H3K4me1 of hair cell enhancers by
supporting cells is not something that is pre-established in E13.5 progenitors, rather
they are largely de novo. This suggests that there is a priming factor that has yet to be
identified, that can prime the hair cell enhancer network without activating it. Through
comparative motif analysis of cell-type specfic de novo priming and common de novo
priming, we have identified potential regulators of these regulatory elements. While
some of these implicated factors, such as SoxC, Sox2, Prdm16, have been studied in
the inner ear, the role of the top enriched factors in the inner ear, Ebf1 and Nfi, is poorly
understood. Single-cell RNA-seq shows that Ebf1 and Nfi factors are in the right time
and place to be potential regulators for establishing de novo H3K4me1 of hair cell
enhancers in supporting cells. Identifying what factors prime hair cell enhancers could
offer a way to increase the transdifferentiation capacity of mature supporting cells.
Despite finding that the primed hair cell enhancers in supporting cells are de novo, it
remains unclear if there are other epigenetic factors for supporting cell plasticity. For
example, we have found that E17.5 supporting cells maintain many progenitor
enhancers in a primed state. We also find that supporting cells keep many progenitor
enhancers active, unlike hair cells. This raises the potential that supporting cell plasticity
may arise from the lack of progenitor enhancer decommissioning at E17.5.
Understanding what enhancers are associated with transdifferentiation potential and
40

when they arise in development, is critical to identifying and utilizing the mechanisms
that establish them.
Chapter 4: A role for Prc1 and Prc2 in cell-fate plasticity
Atoh1 is epigenetically de-repressed in early supporting cells.
Given our observation of low Atoh1 expression and priming of hair cell enhancers
in supporting cells, we next sought to understand the epigenetic repressive state of
Atoh1 in supporting cells. Previous studies from our lab have found the Atoh1 locus to
be marked by repressive H3K27me3 in supporting cells at P1 and onward (Stojanova,
Kwan, and Segil 2015; Tao et al. 2021). By P1, Atoh1 lineage tracing studies no long
observe labeling of supporting cells (Driver et al. 2013; S. Li et al. 2022). At P1, notch
inhibition induced transdifferentiation potential has already begun rapidly declining (Tao
et al. 2021). We sought to characterize H3K27me3 in E13.5 progenitors and E17.5
supporting cells to see if changes in H3K27me3 at Atoh1 and Hes5 could explain their
broad expression observed in our scRNA-seq data at E14.5 during onset of
differentiation.
As described before, we collected H3K27me3 by CUT&RUN for E13.5
progenitors, E17.5 supporting cells, and E17.5 hair cells (Fig. 1.8) (Skene and Henikoff
2017). Hair cells and supporting cells cluster together, away from progenitors, on the
major PC1 with 75% variance explained (Fig. 4.1A). A differential analysis between the
three cell-types was performed with ChromStar (see methods) (Fig 4.1B). For both hair
cells and supporting cells, the gain of H3K27me3 is much more common than its loss.
Hair cells lost H3K27me3 at 1,863 regions while gaining H3K27me3 at 3,175 regions
Supporting cells lose H3K27me3 at 908 regions while gaining H3K27me3 at 2,976
41

regions (Fig. 4.1B). This is consistent with our previous observations that H3K27me3
increases between P1 and P6 (Tao et al. 2021). Interestingly, only 263 regions lost
H3K27me3 only in supporting cells, the other 645 regions lost H3K27me3 in both hair
and supporting cells. To gain a better understanding of the function of these regions we
performed an ontology analysis with GREAT (Fig. 4.1C). The top enriched terms are
neuronal and hair cell related (Fig. 4.1C). This suggests that in addition to priming hair
cell enhancers, supporting cells are also de-repressing hair cell regulatory elements.
One of the genes in this set of regions is Atoh1. Consistent with the observed low
expression of Atoh1 in E17.5 supporting cells, the Atoh1 locus is completely devoid of
H3K27me3 in E17.5 supporting cells while lacking H3K27ac (Fig. 4.1). Another gene
part of these common de-methylated regions is Hes5. Like Atoh1, Hes5 was found to be
broadly expressed in our E14.5 scRNA-seq data (Fig. 2.6B). Also, like Atoh1, the entire
Hes5 gene is marked by H3K27me3 in E13.5 progenitors. By E17.5 both supporting
cells and hair cells are devoid of the repressive mark (Fig. 4.1D).
H2AK119ub in maintain in absence of H3K27me3
As H3K27me3 is lost at the Atoh1 locus in supporting cells, we next sought to
identify the epigenetic basis that keeps Atoh1 expression at bay in E17.5 supporting
cells. Non-canonical Prc1 can recruit Prc2 for establishment of H3K27me3 through its
enzymatic subunit, Rnf2, which mono-ubiquitinates H2AK119 (H2AK119ub) ( Wang et
al. 2004; de Napoles et al. 2004; Blackledge et al. 2014). In Zebrafish, prior to zygotic
genome activation, Prc1 maintains silencing in absence of H3K27me3. Post zygotic
genome activation, Prc2 recognizes H2AK119ub and deposits H3K27me3 at these
sites. If Prc1 is inhibited, H3K27me3 does not get established, and precocious
42

transcription of its target genes occurs after zygotic genome activation (Hickey et al.
2022). Previous studies from our lab found H3K27me3 returns at Atoh1 by P1 and
increases at P6 (Tao et al. 2021). If Prc1 is involved in maintaining repression of Atoh1,
we would expect to see H2AK119ub covering Atoh1, despite absence of H3K27me3. To
test this, we performed CUT&RUN for H2AK119ub for E13.5 progenitors and E17.5
supporting cells. As expected, H2AK119ub covers the entirety of the Atoh1 locus in both
progenitors and supporting cells (Fig. 4.1E). This would suggest H2AK119ub may be
responsible for keeping Atoh1 expression at bay before H3K27me3 returns by P1.
Loss of H3K27me3 at Atoh1 occurs during onset of differentiation
While the loss of H3K27me3 at Atoh1 and Hes5 in both cell types is consistent
with the broad expression seen in our E14.5 scRNA-seq data, the H3K27me3 was
collected at E17.5. To address whether H3K27me3 is lost at the onset of differentiation,
when broad expression of Atoh1 begins, we co-profiled single nuclei RNA-seq and
H3K27me3 using Paried-Tag on E14.5-E15.5 Epcam+ cochlear epithelium (C. Zhu et
al. 2021). After quality control filtering, 9,392 single nuclei remained. PAGA analysis and
cluster annotation was performed as previously described with our scRNA-seq (Fig
4.3A). As Paired-Tag is single-nuclei RNA-seq rather than single-cell, the Paired-Tag
dataset had much higher dropout than our E14.5 single-cell RNA-seq dataset. The
technique is also biased against small transcripts due to stringent SPRI bead clean-up
of unwanted ligation products and oligos. However, regional marker genes and cell-type
markers were still distinguishable between clusters, allowing us to identify presumptive
medial-lateral and apex-base clusters (Fig 4.2A & C). This data set represents the entire
epithelium of the cochlea, which includes the Oc90 expressing roof domain, Otx2
43

expressing future Reisner’s membrane, and the Bmp4 expressing future outer sulcus
(Fig 4.2B & C). These clusters were removed, and only Sox2+/Cdkn1b+,
Pou4f3+/Jag2+, and Prox1+ clusters were kept. (Fig. 4.3B). These remaining clusters
represent the sensory epithelium. Expression of apical markers Hmga2 and Nr2f2, was
used to annotate the presumptive apical clusters, which was consistent with mutually
exclusive expression of Prox1 and Pou4f3 in presumptive base clusters (Fig. 4.2A, B,
C). The histone profiling data for each cluster was then separated out and loaded as
individual IGV tracks along with bulk CUT&RUN tracks for E13.5 progenitors, E17.5 hair
cells, and E17.5 supporting cells for reference. Prox1+, Sox2+, and Cdkn1b+ clusters 3
& 4 were annotated as supporting cells. To ensure there was sufficient data from
supporting cell cluster, the locus for Hox clusters and Pou4f3 were examined. Hox
clusters and Pou4f3 locus are heavily marked by H3K27me3 in bulk CUT&RUN data of
E17.5 supporting cells (Fig 4.3C & D). The Paired-Tag H3K27me3 data from the
supporting cell cluster recapitulates the bulk level data for these loci (Fig 4.3C & D). At
the Atoh1 locus, there is a complete absence of any Paired-Tag H3K27me3 signal from
the supporting cell cluster (Fig. 4.3E). This shows the loss of H3K27me3 is occurring at
the earliest stages of differentiation when broad expression of Atoh1 is observed.
Together this suggests the loss of H3K27me3 at Atoh1 and Hes5 as a potential
mechanism for their broad expression at the onset of differentiation. It also shows that
embryonic supporting cells are in a unique epigenetic state defined by both having a
primed hair cell enhancer network and Atoh1 in a de-repressed state. This
epigenetically permissive state is not seen in either E13.5 progenitors or in P1
supporting cells.
44

Figure 4.1: Repressive chromatin dynamics (A) Principal component analysis (PCA) on H3K27me3
regions from FACS-purified E17.5 hair cells (HC, green), E17.5 supporting cells (SC, orange), and E13.5
progenitors (PG, blue). (B) Heatmap of differential H3K27me3 modified regions identified by ChromstaR.
(C) Gene ontology analysis on PG specific H3K27me3 regions identified in (B), showing enrichment for
hair cell-related terms. (D) IGV snapshot of two PG specific H3K27me3 regions at the Atoh1 and Hes5
locus with H3K27me3 (red) signal tracks for PG, SC, and HC. While Atoh1 and Hes5 loci are H3K27
trimethylated in PG, H3K27me3 is absent for both SC and HC. (E) IGV snapshot of Atoh1 locus with
signal tracks for H2AK119ub (orange) and H3K27me3 (red), showing repressive H2AK119ub is retained
in absence of H3K27me3 in supporting cells.
45

Figure 4.2: Regional marker expression in single-cell Paired-Tag (A-D) UMAP plots with regional
marker gene expression overlayed (A) Apex and base marker genes. (B) Roof markers, for Reissner’s
membrane (C) Neural and abneural domain markers. Prdm16 for Kolliker’s organ, Bmp4 & Gata2 for
future outer sulcus, and Ebf1 for broadly medial. (D) Hair cell and supporting cell marker genes.

46

Figure 4.3: E14.5-E15.5 supporting cell de-repress Atoh1 locus (A) UMAP of E14.5-E15.5 scRNA-seq
from Paired-Tag experiment. (B) UMAP of (A) with only the sensory epithelium kept (C) IGV snapshot of
the HoxC cluster. (C) IGV snapshot of Pou4f3, which never loses H3K27me3 in E17.5 bulk supporting
cells. The single-cell Paired-Tag supporting cells cluster show Pou4f3 is H3K27 tri-methylated. (E) IGV
snapshot of the Atoh1 locus with H3K27me3 signal tracks from Paired-Tag and CUT&RUN. The Paired-
Tag H3K27me3 clusters shown are Kolliker’s organ (KO), Oc90+ cluster, (Roof), and supporting cell
cluster (SC). H3K27me3 is already absent in E14.5-E15.5 supporting cell cluster like what is seen in
E17.5 supporting cells.
47

Summary and Conclusion
Previous studies from our lab had found that at postnatal day 1, the Atoh1 locus
in supporting cells was marked by H3K27me3 and that the levels of H3K27me3
increased by postnatal day 6 (Tao et al. 2021). Further, the Atoh1 locus at E14.5 in
Cdkn1b+ progenitors was also marked by H3K27me3 (Stojanova, Kwan, and Segil
2015). Here, we document a previously undescribed state in E14.5-E17.5 supporting
cells in which the Atoh1 locus is devoid of H3K27me3. In addition, hair cells are devoid
of H3K27me3 at the Hes5 locus, a direct repressor of Atoh1. It may be that the
derepression of these competing transcription factors is a feature of Notch-mediated
lateral inhibition, allowing every cell to ‘participate’ in the patterning process.

