University of Southern California Dissertations and Theses

Epigenetic and genetic reprogramming during embryonic chicken feather bud morphogenesis, hair morphogenesis, and de novo hair regeneration

(USC Thesis Other)

Epigenetic and genetic reprogramming during embryonic chicken feather bud morphogenesis, hair morphogenesis, and de novo hair regeneration

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
EPIGENETIC AND GENETIC REPROGRAMMING DURING EMBRYONIC
CHICKEN FEATHER BUD MORPHOGENESIS, HAIR MORPHOGENESIS, AND
DE NOVO HAIR REGENERATION

by
Michael W. Hughes

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
(PATHOBIOLOGY)

May 2012

Copyright 2012 Michael W. Hughes

ii

Dedication
I dedicate this body of research work to my family; my great-grandfathers
and grandfathers because they built a foundation for me to stand on, my mother,
father, brothers and sisters for their endearing support, and my children and
grandchildren for their inspiration! I love you all.

iii

Acknowledgments
I thank the faculty and staff of the University of Southern California for the
privilege and opportunity to work together in the field of science. Their patience,
support, and guidance enabled our hard work to become a successful endeavor.
For this, I am most grateful.
The administrators in the Department of Pathology have provided
tremendous support. Mrs. Lisa Doumak with her knowledge and skills
successfully navigated the endless lengths of red tape to ensure all requirements
were met. Thank you Lisa!
The Chuong laboratory members provided countless hours of professional
interaction. This guided and inspired this piece of research work. Dr. John Jiang
and Dr. Randy Widelitz were essential to this research because their expertise
made the more difficult assays possible.
I want to thank my committee members for giving one of the most
precious commodities and that is their time. I understand their respective careers
are time constrained with research necessities. However, by giving their time
they ensure the next generation of scientists is prepared to meet the future
challenges.
I want to thank Dr. Cheng-Ming Chuong. Our countless discussions
regarding the complexities of nature gave me the inspiration for this work. His

iv

support was essential because this work would be vastly different without it. I
hope I can uphold the standards that have been set by his laboratory.

v

Table of Contents
Dedication ............................................................................................................ ii
Acknowledgments .............................................................................................. iii
List of Tables ..................................................................................................... vii
List of Figures ................................................................................................... viii
Abstract ............................................................................................................... x
Chapter One: Introduction
1.1 Skin Organ Biology ......................................................................................... 1
1.2 Skin Development .......................................................................................... 2
1.3 Wound Healing ............................................................................................... 4
1.4 Hair Follicles ................................................................................................... 5
1.5 Epigenetic Mechanism.................................................................................... 7
1.6 Reprogramming and regulation of hair regeneration .................................... 11
1.7 Msx2 homeobox genes and regeneration..................................................... 12
1.8 HDAC1 epigenetic enzymes and progenitor cell differentiation .................... 13
1.9 Purpose of study and preliminary data ......................................................... 14

Chapter Two: Epigenetic modulation of embryonic chicken skin feather
bud morphogenesis
2.1 Introduction ................................................................................................... 17
2.2 Materials and Methods ................................................................................. 18
2.3 Results.......................................................................................................... 21
2.4 Discussion .................................................................................................... 46

Chapter Three: Disrupted ectodermal organ morphogenesis in mice
with conditional deletion of histone deacetylase in the epidermis:
dermal cysts, reduced hair variations, supernumerary nails, and hyper-
pigmentation
3.1 Introduction ................................................................................................... 51
3.2 Materials and Methods ................................................................................. 54
3.3 Results.......................................................................................................... 56
3.4 Discussion .................................................................................................... 80

vi

Chapter Four: Repair versus regeneration: Msx2 homeobox gene is
required to generate epidermal stem cells for de novo hair regeneration
after wounding
4.1 Introduction ................................................................................................... 94
4.2 Materials and Methods ................................................................................. 97
4.3 Results........................................................................................................ 101
4.4 Discussion .................................................................................................. 120

Chapter Five: Conclusions and Perspectives
5.1 Conclusion .................................................................................................. 125
5.2 Perspective ................................................................................................. 125
5.3 Future Directions ........................................................................................ 126
Bibliography ...................................................................................................... 128

vii

List of Tables
Table 2-1 6Way ANOVA with interactions exhibits differential gene expression 40
Table 2-2 CSO Microarray data table ................................................................. 44

viii

List of Figures
Figure 1-1 Immunofluorescent staining of histone methylation in ectodermal
organs .................................................................................................................. 9

Figure 1-2 HDAC1 expression pattern in human skin ......................................... 15

Figure 2-1 Expression patterns of epigenetic enzymes in chicken skin .............. 23

Figure 2-2 Inhibiting HMTs or HDACs causes abnormal chick skin organ
development ....................................................................................................... 27

Figure 2-3 Inhibiting HMTs or HDACs causes abnormal feather bud
morphogenesis ................................................................................................... 31

Figure 2-4 Inhibiting HMTs or HDACs modulates BMP and FGF activities on
chick skin organ morphogenesis ........................................................................ 35
Figure 2-5 2D Hierarchical Cluster of 55 Chicken Tissue Samples .................... 38

Figure 2-6 Microarray analysis of CSO development ......................................... 42

Figure 2-7 Model of histone acetylation and methylation during CSO
morphogenesis ................................................................................................... 49

Figure 3-1 HDAC1 Immunostaining Expression Pattern in Wild Type Mouse
Hair ..................................................................................................................... 58

Figure 3-2 Genotyping Assay of K14creHDAC
fl/fl
Mice ....................................... 61

Figure 3-3 Gross morphological phenotype of the K14creHDAC
fl/fl
Mice ............ 66

Figure 3-4 Histology of skin homestasis and tail follicle morphology in 5-6
month old K14creHDAC1
fl/fl
mice ........................................................................ 69

Figure 3-5 Molecular markers of skin differentiation in 2-3 month and 3week
old K14creHDAC1fl/fl mice ................................................................................. 77

Figure 4-1 Altering Msx2 Levels Inhibits wound closure and de novo Hair
regeneration ..................................................................................................... 103

ix

Figure 4-2 Msx2 promoter reporter mice suggested Msx2 is expressed at the
wound margin in early stage regeneration and punctate in late stage
regeneration ..................................................................................................... 107

Figure 4-3 Epithelial specific reduction of Msx2 activity results in poor tissue
regeneration and suppression of de novo hair regeneration ............................ 111

Figure 4-4 The Mad/Smad factor and Homeodomain protein binding sites
regulate Msx2 expression during de novo hair regeneration ............................ 115

Figure 4-5 Inhibiting BMP in the basal layer does not affect de novo hair
regeneration ..................................................................................................... 118

x

Abstract

The ability to engineer a tissue to replace diseased organs is a goal that
when achieved will provide an incredible asset to clinicians. In order to engineer
a tissue, one must understand how the specific tissue develops, maintains
homeostasis, and regenerates. To gain insight into this process, we utilized the
embryonic chicken skin organ (CSO) model to study the effect of histone
acetylation and methylation on feather bud development. In the CSO assay,
Chaetocin, a histone methyltransferase inhibitor, acts upstream of FGF and BMP
to promote feather bud growth. MS275, a histone deacetylase inhibitor, acts
upstream of FGF and BMP to inhibit feather bud growth. Next, we employed a
tissue specific transgenic mouse model to investigate the effect of HDAC1 on
mouse skin morphogenesis. Immunostaining of wild type C57BL/6 skin showed
HDAC1 is predominantly located in the epidermal basal layer, sebaceous glands,
inner and outer root sheaths, and matrix of pelage and vibrissae hair follicles.
K14creHDAC1
fl/fl
mice exhibited a reduction in size and weight, hyperkeratosis,
alopecia, dermal cysts, hyperpigmentation, sebaceous hyperplasia and nail
pathologies including; supernumeray nails, elongated nails, and pigmented nail
plates. Finally, we studied hair follicle regeneration to examine how Msx2
modulates de novo hair growth. Full thickness, large wounds (> 1.5cm in
diameter) were created on the dorsum of wild type (C57bl/6), Msx2LacZ (1.7KB,
560bp and 52bp), Msx2
-/-
, and K14creMsx2
fl/fl
mice. Msx2 expression occurred in

xi

two phases: early in the wound margin and later, as a punctate pattern in the
center wound bed. Genetic ablation of Msx2 inhibits de novo hair regeneration in
both totally null (Msx2
-/-
) or tissue specific (K14creMsx2
fl/fl
) null mice. An Msx2
promoter with a mutation of the Mad/Smad factor binding site failed to induce
Msx2 expression in the wound. However, K14noggin mice showed only a slight
increase in de novo hair regeneration, suggesting BMP alone is insufficient and
more complex regulation is used to control Msx2. Taken together, these data
suggest a hierarchy of gene expression mechanisms that govern tissue specific
structure and function. Epigenetic and genetic mechanisms modulate gene
expression patterns resulting in spacio-temporal control of tissue competence
and morphogenesis.

1

Chapter One: Introduction
1.1 Skin Organ Biology
Regenerative medicine is at the forefront of 21st century medicine. In the
skin field, scientists strive to develop new methods and identify genes that can
enhance the regenerative ability of the skin, following injury or the aging process.
The goal of tissue engineering is to reconstruct a specific organ ex vivo in order
to replace a diseased organ in vivo. In order to accomplish this highly complex
process of tissue reconstruction, the scientist must first understand the molecular
mechanisms of tissue specific development and regeneration.
Skin biology has been studied for centuries (Pott, 1775). The largest organ
of most vertebrate species is the skin (Alibardi, 2003). The skin organ provides
protection from the environment, prevents dehydration, and is essential for life.
These properties, along with others, make skin biology one of the most important
systems to understand. Recently, the skin organ has become a model for
regenerative medicine research (M. Ito, Yang, Andl, Cui, Kim, Millar, &
Cotsarelis, 2007a).
The chicken embryo is the classical model for studying tissue
development. The ability to manipulate tissues in vivo with ease, the relatively
short developmental time period, and availability make the chicken embryo very
appealing to researchers. Feather bud morphogenesis is highly complex.
Interactions between epidermal, muscular, and neural tissues are required for

2

correct feather development. These developmental events must occur in a highly
specific spacio-temporal manner.
The hair follicle has become a strong model for studying tissue
regeneration (M. Ito, Yang, Andl, Cui, Kim, Millar, & Cotsarelis, 2007a)(M. Ito,
Yang, Andl, Cui, Kim, Millar, & Cotsarelis, 2007a; M. Ito, Yang, Andl, Cui, Kim,
Millar, & Cotsarelis, 2007a; M. Ito, Yang, Andl, Cui, Kim, Millar, & Cotsarelis,
2007b). It can regenerate hundreds of times with no loss of tissue integrity or
function. Hair follicle regeneration is one example of scarless wound healing.
Hair follicles develop through reciprocal signaling pathways between embryonic
epithelium and mesenchyme (Dhouailly, 1975). Just like in feather bud
development, hair follicle development depends on highly specific spacio-
temporal gene expression patterns.
By studying embryonic feather bud morphogenesis and hair follicle
morphogenesis we hope to elucidate core biological principles that can aid in the
engineering of a skin organ.