48

Chapter 5: Discussion & Future Directions
We have found that in comparing E13.5 progenitors to early supporting cells,
genes specifying hair cell and supporting cell fates, such as Atoh1, lose repressive
H3K27me3 (Fig 4.1D). In supporting cells, despite loss of H3K27me3, the repressive
Prc1 mark, H2K119ub, is maintained (Fig. 4.1E). At enhancers, supporting cells prime
hair cell associated enhancers by H3K4me1 (Fig 3.1A-C), but do not activate by
H3K27ac as in hair cells (Fig 3.2C-F). Most hair cell enhancers are de novo, while
supporting cells enhancers are already primed in progenitors (Fig. 4.1E). Additionally,
supporting cells maintain progenitor enhancers in a primed state. In contrast, hair cells
decommission progenitor associated enhancers by H3K4me1 removal (Fig 3.1B). In
summary, early supporting cells appear to be more like uncommitted progenitors that
have been specified towards hair cell fate through priming of the hair cell enhancer
network and derepression of Atoh1 (Fig. 5.1). We propose that the capacity for
transdifferentiation arises from the de novo establishment of the hair cell enhancer
network in early differentiating supporting cells. Understanding how these enhancers
are primed could lead to novel regenerative approaches that restore the capacity to
respond regeneratively.
A major missing link in developmental biology is connecting signaling pathways
to transcription factors to epigenetic mechanisms. Epigenetic readers, erasers, and
writers are much more druggable than transcription factors, as are signaling pathways.
Being able to understand how a signaling pathway modulates the behavior of a
transcription factor and the epigenetic mechanisms is acts through would provide a
much greater diversity of therapeutic targets for modulating gene regulatory programs.
49

One of the most exciting developments to address this issue is the use of multi-omics to
co-profile single cell for epigenetic features and RNA (Ma et al. 2020; C. Zhu et al.
2021; Rooijers et al. 2019; Xu et al. 2022; S. Chen, Lake, and Zhang 2019; Clark et al.
2018; Angermueller et al. 2016; C. Zhu et al. 2019; L. Liu et al. 2019; Hou et al. 2016;
Hu et al. 2016; Macaulay et al. 2015; Dey et al. 2015). These techniques can utilize the
statistical power of thousands of single-cell RNA-seq profiles for building gene
regulatory networks, and now incorporate epigenetic features as both regulators and
targets (Kamimoto et al. 2023; Fiers et al. 2018; Aibar et al. 2017; González-Blas et al.
2022; Iacono, Massoni-Badosa, and Heyn 2019; Fleck et al. 2022; Kartha et al. 2022).
What this means is that a transcription factor expression or even epigenetic state, can
be correlated with epigenetic outcomes. In one example, pseudo-temporal ordering was
used to see the sequence of accessibility and RNA changes during differentiation (Ma
et al. 2020). A particular genes expression can be spotted as upstream to a particular
enhancer network getting primed. Knowing the order allows directionality to be applied
to gene regulatory network building.
Factor X
The transcription factor that initiates Atoh1 expression has yet to be identified. A
previous study from our lab has found evidence for activator insufficient for early Atoh1
activation (Abdolazimi, Stojanova, and Segil 2016). In the proposed model, at onset of
differentiation, low levels of Atoh1 expression occurs across prosensory domain, but the
repressive action of Hes & Hey factors keeps Atoh1 expression from reaching threshold
needed for Atoh1 to autoregulate itself (Abdolazimi, Stojanova, and Segil 2016). In other
words, removal of repression is sufficient to upregulate Atoh1 in perinatal supporting
50

cells, but not mature supporting cells. As the authors state, this suggest there is a
positive regulator targeting Atoh1, which they call factor x (Abdolazimi, Stojanova, and
Segil 2016). The broad low level of Atoh1 expression and the removal of H3K27me3 at
the Atoh1 locus in early supporting cells only, supports this model. In addition to Atoh1,
Hes5 may also be a target of factor x, due its similar expression and H3K27me3 de-
repression dynamics. Identifying the transcription factors or signaling pathways leading
to loss of H3K27me3 at Hes5 and Atoh1 may be key to identifying factor x. Now that
single-cell RNA-seq and histone profiling is possible it may be worthwhile to
characterize the epigenetic effects of mutants. For example, single-cell RNA-seq and
H3K27me3co- profiling on a Atoh1 K/O could be used to perform a gene network
analysis to identify what transcription factor correlates with H3K27me3 removal.
Does Prc1 maintain repression at early stages and later recruit H3K27me3?
In absence of H3K27me3 at the Atoh1 locus in supporting cells, the Prc1 mark
H2AK119ub is retained (Fig 4.1E). This raises the possibility that Prc1 is maintaining
Atoh1 in a repressed state. Independently, it could also be serving to recruit Prc2 and
restore H3K27me3 levels seen at P6 (Fig 5.1). This type of Prc1 mediated recruitment
has been observed during development (Hickey et al. 2022). Removing Prc1 from the
Atoh1 locus may either derepress Atoh1 and/or prevent reestablishment of H3K27me3.
While there are E3 ubiquitin ligase inhibitors for Prc1, it is not clear that the enzymatic
activity is the primary mechanism for repression and H3K27me3 recruitment. A new
inhibitor, RB-3 bypasses this concern by inhibiting the interaction of Prc1’s enzymatic
core with nucleosomes. This inhibitor has been shown to better recapitulate a Prc1
51

knockout (Shukla et al. 2021). The predicted effect of this inhibitor would be extension
of the window of plasticity and/or Atoh1 upregulation.
We hypothesize that the capacity of supporting cell transdifferentiation potential
is a byproduct of its simultaneous differentiation with hair cells. Identifying the regulators
that establish this plasticity between E12.5-E14.5 would offer a novel approache to
restoring plasticity to adult supporting cells. If there are transcription factors establishing
transdifferentiation potential through priming, could we not reprogram plasticity back into
cells?

Figure 5.1: The window of plasticity A working model. Window of plasticity represents timeframe
supporting cells are capable of transdifferentiation. From left to right; E13.5 progenitors have already
primed the supporting cell enhancer network. At E14, derepression and upregulation of Atoh1 and Hes5
sets the stage for Notch-mediated lateral inhibition to resolve cell-fate selection. At the same time, all cells
in the prosensory domain prime a hair cell enhancer network. Supporting cell maintain this primed
signature until P6 when they lose their capacity to transdifferentiate.

52

References
Abdolazimi, Y., Z. Stojanova, and N. Segil. 2016. “Selection of Cell Fate in the Organ of
Corti Involves the Integration of Hes/Hey Signaling at the Atoh1 Promoter.”
Development 143 (5): 841–50. https://doi.org/10.1242/dev.129320.
Abelló, Gina, and Berta Alsina. 2007. “Establishment of a Proneural Field in the Inner
Ear.” International Journal of Developmental Biology 51 (6–7): 483–93.
https://doi.org/10.1387/ijdb.072343ga.
Abraham, L., and A. L. Kierszenbaum. 2007. “Histology and Cell Biology.” An
Introduction to Pathology. In, Spermatogenez, 569–94.
Aibar, Sara, Carmen Bravo González-Blas, Thomas Moerman, Vân Anh Huynh-Thu,
Hana Imrichova, Gert Hulselmans, Florian Rambow, et al. 2017. “SCENIC:
Single-Cell Regulatory Network Inference and Clustering.” Nature Methods 14
(11): 1083–86. https://doi.org/10.1038/nmeth.4463.
Angermueller, Christof, Stephen J. Clark, Heather J. Lee, Iain C. Macaulay, Mabel J.
Teng, Tim Xiaoming Hu, Felix Krueger, et al. 2016. “Parallel Single-Cell
Sequencing Links Transcriptional and Epigenetic Heterogeneity.” Nature
Methods 13 (3): 229–32. https://doi.org/10.1038/nmeth.3728.
Assmus, Adrienne M., John J. Mullins, Cara M. Brown, and Linda J. Mullins. 2020.
“Cellular Plasticity: A Mechanism for Homeostasis in the Kidney.” Acta
Physiologica 229 (1): e13447. https://doi.org/10.1111/apha.13447.
“Babraham Bioinformatics - FastQC A Quality Control Tool for High Throughput
Sequence Data.” n.d. Accessed March 13, 2023.
https://www.bioinformatics.babraham.ac.uk/projects/fastqc/.
Basch, Martin L, Rogers M Brown II, Hsin-I Jen, Fatih Semerci, Frederic Depreux,
Renée K Edlund, Hongyuan Zhang, et al. 2016. “Fine-Tuning of Notch Signaling
Sets the Boundary of the Organ of Corti and Establishes Sensory Cell Fates.”
Edited by Tanya T Whitfield. ELife 5 (December): e19921.
https://doi.org/10.7554/eLife.19921.
Bergen, Volker, Marius Lange, Stefan Peidli, F. Alexander Wolf, and Fabian J. Theis.
2020. “Generalizing RNA Velocity to Transient Cell States through Dynamical
Modeling.” Nature Biotechnology 38 (12): 1408–14.
https://doi.org/10.1038/s41587-020-0591-3.
Bermingham, Nessan A., Bassem A. Hassan, Steven D. Price, Melissa A. Vollrath,
Nissim Ben-Arie, Ruth Anne Eatock, Hugo J. Bellen, Anna Lysakowski, and
Huda Y. Zoghbi. 1999. “Math1: An Essential Gene for the Generation of Inner
Ear Hair Cells.” Science 284 (5421): 1837–41.
https://doi.org/10.1126/science.284.5421.1837.
Bermingham-McDonogh, Olivia, Elizabeth C. Oesterle, Jennifer S. Stone, Clifford R.
Hume, Huy M. Huynh, and Toshinori Hayashi. 2006. “The Expression of Prox1
during Mouse Cochlear Development.” The Journal of Comparative Neurology
496 (2): 172–86. https://doi.org/10.1002/cne.20944.
Blackledge, Neil P., Anca M. Farcas, Takashi Kondo, Hamish W. King, Joanna F.
McGouran, Lars L. P. Hanssen, Shinsuke Ito, et al. 2014. “Variant PRC1
Complex-Dependent H2A Ubiquitylation Drives PRC2 Recruitment and
53

Polycomb Domain Formation.” Cell 157 (6): 1445–59.
https://doi.org/10.1016/j.cell.2014.05.004.
Bleckwehl, Tore, Giuliano Crispatzu, Kaitlin Schaaf, Patricia Respuela, Michaela
Bartusel, Laura Benson, Stephen J. Clark, et al. 2021. “Enhancer-Associated
H3K4 Methylation Safeguards in Vitro Germline Competence.” Nature
Communications 12 (1): 5771. https://doi.org/10.1038/s41467-021-26065-6.
Bok, Jinwoong, Colleen Zenczak, Chan Ho Hwang, and Doris K. Wu. 2013. “Auditory
Ganglion Source of Sonic Hedgehog Regulates Timing of Cell Cycle Exit and
Differentiation of Mammalian Cochlear Hair Cells.” Proceedings of the National
Academy of Sciences 110 (34): 13869–74.
https://doi.org/10.1073/pnas.1222341110.
Boller, Sören, Senthilkumar Ramamoorthy, Duygu Akbas, Robert Nechanitzky, Lukas
Burger, Rabih Murr, Dirk Schübeler, and Rudolf Grosschedl. 2016. “Pioneering
Activity of the C-Terminal Domain of EBF1 Shapes the Chromatin Landscape for
B Cell Programming.” Immunity 44 (3): 527–41.
https://doi.org/10.1016/j.immuni.2016.02.021.
Bramhall, Naomi F., Fuxin Shi, Katrin Arnold, Konrad Hochedlinger, and Albert S. B.
Edge. 2014. “Lgr5-Positive Supporting Cells Generate New Hair Cells in the
Postnatal Cochlea.” Stem Cell Reports 2 (3): 311–22.
https://doi.org/10.1016/j.stemcr.2014.01.008.
Brigande, John V., and Stefan Heller. 2009. “Quo Vadis, Hair Cell Regeneration?”
Nature Neuroscience 12 (6): 679–85. https://doi.org/10.1038/nn.2311.
Calo, Eliezer, and Joanna Wysocka. 2013. “Modification of Enhancer Chromatin: What,
How, and Why?” Molecular Cell 49 (5): 825–37.
https://doi.org/10.1016/j.molcel.2013.01.038.
Carlson, M. 2019. “Org. Mm. Eg. Db.” Genome Wide Annotation for Mouse.
Castro-Mondragon, Jaime A, Rafael Riudavets-Puig, Ieva Rauluseviciute, Roza
Berhanu Lemma, Laura Turchi, Romain Blanc-Mathieu, Jeremy Lucas, et al.
2022. “JASPAR 2022: The 9th Release of the Open-Access Database of
Transcription Factor Binding Profiles.” Nucleic Acids Research 50 (D1): D165–
73. https://doi.org/10.1093/nar/gkab1113.
Chai, Renjie, Anping Xia, Tian Wang, Taha Adnan Jan, Toshinori Hayashi, Olivia
Bermingham-McDonogh, and Alan Gi-Lun Cheng. 2011. “Dynamic Expression of
Lgr5, a Wnt Target Gene, in the Developing and Mature Mouse Cochlea.”
Journal of the Association for Research in Otolaryngology 12 (4): 455–69.
https://doi.org/10.1007/s10162-011-0267-2.
Chen, Ping, and Neil Segil. 1999. “P27Kip1 Links Cell Proliferation to Morphogenesis in
the Developing Organ of Corti.” Development 126 (8): 1581–90.
https://doi.org/10.1242/dev.126.8.1581.
Chen, Song, Blue B. Lake, and Kun Zhang. 2019. “High-Throughput Sequencing of the
Transcriptome and Chromatin Accessibility in the Same Cell.” Nature
Biotechnology 37 (12): 1452–57. https://doi.org/10.1038/s41587-019-0290-0.
Clark, Stephen J., Ricard Argelaguet, Chantriolnt-Andreas Kapourani, Thomas M.
Stubbs, Heather J. Lee, Celia Alda-Catalinas, Felix Krueger, et al. 2018.
“ScNMT-Seq Enables Joint Profiling of Chromatin Accessibility DNA Methylation
54

and Transcription in Single Cells.” Nature Communications 9 (1): 781.
https://doi.org/10.1038/s41467-018-03149-4.
Corwin, Jeffrey T., and Douglas A. Cotanche. 1988. “Regeneration of Sensory Hair
Cells After Acoustic Trauma.” Science 240 (4860): 1772–74.
https://doi.org/10.1126/science.3381100.
Cox, Brandon C., Renjie Chai, Anne Lenoir, Zhiyong Liu, LingLi Zhang, Duc-Huy
Nguyen, Kavita Chalasani, et al. 2014. “Spontaneous Hair Cell Regeneration in
the Neonatal Mouse Cochlea in Vivo.” Development 141 (4): 816–29.
https://doi.org/10.1242/dev.103036.
Creyghton, M. P., A. W. Cheng, G. G. Welstead, T. Kooistra, B. W. Carey, E. J. Steine,
J. Hanna, et al. 2010. “Histone H3K27ac Separates Active from Poised
Enhancers and Predicts Developmental State.” Proceedings of the National
Academy of Sciences 107 (50): 21931–36.
https://doi.org/10.1073/pnas.1016071107.
Cruickshanks, Karen J., Ted S. Tweed, Terry L. Wiley, Barbara E. K. Klein, Ronald
Klein, Rick Chappell, David M. Nondahl, and Dayna S. Dalton. 2003. “The 5-Year
Incidence and Progression of Hearing Loss: The Epidemiology of Hearing Loss
Study.” Archives of Otolaryngology –Head & Neck Surgery 129 (10): 1041–46.
https://doi.org/10.1001/archotol.129.10.1041.
Cunningham, Lisa L., and Debara L. Tucci. 2017. “Hearing Loss in Adults.” New
England Journal of Medicine 377 (25): 2465–73.
Dey, Siddharth S., Lennart Kester, Bastiaan Spanjaard, Magda Bienko, and Alexander
van Oudenaarden. 2015. “Integrated Genome and Transcriptome Sequencing of
the Same Cell.” Nature Biotechnology 33 (3): 285–89.
https://doi.org/10.1038/nbt.3129.
Dobin, Alexander, Carrie A. Davis, Felix Schlesinger, Jorg Drenkow, Chris Zaleski,
Sonali Jha, Philippe Batut, Mark Chaisson, and Thomas R. Gingeras. 2013.
“STAR: Ultrafast Universal RNA-Seq Aligner.” Bioinformatics 29 (1): 15–21.
https://doi.org/10.1093/bioinformatics/bts635.
Dodonova, Svetlana O., Fangjie Zhu, Christian Dienemann, Jussi Taipale, and Patrick
Cramer. 2020. “Nucleosome-Bound SOX2 and SOX11 Structures Elucidate
Pioneer Factor Function.” Nature 580 (7805): 669–72.
https://doi.org/10.1038/s41586-020-2195-y.
Doetzlhofer, Angelika, Martin L. Basch, Takahiro Ohyama, Manfred Gessler, Andrew K.
Groves, and Neil Segil. 2009. “Hey2 Regulation by FGF Provides a Notch-
Independent Mechanism for Maintaining Pillar Cell Fate in the Organ of Corti.”
Developmental Cell 16 (1): 58–69. https://doi.org/10.1016/j.devcel.2008.11.008.
Driver, Elizabeth C., and Matthew W. Kelley. 2009. “Specification of Cell Fate in the
Mammalian Cochlea.” Birth Defects Research Part C: Embryo Today: Reviews
87 (3): 212–21. https://doi.org/10.1002/bdrc.20154.
Driver, Elizabeth Carroll, and Matthew W. Kelley. 2020. “Development of the Cochlea.”
Development 147 (12): dev162263. https://doi.org/10.1242/dev.162263.
Driver, Elizabeth Carroll, Laura Sillers, Thomas M. Coate, Matthew F. Rose, and
Matthew W. Kelley. 2013. “The Atoh1-Lineage Gives Rise to Hair Cells and
Supporting Cells within the Mammalian Cochlea.” Developmental Biology 376
(1): 86–98. https://doi.org/10.1016/j.ydbio.2013.01.005.
55

Dror, Amiel A., and Karen B. Avraham. 2010. “Hearing Impairment: A Panoply of Genes
and Functions.” Neuron 68 (2): 293–308.
https://doi.org/10.1016/j.neuron.2010.10.011.
Duncan, Luke J., Dominic A. Mangiardi, Jonathan I. Matsui, Julia K. Anderson, Kate
MClaughlin-Williamson, and Douglas A. Cotanche. 2006. “Differential Expression
of Unconventional Myosins in Apoptotic and Regenerating Chick Hair Cells
Confirms Two Regeneration Mechanisms.” Journal of Comparative Neurology
499 (5): 691–701. https://doi.org/10.1002/cne.21114.
Durinck, Steffen, Yves Moreau, Arek Kasprzyk, Sean Davis, Bart De Moor, Alvis
Brazma, and Wolfgang Huber. 2005. “BioMart and Bioconductor: A Powerful Link
between Biological Databases and Microarray Data Analysis.” Bioinformatics 21
(16): 3439–40. https://doi.org/10.1093/bioinformatics/bti525.
Durrant, J. D., and J. H. Lovrinic. 1995. “Bases of Hearing Science. Balitmore.” MD:
Williams & Wilkins.
Ebeid, Michael, Kathy Barnas, Hongji Zhang, Amal Yaghmour, Gabriele Noreikaite, and
Bryan C. Bjork. 2022. “PRDM16 Expression and Function in Mammalian
Cochlear Development.” Developmental Dynamics 251 (10): 1666–83.
https://doi.org/10.1002/dvdy.480.
Fiers, Mark W. E. J., Liesbeth Minnoye, Sara Aibar, Carmen Bravo González-Blas,
Zeynep Kalender Atak, and Stein Aerts. 2018. “Mapping Gene Regulatory
Networks from Single-Cell Omics Data.” Briefings in Functional Genomics 17 (4):
246–54. https://doi.org/10.1093/bfgp/elx046.
Fleck, Jonas Simon, Sophie Martina Johanna Jansen, Damian Wollny, Fides Zenk,
Makiko Seimiya, Akanksha Jain, Ryoko Okamoto, et al. 2022. “Inferring and
Perturbing Cell Fate Regulomes in Human Brain Organoids.” Nature, October, 1–
8. https://doi.org/10.1038/s41586-022-05279-8.
Fritzsch, B., K. W. Beisel, K. Jones, I. Fariñas, A. Maklad, J. Lee, and L. F. Reichardt.
2002. “Development and Evolution of Inner Ear Sensory Epithelia and Their
Innervation.” Journal of Neurobiology 53 (2): 143–56.
https://doi.org/10.1002/neu.10098.
Gk, Michalopoulos, Barua L, and Bowen Wc. 2005. “Transdifferentiation of Rat
Hepatocytes into Biliary Cells after Bile Duct Ligation and Toxic Biliary Injury.”
Hepatology (Baltimore, Md.) 41 (3). https://doi.org/10.1002/hep.20600.
Golden, Erin J., Ana Benito-Gonzalez, and Angelika Doetzlhofer. 2015. “The RNA-
Binding Protein LIN28B Regulates Developmental Timing in the Mammalian
Cochlea.” Proceedings of the National Academy of Sciences 112 (29): E3864–
73. https://doi.org/10.1073/pnas.1501077112.
Goldman, Joseph A., and Kenneth D. Poss. 2020. “Gene Regulatory Programmes of
Tissue Regeneration.” Nature Reviews Genetics 21 (9): 511–25.
https://doi.org/10.1038/s41576-020-0239-7.
González-Blas, Carmen Bravo, Seppe De Winter, Gert Hulselmans, Nikolai Hecker,
Irina Matetovici, Valerie Christiaens, Suresh Poovathingal, Jasper Wouters, Sara
Aibar, and Stein Aerts. 2022. “SCENIC+: Single-Cell Multiomic Inference of
Enhancers and Gene Regulatory Networks.” bioRxiv.
https://doi.org/10.1101/2022.08.19.504505.
56