1.2 Skin Development
The skin organ is a keratinized barrier that covers most of the body and is
derived from non-neural ectodermal tissue. There are many different types and
regions of skin. Examples are dermal skin, oral epithelia, palmer or plantar skin,
and cranial skin. The skin is composed of an epithelial component and a

3

mesenchymal component. The epithelial tissue comprises the epidermis that is
composed of keratinocytes. Basal keratinocytes sit on the basement membrane
and proliferate to maintain this population. To replenish the skin organ during
normal homeostasis, the basal keratinocytes detach from the basement
membrane and differentiate to the surface of the skin. The mesenchymal tissue
comprises the dermis and is composed of fibroblasts. The ectodermal
appendages of hair, teeth, glands…etc., derive from these different ecotodermal
organs. These appendages are subject to degradation during normal wear and
tear, and are susceptible to injury that can be repaired or regenerated depending
on the extent of injury. The skin appendages are the products of epithelial-
mesenchymal interactions. Ectodermal dysplasia syndromes occur when these
interaction processes go awry.
There are multiple molecular and cellular mechanisms that occur during
development and regeneration of skin appendages. Different ectodermal organs
may have different homeostasis of progenitor or differentiation states, which may
change dynamically depending on physiological (continuous versus episodic
regeneration, change of hormone status, etc.) and pathological (e.g., wounding)
conditions. The need for activation of progenitor cells is different depending on
how often cells are replenished in different organ systems. In the stem cell
scenario, this is regulated through cellular homeostasis utilizing feedback loops.
However, a cell’s status can be dramatically altered by local or systemic

4

conditions. Although cell entities and molecular markers are important, so too is
the functional states of cell groups and the force that drive this process.

1.3 Wound Healing
Wound healing is an important physiological process that repairs injured
tissues to restore functionality. Wound healing is dependent on multiple factors
such as the site of occurrence, type of wound, age of the host, and ethnicity
(Muneoka, Allan, Yang, Lee, & Han, 2008; Muneoka, Han, & Gardiner, 2008).
This process is dynamic and not perfect. Problems of scarring or malformation
often occur. Scarring is the result of too much collagen matrix deposition in the
wound site. Malformation is the result of the regeneration process gone awry.
The amount of scarring and malformation determines the final tissue
functionality. No scarring or malformation leads to complete restoration of
function and thus total regeneration. Too much scarring or malformation can
cause complete loss of function. Tissue regeneration is the complete repair and
restoration of tissue structure and function in total. Amphibians possess the
ability to regenerate their amputated limbs completely (Yokoyama, 2008). This
ability does not seem to be present in adult mammals. However, a human fetus
up to the second trimester is thought to possess the ability to regenerate an
amputated finger based on data from animal models (Allan et al., 2006;
Illingworth, 1974). An adolescent child can regenerate an amputated digit tip

5

including the bone (Illingworth, 1974). During puberty, this ability is lost. It is
evident that we lose these abilities as we age. We now know that new hairs can
regenerate in a wound bed if the right environment is provided in a process
termed de novo hair regeneration (M. Ito, Yang, Andl, Cui, Kim, Millar, &
Cotsarelis, 2007a). If we can learn the lessons of de novo hair regeneration then
we might be able to apply it to total skin regeneration, and eventually to digit or
limb regeneration.
Following tissue injury, the wound healing process begins. A wide
spectrum of repair possibilities can occur. On one end of the extreme, the wound
reepitheliazes, and connective tissue is deposited to form scar. This would be an
example of the ‘repair ‘mechanism. On the opposite end of the spectrum, the
wound can regenerate from stem cells or a blastema to form functional tissue.
This would be the ‘regeneration’ mechanism. We need to understand the
different cellular and molecular mechanisms that modulate normal skin
morphogenesis, wound healing scar formation, or tissue regeneration.

1.4 Hair Follicles
The pilo-sebaceous unit is comprised of a hair follicle, the sebaceous
gland, and an erector pilli muscle. The hair follicle has many regions or domains.
The lower region of the hair follicle contains the dermal papilla. The dermal
papilla is the signaling center responsible for regenerating a new hair shaft during

6

the hair cycle. The matrix area of the hair follicle is a layer of keratinocytes
surrounding the dermal papilla in the lower region of the hair follicle. The matrix
cells generate the hair fiber and inner root sheath. These two regions, the dermal
papilla and the matrix area, undergo reciprocal molecular signaling during the
hair cycle. Hair follicles play a significant role in the wound healing process. Cells
from the hair bulge, collar region, and the outer root sheath contribute to re-
epithelialize the skin. This re-epitheliazation process is guided by a coordinated
signaling event occurring between the dermis and epidermis. Many genes have
been identified that regulate the epithelial-mesenchymal interaction pathways
that control skin appendage formation. One such gene is the Muscle Segment
Homeobox gene. Msx2 is a member of the muscle segment homeobox gene
family and encodes a transcription factor. Msx2 has been shown to be involved in
the development of the skin, the digits, and craniofacial development (Liu et al.,
1994; Ma, Golden, Wu, & Maxson, 1996; Stelnicki et al., 1997). The encoded
protein also promotes cell growth under certain conditions and may be an
important target for the RAS signaling pathways. Mutations in this gene are
associated with parietal foramina 1 and Boston Type Craniosynostosis (Jabs et
al., 1993; Ma et al., 1996). Does Msx2 play a role in normal re-epithelization of
the skin or de novo hair regeneration?

7

1.5 Epigenetic Mechanism
Every somatic cell in an individual has identical genetic composition or
DNA sequence. These somatic cells comprise the different tissues of the body.
Each tissue has different abilities to repair or regenerate. The differences in
these abilities have been attributed to the presence or absence of tissue
‘competence’. Competence is the ability of a cell or tissue to respond to a specific
molecular signal. If a tissue cannot respond to a molecular stimulus then the
tissue has become ‘determined’. The molecular basis of this competence
remains elusive and could be one of the keys to regenerative medicine. A
potential control mechanism of tissue regeneration is epigenetic modulation.
Recent progress in epigenetics suggests that chromatin modifications may play a
role in stem cell progression (Fischle, Wang, & Allis, 2003; Kubicek et al., 2006).
Perturbation of chromatin modifiying enzymes leads to the arrest of early mouse
embryo development (Li, Zheng, & Dean, 2010), or alters neurosphere
differentiation (Schmittwolf et al., 2005). The role of epigenetic modifications in
skin development, repair and regeneration needs to be explored. The epigenetic
concept helps explain why there is only one progenitor cell group in the mouse
tail epidermis without the slow cycling and transient amplifying (TA) cells seen in
the human epidermis (Jones, Simons, & Watt, 2007). It also helps to explain why
different types of hair follicles may or may not have hair germs as well as bulge
regions (Greco et al., 2009).

8

We hypothesize that epigenetic changes in chromatin play a central role in
maintaining and or restricting the multi-potentiality of epidermal precursor cells,
and play a role in the ability of tissue to induce either appendage repair or
regeneration. The corollary is, if we can reverse these modifications, then we
may achieve some level of de-differentiation and be able to modulate the course
of wound healing. The ultimate goal is to guide the wound healing process away
from scar formation and towards tissue regeneration, through the reprogramming
of cells by epigenetic modulation. In preliminary experiments, we wondered if the
epigenetic pattern could be visualized, if the epigenetic pattern was tissue or cell
specific, if this pattern changed during development or maturation, and if this
pattern was evolutionarily conserved. We utilized immunofluorescense to
visualize specific nuclear patterns in the ectodermal organs of various species
(Figure 1-1). We observe nuclear specific staining of epigenetic marks. This
staining is present in all species studied. We observe different nuclear patterns in
different tissues. Interestingly, the histone modification marks exhibit specific
nuclear patterns. There is punctate, nebulous, and low level staining patterns.
These patterns were highly dynamic and we surmise they depend on the cell
state.

9

Figure 1-1 Immunofluorescent staining of histone methylation in
ectodermal organs Anole scale epithelium, chicken skin organ, and mouse tail
hair follicles were immunostained with an antibody directed against H3K9mCH3
or H3K9triCH3. The H3K9mCH3 stained ectodermal organs exhibited
predominantly 3 nuclear staining patterns; punctate, diffuse, and little to no
staining. The hair follicle exhibited different nuclear patterns in each of the hair
follicle compartments. The sebaceous gland demonstrated a very dynamic
pattern. (green = antibody staining, blue = DAPI)

10

Figure 1-1 Continued:

11

Secreted morphogen BMP suppresses hair regeneration, while the BMP
inhibitor noggin enhances hair regeneration. Homeobox gene Msx2 null mice
exhibit a suppression of hair regeneration but accelerated epidermal
differentiation. The epigenetic enzyme HDAC1 is involved in skin morphogenesis
(Longworth, Wilson, & Laimins, 2005). During physiological hair cycling and
wound healing, Msx2 and HDAC work may together to keep epidermal cells in a
multi-potential progenitor status; while BMP favors epidermal differentiation and
wound closure. A shift in the balance between MSX and BMP might have a
consequence on repair versus regeneration. This hypothesis was tested by
studying their expression patterns, and by using targeted gene disruption in
physiological regenerative hair waves and in regenerative wounds.

1.6 Reprogramming and regulation of hair regeneration
Reprogramming of skin cells by iPS to ESC is feasible (Takahashi &
Yamanaka, 2006), but in this case we only need the cells to go back to the multi-
potential status of ectodermal organs. How can this be done? One remarkable
phenomenon during de novo hair formation is the growth of new hairs is only
seen in the center of the wound. This occurs when a larger than 1cm full
thickness wound is opened in the adult mouse skin (M. Ito, Yang, Andl, Cui, Kim,
Millar, & Cotsarelis, 2007a). This is remarkable because it represents a

12

"physiological reprogramming" and no exogenous factors are required! We
stated that BMP, whether present in the micro-environment in the stem cell niche
or in the dermal macro-environment, can suppress hair regeneration. There is a
link between BMP activity and epidermal differentiation. We also demonstrated
that when more hairs were plucked,the hairs regenerate faster (M. V. Plikus,
Widelitz, Maxson, & Chuong, 2009). Therefore reduction of BMP activity shifts
the equilibrium toward the progenitor cell status and favors hair regeneration (M.
Plikus et al., 2004).

1.7 Msx2 homeobox genes and regeneration
Msx2 null mice exhibited a delay in anagen re-entry and required a longer
time for hair regeneration after hair plucking (Ma, Liu, Wu, Plikus, Jiang, Bi, Liu,
Muller-Rover, Peters, Sundberg, Maxson, Maas, & Chuong, 2003a). When full
thickness wounds were made, surprisingly we found that Msx2 is transiently
induced in the wound margin. If the wound is small (< 3mm wound), Msx2
disappears in 7 days. The closure of these wounds in Msx2 null mice is faster
than the control (Yeh et al., 2009). Therefore we formulate the hypothesis that
the sustained presence of Msx genes help maintain the progenitor status and
modulates full skin regeneration (including appendage formation), while depletion
of Msx 2 genes may favor differentiation and faster repair (forming epidermis
only). This is also inspired by the work in limb regeneration, in which the

13

presence of Msx is required for newborn mouse digit regeneration (Han, Yang,
Farrington, & Muneoka, 2003) and salamander limb regeneration (Yokoyama,
2008).

1.8 HDAC1 epigenetic enzymes and progenitor cell differentiation
The role of HDAC1 in skin morphogenesis is not well studied. HDAC1
deacetylates histones, promoting the formation of heterochromatin, and this
tends to suppress gene transcription (Fischle et al., 2003). In preliminary
experiments we observe HDAC1 expression patterns in human skin to be
predominantly in the basal layer of interfollicular and sweat gland duct epithelia
(Figure 1-2).
In thinking about how Msx2 may work to regulate cell differentiation,
Yoshizawa et al demonstrated that Msx2 can recruit HDAC1 and together
prevent ossification of fibroblasts (Yoshizawa et al., 2004). These findings lead
us to hypothesize HDAC1 and Msx2 may work together to keep cells in
progenitor states in physiological hair regeneration (hair waves), and in wound
healing. In many cases, Msx2 is transiently expressed, and reparative wound
healing takes place (Yeh et al., 2009). If the wound environment is able to
sustain the expression of HDAC1 and Msx2 longer, then regenerative wound
healing with the formation of hairs may be enhanced.

14

1.9 Purpose of the study
Why study tissue regeneration?
In recent years, major progress in stem cell biology has highlighted the
promise of regenerative medicine. Molecules involved in regulating stem cell
multi-potentiality and differentiation are being elucidated daily. To identify and
mimic the micro-environment signals is the first step in directly guiding human
embryonic stem cells to the non-neural ectodermal lineage, and then to the
dermis, and then to different ectodermal appendages. The role of these
pathways in the skin organ may lead to new treatments of various forms of
diseases such as ectodermal dysplacias, alopecias, scarring, and wound healing
scenarios.

15

Figure 1-2 HDAC1 expression pattern in human skin HDAC1 expression
patterns in full thickness human skin was visualized using an antibody directed
against the C-terminus of the HDAC1 protein. (A) Human skin demonstrates
HDAC1 staining overlaps with the basal layer marker cytokeratin 14 (K14)
staining. (dark red = antibody staining) (B) The dermal side of the human skin
sample demonstrates HDAC1 staining in the basal layer of interfollcular and
sweat gland duct epithelia. (green = antibody staining)

16

Figure 1-2 Continued:

17

Chapter 2: Epigenetic modulation of embryonic chicken skin feather bud
morphogenesis
2.1 Introduction
Limb and tissue regeneration has been the dream of medical scientists for
centuries. Countless investigators have observed the regenerative ability of
amphibians and were brave enough to ask the question, “What if mammals
possessed this ability?” The axolotl demonstrates the ability to regenerate limbs
and this process has been studied extensively (Yokoyama, 2008). After limb
amputation, the axolotl develops a blastema through which bone, muscle, nerve,
and skin are regenerated in a spacio-temporal manner to restore structure and
function. Bone morphogenic proteins (BMPs) and fibroblast growth factors
(FGFs) have been shown to play a significant role in tissue and limb regeneration
(Yu et al., 2010). BMPs are members of the transforming growth factor (TGF )
superfamily (Wu & Hill, 2009). They bind to serine/threonine kinase receptors
that initiate a signal cascade resulting in the translocation of transcription factor
complexes into the nucleus (Wu & Hill, 2009). FGFs bind FGF receptor (FGFr)
tyrosine kinases that in turn activate the Mitogen Activated Phosphorylation
Kinase (MAPK) signaling pathway. In turn, the BMP and FGF signaling pathways
modulate tissue morphogenesis (Guo & Wang, 2009). The many different BMPs,
BMPrs, FGFs, and FGFrs, permit a multitude of diverse signaling modalities and
create great mechanistic complexity.

18

To gain insight into this process, we utilized the embryonic chicken skin
organ (CSO) model to study feather bud morphogenesis. The development of a
feather from epithelial and mesenchymal interactions is a complex process. We
manipulated the epigenetic environment to better understand how morphogens
are modulated. CSOs were treated with epigenetic enzyme inhibitors. Chaetocin,
a histone methyltransferase inhibitor, acts upstream of FGF4 and BMP2 to
promote feather bud growth. MS275, a histone deacetylase inhibitor, acts
upstream of FGF4 and BMP2 to inhibit feather bud growth. Taken together, these
data suggest that histone methylation and histone acetylation modulate feather
bud morphogenesis via gene expression changes. More specifically, decreasing
histone methylation and increasing histone acetylation changes the chromatin
compaction pattern and modulates the activities of FGF4 and BMP2.

2.2 Materials & Methods
2.2.1 Chicken Skin Organ Preparations and Culture
Embryonic chickens at day 7 were harvested. The dorsal skins were
surgically removed. They were cultured on cell culture inserts in DMEM fortified
with 10% chicken serum. CSOs were treated with increasing concentrations of
Chaetocin (Sigma) and or MS275 (Sigma), or Chaetocin and MS275 soaked
beads, or BMP2 and FGF4 soaked beads, for 2 to 5 days.

19

2.2.2 Histological preparations
The CSOs were fixed in 4% PFA and dehydrated in a graded alcohol
series. The tissue was cleared in Xylene and embedded in paraffin wax. Seven
micron sections were cut on a microtome. H&E sections were done according to
an accepted protocol. Whole mount tissues were fixed in 4% PFA and then
stored at 4
o
C in PBS with NaAzide.

2.2.3 Immunohistochemical (IHC) and In situ hybridization (ISH) procedures
Whole mount and section IHC was performed as previously mentioned.
Briefly, tissues were permeabilized with methanol and blocked with 3% H2O2 for
30 min, and then blocked with mouse serum for 1hr. The primary antibody was
added and incubated over night at 4
o
C. The tissue was washed with Tris buffered
saline with tween-20 (TBST) and the secondary antibody was added for 1hr at
room temperature. The tissue was washed with TBST and if utilized, a tertiary
antibody was added for 1 hr at room temperature. The tissue was washed and
color was developed using the AEC kit or fluorescence was visualized. The
fluorescent secondary was Alexa 488 from Invitrogen (Carlsbad, CA). -catenin
antibody is from Sigma (St. Louis, MO). The HDAC1 antibody is from Lab Vision
(Fremont, CA). Whole mount and section ISH was performed as previously
mentioned. Briefly, tissues were fixed over night at 4
o
C in 4% PFA and then

20

dehydrated in a graded methanol series diluted in phosphate buffered saline with
tween-20 (PBT). The samples were rehydrated, bleached with 6% H
2
O
2
, and
digested with proteinase K. The samples were incubated with mRNA antisense
probe labeled with DIG over night at 65
o
C and then washed the next day with
SSC. The samples were incubated with anti-DIG antibody over night at 4
o
C and
color was developed using NBT and BCIP.

2.2.4 Alkaline Phosphatase Staining
To detect feather bud cell condensations, alkaline phosphatase staining
was performed as mentioned in Ito 2007 (M. Ito, Yang, Andl, Cui, Kim, Millar, &
Cotsarelis, 2007a). Briefly, dorsal skins were excised and epidermis separated
from the dermis. The dermis was fixed in acetone at 4
o
C and then incubated in
NBT/BCIP to exhibit alkaline phosphatase staining.

2.2.5 Microarray Analysis
CSOs were cultured as previously mentioned and then treated with a
known Class I HDAC inhibitor, Trichostatin A (TSA). CSOs were cultured for 2
days in 100 M TSA and then harvested. Total RNA was extracted, labeled with a
fluorophore, and hybridized to an oligonucleotide chip. The Affymetrix Chicken
Genome Chip was utilized to examine global gene expression changes. Partek
Genomic Suite and Ingenuity Pathway Analysis software were used to analyze

21

the microarray data. Analysis of Variance (ANOVA) with a False Discovery Rate
of 0.5 was used to determine significant differences in gene expression levels.
2D Agglomerative Hierarchical Clustering analysis was utilized to rapidly
visualize sets of differentially expressed genes.

2.3 Results
2.3.1 Expression patterns of epigenetic enzymes
In order to discover if chromatin compaction patterns were playing a role
in feather bud morphogenesis we conducted a panel of gene expression pattern
studies. The first question we asked was, “Are epigenetic enzymes present in the
chicken skin?”
Embryonic chickens at day E7.5, E8.5 and adult chicken skin were utilized
to study the expression patterns of representative histone methyltrasnferase
(HMT) and histone deacetylase (HDAC) epigenetic enzymes. In situ hybridization
for Suv39h1 exhibited expression in the epithelium and the mesenchyme of
embryonic chicken skin (Figure 2-1A). This expression became restricted to the
epithelium of the skin in the adult chicken skin (Figure 2-1A). At E7.5 and E8.5
the Suv39h1 expression is stronger in the feather bud region when compared to
the interbud region (Figure 2-1A). Immunohistochemistry (IHC) of histone
deacetylase 1 (HDAC1) exhibits expression in the epithelium and mesenchyme

22

of the feather bud and interbud regions (Figure 2-1B). We conclude that
epigenetic enzymes are present in the chicken skin organ.

23

Figure 2-1 Expression patterns of epigenetic enzymes in chicken skin.
(A) In situ hybridization of Suv39h1 (histone methyltransferase) in embryonic
chicken skin exhibits increased expression in the bud forming regions. (B)
Immunohistochemistry of HDAC1 (histone deacetylase 1) exhibits expression in
the bud and interbud epithelium and mesenchyme.

24

Figure 2-1 Continued:

25

2.3.2 Does perturbing epigenetic enzymes affect skin organ development?
In order to determine if HMTs or HDACs were playing a role in chicken
skin organ morphogenesis, we treated the culture media of chicken skin organ
(CSO) cultures with an HMT inhibitor (chaetocin) or an HDAC inhibitor (MS275)
(Figure 2-2). The time course and dose response assays exhibit a delay in
cellular condensations, feather bud formation, and feather bud elongation with
increasing concentrations of each inhibitor (Figure 2-2A). However, Chaetocin
and MS275 demonstrate marginally different inhibitory effects. Chaetocin exhibits
a more general inhibition of feather bud morphogenesis demonstrating delayed
feather bud development. MS275 exhibits suppression of feather bud elongation
in addition to the general inhibition of feather bud morphogenesis (Figure 2-2A).
MS275 treated CSOs exhibit a moderately normal feather bud diameter but the
feather bud length is shorter when compared to controls. Interestingly, both
Chaetocin and MS275 demonstrate the ability to change -catenin and sonic
hedgehog (SHH) gene expression patterns (Figure 2-2B). Increasing
concentrations of MS275 exhibit a suppression of and change in gene
expression pattern of -catenin and SHH (Figure 2-2B). The SHH gene
expression is not observed at the feather bud tip in high level MS275 treated
CSOs when compared to the control CSOs. Increasing concentrations of MS275
also demonstrate a gradation of -catenin gene expressions from regions of low
levels where putative feather buds would normally form, to small halos of

26

immature feather bud regions, and finally to reduced levels in more mature
feather bud regions when compared to control CSOs. Chaetocin treatment
increases the feather bud region expressing -catenin and SHH when compared
to the control CSOs (Figure 2-2B). The Chaetocin treated CSOs exhibit feather
buds that have changed their feather bud shape from conical to more flattened.
The distal region of these flattened feather buds exhibit an increase in -catenin
and SHH when compared to the distal region of the control feather buds with a
conical shape (Figure 2-2B).

27

Figure 2-2 Inhibiting HMTs or HDACs causes abnormal chick skin organ
development.
(A) Treating chicken skin organs at E7 with Chaetocin (HMT inhibitor) or MS275
(HDAC inhibitor) causes a delay in cell condensation, feather bud formation, and
feather bud elongation after 2 days in culture. (B) High doses of Chaetocin
delays cell condensation, bud formation, and bud elongation. Chaetocin
increases the area of -catenin and SHH expression in feather buds. (B) High
doses of MS275 delays cell condensation, bud formation, and inhibits
bud elongation. MS275 decreases -catenin and SHH expression. Interestingly,
moderate doses of Chaetocin and MS275 cause feather bud fusion and a
change in feather bud shape, and the -catenin and SHH expression patterns.