Gu, Rende, Rogers M. Brown, Chih-Wei Hsu, Tiantian Cai, Alyssa L. Crowder, Victor G.
Piazza, Tegy J. Vadakkan, Mary E. Dickinson, and Andrew K. Groves. 2016.
“Lineage Tracing of Sox2-Expressing Progenitor Cells in the Mouse Inner Ear
Reveals a Broad Contribution to Non-Sensory Tissues and Insights into the
Origin of the Organ of Corti.” Developmental Biology 414 (1): 72–84.
https://doi.org/10.1016/j.ydbio.2016.03.027.
Habener, Joel F., and Violeta Stanojevic. 2012. “α-Cell Role in β-Cell Generation and
Regeneration.” Islets 4 (3): 188. https://doi.org/10.4161/isl.20500.
Haddon, Catherine, and Julian Lewis. 1996. “Early Ear Development in the Embryo of
the Zebrafish, Danio Rerio.” Journal of Comparative Neurology 365 (1): 113–28.
https://doi.org/10.1002/(SICI)1096-9861(19960129)365:1<113::AID-
CNE9>3.0.CO;2-6.
Harms, Matthew J., Hee-Woong Lim, Yugong Ho, Suzanne N. Shapira, Jeff Ishibashi,
Sona Rajakumari, David J. Steger, Mitchell A. Lazar, Kyoung-Jae Won, and
Patrick Seale. 2015. “PRDM16 Binds MED1 and Controls Chromatin Architecture
to Determine a Brown Fat Transcriptional Program.” Genes & Development 29
(3): 298–307. https://doi.org/10.1101/gad.252734.114.
Hartman, Byron, Robert Durruthy-Durruthy, Roman Laske, Steven Losorelli, and Stefan
Heller. 2015. “Identification and Characterization of Mouse Otic Sensory Lineage
Genes.” Frontiers in Cellular Neuroscience 9 (March): 79.
https://doi.org/10.3389/fncel.2015.00079.
Hemberger, Myriam, Wendy Dean, and Wolf Reik. 2009. “Epigenetic Dynamics of Stem
Cells and Cell Lineage Commitment: Digging Waddington’s Canal.” Nature
Reviews Molecular Cell Biology 10 (8): 526–37. https://doi.org/10.1038/nrm2727.
Hickey, Graham JM, Candice L Wike, Xichen Nie, Yixuan Guo, Mengyao Tan, Patrick J
Murphy, and Bradley R Cairns. 2022. “Establishment of Developmental Gene
Silencing by Ordered Polycomb Complex Recruitment in Early Zebrafish
Embryos.” Edited by Xin Chen, Richard M White, and Charles K Kaufman. ELife
11 (January): e67738. https://doi.org/10.7554/eLife.67738.
Hj, Adler, and Raphael Y. 1996. “New Hair Cells Arise from Supporting Cell Conversion
in the Acoustically Damaged Chick Inner Ear.” Neuroscience Letters 205 (1).
https://doi.org/10.1016/0304-3940(96)12367-3.
Hou, Yu, Huahu Guo, Chen Cao, Xianlong Li, Boqiang Hu, Ping Zhu, Xinglong Wu, et
al. 2016. “Single-Cell Triple Omics Sequencing Reveals Genetic, Epigenetic, and
Transcriptomic Heterogeneity in Hepatocellular Carcinomas.” Cell Research 26
(3): 304–19. https://doi.org/10.1038/cr.2016.23.
Hu, Youjin, Kevin Huang, Qin An, Guizhen Du, Ganlu Hu, Jinfeng Xue, Xianmin Zhu,
Cun-Yu Wang, Zhigang Xue, and Guoping Fan. 2016. “Simultaneous Profiling of
Transcriptome and DNA Methylome from a Single Cell.” Genome Biology 17 (1):
1–11. https://doi.org/10.1186/s13059-016-0950-z.
Hume, Maxwell A., Luis A. Barrera, Stephen S. Gisselbrecht, and Martha L. Bulyk.
2015. “UniPROBE, Update 2015: New Tools and Content for the Online
Database of Protein-Binding Microarray Data on Protein-DNA Interactions.”
Nucleic Acids Research 43 (Database issue): D117-122.
https://doi.org/10.1093/nar/gku1045.
57

Iacono, Giovanni, Ramon Massoni-Badosa, and Holger Heyn. 2019. “Single-Cell
Transcriptomics Unveils Gene Regulatory Network Plasticity.” Genome Biology
20 (1): 110. https://doi.org/10.1186/s13059-019-1713-4.
K, Yanger, Zong Y, Maggs Lr, Shapira Sn, Maddipati R, Aiello Nm, Thung Sn, Wells Rg,
Greenbaum Le, and Stanger Bz. 2013. “Robust Cellular Reprogramming Occurs
Spontaneously during Liver Regeneration.” Genes & Development 27 (7).
https://doi.org/10.1101/gad.207803.112.
Kamimoto, Kenji, Blerta Stringa, Christy M. Hoffmann, Kunal Jindal, Lilianna Solnica-
Krezel, and Samantha A. Morris. 2023. “Dissecting Cell Identity via Network
Inference and in Silico Gene Perturbation.” Nature 614 (7949): 742–51.
https://doi.org/10.1038/s41586-022-05688-9.
Kang, Junsu, Jianxin Hu, Ravi Karra, Amy L. Dickson, Valerie A. Tornini, Gregory
Nachtrab, Matthew Gemberling, Joseph A. Goldman, Brian L. Black, and
Kenneth D. Poss. 2016. “Modulation of Tissue Repair by Regeneration Enhancer
Elements.” Nature 532 (7598): 201–6. https://doi.org/10.1038/nature17644.
Kartha, Vinay K., Fabiana M. Duarte, Yan Hu, Sai Ma, Jennifer G. Chew, Caleb A.
Lareau, Andrew Earl, et al. 2022. “Functional Inference of Gene Regulation
Using Single-Cell Multi-Omics.” Cell Genomics 2 (9): 100166.
https://doi.org/10.1016/j.xgen.2022.100166.
Kassambara, Alboukadel, and Maintainer Alboukadel Kassambara. 2020. “Package
‘Ggpubr.’” R Package Version 0.1 6 (0).
Kelley, Matthew W. 2006. “Regulation of Cell Fate in the Sensory Epithelia of the Inner
Ear.” Nature Reviews Neuroscience 7 (11): 837–49.
https://doi.org/10.1038/nrn1987.
———. 2007. “Cellular Commitment and Differentiation in the Organ of Corti.”
International Journal of Developmental Biology 51 (6–7): 571–83.
https://doi.org/10.1387/ijdb.072388mk.
Kelly, Michael C., Qing Chang, Alex Pan, Xi Lin, and Ping Chen. 2012. “Atoh1 Directs
the Formation of Sensory Mosaics and Induces Cell Proliferation in the Postnatal
Mammalian Cochlea In Vivo.” Journal of Neuroscience 32 (19): 6699–6710.
https://doi.org/10.1523/JNEUROSCI.5420-11.2012.
Kulakovskiy, Ivan V, Ilya E Vorontsov, Ivan S Yevshin, Ruslan N Sharipov, Alla D
Fedorova, Eugene I Rumynskiy, Yulia A Medvedeva, et al. 2018. “HOCOMOCO:
Towards a Complete Collection of Transcription Factor Binding Models for
Human and Mouse via Large-Scale ChIP-Seq Analysis.” Nucleic Acids Research
46 (Database issue): D252–59. https://doi.org/10.1093/nar/gkx1106.
Lawrence, Michael, Robert Gentleman, and Vincent Carey. 2009. “Rtracklayer: An R
Package for Interfacing with Genome Browsers.” Bioinformatics 25 (14): 1841–
42.
Lawrence, Michael, Wolfgang Huber, Hervé Pagès, Patrick Aboyoun, Marc Carlson,
Robert Gentleman, Martin T. Morgan, and Vincent J. Carey. 2013. “Software for
Computing and Annotating Genomic Ranges.” PLOS Computational Biology 9
(8): e1003118. https://doi.org/10.1371/journal.pcbi.1003118.
Le Douarin, Nicole M. 1984. “Cell Migrations in Embryos.” Cell 38 (2): 353–60.
https://doi.org/10.1016/0092-8674(84)90490-2.
58

Lee, Kihyun, Hyunwoo Cho, Robert W. Rickert, Qing V. Li, Julian Pulecio, Christina S.
Leslie, and Danwei Huangfu. 2019. “FOXA2 Is Required for Enhancer Priming
during Pancreatic Differentiation.” Cell Reports 28 (2): 382-393.e7.
https://doi.org/10.1016/j.celrep.2019.06.034.
Lee, Yun-Shain, Feng Liu, and Neil Segil. 2006. “A Morphogenetic Wave of P27Kip1
Transcription Directs Cell Cycle Exit during Organ of Corti Development.”
Development 133 (15): 2817–26. https://doi.org/10.1242/dev.02453.
Li, Bo, and Colin N. Dewey. 2011. “RSEM: Accurate Transcript Quantification from
RNA-Seq Data with or without a Reference Genome.” BMC Bioinformatics 12
(August): 323. https://doi.org/10.1186/1471-2105-12-323.
Li, Heng, Bob Handsaker, Alec Wysoker, Tim Fennell, Jue Ruan, Nils Homer, Gabor
Marth, Goncalo Abecasis, Richard Durbin, and 1000 Genome Project Data
Processing Subgroup. 2009. “The Sequence Alignment/Map Format and
SAMtools.” Bioinformatics 25 (16): 2078–79.
Li, Qunhua, James B. Brown, Haiyan Huang, and Peter J. Bickel. 2011. “Measuring
Reproducibility of High-Throughput Experiments.” The Annals of Applied
Statistics 5 (3): 1752–79. https://doi.org/10.1214/11-AOAS466.
Li, Shuting, Ting Fan, Chao Li, Yunfeng Wang, Jian Li, and Zhiyong Liu. 2022. “Fate-
Mapping Analysis of Cochlear Cells Expressing Atoh1 MRNA via a New
Atoh13*HA-P2A-Cre Knockin Mouse Strain.” Developmental Dynamics 251 (7):
1156–74. https://doi.org/10.1002/dvdy.453.
Liu, Longqi, Chuanyu Liu, Andrés Quintero, Liang Wu, Yue Yuan, Mingyue Wang,
Mengnan Cheng, et al. 2019. “Deconvolution of Single-Cell Multi-Omics Layers
Reveals Regulatory Heterogeneity.” Nature Communications 10 (1): 470.
https://doi.org/10.1038/s41467-018-08205-7.
Liu, Zhiyong, Thomas Owen, Jie Fang, R. Sathish Srinivasan, and Jian Zuo. 2012. “In
Vivo Notch Reactivation in Differentiating Cochlear Hair Cells Induces Sox2 and
Prox1 Expression but Does Not Disrupt Hair Cell Maturation.” Developmental
Dynamics 241 (4): 684–96. https://doi.org/10.1002/dvdy.23754.
Lo, Yuan-Hung, Eunah Chung, Zhaohui Li, Ying-Wooi Wan, Maxime M. Mahe, Min-
Shan Chen, Taeko K. Noah, et al. 2017. “Transcriptional Regulation by ATOH1
and Its Target SPDEF in the Intestine.” Cellular and Molecular Gastroenterology
and Hepatology 3 (1): 51. https://doi.org/10.1016/j.jcmgh.2016.10.001.
Love, Michael I., Wolfgang Huber, and Simon Anders. 2014. “Moderated Estimation of
Fold Change and Dispersion for RNA-Seq Data with DESeq2.” Genome Biology
15 (12): 550. https://doi.org/10.1186/s13059-014-0550-8.
Ma, Sai, Bing Zhang, Lindsay M. LaFave, Andrew S. Earl, Zachary Chiang, Yan Hu,
Jiarui Ding, et al. 2020. “Chromatin Potential Identified by Shared Single-Cell
Profiling of RNA and Chromatin.” Cell 183 (4): 1103-1116.e20.
https://doi.org/10.1016/j.cell.2020.09.056.
Maass, Juan C., Rende Gu, Martin L. Basch, Joerg Waldhaus, Eduardo Martin Lopez,
Anping Xia, John S. Oghalai, Stefan Heller, and Andrew K. Groves. 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. https://www.frontiersin.org/article/10.3389/fncel.2015.00110.
59