28

Figure 2-2 Continued:

29

2.3.3 Inhibiting HMTs or HDACs causes abnormal feather bud
morphogenesis.
Because Chaeticon and MS275 exhibited different effects on CSO
development, we decided to further investigate the inhibitor assay by applying
beads soaked with inhibitor directly on developing CSOs (Figure 2-3). The CSOs
treated with MS275 soaked beads exhibited a complete inhibition of feather bud
development (Figure 2-3A). Cellular condensations fail to form resulting in no
feather bud morphogenesis. A circular zone of inhibition forms around the MS275
soaked beads. At the perimeter of this zone, cellular condensations are present.
Farther away and outside of this zone, feather buds undergo normal
development. The CSOs treated with Chaetocin soaked beads exhibited a
circular zone of initial delay of feather bud formation (Figure 2-3A). However, this
delay is alleviated and feather bud development appears normal at the later time
points (Figure 2-3A). The feather bud orientation is altered in the Chaetocin
soaked bead CSOs. The feather buds change their orientation from the anterior-
posterior direction to an orientation directed towards the beads (Figure 2-3A). In
addition, the feather bud patterning is altered in CSOs treated with Chaetocin
soaked beads. The spacing between each feather bud is decreased resulting in
feather buds developing closer together (Figure 2-3A). The expression patterns
of -catenin and SHH were altered in the HDAC and HMT inhibitor treated zones
(Figure 2-3B). MS275 bead treated CSOs exhibited a complete suppression of

30

both -catenin and SHH gene expression inside the circular zone. Outside of this
zone, -catenin and SHH gene expressions appeared normal (Figure 2-3B).
Chaetocin bead treated CSOs exhibit a suppression of SHH and an increase in
-catenin inside the circular zone. Outside of this zone, -catenin and SHH gene
expressions appeared normal (Figure 2-3B). Furthermore, the SHH gene
expression pattern changed in the Chaetocin bead treated CSOs. SHH is
predominantly expressed at the tip of a normal feather bud while the Chaetocin
treated CSOs exhibited a more diffuse pattern (Figure 2-3B). Chaetocin bead
treated CSOs also exhibit a fusion of feather buds (Figure 2-3C). Confocal
microscopy demonstrates the -catenin expression pattern of normal and
abnormal feather bud morphology. Feather buds nearest to the Chaetocin beads
exhibit fusion and abnormal -catenin expression patterns. Feather buds farther
away exhibit normal morphology and -catenin expression patterns (Figure 2-
3C).

31

Figure 2-3 Inhibiting HMTs or HDACs causes abnormal feather bud
morphogenesis.
(A) Treating chicken skin organs at E7 with Chaetocin (HMT inhibitor) beads
causes a change in feather bud orientation and feather bud patterning. (A)
Treating chicken skin organs at E7 with MS275 (HDAC inhibitor) beads inhibits
cell condensation, feather bud formation, and feather bud elongation after 2 days
in culture. (B) Chaetocin causes abnormal -catenin and abnormal SHH
expression patterns. (B) MS275 suppresses -catenin and SHH expression. (C)
Chaetocin bead treated CSOs exhibit abnormal -catenin protein expression
patterns and fusion of feather buds. (* = beads, blue box outlines magnified area)

32

Figure 2-3 Continued:

33

2.3.4 Inhibiting HMTs or HDACs causes abnormal chick skin organ
development.
How does perturbing epigenetic enzyme activity affect BMP or FGF
activities on skin organ development?
HMTs and HDACs act upstream of morphogens BMP and FGF.
Inhibiting HMTs with Chaetocin or HDACs with MS275 causes abnormal
chick skin organ development. Because Chaetocin and MS275 exhibit different
effects on feather bed morphogenesis, we decided to further investigate the
inhibitor assay by applying morphogen soaked beads directly on developing
CSOs (Figure 2-4). Previously, we have shown the effect of BMPs and FGFs on
chicken feather bud development (Jung, Oropeza, & Thesleff, 1999). BMP
demonstrated a suppression of feather bud development (Jung et al., 1998).
When we treat CSO’s with both BMP2 and MS275 we observe an increase of
inhibition in feather bud development (Figure 2-4). The zone of inhibition of
BMP2 was dramatically increased when MS275 was added. When we treat
CSO’s with both BMP2 and Chaeatocin we observe a relief of inhibition of feather
bud development. The inhibitory action of BMP2 on feather bud development
was reveresed by treatment with Chaetocin and normal feather bud development
occurs (Figure 2-4). FGF4 demonstrated a promotion of feather bud development
(Jung et al., 1998). When we treat the CSO’s with FGF4 and MS275 we observe

34

a complete inhibition of feather bud development (Figure 2-4). The promotion of
feather bud development by FGF4 was completely reversed and feather bud
development was prevented.

35

Figure 2-4 Inhibiting HMTs or HDACs modulates BMP and FGF activities on
chick skin organ morphogenesis.
Treating chicken skin organs simultaneously with Chaetocin and BMP permits
feather bud development. Treating chicken skin organs with MS275
simultaneously with BMP exacerbates feather bud inhibition synergistically.
Treating chicken skin organs with FGF promotes feather bud development.
Treating chicken skin organs with Chaetocin simultaneously with FGF
exacerbates feather bud development synergistically. Treating chicken skin
organs with MS275 simultaneously with FGF inhibits feather bud development.

36

Figure 2-4 Continued:

37

2.3.5 Cellular differentiation is suppressed by the HDAC inhibitor MS275
What gene expression changes occur during normal CSO morphogenesis,
and are these expression patterns altered when HDAC activity is inhibited?
Microarray analysis was performed to measure gene expression changes
during CSO morphogenesis. CSOs were cultured for 2 days with or without
HDAC inhibitor treatment, and their respective RNAs were harvested. These
RNAs were probed onto Affymetrix Chicken Genome chips. The 5 CSO samples
were uploaded with 46 other chicken tissue samples into Partek Genomic Suite
(PGS) software. The microarray data demonstrated multiple gene expression
changes between samples (Figure 2-5). Two dimensional hierarchical clustering
exhibited unique gene expression profiles for the E7 control, E7+2days in culture,
and E7+2days in culture TSA treated samples (Figure 2-5). We next looked at
the differential gene expression changes that occur during CSO morphogenesis
in culture for 2 days (Table 2-1). Epigenetic genes (MYST), transcription factors
(Msx2), and cellular differentiation genes (keratins) all demonstrated changes in
gene expression (Table 2-1). Some of these major biological groups, like the
keratin gene expression changes, were inhibited when treated with the HDAC
inhibitor (Table 2-1). Interestingly, cytokeratin 23 (Krt23) expression, an HDAC
inducible keratin, was not significantly changed.

38

Figure 2-5 2D Hierarchical Cluster of 51 Chicken Tissue Samples. 51
chicken tissue samples comprised of adult and embryo stages were probed onto
Affymetrix Chicken Genome chips and clustered according to gene expression
profiles. The samples clustered according to similarity. All of the adult epithelial
tissue samples clustered together (A), all of the adult mesenchymal tissue
samples clustered together (B), all of the embryonic epithelial tissue samples
clustered together (D), and all of the embryonic mesenchymal samples clustered
together (C). All of the CSO samples clustered within the embryonic
mesenchymal sample group as a unique sub-cluster (C*).

39

Figure 2-5 Continued:

40

Table 2-1 6Way ANOVA with interactions exhibits differential gene
expression. 51 chicken tissue samples comprised of adult and embryo stages
were probed onto Affymetrix Chicken Genome chips and analyzed for differential
gene expression by 6Way ANOVA. Approximately 8,700 genes were analyzed.
Some genes are represented by multiple probe sets. Due to space limitations
only 81 tests are shown.

41

Table 2-1 Continued:

42

Figure 2-6 Microarray analysis of CSO development.
CSOs were cultured for 0 to 2days and the RNA were harvested and probed onto
oligonucleotide microarray chips. 2D hierarchical clustering exhibits the CSO
samples clustering together according to similar gene expression profiles. UO126
(green) is a map kinase inhibitor and TSA (purple) is an HDAC inhibitor. (pink =
E7+2days in culture, brown = E7 with 1hr culture)

43

Figure 2-6 Continued:

44

Table 2-2 CSO Microarray data table. Control versus TSA treated CSO
microarray data table listing differentially expressed genes. Differential gene
expression analysis identified specific gene changes occurring during CSO
development. A biological subset of these gene expression patterns were altered
when the CSOs were treated with the HDAC inhibitor (TSA).

45

Table 2-2 Continued:

46

2.4 Discussion
2.4.1 Hierarchy of gene expression mechanisms modulate feather bud
morphogenesis
Feather bud formation requires the spacio-temporal expression of specific
genes. First, an incompetent epithelial field becomes competent. Second, this
field is patterned into bud or interbud regions. Finally, the feather bud primordia
are formed. -catenin, Bmp, FGF, and SHH have all been demonstrated to play a
role in this process (Jung et al., 1998; Widelitz, Jiang, Noveen, Chen, & Chuong,
1996). The chromatin must be in the correct formation to allow induction of these
molecules. Histone acetylation and histone methylation statuses must coordinate
with the spacio-temporal expression of these genes. Here we investigate the role
of global histone acetylation and methylation during feather bud morphogenesis.
Perturbing histone acetylation led to the inhibition of feather bud formation and
perturbing histone methylation led to the disruption of feather bud patterning
(Figure 2-7).

2.4.2 Previous work with BMP and FGF
Jung et al demonstrated the effects of inhibiting Bmp and FGF during
feather bud morphogenesis (Jung et al., 1998). FGF promoted feather bud
formation and caused feather bud fusion. In this work, we demonstrate that
Chaetocin promotes feather bud formation and feather bud fusion, and increases

47

the SHH expression. Furthermore, Chaetocin seems to act synergistically with
FGF to promote feather bud formation (Figure 2-3 and Figure 2-4). This suggests
that histone methylation may be modulating the FGF pathway during feather bud
patterning. Jung et al also demonstrated that Bmps inhibit feather bud formation.
In this work we demonstrate that MS275 inhibits feather bud formation by
inhibiting -catenin expression. Finally, MS275 seems to act synergistically with
Bmp2 to inhibit feather bud formaiton (Figure 2-3 and Figure 2-4). This suggests
histone acetylation may be modulating the Bmp pathway during feather bud
morphogenesis.

2.4.3 Tissue Competency
All ectodermal organs are derived from reciprocal epithelial-mesenchymal
interactions (C. M. Chuong, Widelitz, Ting-Berreth, & Jiang, 1996; Dhouailly,
1975; Fuchs & Horsley, 2008). If each skin appendage is generated from tissues
with identical genetic content, then how are the varieties of appendages
possible? One mechanism could be epigenetic modulation of cellular
competency. By adjusting a tissue’s competency for a specific morphogen, a
variety of skin appendages is possible. Depending on which inhibitor we treat the
CSOs with; MS275, Chaetocin, or TSA, we observe an inhibition of feather bud
induction or an increase in feather bud inducing gene expression of -catenin or
SHH (Figures 2-2 and 2-3). This suggests that tissue competency is being

48

modulated. Furthermore, when we treat CSOs with TSA we observe a down
regulation of keratins which are differentiation markers (Table 2-1). This further
suggests a modulation of tissue competency by HDACs. However, because the
inhibitors utilized in our assays are not gene specific, we may be observing non-
specific affects causing the phenotypes and gene expression changes observed.
One possible example is that K23 expression levels do not change when CSOs
are treated with TSA. K23 has been shown to be HDAC inducible (Zhang, Wang,
Huang, Nelson, & Smith, 2001). By inhibiting HDACs with TSA, we would predict
to measure a change in K23 gene expression. We did not observe this (Table 2-
1). However, K23 expression levels did not change from E7 to E7+2days in
culture from the control group (Table 2-1). This suggests that K23 may not be
involved in feather morphogenesis or that our time points may not be correct for
measuring changes in K23 expression for this process.