Macaulay, Iain C., Wilfried Haerty, Parveen Kumar, Yang I. Li, Tim Xiaoming Hu, Mabel
J. Teng, Mubeen Goolam, et al. 2015. “G&T-Seq: Parallel Sequencing of Single-
Cell Genomes and Transcriptomes.” Nature Methods 12 (6): 519–22.
https://doi.org/10.1038/nmeth.3370.
Maintainer, Maintainer Bioconductor Package, Biostrings GenomicRanges, and
BiocParallel Rsamtools. 2015. “Package ‘GenomicAlignments.’”
Martin, Marcel. 2011. “Cutadapt Removes Adapter Sequences from High-Throughput
Sequencing Reads.” EMBnet.Journal 17 (1): 10–12.
https://doi.org/10.14806/ej.17.1.200.
McLean, Cory Y, Dave Bristor, Michael Hiller, Shoa L Clarke, Bruce T Schaar, Craig B
Lowe, Aaron M Wenger, and Gill Bejerano. 2010. “GREAT Improves Functional
Interpretation of Cis-Regulatory Regions.” Nature Biotechnology 28 (5): 495–501.
https://doi.org/10.1038/nbt.1630.
McLeay, Robert C., and Timothy L. Bailey. 2010. “Motif Enrichment Analysis: A Unified
Framework and an Evaluation on ChIP Data.” BMC Bioinformatics 11: 165.
https://doi.org/10.1186/1471-2105-11-165.
Mizutari, Kunio, Masato Fujioka, Makoto Hosoya, Naomi Bramhall, Hirotaka James
Okano, Hideyuki Okano, and Albert S. B. Edge. 2013. “Notch Inhibition Induces
Cochlear Hair Cell Regeneration and Recovery of Hearing after Acoustic
Trauma.” Neuron 77 (1): 58–69. https://doi.org/10.1016/j.neuron.2012.10.032.
Montcouquiol, Mireille, and Matthew W. Kelley. 2003. “Planar and Vertical Signals
Control Cellular Differentiation and Patterning in the Mammalian Cochlea.”
Journal of Neuroscience 23 (28): 9469–78.
https://doi.org/10.1523/JNEUROSCI.23-28-09469.2003.
Morsli, Hakim, Daniel Choo, Allen Ryan, Randy Johnson, and Doris K. Wu. 1998.
“Development of the Mouse Inner Ear and Origin of Its Sensory Organs.” Journal
of Neuroscience 18 (9): 3327–35. https://doi.org/10.1523/JNEUROSCI.18-09-
03327.1998.
Morton, Cynthia C., and Walter E. Nance. 2006. “Newborn Hearing Screening — A
Silent Revolution.” New England Journal of Medicine 354 (20): 2151–64.
https://doi.org/10.1056/NEJMra050700.
Napoles, Mariana de, Jacqueline E. Mermoud, Rika Wakao, Y. Amy Tang, Mitusuhiro
Endoh, Ruth Appanah, Tatyana B. Nesterova, et al. 2004. “Polycomb Group
Proteins Ring1A/B Link Ubiquitylation of Histone H2A to Heritable Gene
Silencing and X Inactivation.” Developmental Cell 7 (5): 663–76.
https://doi.org/10.1016/j.devcel.2004.10.005.
Nicetto, Dario, and Kenneth S. Zaret. 2019. “Role of H3K9me3 Heterochromatin in Cell
Identity Establishment and Maintenance.” Current Opinion in Genetics &
Development, Genome Architecture and Expression, 55 (April): 1–10.
https://doi.org/10.1016/j.gde.2019.04.013.
Ong, Chin-Tong, and Victor G Corces. 2012. “Enhancers: Emerging Roles in Cell Fate
Specification.” EMBO Reports 13 (5): 423–30.
https://doi.org/10.1038/embor.2012.52.
Ou, Jianhong, Michael Brodsky, Scot Wolfe, and Lihua Julie Zhu. 2023. “MotifStack:
Plot Stacked Logos for Single or Multiple DNA, RNA and Amino Acid Sequence.”
60

Bioconductor version: Release (3.16).
https://doi.org/10.18129/B9.bioc.motifStack.
Pages, Herve, Marc Carlson, Seth Falcon, Nianhua Li, and Maintainer Bioconductor
Package Maintainer. 2017. “Package ‘AnnotationDbi.’” Bioconductor Packag.
Maint.
Phippard, Deborah, Ahlke Heydemann, Mark Lechner, Lihui Lu, Daniel Lee, Tim Kyin,
and E. Bryan Crenshaw. 1998. “Changes in the Subcellular Localization of the
Brn4 Gene Product Precede Mesenchymal Remodeling of the Otic Capsule.”
Hearing Research 120 (1): 77–85. https://doi.org/10.1016/S0378-5955(98)00059-
8.
“Picard Toolkit.” 2019. Broad Institute, GitHub Repository. Broad Institute.
https://broadinstitute.github.io/picard/.
Prajapati-DiNubila, Meenakshi, Ana Benito-Gonzalez, Erin Jennifer Golden, Shuran
Zhang, and Angelika Doetzlhofer. 2019. “A Counter Gradient of Activin A and
Follistatin Instructs the Timing of Hair Cell Differentiation in the Murine Cochlea.”
Edited by Marianne E Bronner, Tanya T Whitfield, and Matthew W Kelley. ELife 8
(June): e47613. https://doi.org/10.7554/eLife.47613.
Purves, D., G. Augustine, D. Fitzpatrick, L. C. Katz, A. S. LaMantia, J. O. McNamara,
and S. M. Williams. 2001. “Hair Cells and the Mechanoelectrical Transduction of
Sound Waves.” Neuroscience, 294–99.
Ramírez, Fidel, Devon P. Ryan, Björn Grüning, Vivek Bhardwaj, Fabian Kilpert, Andreas
S. Richter, Steffen Heyne, Friederike Dündar, and Thomas Manke. 2016.
“DeepTools2: A next Generation Web Server for Deep-Sequencing Data
Analysis.” Nucleic Acids Research 44 (W1): W160–65.
Roberson, David W., Julie A. Alosi, and Douglas A. Cotanche. 2004. “Direct
Transdifferentiation Gives Rise to the Earliest New Hair Cells in Regenerating
Avian Auditory Epithelium.” Journal of Neuroscience Research 78 (4): 461–71.
https://doi.org/10.1002/jnr.20271.
Robinson, Mark D., Davis J. McCarthy, and Gordon K. Smyth. 2010. “EdgeR: A
Bioconductor Package for Differential Expression Analysis of Digital Gene
Expression Data.” Bioinformatics 26 (1): 139–40.
https://doi.org/10.1093/bioinformatics/btp616.
Rooijers, Koos, Corina M. Markodimitraki, Franka J. Rang, Sandra S. de Vries, Alex
Chialastri, Kim L. de Luca, Dylan Mooijman, Siddharth S. Dey, and Jop Kind.
2019. “Simultaneous Quantification of Protein–DNA Contacts and
Transcriptomes in Single Cells.” Nature Biotechnology 37 (7): 766–72.
https://doi.org/10.1038/s41587-019-0150-y.
Rubel, Edwin W., and Bernd Fritzsch. 2002. “Auditory System Development: Primary
Auditory Neurons and Their Targets.” Annual Review of Neuroscience 25 (1):
51–101. https://doi.org/10.1146/annurev.neuro.25.112701.142849.
Ryals, Brenda M., and Edwin W. Rubel. 1988. “Hair Cell Regeneration after Acoustic
Trauma in Adult Coturnix Quail.” Science 240 (4860): 1774–76.
Sánchez-Calderón, Hortensia, Marta Milo, Yolanda León, and Isabel Varela-nieto. 2007.
“A Network of Growth and Transcription Factors Controls Neuronal Differentation
and Survival in the Developing Ear.” International Journal of Developmental
Biology 51 (6–7): 557–70. https://doi.org/10.1387/ijdb.072373hs.
61

Satoh, Takunori, and Donna M. Fekete. 2005. “Clonal Analysis of the Relationships
between Mechanosensory Cells and the Neurons That Innervate Them in the
Chicken Ear.” Development 132 (7): 1687–97. https://doi.org/10.1242/dev.01730.
Shukla, Shirish, Weijiang Ying, Felicia Gray, Yiwu Yao, Miranda L. Simes, Qingjie Zhao,
Hongzhi Miao, et al. 2021. “Small Molecule Inhibitors Targeting Polycomb
Repressive Complex 1 RING Domain.” Nature Chemical Biology 17 (7): 784–93.
https://doi.org/10.1038/s41589-021-00815-5.
Skene, Peter J, and Steven Henikoff. 2017. “An Efficient Targeted Nuclease Strategy
for High-Resolution Mapping of DNA Binding Sites.” Edited by Danny Reinberg.
ELife 6 (January): e21856. https://doi.org/10.7554/eLife.21856.
Smith, Tom, Andreas Heger, and Ian Sudbery. 2017. “UMI-Tools: Modeling Sequencing
Errors in Unique Molecular Identifiers to Improve Quantification Accuracy.”
Genome Research 27 (3): 491–99. https://doi.org/10.1101/gr.209601.116.
Stojanova, Zlatka P., Tao Kwan, and Neil Segil. 2015. “Epigenetic Regulation of Atoh1
Guides Hair Cell Development in the Mammalian Cochlea.” Development
(Cambridge, England) 142 (20): 3529–36. https://doi.org/10.1242/dev.126763.
Takebayashi, Shinji, Norio Yamamoto, Daisuke Yabe, Hitoshi Fukuda, Ken Kojima,
Juichi Ito, and Tasuku Honjo. 2007. “Multiple Roles of Notch Signaling in
Cochlear Development.” Developmental Biology 307 (1): 165–78.
https://doi.org/10.1016/j.ydbio.2007.04.035.
Tang, Louisa S., Heather M. Alger, Feng Lin, and Fred A. Pereira. 2005. “Dynamic
Expression of COUP-TFI and COUP-TFII during Development and Functional
Maturation of the Mouse Inner Ear.” Gene Expression Patterns 5 (5): 587–92.
https://doi.org/10.1016/j.modgep.2005.03.012.
Tao, Litao, Haoze V. Yu, Juan Llamas, Talon Trecek, Xizi Wang, Zlatka Stojanova,
Andrew K. Groves, and Neil Segil. 2021. “Enhancer Decommissioning Imposes
an Epigenetic Barrier to Sensory Hair Cell Regeneration.” Developmental Cell 56
(17): 2471-2485.e5. https://doi.org/10.1016/j.devcel.2021.07.003.
Taudt, Aaron, Minh Anh Nguyen, Matthias Heinig, Frank Johannes, and Maria Colome-
Tatche. 2016. “ChromstaR: Tracking Combinatorial Chromatin State Dynamics in
Space and Time.” BioRxiv, August, 038612. https://doi.org/10.1101/038612.
Thorel, Fabrizio, Virginie Népote, Isabelle Avril, Kenji Kohno, Renaud Desgraz, Simona
Chera, and Pedro L. Herrera. 2010. “Conversion of Adult Pancreatic α-Cells to β-
Cells after Extreme β-Cell Loss.” Nature 464 (7292): 1149–54.
https://doi.org/10.1038/nature08894.
Tj, Klisch, Xi Y, Flora A, Wang L, Li W, and 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 (8). https://doi.org/10.1073/pnas.1100230108.
Treiber, Thomas, Elizabeth M. Mandel, Sebastian Pott, Ildiko Györy, Sonja Firner,
Edison T. Liu, and Rudolf Grosschedl. 2010. “Early B Cell Factor 1 Regulates B
Cell Gene Networks by Activation, Repression, and Transcription- Independent
Poising of Chromatin.” Immunity 32 (5): 714–25.
https://doi.org/10.1016/j.immuni.2010.04.013.
Van Camp, Guy. 2022. “Hereditary Hearing Loss Homepage.” URL: Http://Dnalab-
Www. Uia. Ac. Be/Dnalab/Hhh/.
62

VandenBosch, Leah S., Stefanie G. Wohl, Matthew S. Wilken, Marcus Hooper, Connor
Finkbeiner, Kristen Cox, Laura Chipman, and Thomas A. Reh. 2020.
“Developmental Changes in the Accessible Chromatin, Transcriptome and Ascl1-
Binding Correlate with the Loss in Müller Glial Regenerative Potential.” Scientific
Reports 10 (1): 13615. https://doi.org/10.1038/s41598-020-70334-1.
Wang, Allen, Feng Yue, Yan Li, Ruiyu Xie, Thomas Harper, Nisha A. Patel, Kayla Muth,
et al. 2015. “Epigenetic Priming of Enhancers Predicts Developmental
Competence of HESC-Derived Endodermal Lineage Intermediates.” Cell Stem
Cell 16 (4): 386–99. https://doi.org/10.1016/j.stem.2015.02.013.
Wang, Chaochen, Ji-Eun Lee, Binbin Lai, Todd S. Macfarlan, Shiliyang Xu, Lenan
Zhuang, Chengyu Liu, Weiqun Peng, and Kai Ge. 2016. “Enhancer Priming by
H3K4 Methyltransferase MLL4 Controls Cell Fate Transition.” Proceedings of the
National Academy of Sciences 113 (42): 11871–76.
https://doi.org/10.1073/pnas.1606857113.
Wang, Hengbin, Liangjun Wang, Hediye Erdjument-Bromage, Miguel Vidal, Paul
Tempst, Richard S. Jones, and Yi Zhang. 2004. “Role of Histone H2A
Ubiquitination in Polycomb Silencing.” Nature 431 (7010): 873–78.
https://doi.org/10.1038/nature02985.
White, Patricia M., Angelika Doetzlhofer, Yun Shain Lee, Andrew K. Groves, and Neil
Segil. 2006. “Mammalian Cochlear Supporting Cells Can Divide and Trans-
Differentiate into Hair Cells.” Nature 441 (7096): 984–87.
https://doi.org/10.1038/nature04849.
Wolf, F. Alexander, Philipp Angerer, and Fabian J. Theis. 2018. “SCANPY: Large-Scale
Single-Cell Gene Expression Data Analysis.” Genome Biology 19 (1): 1–5.
https://doi.org/10.1186/s13059-017-1382-0.
Wolf, F. Alexander, Fiona K. Hamey, Mireya Plass, Jordi Solana, Joakim S. Dahlin,
Berthold Göttgens, Nikolaus Rajewsky, Lukas Simon, and Fabian J. Theis. 2019.
“PAGA: Graph Abstraction Reconciles Clustering with Trajectory Inference
through a Topology Preserving Map of Single Cells.” Genome Biology 20 (1): 1–
9. https://doi.org/10.1186/s13059-019-1663-x.
Xu, Wei, Weilong Yang, Yunlong Zhang, Yawen Chen, Ni Hong, Qian Zhang, Xuefei
Wang, et al. 2022. “ISSAAC-Seq Enables Sensitive and Flexible Multimodal
Profiling of Chromatin Accessibility and Gene Expression in Single Cells.” Nature
Methods 19 (10): 1243–49. https://doi.org/10.1038/s41592-022-01601-4.
Y, Zong, Panikkar A, Xu J, Antoniou A, Raynaud P, Lemaigre F, and Stanger Bz. 2009.
“Notch Signaling Controls Liver Development by Regulating Biliary
Differentiation.” Development (Cambridge, England) 136 (10).
https://doi.org/10.1242/dev.029140.
Yang, K., and J. Kang. 2019. “Tissue Regeneration Enhancer Elements: A Way to
Unlock Endogenous Healing Power.” Developmental Dynamics 248 (1): 34–42.
https://doi.org/10.1002/dvdy.24676.
Yu, Guangchuang, and Qing-Yu He. 2016. “ReactomePA: An R/Bioconductor Package
for Reactome Pathway Analysis and Visualization.” Molecular BioSystems 12 (2):
477–79.
63

Yu, Guangchuang, Li-Gen Wang, Yanyan Han, and Qing-Yu He. 2012. “ClusterProfiler:
An R Package for Comparing Biological Themes among Gene Clusters.” Omics:
A Journal of Integrative Biology 16 (5): 284–87.
Yu, Haoze V., Litao Tao, Juan Llamas, Xizi Wang, John D. Nguyen, Talon Trecek, and
Neil Segil. 2021. “POU4F3 Pioneer Activity Enables ATOH1 to Drive Diverse
Mechanoreceptor Differentiation through a Feed-Forward Epigenetic
Mechanism.” Proceedings of the National Academy of Sciences 118 (29):
e2105137118. https://doi.org/10.1073/pnas.2105137118.
Zaret, Kenneth S., and Susan Mango. 2016. “Pioneer Transcription Factors, Chromatin
Dynamics, and Cell Fate Control.” Current Opinion in Genetics & Development
37 (April): 76–81. https://doi.org/10.1016/j.gde.2015.12.003.
Zentner, Gabriel E., Paul J. Tesar, and Peter C. Scacheri. 2011. “Epigenetic Signatures
Distinguish Multiple Classes of Enhancers with Distinct Cellular Functions.”
Genome Research 21 (8): 1273–83. https://doi.org/10.1101/gr.122382.111.
Zhang, Yong, Tao Liu, Clifford A Meyer, Jérôme Eeckhoute, David S Johnson, Bradley
E Bernstein, Chad Nusbaum, et al. 2008. “Model-Based Analysis of ChIP-Seq
(MACS).” Genome Biology 9 (9): R137. https://doi.org/10.1186/gb-2008-9-9-r137.
Zheng, J. Lisa, Jianyong Shou, Francois Guillemot, Ryoichiro Kageyama, and Wei-
Qiang Gao. 2000. “Hes1 Is a Negative Regulator of Inner Ear Hair Cell
Differentiation.” Development 127 (21): 4551–60.
https://doi.org/10.1242/dev.127.21.4551.
Zhou, Yeqiao, Jelena Petrovic, Jingru Zhao, Wu Zhang, Ashkan Bigdeli, Zhen Zhang,
Shelley L. Berger, Warren S. Pear, and Robert B. Faryabi. 2022. “EBF1 Nuclear
Repositioning Instructs Chromatin Refolding to Promote Therapy Resistance in T
Leukemic Cells.” Molecular Cell, February.
https://doi.org/10.1016/j.molcel.2022.01.015.
Zhu, Chenxu, Miao Yu, Hui Huang, Ivan Juric, Armen Abnousi, Rong Hu, Jacinta
Lucero, M. Margarita Behrens, Ming Hu, and Bing Ren. 2019. “An Ultra High-
Throughput Method for Single-Cell Joint Analysis of Open Chromatin and
Transcriptome.” Nature Structural & Molecular Biology 26 (11): 1063–70.
https://doi.org/10.1038/s41594-019-0323-x.
Zhu, Chenxu, Yanxiao Zhang, Yang Eric Li, Jacinta Lucero, M. Margarita Behrens, and
Bing Ren. 2021. “Joint Profiling of Histone Modifications and Transcriptome in
Single Cells from Mouse Brain.” Nature Methods 18 (3): 283–92.
https://doi.org/10.1038/s41592-021-01060-3.
Zhu, Lihua J., Claude Gazin, Nathan D. Lawson, Hervé Pagès, Simon M. Lin, David S.
Lapointe, and Michael R. Green. 2010. “ChIPpeakAnno: A Bioconductor
Package to Annotate ChIP-Seq and ChIP-Chip Data.” BMC Bioinformatics 11: 1–
10.
Zine, Azel, Alexandre Aubert, Jiping Qiu, Stavros Therianos, Francois Guillemot,
Ryoichiro Kageyama, and Francois de Ribaupierre. 2001. “Hes1 and Hes5
Activities Are Required for the Normal Development of the Hair Cells in the
Mammalian Inner Ear.” The Journal of Neuroscience 21 (13): 4712–20.

64

Appendix I: Materials and Methods
Mice
All experiments were performed at the University of Southern California. All animal
experiments were conducted according to the National Institutes of Health Guide for
Care and Use of Laboratory Animals. Protocols and experiments using animals were
approved by the Institutional Animal Care and Use Committee at the University of
Southern California. Mice were housed with free access to chow and water on a 12hr
day/night cycle. Breeding and genotyping of the mice was performed according to USC
standard procedures.

Sox2-Cre-ERT2 mice for lineage tracing sensory progenitors (JAX #017593)
Ai14(RCL-tdT)-D reporter mice (JAX #007914) with tdTomato fluorescent protein and
Floxed transcriptional STOP cassette.
P27Kip1-GFP mice for marking sensory progenitors in the organ of Corti.
Lfng-GFP mice for marking supporting cells in the organ of Corti.
Atoh1-fGFP for marking hair cells in organ of Corti.

CUT&RUN
CUT&RUN was used for chromatin profiling of histones and transcription factors using
the low-salt option protocol (Skene and Henikoff 2017). For each replicate a minimum of
5,000 and maximum of 10,000 cells collected from FACS were used as input. The input
cell count for samples used in comparisons were within 20% of each other. Paired-end
reads were quality trimmed to 36 bp with cutadapt (v1.18) and aligned to mm10
reference genome (Gencode Mm10v11) with STAR aligner using parameters end-to-
end and alignIntronMax=1 for DNA alignment. PCR duplicates were removed with
STAR. Only autosomal chromosomes were selected and used for downstream analysis.

Antibodies
Anti-H3K27me3 (Active Motif, catalog #39155)
Anti-H3K27ac (Active Motif, catalog #39133)
Anti-H3K4me1 (Active Motif, catalog #61633)
Anti-H2AK119ub (Cell Signaling Technology, catalog #8240S)
Anti-CD326 EpCAM Microbeads (Miltenyi Biotec, catalog #130-105-958)

Paired-Tag
Protocol was follow as described in (C. Zhu et al. 2021) with the following modifications.
Nuclei were filtered with 10-20 micron pluriSelect cell filter on first pool and a 20uM filter
on last pool.

Motif Analysis
Motif enrichment analysis was performed with AMEs using mouse uniprobe,
HOCOMOCO, and JASPAR motif database (Castro-Mondragon et al. 2022;
Kulakovskiy et al. 2018; Hume et al. 2015; McLeay and Bailey 2010). Enriched motifs
were then clustered and grouped by motif similarity with motifStack using a p-value
cutoff of 0.01 (Ou et al. 2023). The lowest p-value within a merged motif group was
65

reported along with associated transcription factors listed in order of lowest p-value.
Only expressed transcription factors were included. Transcription factors were
considered expressed if the total average FPKM across bulk RNA-seq samples used in
this study was greater than or equal to 5. RNA-seq from negative sort populations were
excluded. Genes were also kept if they were expressed in any cluster from single-cell
RNA-seq samples.

Bulk RNA-seq
Total RNA was extracted with Quick-RNA Microprep kit (Zymo Research), quantified by
bioanalyzer, and then processed for libraries with QIAseq FX Single Cell RNA Library
Kit (Qiagen). At least 3 replicates were collected for each cell-type and stage and
sequenced to a depth of >=20 million reads.