49

Figure 2-7 Model of histone acetylation and methylation during CSO
morphogenesis. MS275 inhbits HDAC and thus increases histone acetylation.
High levels of histone acetylation inhibits tissue competency as demonstrated by
-catenin suppression. Chaetocin inhibits HMT thus decreasing histone
methylation. Low levels of histone methylation promote feather bud fusion as
demonstrated by SHH expression.

50

Figure 2-7 Continued:

Adapted from Lin et al., Current Opinion in Cell Biology 2006, 18:730.

51

Chapter 3: Disrupted ectodermal organ morphogenesis in mice with
conditional deletion of histone deacetylase in the epidermis: dermal cysts,
reduced hair variations, supernumerary nails, and hyper-pigmentation
3.1 Introduction
Gene expression is modulated by a hierarchy of mechanisms. The DNA
sequence, transcription machinery, and chromatin status must be coordinated for
gene expression to occur properly. Among the many levels of epigenetic
regulation (Ikegami, Ohgane, Tanaka, Yagi, & Shiota, 2009), histone modification
is one of the major mechanisms. Histone proteins can be acetylated, methylated,
phosphorylated, or ubiquinated. Through these interactions, different regions of
chromatin in developing cells can be in active euchromatin status or inactive
heterochromatin status at different stages of development and regeneration.
Recent progress in epigenetics has helped launch a new phase of studies to
interpret organ development in the context of epigenetic changes (Ikegami et al.,
2009).
Ectodermal organs, consisting of hairs, scales, nails, glands, etc, develop
through epithelial – mesenchymal interactions mediated by the exchange of
molecular signals to achieve progression toward coordinated functional
morphogenesis (C. -. Chuong, 1998; Dhouailly, 1975; Fuchs & Horsley, 2008).
Further, hairs have the unique ability to be episodically lost and replaced
throughout the life of an animal (Stenn & Paus, 2001). Understanding how these

52

ectodermal organs develop, maintain homeostasis, and regenerate at the
epigenetic level is our objectives in this study.
Here we focus on the roles of histone deacetylase in the development and
regeneration of ectodermal organs. Histone deacetlyase 1 (HDAC1) and HDAC2
are predominanty expressed in the basal layer of human foreskin keratinocyte
raft cultures, and are absent in the respective suprabasal layers (Longworth et
al., 2005). The Hdac1 gene encodes an enzyme that removes acetyl groups from
histone proteins. This protein is highly conserved throughout the animal kingdom
and is present in multiple chromatin remodeling complexes. HDAC1 is
ubiquitously expressed in all tissues (Brunmeir, Lagger, & Seiser, 2009;
Haberland, Montgomery, & Olson, 2009). Perturbance of HDAC activity has been
associated with abnormal intestinal epithelial cell differentiation (Tou, Liu, &
Shivdasani, 2004), cardiac muscle development (Montgomery et al., 2007),
Wilm’s Tumor development (Shao et al., 2007), pituitary gland development
(Olson et al., 2006), embryonic stem cell differentiation (E. R. Lee, Murdoch, &
Fritsch, 2007), and mammary epithelial cell structure and function (Le Beyec et
al., 2007).
The functions of HDAC1 and HDAC2 have recently been studied in the
skin by conditionally deleting them from embryonic mouse skin (LeBoeuf et al.,
2010). Loss of both HDAC1/2 activities resulted in the skin becoming thinner and
the failure of epidermal stratification, similar to p63 deletion (Candi et al., 2006;

53

Mills et al., 1999). Hair follicles did not form. Because the phenotype was so
severe, the later events of hair biology were not studied. Therefore the roles of
HDAC in the morphogenesis and regeneration of hair follicles and other skin
ectodermal organs remain unknown.
Here we attempt to address the role of HDAC1 in skin morphogenesis and
post-natal tissue homeostasis utilizing the cytokeratin 14 (Krt14) promoter to
drive tissue specific reduction of HDAC1 activity in a mouse model. We
generated epithelial specific deletion of HDAC1 and dual HDAC1 homozygote
/HDAC2 heterozygote transgenic mice. These transgenic mice exhibit a
spectrum of similar skin phenotypes with different severity. The lesions include
epidermal hyperkeratosis, alopecia, epidermal utricles, dermal cysts, sebaceous
hyperplasia, aberrant nail formation, and hyperpigmentation. Molecular
characterization implies that HDAC in the epidermis is required for hair follicle
morphogenesis and differentiation, and for the proper epithelial interaction with
the dermal papilla. With suppressed HDAC, hair follicles degenerate into dermal
cysts composed of epidermis stuck in abnormal progenitor status. Since this
epidermis cannot assume normal function, dermal papillae fail to form properly,
and melanocytes are increased in number and infiltrate tissues that are normally
unpigmented. The mice present themselves as a animal model for studying the
roles of histone acetylation in the development, regeneration and pathology of
ectodermal organs.

54

3.2 Materials & Methods
3.2.1 Transgenic mice
The Krt14cre mice were from Yang Chai lab at USC (Hosokawa et al.,
2009). The Hdac1loxp and Hdac2loxp mice were a kind gift of Eric Olson
(Montgomery et al., 2007). Krt14cre mice were bred with Hdac1loxp and or
Hdac2loxp mice to generate Krt14creHDAC1
fl/fl
and Krt14creHDAC1
fl/fl
,HDAC2
+/fl

mice. Mice were genotyped as previously described (Montgomery et al., 2007).

3.2.2 Histology and Immunostaining
Paraffin section immunostaining procedures were performed as previously
described (Jiang, Stott, Widelitz, & Chuong, 1998) using the following antibodies;
KRT14 (Lab Vision), KRT10 (Lab Vision), -catenin (Sigma), PCNA (Chemicon),
p63 (Santa Cruz), involucrin (Lab Vision), KRT15 (Lab Vision), Sox2 (Abcam),
Sox9 (Santa Cruz), NCAM (C. M. Chuong & Edelman, 1985), AE13 (Santa
Cruz), AE15 (Santa Cruz) and HDAC1 (Abcam and Invitrogen). Briefly, tissue
was fixed in 4% PFA, dehydrated with a graded ethanol series, and paraffin
embedded. Tissue sections were then rehydrated, blocked with mouse serum,
and primary antibody incubated overnight at 4
o
C. Slides were then washed three
times and secondary antibody incubated for two hours at room temperature.
Slides were washed three times and tertiary antibody incubated for thirty minutes

55

at room temperature. Color was developed with an AEC kit (Vector Labs). Whole
mount immunostaining was previously described (Jiang et al., 1998). Briefly,
mouse dorsal skin or tail skin was incubated in 20mM EDTA at 37
o
C until the
epithelium can be separated from the mesenchyme. The epithelium was fixed in
4% PFA for overnight at 4
o
C. The epithelium was blocked and permeabilized,
then incubated overnight with primary antibody. The next day the samples were
washed five times and then incubated overnight with secondary antibody. The
next day the samples were washed five times, mounted, and then imaged on a
confocal microscope.

3.2.3 Confocal microscopy
A Zeiss LS510 confocal microscope was used to image the fluorescently
labeled cells and nuclei. Z stack images were captured. These images were
processed in the Zeiss LSM Image Browser software.

3.2.4 RT-PCR
mRNA was extracted from mouse skin using the RNeasy Protection kit
(Qiagen, Valencia, CA, USA). AMV reverse transcriptase (Roche Diagnostics,
Palo Alto, CA, USA) was used for reverse transcription (RT) and polymerase
chain reaction (PCR) was performed with Taq polymerase (Invitrogen, Carlsbad,
CA, USA) using the manufacturer’s protocol. The primers for Krt10 (ID

56

12852157a1), Krt14 (ID 21489935a2), involucrin (ID 6680506a2), loricrin (ID
6678708a2), VDR (ID 31543944a2), Axin2 (ID 31982733a1), Blimp1 (ID
6680784a2), Gli2 (ID 21411092a3), and IHH (ID 14149643a3) were utilized
(Wang & Seed, 2003).

3.3 Results
3.3.1 HDAC1 Expression Pattern in Wild Type Mice
The normal pelage hair structure at the early anagen stage is
demonstrated using H&E staining. The distribution of endogenous HDAC1 in
early and late anagen phases of wild type C57BL/6 mouse skin was explored by
immunostaining. In early anagen skin, HDAC1 is widespread predominantly
located within the basal epithelial layer of the dorsal skin and hair follicles (Figure
3-1A). As the keratinocytes differentiate in the interfollicular suprabasal
epithelium, HDAC1 expression is lost. HDAC1 expression in late anagen skin
was explored in a pelage hair follicle (fluorescent staining, Figure 3-1B) and a
vibrissae hair follicle (chromogenic staining, Figure 3-1C). These images produce
similar distribution patterns, enabling us to see that HDAC1 is strongly expressed
in the outer root sheath (ORS), weaker in the inner root sheath (IRS) and absent
in the medulla of these follicles. HDAC1 also is present in the sebaceous glands
of pelage hairs (Figure 3-1B-C). Whole mount immunostaining of mouse tail
epithelium demonstrated that HDAC1 expression is present throughout the pilo-

57

sebaceous units (Figure 3-1D). We next used confocal microscopy to analyze the
distribution of HDAC1 within the mouse tail epithelium at a higher resolution
(Figure 3-1E). The panel on the left demonstrates a compressed view through all
of the optical planes. Of note, HDAC1 (green) is enriched in the matrix epithelium
when compared to other regions of the pelage and tail hair follicle (Figure 3-1B-
E).

58

Figure 3-1 HDAC1 Immunostaining Expression Pattern in Wild Type Mouse
Hair.
(A) H&E stained section from mouse dorsal skin showing the structure of a
pelage hair follicle at early anagen. HDAC1 immunostaining of a parallel section
shows expression predominantly in the basal layer of the epithelium. (B) HDAC1
immunofluorescence staining (green) of an anagen stage pelage hair follicle.
Nuclei are stained with propidium iodide (red). (C) HDAC1 immunostaining (red)
of a vibrissa hair follicle. (D) HDAC1 whole mount immunofluorescence staining
(red) and nuclei (DAPI, blue) of tail hair follicles. HDAC1 expression is in
sebaceous glands, inner and outer root sheaths and matrix areas. (E) Three
confocal Z-stack images of a single tail hair follicle immunostained for HDAC1
(green) and nuclei (DAPI, blue). Compressed view of all sections (left), a single
inner plane (middle) and a single outer plane (right). (arrow = upper hair follicle
epithelium) (arrowhead = hair matrix)

59

Figure 3-1 Continued:

60

3.3.2 Production of Krt14creHDAC1
fl/fl
Mice
Since Hdac1 knockout mice are embryonic lethal (Lagger et al., 2002), we
generated tissue specific HDAC1 functionally null transgenic mice. We utilized
the cre-lox system to inhibit the activity of HDAC1 in the epithelium of mouse
skin. Under the direction of the human Krt14 gene promoter, cre recombinase
was utilized to remove exons 5, 6, and 7 of the mouse Hdac1 gene (Figure 3-2A;
adapted from Montgomery 2007), and or exons 2, 3, and 4 of the mouse Hdac2
gene (not shown). Krt14cre mice were bred to Hdac1loxp or
Hdac1loxp/Hdac2loxp mice to generate Krt14creHDAC1
fl/fl
and Krt14creHDAC1
fl/fl

HDAC2
+/fl
mice. In these examples, mouse A is Krt14creHDAC1
+/fl
HDAC2
fl/fl
,
mouse B is Krt14creHDAC1
fl/fl
HDAC2
+/+
, and mice C and D are
Krt14creHDAC1
+/fl
HDAC2
+/fl
(Figure 3-2B). PCR amplification showed that all
mice demonstrating a hair phenotype possessed the recombined Hdac1 allele
(Figure 3-2C).