Reads were mapped to the mouse reference genome (Gencode Mm10v11) using
STAR. Read counts were quantified by RSEM (B. Li and Dewey 2011). Only autosomal
protein coding genes with polyA tail transcripts annotated were kept. Transcript counts
were collapsed to gene counts. Differentially expressed genes were identified using the
DESeq2 package (Love, Huber, and Anders 2014). Genes with a log fold change
threshold greater than 0.5 and adjusted P-value of less than or equal to 0.05 were
considered significant. Principle component analysis and unsupervised hierarchical
clustering of RNA-seq was performed using counts transformed by DESeq2’s
regularized logarithm (Rlog). MA plots from DESeq2 analysis were plotted with ggpubr
(Kassambara and Kassambara 2020).

Chromatin Assays & Analysis
CUT&RUN libraries were constructed with the Accel-NGS 2S Plus DNA Library Kit
(Swift Biosciences, #21024) using lowest input option. 2S A&B MID indexing Kit was
used for index addition steps (Swift Biosciences, #27396). These indexes add unique
molecular identifier barcodes for PCR deduplication.

Encode ChIP—seq pipeline were adapted for analysis of CUT&RUN data (Dobin et al.
2013). Reads from the raw fastq files were aligned to GENCODE mm10v11 genome
assembly using STAR package (Dobin et al. 2013). PCR duplicates were removed
based on UMI using UMItools, and peaks were called by Model-based analysis of ChIP-
Seq (MACS2) with FDR< 0.01 and the dynamic lambda (--nolambda) option for
individual replicates (Smith, Heger, and Sudbery 2017; Zhang et al. 2008). For each
sample, irreproducible discovery rate (IDR) peaks, overlap peaks, and pooled peaks
were identified between the biological replicates (Q. Li et al. 2011). Peaks from MACS2
and IDR analysis were only used for assessment of quality. All samples in study were
rated reproducible, had expected 150-300bp fragment size, and were free of PCR bottle
necking, as assessed by Encode quality control metrics.

H3K4me1, H3K27me3, and H3K27ac regions were identified using the R package
ChromstaR (parameters: binsize = 500 bp, stepsize = 250 bp, mode = full). For
comparison of paired Sox2-CreER TdT+ apex and base cochlea, H3K27me3 and
H2AK119ub were randomly sample to 20 million reads for each replicate. For all other
datasets reads were randomly sampled for each replicate to 10 million. For H3K4me1
66

regions, posterior cutoffs used were adjusted to be more conservative than default
cutoffs defined by program, while producing expected percent of genome modified for
the respective epigenetic mark. For H3K4me1 regions, posterior probability cutoffs of
0.995, 0.999999, 0.995 were used for 13.5 progenitors, E17.5 hair cells, and E17.5
supporting cells, respectively. H3K27me3 and H3K27ac were called together, with
model set to make marks mutually exclusive. Regions with a mean read count less than
5 across comparison samples were removed for differential analysis.

Promoter regions were defined as all regions within 500bp upstream and 200bp downstream of
mRNA coding transcription start sites. All region outside of promoter regions were considered
distal. Regions from Encode’s mm10 blacklist were removed from all chromatin analysis.

To generate signal tracks for CUT&RUN data, scale factors were first computed for each
sample using edgeR’s calcNormFactors function with total reads counted regions modified in all
samples identified by chromstaR as input (Robinson, McCarthy, and Smyth 2010). Coverage
files were generated from aligned bam files using Deeptools’ bamCoverage function with
previously calculated scale factors (additional parameters: normalizeUsing=CPM, binSize=30).
Coverage tracks of replicates were averaged using Deeptools (Ramírez et al. 2016). Heatmaps
were generated with Deeptools using 6kb windows for H3K27ac and H3K4me1 and 50kb
window for H3K27me3 centered on region.

GO & GREAT Analysis
Gene ontology analysis was performed on categorized gene sets using R clusterProfiler
package. GO results were visualized using the R enrichplot package. For ontology
analysis of enhancer regions, GREAT version 4.0.4 was used with mm10 genome
annotation (McLean et al. 2010). For GREAT analysis, the default basal plus extension
option was chosen with distal set to maximum of 500kb.

Flow cytometry preparations
Cochleae were incubated in 0.25% trypsin for 8 minutes at 37 degrees Celsius, then
transferred to ice and gently titrated to single cell suspension. Media (DMEM with 10%
FBS) was added to the dissociated cells and then passed through a 70um cell strainer
and then FACS-purified).

Cell purification
FACS-purification of inner ear cell types was performed as described (White et al.
2006). Fluorescence-activated cell sorting (FACS) of cells or nuclei was performed on
either an Aria I or Aria II (BD Biosciences). For sorting inner ear cell types, mouse
cochleae were dissected, dissociated with 0.05% Trypsin and 1 mg/ml collagenase
(Worthington) into single cell suspension and purified by FACS using a 100 μm nozzle.
Only sorts with more than 96% cell purity were used for analysis of gene expression
and genomic assay. For CUT&RUN, cells sorted into PBS+5%FBS. Cells were then
immediately pelleted at 500g for 3min at 4°C. The published CUT&RUN protocol was
then performed from this step following the low-salt option for digestion.

For Paired-Tag, whole E14.5-E15.5 cochlea were enzymatically dissociated with a
cocktail of 400µl of 0.25% Trypsin, 50µl of 1 mg/ml Dispase, and 50µl of 1 mg/ml
collagenase for 10 minutes at 37°C and then quenched with PBS + 5% Fetal Bovine
67

Serum (FBS). The digested tissue was triturated 100 times using a wide-bore 200µl
pipette tip. Cells were pelleted at 400g for 5 minutes at 4°C and resuspended in 90 uL
of buffer (PBS with 0.5% BSA and 2mM EDTA). 10ul of CD326 (EPCAM) MicroBeads
(Miltenyi Biotec 130-105-958) were added to the cell suspension and incubated on ice
for 10 minutes. To wash, 2mL of buffer was added to cell suspension, cells were then
pelleted at 400g for 5min at 4°C and resuspended in 500uL of buffer. Epcam+ cells
were isolated on MACS MS Columns (130-042-201) following manufacturer’s protocol.
Cells were then pelleted and resuspended in CUT&RUN wash buffer with 10% DMSO,
transferred to freezing vials, placed in Mr. Frosty container, and banked in -80°C
storage. Paired-Tag experiments were performed within 3 weeks of collection and
freezing.

Contributions
Juan Llamas performed all dissections. All analysis was performed by Talon Trecek. All
H3K27me3, H3K27ac, H2AK119ub, single-cell RNA-seq, E15.5 Atoh1-fGFP RNA-seq
data sets were assayed by Talon Trecek. E17.5 and P1 Atoh1-fGFP RNA-seq data sets
were collected by Dr. Vincent Haoze, Litao Tao, and Talon Trecek. H3K4me1
CUT&RUN data sets were assayed by Dr. Litao Tao (Tao et al. 2021). ATAC-seq data
sets were collected by Dr. Yassan Abdolazimi and Dr. Vincent Haoze (H. V. Yu et al.
2021).

NGS Data sets

Data sets collected in this work by Talon Trecek:

Sample Name Description Stage Marker Antibody
E15_Atoh1_High_fGFP_RNAs
eq_rep1
RNAseq from high Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E15_Atoh1_High_fGFP_RNAs
eq_rep2
RNAseq from high Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E15_Atoh1_High_fGFP_RNAs
eq_rep3
RNAseq from high Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E15_Atoh1_low_fGFP_RNAse
q_rep1
RNAseq from low Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E15_Atoh1_low_fGFP_RNAse
q_rep2
RNAseq from low Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E15_Atoh1_low_fGFP_RNAse
q_rep3
RNAseq from low Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E15_Atoh1_Neg_fGFP_RNAs
eq_rep1
RNAseq from negative Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E15_Atoh1_Neg_fGFP_RNAs
eq_rep2
RNAseq from negative Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E15_Atoh1_Neg_fGFP_RNAs
eq_rep3
RNAseq from negative Atoh1 fGFP cells E15.5
Atoh1
fusion
GFP
Not applicable
E17_SC_RNAseq_rep1 RNAseq from supporting cells E17.5
Lfng
GFP
Not applicable
E17_SC_RNAseq_rep2 RNAseq from supporting cells E17.5
Lfng
GFP
Not applicable
E13_PG_CUTRUN_H3K27ac_
rep1
H3K27ac CUT&RUN prosensory progentiors E13.5
P27kip1
GFP
Active Motif,
catalog #39133
68

E13_PG_CUTRUN_H3K27ac_
rep2
H3K27ac CUT&RUN prosensory progentiors E13.5
P27kip1
GFP
Active Motif,
catalog #39133
E17_SC_CUTRUN_H3K27ac_
rep1
H3K27ac CUT&RUN supporting cells E17.5
Lfng
GFP
Active Motif,
catalog #39133
E17_SC_CUTRUN_H3K27ac_
rep2
H3K27ac CUT&RUN from supporting cells E17.5
Lfng
GFP
Active Motif,
catalog #39133
E17_HC_CUTRUN_H3K27ac_
rep1
H3K27ac CUT&RUN from cells E17.5
Atoh1
fusion
GFP
Active Motif,
catalog #39133
E17_HC_CUTRUN_H3K27ac_
rep2
H3K27ac CUT&RUN from cells E17.5
Atoh1
fusion
GFP
Active Motif,
catalog #39133
E13_PG_CUTRUN_H3K27me
3_rep1
H3K27me3 CUT&RUN from prosensory progentiors E13.5
P27kip1
GFP
Active Motif,
catalog #39155
E13_PG_CUTRUN_H3K27me
3_rep2
H3K27me3 CUT&RUN from prosensory progentiors E13.5
P27kip1
GFP
Active Motif,
catalog #39155
E17_SC_CUTRUN_H3K27me
3_rep1
H3K27me3 CUT&RUN from supporting cells E17.5
Lfng
GFP
Active Motif,
catalog #39155
E17_SC_CUTRUN_H3K27me
3_rep2
H3K27me3 CUT&RUN from supporting cells E17.5
Lfng
GFP
Active Motif,
catalog #39155
E17_HC_CUTRUN_H3K27me
3_rep1
H3K27me3 CUT&RUN from hair cells E17.5
Atoh1
fusion
GFP
Active Motif,
catalog #39155
E17_HC_CUTRUN_H3K27me
3_rep2
H3K27me3 CUT&RUN from hair cells E17.5
Atoh1
fusion
GFP
Active Motif,
catalog #39155
E14_Apex_CUTRUN_H3K27m
e3_rep1
H3K27me3 CUT&RUN from Sox2-CreER lineage traced apex
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Active Motif,
catalog #39155
E14_Apex_CUTRUN_H3K27m
e3_rep2
H3K27me3 CUT&RUN from Sox2-CreER lineage traced apex
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Active Motif,
catalog #39155
E14_Apex_CUTRUN_H3K27m
e3_rep3
H3K27me3 CUT&RUN from Sox2-CreER lineage traced apex
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Active Motif,
catalog #39155
E14_Base_CUTRUN_H3K27m
e3_rep1
H3K27me3 CUT&RUN from Sox2-CreER lineage traced base
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Active Motif,
catalog #39155
E14_Base_CUTRUN_H3K27m
e3_rep2
H3K27me3 CUT&RUN from Sox2-CreER lineage traced base
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Active Motif,
catalog #39155
E14_Base_CUTRUN_H3K27m
e3_rep3
H3K27me3 CUT&RUN from Sox2-CreER lineage traced base
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Active Motif,
catalog #39155
E14_Apex_CUTRUN_H2AK11
9ub_rep1
H2AK119ub CUT&RUN from Sox2-CreER lineage traced apex
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Cell Signaling
Technology,
catalog #8240S
E14_Apex_CUTRUN_H2AK11
9ub_rep2
H2AK119ub CUT&RUN from Sox2-CreER lineage traced apex
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Cell Signaling
Technology,
catalog #8240S
E14_Apex_CUTRUN_H2AK11
9ub_rep3
H2AK119ub CUT&RUN from Sox2-CreER lineage traced apex
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Cell Signaling
Technology,
catalog #8240S
E14_Base_CUTRUN_H2AK11
9ub_rep1
H2AK119ub CUT&RUN from Sox2-CreER lineage traced base
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Cell Signaling
Technology,
catalog #8240S
E14_Base_CUTRUN_H2AK11
9ub_rep2
H2AK119ub CUT&RUN from Sox2-CreER lineage traced base
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Cell Signaling
Technology,
catalog #8240S
E14_Base_CUTRUN_H2AK11
9ub_rep3
H2AK119ub CUT&RUN from Sox2-CreER lineage traced base
cochlear epithelium cells
E14.5
Sox2-
CreER;
TdT-fl
Cell Signaling
Technology,
catalog #8240S
E13_PG_CUTRUN_H2K119ub
_rep1
H2AK119ub CUT&RUN prosensory progenitors E13.5
P27kip1
GFP
Cell Signaling
Technology,
catalog #8240S
E13_PG_CUTRUN_H2K119ub
_rep2
H2AK119ub CUT&RUN from prosensory progenitors E13.5
P27kip1
GFP
Cell Signaling
Technology,
catalog #8240S
E17_SC_CUTRUN_H2K119ub
_rep1
H2AK119ub CUT&RUN from supporting cells E17.5
Lfng
GFP
Cell Signaling
Technology,
catalog #8240S
69