61

Figure 3-2 Genotyping Assay of Krt14creHDAC
fl/fl
Mice.
(A) Schematic figure demonstrating the structure of the Hdac1 construct and cre
mediated excision (adapted from Montgomery 2007). (B) PCR amplification of
Hdac alleles from genomic DNA isolated from the tails of transgenic mice
denotes the genotypes. Lanes 4, 5, and 6 demonstrate the genotype of animal B;
homozygous state of Hdac1loxp, the wild type state of Hdac2, and the presence
of Krt14cre. (C) PCR amplification demonstrates the recombined Hdac1 allele of
Krt14creHDAC
fl/fl
in all phenotype exhibiting mice.

62

Figure 3-2 Continued:

63

3.3.3 Gross Phenotype of Krt14creHDAC1
fl/fl

Gross Pathology
The Krt14creHDAC1
fl/fl
mice demonstrated a reduction in size, weight, and
lifespan compared to wild type littermates (Figure 3-3A). Here we focus on
characterizing changes in their skin, hair and nails.

Craniofacial ectodermal organs
The Krt14creHDAC1
fl/fl
mouse eye morphology appeared normal (data not
shown). However, the eyelids exhibit an increase in size (Figure 3-3B), resulting
in a smaller eye opening than that of the wild type mouse. Krt14creHDAC1
fl/fl

mice exhibit hypotrichosis or madarosis, the loss of eyelashes. The size and
number of eyelashes, and hairs in regions surrounding the eyes are decreased
(Figure 3-3B). The pelage hair surrounding the eye opening of the wild type
mouse exhibits a specific hair pattern. This pattern is disrupted in the
Krt14creHDAC1
fl/fl
mice (Figure 3-3B). These traits were observed and
exacerbated in the double transgenic Krt14creHDAC1
fl/fl
,HDAC2
+/fl
mice (Figure
3-3B).

Vibrissae
The vibrissae of the Krt14creHDAC1
fl/fl
mice exhibit a loss and a reduction
in size of vibrissae hair filaments (Figure 3-3B). The length and diameter of the

64

vibrissae hair fiber is greatly reduced when compared to the wild type (Figure 3-
3B and data not shown). The number of vibrissae follicles appeared normal.
These traits were observed and exacerbated in the double transgenic
Krt14creHDAC1
fl/fl
,HDAC2
+/fl
mice (Figure 3-3B).

Skin and hairs
The dorsal skin of Krt14creHDAC1
fl/fl
mice exhibits extensive and
progressive alopecia, hyperkeratosis, pigmented pouches called utricles and
dermal cysts (Fig 3-3C). Note, hairs were shaved to reveal the underlying wild
type skin, while it was not necessary to shave the Krt14creHDAC1fl/fl mice.
Examination of the dermal side of Krt14creHDAC1
fl/fl
dorsal skin exhibits dermal
cysts, disrupted hair patterns, coiled hair follicles, and abnormal hair alignment
compared to wild type littermates.
Wild type mouse skin possesses 4 types of pelage hairs: guard, awl,
auchen, and zigzag (Schmidt-Ullrich et al., 2006). These can not be differentiated
by gross examination in Krt14creHDAC1
fl/fl
skin, and the skin exhibits
predominantly one type of hair morphology. The sparse hair fibers were slightly
shorter, thinner, and sometimes coiled (Figure 3-3D). The hair medulla cells
appear as a single row, are smaller, misshapen and frequently misaligned
(Figure 3-3D). These traits were observed and exacerbated in the double
transgenic Krt14creHDAC1
fl/f
,HDAC2
+/fl
mice (Figure 3-3D).

65

Nails and Paws
The paws and digits exhibited a reduction in overall size, hyperkeratosis,
supernumerary nails and hyperpigmentation (Figure 3-3E). Pigmentation was
increased in the nails, fat pads, and epidermis of the paw.
The front paws demonstrated a unique outgrowth of the hyponychium that
is greatly extended beyond the nail proper, and the nail bed is often widened
(data not shown). The rear paws contained pigmented supernumerary nails. The
pigmentation in these nails is due to the presence of ectopic pigmented
melanocytes (Figure 3-3E). Each trait described above is observed and
exacerbated in the double transgenic Krt14creHDAC1
fl/fl
HDAC2
+/fl
mice (Figure 3-
3E).

66

Figure 3-3 Gross morphological phenotype of the Krt14creHDAC
fl/fl
Mice.
(A) Adult Krt14creHDAC1
fl/fl
mice are smaller than control littermates and exhibit
alopecia. (B) Eye openings are reduced with regions of hypotrichia on the eyelids
and regions surrounding the eyes. The hair pattern surrounding the eyes is
disrupted. The vibrissae exhibit hypotrichia and the interfollicular epithelium
exhibits hyperpigmentation. (C) The Krt14creHDAC
fl/fl
dorsal skin exhibits
prominent pigmented cysts (arrow). Removing the skin and viewing the dermal
side exhibits abnormal hair patterning, malformation of hair follicles and dermal
cysts (boxes = magnified area). (D) Krt14creHDAC1
fl/fl
mice have predominantly
‘zigzag-like’ hairs that are thin and abnormally formed. (E) The paws exhibit
hyperkeratosis and hyperpigmentation. Supernumerary pigmented claws form
from the rear digits. H&E of the nail bed demonstrated extra nails developed on
the lateral sides of the digit. Melanocytes were present at the dermal-epidermal
junction of the supernumerary nails. These phenotypes are more severe in
Krt14creHDAC1
fl/fl
, HDAC2
+/fl
dual transgenic mice.

67

Figure 3-3 Continued:

68

Tail
The tail of the Krt14creHDAC1
fl/fl
mice exhibited hyperkeratosis,
hyperpigmentation, sebaceous hyperplasia, and a reduction of hair when
compared to the wild type (Figure 3-4B). The tail was smaller in diameter and
length. These traits were exacerbated in the double transgenic
Krt14creHDAC1
fl/fl
HDAC2
+/fl
mice (Figure 3-4B). The size of the sebaceous
glands was increased (Figure 3-4B). The pigmentation of the pilosebaceous unit
was disrupted (Figure 3-4B). Normally, the pigmented sections of the tail
epithelium are restricted to the pilosebaceous units. The Krt14creHDAC1
fl/fl
mice
tail epithelium exhibited an increase in the amount of pigmentation in the
interfollicular and follicular tissue. These traits are observed and exacerbated in
the double transgenic Krt14creHDAC1
fl/fl
,HDAC2
+/fl
mice (Figure 3-4B).

69

Figure 3-4 Histology of skin homeostasis and tail follicle morphology in 5-6
month old Krt14creHDAC1
fl/fl
mice. (A) H&E staining demonstrates a reduction
in hair follicles, the prevalence of hair cysts, and a thicker interfollicular epidermis
in Krt14creHDAC1
fl/fl
mouse skin. PCNA staining was increased in the
interfollicular, infindibular, and dermal cyst epithelia. KRT15 and TUNEL staining
was increased in the dermal cyst epithelium. KRT14 expression is increased and
expands beyond the basal epithelial layer, and in the dermal cyst wall. p63
expression appears normal and is present in the cyst epithelia. KRT10 and AE13
(IRS) staining is greatly diminished. However, involucrin is increased in the
dermal cyst epithelium (data not shown). B) Tail skin of Krt14creHDAC1
fl/fl
and
Krt14creHDAC1
fl/fl
, HDAC2
+/fl
mice exhibited hyperpigmentation, a reduction of
hair follicles, and a disruption of ectodermal organ patterning. Oil Red O staining
exhibits an increase in sebocytes. These phenotypes are more severe in
Krt14creHDAC1
fl/fl
, HDAC2
+/fl
dual transgenic mice.

70

Figure 3-4 Continued:

71

3.3.4 Histology of Krt14creHDAC1
fl/fl

Dorsal Skin
We next examined sections of dorsal skin from 5-6 month old
Krt14creHDAC1
fl/fl
mice for histological changes. Sections clearly demonstrate a
loss of hair follicles and the presence of utricles and dermal cysts (Figure 3-5).
H&E staining demonstrated abnormal skin morphology when compared to wild
type skin architecture (Figure 3-5A). In addition to moderate hyperkeratosis,
Krt14creHDAC1
fl/fl
mice have a thickened interfollicular epidermis with highly
variable cell size and shape between the basal and suprabasal layers. The
sebaceous glands have normal maturation with abundant sebum containing
cytoplasm, but there is significant sebaceous hyperplasia. A few hair follicles
appear to be normal, however, most seem to have degenerated, leaving utricles
with enlarged infindibular regions and dilated epithelial cysts of varied sizes.
These epithelial cysts contain multiple layers of squamous keratinocytes that
express KRT14 but not KRT10. Some cysts do not contain pigmented keratin
fibers, others have a few of these fibers with some arranged in hair ‘shaft-like’
keratin columns, while others are completely filled with pigmented fibrous
material. There is also abundant pigment infiltrated throughout the dermis. Some
sebaceous glands can be seen connected to cysts located at a more superficial
level of the reticular dermis or connected to the bottom of utricles.

72

We characterized changes in the expression of genes associated with
proliferation and differentiation in the skin of 5-6 months old Krt14creHDAC1
fl/fl

mice (Figure 3-5A). Proliferation was detected by staining for proliferating cell
nuclear antigen (PCNA). The dorsal skin of Krt14creHDAC1
fl/fl
mice exhibits an
increase in PCNA expression in follicular and interfollicular basal cell
keratinocytes. Cytokeratin 15 (KRT15), a marker for hair follicle stem cells,
produced a normal staining pattern in ‘normal-like’ hair follicles. The ‘less normal’
hair follicles exhibited an increase in KRT15 expression. Finally, the dermal cysts
exhibited greatly reduced KRT15 expression compared to wild type. The
distribution of p63, a transcription factor in basal epithelial cells modulating
epithelial stratification, was similar to wild type expression patterns. The dorsal
skin of Krt14creHDAC1
fl/fl
mice exhibits increased apoptosis revealed by TUNEL
staining. Interestingly, regions of apoptosis and proliferation co-localized within
the follicular and interfollicular basal layer keratinocytes.
Subsequently, we examined the status of various cell populations in the
epithelium to determine their differentiation status (Figure 3-5A). Expression of
cytokeratin 14 (KRT14), a marker for the basal epithelium, is present in both
basal and suprabasal layers of Krt14creHDAC1
fl/fl
follicular and interfollicular skin.
The KRT14 expression pattern of Krt14creHDAC1
fl/fl
dorsal skin demonstrates
staining of epithelial cysts throughout the dermis (figure 3-5A). The dermal cyst
walls exhibit increased levels of KRT14. While, cytokeratin 10 (KRT10), a marker

73

of suprabasal skin epithelium differentiation, demonstrated expansion in the
interfollicular epithelium, yet it is not present in dermal cyst epithelia. Involucrin is
a cell envelope protein, a marker for suprabasal layer of skin epithelium, and IRS
of the hair follicle. The epithelia of utricles and dermal cysts of Krt14creHDAC1
fl/fl

mice exhibit an increase in expression of involucrin when compared to wild type
(data not shown). AE13 stains for keratins present in the IRS. The dorsal skin of
Krt14creHDAC1
fl/fl
mice exhibit reduced AE13 staining (Figure 3-5A). Some cysts
exhibited very minor positive staining of their fibrous contents. AE15 stains hair
keratins present in the ORS and medulla. Krt14creHDAC1
fl/fl
mice exhibit little to
no AE15 staining (data not shown).
We wondered whether earlier developmental stages of hair loss and cyst
development would be evident in skin from younger mice. Skin from 2-3 month
old mice were sectioned and stained for markers involved in proliferation and
differentiation (Figure 3-5A). Here we only show expression patterns of
Krt14creHDAC1
fl/fl
mice, since wild type expression patterns have been well
documented in the literature. KRT14 and Neural Cell Adhesion Molecule (NCAM)
were used to provide an overview of skin structure. A similar distribution of
'normal-like’ hair follicles, ‘abnormal hair follicles', utricles and various cysts types
of those that are observed in older mice (Figure 3-4 and 3-5A) are present.
KRT14 was highly expressed in the cysts and showed a similar distribution as
described previously. NCAM normally stains the dermal papilla of hair follicles. In

74

Krt14creHDAC1
fl/fl
mouse skin, NCAM shows a more widely dispersed pattern
when compared to wild type (Figure 3-5A).
To evaluate proliferation in the dermal cysts we stained sections from 2-3
month old mice with antibodies to PCNA. PCNA immunostaining exhibited an
increase in the hair follicle matrix region and is present in the cyst wall epithelia
(Figure 5A). Some cysts have very high levels of PCNA in the epithelium
surrounding the cyst, while PCNA is more sparsely distributed in other cysts.
Similar to the older mice, KRT10 exhibited no expression in cyst wall epithelia
while Involucrin expression is increased in cyst wall epithelia (Figure 5A). p63 is
present and exhibits a normal expression pattern. However, -catenin expression
is increased and present in the cell membranes. -catenin regulates the
conversion of stem cell to transient amplifying cells and is present in the basal
layer of interfollicular epithelium, the ORS, and the hair follicle bulge. -catenin is
widely dispersed in the membranes of ketratinocytes within the cyst epithelium
(Figure 5A). Interestingly, Sox9 is present in the cyst wall epithelia (Figure 5A).
Sox9 is a marker for the basal layer, ORS, and activated stem cells that are
migrating out of the bulge to populate the hair matrix. Krt14creHDAC1
fl/fl
mouse
skin exhibited an increase in expression of Sox9 in the basal layer of the
epithelium (Figure 5A) and the outer layers of the dermal cyst walls. Sox2 is a DP
marker for guard, awl, and auchene hair follicles (Driskell, Giangreco, Jensen,
Mulder, & Watt, 2009). Sox2 immunostaining exhibits a decrease in positive DP

75

when compared to wildtype, and is present in some dermal cysts (data not
shown). These overall structures observed at 2-3 months do not appear to be
significantly different from those observed in the older mice.
We further examined the skin of 3 week old Krt14creHDAC1
fl/fl
mouse skin
and similar phenotypes were observed (Figure 5B). There are more regions with
normal appearing hair follicles. There are also follicles that appear to be in
transitional stages between a ‘ballooned’ hair follicle and a dermal cyst. There
are several samples that exhibit abnormal hair follicles with a DP connected by a
stalk of epithelial cells to a disorganized matrix area, that in turn, is connected to
an enlarged infundibulum filled with fibrous material. These samples suggest that
HDAC1 null matrix cells are unable to function properly and give rise to poorly
formed cortex cells and IRS. AlkP positive sebaceous gland cells appear to be
surrounding the bottom of specific utricles (Figure 5B). The 3 week old
Krt14creHDAC1
fl/fl
mouse data suggests there may be a difference in severity
compared to the older mice, and hair follicles may degenerate into cysts as the
mouse ages.
Immunostaining for proliferation and differentiation markers exhibits similar
patterns as the 2 month and 5 month old mice (Figure 4 and 5). PCNA
immunostaining exhibited an increase in basal layer epithelia, infundibular
epithelia, and cyst wall epithelia (Figure 5B). KRT14 expression is increased in
basal layer epithelia and is present in cyst wall epithelia. The KRT10 expression

76

is expanded in the thickened interfollicular epithelium but is absent from the
dermal cysts, similar to the 2-3 month and 5-6 month old mice. Involucrin exhibits
increased expression from basal to suprabasal layers, and throughout utricles
and dermal cysts. AE13 immunostaining is present in the normal hair fibers and
in specific cysts that contain fibrous material. Cysts within the dorsal skin of
Krt14creHDAC1
fl/fl
mice exhibited no AE15 expression while expression could be
seen in the adjacent less affected hair follicles.

77

Figure 3-5 Molecular markers of skin differentiation in 2-3 month and 3week
old Krt14creHDAC1
fl/fl
mice. (A) KRT14 expression is increased and expands
beyond the basal epithelial layer, and in the dermal cyst wall. Dermal NCAM is
more dispersed than in wild type mouse skin. KRT10, AE13 and AE15 are
decreased in the cyst epithelium. Involucrin is increased in the infundibular and
dermal cyst epithelia. PCNA staining was increased in the interfollicular,
infindibular, and dermal cyst epithelia. p63 expression appears normal and is
present in the cyst epithelia. -catenin expression is elevated in all epithelia and
localized to the cell membranes. Sox9, a hair follicle ORS and hair bulge marker,
is present in the cyst epithelium. Sox 2 is a guard/awl/auchene, but not zig-zag,
dermal papilla marker and is present in a few relatively formed hair dermal
papillae and some cysts. The Sox2 positive cells in cysts are not well organized.
(B) H&E stained section of 3week old dorsal skin exhibits

78

Figure 3-5 Continued:

79

Tail
The shorter and thinner tail of the Krt14creHDAC1
fl/fl
mice exhibited
hyperkeratosis, hyperpigmentation, sebaceous hyperplasia, and a reduction of
hair number compared to the wild type (Figure 4B). The size of the sebaceous
glands is increased in a dose dependent manner. The pigmentation of the pilo-
sebaceous unit was disrupted. Normally, the pigmented sections of the tail
epithelium are restricted to the pilo-sebaceous units but the Krt14creHDAC1
fl/fl

mice tail exhibited increased interfollicular and follicular pigmentation in the
epidermis (Figure 4B)
To verify that the enlarged structures seen when examining pilosebacous
units were indeed sebaceous glands, we used Oil Red O to stain for oils present
within the sebaceous glands (Figure 4B). This staining is increased tremendously
in the sebaceous glands of Krt14creHDAC1
fl/fl
mice and is increased in a dose
dependent manner. Pigmentation and sebaceous gland phenotypes were
exacerbated in the dual transgenic Krt14creHDAC1
fl/fl
HDAC2
+/fl
mice (Figure 4B).
Nails and Paws
To further characterize the nails of Krt14creHDAC1
fl/fl
mice, we used H&E
staining which revealed differences between the transgenic and wild type nails
(Figure 3E). We highlighted three areas of the nails. One is responsible for hard
keratin production (green) and two produce soft keratins (blue and red). The
regions that normally make soft keratins in wild type mouse nails make hard

80

keratins in the supernumery nails of Krt14creHDAC1
fl/fl
mice. The
hyperpigmentation of the supernumerary nails is due to the presence of
melanocytes in the supernumerary nail bed (Figure 3E).

3.4 Discussion
3.4.1 Epigenetic enzyme expression pattern detection.
Although HDAC1 is thought to be ubiquitously expressed, more studies
are implicating more complex tissue specific expression patterns. HDAC1 and
HDAC2 have been shown to be expressed predominantly in the basal layer of
human foreskin keratinocyte rafts (Longworth et al., 2005). Here we demonstrate
an enrichment of HDAC1 expression in the mouse skin basal layer and the hair
matrix epithelia (Figure 2-1). However, these visualized expression patterns rely
on the specificity of the HDAC1 antibody utilized. Longworth et al acquired
HDAC1 antibody from Cell Signaling Technologies (Beverly, MA.), Le Bouef et al
from Invitrogen (Carlsbad CA.), and this study from Abcam (Cambridge, MA.)
and Invitrogen (Carlsbad, CA.). If the antibody for HDAC1 is not specific, and
potentially crossreacts with other HDACs, then the expression patterns visualized
are not specific to HDAC1. This is a significant possibility because of HDACs are
highly conserved and class I HDACs have very similar structures. Thus there is a
distinct possibility that the observed HDAC1 expression patterns are due to
detection of all class I HDACs. If this were the case, then our expression pattern

81

studies would be affected. Our results would then be expression patterns of class
I HDACs and not specific to HDAC1. This would suggest an enrichment of class I
HDACs in the in the mouse skin basal layer and the hair matrix epithelia (Figure
2-1). The decision to conditionally perturb HDAC1 from KRT14 expressing
tissues would have been based on this data. The resulting phenotype observed
is due to the specific reduction of HDAC1 and not HDAC2 or HDAC3 (HDAC3
genotype data not shown). However, we feel the literature demonstrates the
specificity of HDAC1 antibodies to be valid. The antibodies utilized in our study
were chosen because they were validated by other researchers in their
respective studies.

3.4.2 Defective morphogenesis and differentiation in the skin of
Krt14creHDAC1
fl/fl
adult mice.
Our hypothesis is the presence of HDAC1 in the basal layer of skin
epithelium is playing a role in skin morphogenesis. We tested this hypothesis by
creating tissue specific, functionally null HDAC1 mice. Here we demonstrate the
expression pattern of HDAC1 in normal adult mouse skin and disruption of
HDAC1 activity leads to abnormal skin morphogenesis. The special-temporal
expression of HDAC1 in the basal layer, hair follicle ORS, and hair follicle matrix
suggests a role in keratinocyte differentiation, hair fiber formation, and epithelial
barrier formation.