E17_SC_CUTRUN_H2K119ub
_rep2
H2AK119ub CUT&RUN from supporting cells E17.5
Lfng
GFP
Cell Signaling
Technology,
catalog #8240S
E13_PG_CUTRUN_H3K4me1
_rep1
H3K4me1 CUT&RUN from prosensory progenitors E13.5
P27kip1
GFP
Active Motif,
catalog #61633
E13_PG_CUTRUN_H3K4me1
_rep2
H3K4me1 CUT&RUN from prosensory progenitors E13.5
P27kip1
GFP
Active Motif,
catalog #61633
E14_Sox2CreER_TdTfl_scRN
Aseq
single-cell RNAseq on TdT positive cochlear epithelium, E14.5 E14.5
Sox2Cre
ER; TdT-
fl
Not applicable
E14_E15_Epcam_PairedTag_
cDNA
Paired-Tag on Epcam positive cochlear epithelium
E14.5-
E15.5
Wildtype
Miltenyi Biotec,
catalog #130-105-
958
E14_E15_Epcam_PairedTag_
H3K27me3
Paired-Tag on Epcam positive cochlear epithelium
E14.5-
E15.5
Wildtype
Miltenyi Biotec,
catalog #130-105-
958; Active Motif,
catalog #39155

CUT&RUN sequencing data sets:
E13.5 Cdkn1b-GFP Progenitor cells H3K27ac x2
E17.5 Lfng-GFP Supporting cells H3K27ac x2
E17.5 Atoh1-fGFP Hair cells H3K27ac x2
E13.5 Cdkn1b-GFP Progenitor cells H3K27me3 x2
E17.5 Lfng-GFP Supporting cells H3K27me3 x2
E17.5 Atoh1-fGFP Hair cells H3K27me3 x2
E14.5 Sox2CreER traced at E13.5; Apex H3K27me3 x3
E14.5 Sox2CreER traced at E13.5; Base H3K27me3 x3
E14.5 Sox2CreER traced at E13.5; Apex H2AK119ub x3
E14.5 Sox2CreER traced at E13.5; Base H2AK119ub x3

RNA sequencing data sets:
E14.5 Sox2CreER traced at E13.5 single-cell RNA-seq with 10x 3’ RNA
E14.5-15.5 wild-type Epcam+ cochlear epithelium Paired-Tag RNA+ H3K27me3
E17.5 Atoh1-fGFP Hair cells RNA-seq x3
E15.5 Atoh1-fGFP Hair cells RNA-seq x3
E15.5 Atoh1-fGFP low GFP RNA-seq x3
E15.5 Atoh1-fGFP negative GFP RNA-seq x3
P1 Atoh1-fGFP Hair cells RNA-seq x3

Previously published:
E13.5 Cdkn1b-GFP Progenitor cells H3K4me1 x2 (Tao et al. 2021).
E17.5 Lfng-GFP Supporting cells H3K4me1 x2 (Tao et al. 2021).
E17.5 Atoh1-fGFP Hair cells H3K4me1 x2 (Tao et al. 2021).
E13.5 Cdkn1b-GFP Progenitor cells ATAC x2 (H. V. Yu et al. 2021).
E17.5 Lfng-GFP Supporting cells ATAC x2 (H. V. Yu et al. 2021).
E17.5 Atoh1-fGFP Hair cells ATAC x2 (H. V. Yu et al. 2021).

70

Software Source Link
GenomicRanges (Lawrence et al. 2013) https://www.bioconductor.org/packages/release/bioc/html/GenomicRanges.html
DEseq2 (Love, Huber, and Anders 2014) https://www.bioconductor.org/packages/release/bioc/html/DESeq2.html
biomaRt (Durinck et al. 2005) https://www.bioconductor.org/packages/release/bioc/html/biomaRt.html
STAR (v2.5.4b) (Dobin et al. 2013) https//github.com/alexdobin/STAR
cutadapt (Martin 2011) https://cutadapt.readthedocs.io/
FastQC (Babraham Bioinformatics ) https://www.bioinformatics.babraham.ac.uk/projects/fastqc/
ggpubr (Kassambara 2020) https://github.com/kassambara/ggpubr
clusterProfiler (G. Yu et al. 2012) https://bioconductor.org/packages/release/bioc/html/clusterProfiler.html
org.Mm.eg.db (Carlson 2019) http://bioconductor.org/packages/release/data/annotation/html/org.Mm.eg.db.html
ReactomePA (G. Yu and He 2016) https://bioconductor.org/packages/release/bioc/html/ReactomePA.html
ChromstaR (Taudt et al. 2016) https://bioconductor.org/packages/release/bioc/html/chromstaR.html
deeptools 3.2 (Ramírez et al. 2016) https://deeptools.readthedocs.io/
GenomicFeatures (Lawrence et al. 2013) https://bioconductor.org/packages/release/bioc/html/GenomicFeatures.html
AnnotationDbi (Pages et al. 2017) https://bioconductor.org/packages/release/bioc/html/AnnotationDbi.html
ChIPpeakAnno (L. J. Zhu et al. 2010) https://bioconductor.org/packages/release/bioc/html/ChIPpeakAnno.html
SAMtools (H. Li et al. 2009) http://www.htslib.org/
Picard (“Picard Toolkit” 2019) https://github.com/broadinstitute/picard
RSEM (B. Li and Dewey 2011) https://github.com/deweylab/RSEM
motifStack (Ou et al. 2023) https://bioconductor.org/packages/release/bioc/html/motifStack.html
MEME suite - AME (McLeay and Bailey 2010) https://meme-suite.org/meme/doc/ame.html
rtracklayer (Lawrence, Gentleman, and Carey
2009)
https://bioconductor.org/packages/release/bioc/html/rtracklayer.html
GenomicAlignments (Maintainer, GenomicRanges, and
Rsamtools 2015)
https://bioconductor.org/packages/release/bioc/html/GenomicAlignments.html
SCANPY (Wolf, Angerer, and Theis 2018) https://scanpy.readthedocs.io/en/stable/index.html
scVelo (Bergen et al. 2020) https://scvelo.readthedocs.io/en/stable/
UMI-tools (Smith, Heger, and Sudbery 2017) https://github.com/CGATOxford/UMI-tools
IDR (Q. Li et al. 2011) https://github.com/nboley/idr
MACS2 (Zhang et al. 2008) https://github.com/macs3-project/MACS

Abstract (if available)

Abstract Embryonic and perinatal supporting cells of the inner ear harbor a latent capacity for regeneration of sensory hair cells, but this capacity is lost after birth. To understand the epigenetic basis for how and when this plasticity is established, we mapped the epigenetic and transcriptional changes occurring as cochlear progenitors differentiate into hair cells and supporting cells. We purified both differentiating cell types and found that initially, each cell type epigenetically de-represses regulators of both hair cells and supporting cells by removing inhibitory H3K27me3 histone marks. This de-repression occurs with co-expression of both hair cell and supporting cell fate factors in both cell types. Both cell types acquire a common primed (H3K4me1) enhancer landscape that is associated with hair cell fate and the transdifferentiation potential of supporting cells. While supporting cells prime the hair cell enhancer landscape, they fail to epigenetically activate (H3K27ac) these enhancers. Additionally, while early supporting cells remove H3K27me3 from the Atoh1 locus, they maintain the repressive Prc1 mark H2K119ub1. In contrast to hair cells, supporting cell maintain progenitor enhancers in either an active or primed state. Understanding how early supporting cells prime the hair cell epigenetic program without activation and its relation to transdifferentiation stands to provide a novel regenerative approach to restoring sensory hair 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

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

PDF

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

PDF

Transcriptional regulation in nephron progenitor cells

PDF

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

PDF

Neural fate regulated by extrinsic signaling and epigenetics

PDF

Molecular programs in epithelial morphogenesis in mammalian organ systems

PDF

Dissecting the heterogeneity of mouse hematopoietic stem cells in vivo

PDF

Using epigenetic toggle switches to repress tumor-promoting gene expression

PDF

Understanding human nephrogenesis and scaling synthesis of organoids facilitate modeling of kidney development and disease

PDF

Skeletal cell fate plasticity in zebrafish bone development and regeneration

PDF

Epigenetic plasticity of cultured female human embryonic stem cells and regulation of gene expression and chromatin by PR-SET7 mediated H4K20me1

PDF

Nr5a2 balances lineage decisions to promote diverse connective tissue fates in the jaw

PDF

Characterization of metabolically distinct muscle resident fibro-adipogenic subpopulations reveals a potentially exploitable mechanism of skeletal muscle aging and disease

PDF

β-catenin couples self-renewal, induction and epithelial morphogenesis and patterning at the outset of mammalian nephrogenesis

PDF

Investigating molecular roadblocks to enhance direct cellular reprogramming

PDF

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

PDF

Molecular role of EZH2 overexpression in Colorectal Cancer progression

PDF

Fibroblastic connective tissue cells: the blastema stem cells and source of large-scale chondrogenesis in the regenerating lizard tail

PDF

Generation and characterization of an in vitro model of inner ear sensory hair cells using transcription factor mediated cellular reprogramming

Asset Metadata

Creator Trecek, Talon (author)

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

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Development, Stem Cells and Regenerative Medicine

Degree Conferral Date 2023-05

Publication Date 11/11/2023

Defense Date 05/11/2023

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

Tag Development,epigenetics,gene regulation,genomics,inner ear,OAI-PMH Harvest,plasticity,Regeneration,single cell,stem cells,Transcription,transcription factors

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Crump, Gage (committee chair), Farnham, Peggy (committee member), Groves, Andrew (committee member), McMahon, Andrew (committee member)

Creator Email talontrecek@gmail.com,trecek@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113121768

Unique identifier UC113121768

Identifier etd-TrecekTalo-11830.pdf (filename)

Legacy Identifier etd-TrecekTalo-11830

Document Type Dissertation

Format theses (aat)

Rights Trecek, Talon

Internet Media Type application/pdf

Type texts

Source 20230512-usctheses-batch-1043 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

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

Repository Email cisadmin@lib.usc.edu