82

Histone acetylation is generally thought to favor euchromatin in which
RNA polymerase and its coactivators can associate with their target promoters
and initiate transcription. Histone deacetylation promotes heterochromatin
formation and inactivation of genes in these regions. Suppression of histone
deacetylation prevents the removal of acetyl groups on histones, leaving these
DNA regions in active states, i.e. euchromatin. Overall, our results suggest that
HDAC helps maintain the basal layer epithelia, ORS, hair follicle stem cells, and
hair follicle matrix cells. The reduction of HDAC activities promotes cells to retain
their progenitor status, and inhibits the proper differentiation.
HDAC1 is expressed in the basal layer, hair follicle ORS, is enriched in the
hair follicle matrix (Figure 1), and suggests a role in keratinocyte differentiation,
ORS formation, IRS formation, and hair fiber formation. Mice with Krt14creHDAC
deletion exhibit regions of hyperkeratosis, alopecia, hyperpigmentation, abnormal
hair follicles, presence of utricles, epithelial cysts in the dermis, sebaceous
hyperplasia, and abnormal skin tissue organization. Some epidermis of the
interfollicular epithelium was thickened. In the interfollicular epidermis the KRT14
expression expanded into the suprabasal layers and the KRT10 expression is
increased. Proliferation, apoptosis, and differentiation were all up-regulated in
the interfollicular epithelium.
Interestingly, the hair stem cell marker data demonstrated a spectrum of
expression patterns. The relatively normal appearing K14creHDAC1
fl/fl
hair

83

follicles exhibited normal levels of KRT15. However, abnormal follicles exhibited
increased levels of KRT15+ cells. Finally, the dermal cysts of HDAC1 null mice
expresses little to no KRT15. This suggests HDAC1 aids in the maintenance of
the KRT15 population of cells and the ability of KRT15 positive cells to
differentiate. Sox9 has been shown to be important in interfollicular and follicular
epithelial stem cells (Nowak, Polak, Pasolli, & Fuchs, 2008). Here we show the
Krt14creHDAC1
fl/fl
Sox9 expression pattern is perturbed exhibiting an increase in
Sox9 expression in the dermal cysts and a general increase in the dermis. -
catenin exhibits a similar increase in expression throughout the dermal
epithelium. The epithelium surrounding the dermal cysts is strongly positive for
p63, Sox9, -catenin, and KRT14. All of these genes have been shown to
modulate skin stem cell populations. This suggests that HDAC1 is helping to
modulate the skin stem cell populations and suppressing HDAC1 activity results
in an increase in skin stem cell populations.
In the follicular regions, phenotypes range from ‘normal’ hair follicles to a
spectrum of abnormal hair follicles, which may arise from either a failure to form
normal hair follicles or to degenerate from hair follicles, to utricles and dermal
cysts. In the follicles with generally normal morphology, KRT15 is expressed
beyond the follicle bulge, but p63 expression appears normal. Among these
abnormal structures, we observe utricles, which have a widened infundibulum
region with extended lumens and wide hair canals, with additional epithelial

84

layers (Figure 4 and 5). We observe multiple cysts in the dermis with variations in
size, shape and molecular composition (Figure 6). In the cyst wall epithelia,
KRT14 positive layers were expanded. Interestingly, KRT10 expression is absent
in the cysts suggesting that these structures may possess a more progenitor cell
state, and may be directed towards a hair follicle fate versus an interfollicular
fate. Hair keratins recognized by monoclonal antibody AE13 (IRS marker) and
AE15 (ORS marker) were negative in almost all the cyst epithelium and lumen.
Since the IRS is derived from matrix cells, the result suggests that a matrix
doesn’t form, or that matrix cells are unable to differentiate towards these fates in
Krt14creHDAC1
fl/fl
mice. The younger mice (Figure 5B) exhibit an increase in the
number of normal hair follicles and less mature cysts. This suggests the cyst
phenotype may increase with age due to the failure of proper hair follicle
regeneration.
The enrichment of HDAC1 in the matrix area of the hair follicle suggests a
role for HDAC1 in hair fiber formation because matrix cells differentiate to
become Henle’s layer, Huxley’s layer, and the inner root sheath cuticle.
Perturbing HDAC1 in the hair follicle epithelium causes abnormal hair fibers with
reduced AE13 (IRS) and AE15 (ORS) expression. This data suggests a similar
mechanism as that in the epithelium of the skin; HDAC1 is promoting cellular
differentiation in the basal epithelium. In the hair matrix case, HDAC1 is
modulating the ability of matrix cells to differentiate and form the respective

85

layers of the IRS. Failure of AE15 expression is most likely attributed to the
inability of the KRT14+ basal layer to differentiate as previously mentioned.
Furthermore, these matrix cells may be unable to perform in the reciprocal
signaling cascade that occurs during hair development and regeneration with the
DP. The normal appearing Krt14creHDAC1
fl/fl
hair follicles exhibited normal
KRT15 levels, the abnormal follicles exhibited increased levels of KRT15+ cells,
and mature dermal cysts expressed no KRT15 (Figure 4 and 5). We observed
utricles, dermal cysts, and abnormal hair follicles. These abnormal hair follicles
exhibited large utricles distally, and abnormal matrix areas connecting to a DP
through a thin epithelial stalk. This abnormal hair follicle may be an intermediate
stage of utricle and cyst formation. If this is the case, then a possible mechanism
for utricle and dermal cyst formation is the failure of matrix cells to function
properly. The matrix is responsible for forming the IRS and cuticle cells. If HDAC
null matrix cells produced unorganized fibrous material, then a lumen filled utricle
could develop. This utricle would be large if fibrous material production from
matrix cells were abundant. Furthermore, the matrix needs to respond to signals
from the DP. Malfunctioning matrix cells could respond incorrectly, and abnormal
morphogenesis could occur. This miscommunication between the matrix and the
DP could lead to dermal cyst formation. The cyst epidermis exhibit high levels of
proliferation and apoptosis suggesting a loss of homeostasis control. We found
few signs of normal dermal papillae in the dermis immediately surrounding the

86

cysts. In most extreme cyst examples, there are fragments of dermal cell clusters
that are positive of DP markers scattered along the outside of cyst wall or
throughout the dermis.
The skin also exhibits sebaceous hyperplasia, particularly in the tail. This
phenotype exhibits a clear dependence on HDAC level, depending on the loss of
HDAC1 or loss of HDAC1 with heterozygote HDAC2 alelles. The sebaceous
glands are located through out the dermis and are not always associated with a
hair follicle or cyst. This suggests a separation of the pilo-sebaceous unit (Figure
4 and 5).
The reduction of HDAC1 activity results in an increase in sebaceous
glands, epidermis, expanded infundibulum, and dermal cysts. All together, these
results suggest that HDAC1 plays a role in epidermal basal layer - suprabasal
layer cell homeostasis, hair follicle stem cell -TA - differentiated cell homeostasis,
matrix cell function, and matrix - dermal papilla interactions.

3.4.3 Context-dependent roles of HDAC
Although HDAC1 and HDAC2 have overlapping expression patterns
during development and tissue homeostasis, and seem to be able to compensate
for one another in vitro, HDAC 1 and 2 have distinct functions in vivo where they
modulate different sets of genes (Brunmeir et al., 2009; Haberland et al., 2009).
Specifically, HDAC1 modulates Wnt signaling during skin epithelium

87

development (Alonso & Fuchs, 2003; Brunmeir et al., 2009; Chen, Fernandez,
Mische, & Courey, 1999). The role of HDAC1 and HDAC2 in the skin and hair
follicle development was previously explored utilizing HDAC1/HDAC2 functionally
null skin epithelium during mouse skin development (LeBoeuf et al., 2010). They
found that a single deletion of HDAC1 or of HDAC2 did not produce a significant
phenotypic change. No phenotypes were observed for mice with
Krt14creHDAC1
fl/fl
, Krt14creHDAC2
fl/fl
or Krt14creHDAC1
+/fl
,HDAC2
fl/fl
, or
KrtcreHDAC1
fl/fl
,HDAC2
+/fl
. However, Krt14creHDAC1
fl/fl
,HDAC2
fl/fl
mice were
embryonic lethal at approximately E18.5 to E19 due to the failure of the skin
organ to develop correctly. Therefore, they focused on the effects of HDAC on
embryonic development. Levels of H3K9Ac were elevated in the embryonic skin
of these double knockout mice. At a gross level, these embryonic mice displayed
thin and smooth skin in which the epidermis did not stratify. Hair follicle
development did not initiate. Eyelids did not fuse and digits did not separate.
Taste bud and tooth defects were also noted. These defects suggest that the
mouse may be stopped at an early stage of tissue development. Molecular
markers confirm this possibility. At the cellular level, there is a progressive loss of
proliferation suggesting that progenitor cells are not maintained appropriately.
There is also a concomitant progressive increase in apoptosis exacerbating the
defects. They noted that the HDAC1
fl/fl
,HDAC2
fl/fl
phenotypes were similar in
many respects to those seen in p63 mutants. In particular, decreases in

88

p16/Ink4a produced by Np63 required functional HDAC1 and HDAC2. Using
ChIP analysis they found that HDAC activity was elevated in promoters of Np63
repressed genes. Furthermore, p53 expression is inhibited by HDAC1 and
HDAC2 activity. In the double knockout mice, p53 expression is increased and
may lead to decreased cell cycle progression and increased apoptosis.
In contrast to their findings, our Krt14creHDAC1
fl/fl
, Krt14creHDAC2
fl/fl
, or
Krt14creHDAC1
fl/fl
,HDAC2
+/fl
, transgenic mice did survive to adulthood and these
adult mice exhibited extreme skin phenotypes. Hence, we focused on these later
defects in skin development. We noted defects in the skin, claws, paws, eyes
and pigmentation. LeBoeuf et al (LeBoeuf et al., 2010) described a decrease in
expression of PCNA during dual HDAC1/HDAC2 functionally null embryonic skin
epithelium development. In contrast we see increased PCNA expression in the
adult epithelium. Skin development and adult skin epithelium homeostasis are
biologically different processes, so these seemingly contradictory results are not
surprising. We observed an increase in TUNEL staining in the adult mouse skin
epithelium and LeBoeuf et al (LeBoeuf et al., 2010) found increased TUNEL
staining in E16.5 mouse skin suggesting an increase in apoptosis in the basal
layer during skin epithelium development and in adult skin epithelial
homeostasis.
One possible source of difference is the mouse strain used. Both groups
used the same HDAC floxed lines from Olson’s lab (Montgomery et al., 2007),

89

yet these were crossed to mice containing different K14 constructs in different
mouse strains. Our Krt14cre mice came from Dr. Chai (Y. Ito et al., 2001) while
Millar's mice were from different source (Andl, Reddy, Gaddapara, & Millar,
2002). Millar's Krt14cre construct was in B6SJLF1/J while ours was in a
background of C57BL6/J x CBA. Many researchers utilizing transgenic mice to
investigate the role of HDACs in developmental processes have observed
contradictory results and this has been attributed to inter-strain differences
(Brunmeir et al., 2009).

3.4.4 Comparison with dermal cysts from Hairless mice
In many aspects, the phenotype of our Krt14creHDAC1
fl/fl
and
Krt14creHDAC1
fl/fl
,HDAC2
+/fl
mice is similar to that observed in hairless, rhino,
and rhino Yurlovo mice (Panteleyev et al., 1998; Panteleyev et al., 1998;
Panteleyev, Paus, Ahmad, Sundberg, & Christiano, 1998). The hairless gene is
expressed in the epidermis and hair follicle, but not in the dermis. In humans,
mutations in the hairless gene were found to underlie atrichia with papular
eruption (Ahmad et al., 1999; Balighi et al., 2009; Michailidis, Theos, Zlotogorski,
Martinez-Mir, & Christiano, 2007). Hr
-/-
mice exhibit no hair at 3 to 5 months of
age, when mature they weigh more than wild type mice, the sebaceous glands
are enlarged and are always associated with an utricle, their skin easily
ulcerates, and they possess elongated and excessively curved claws. In contrast,

90

Krt14creHDAC1
fl/fl
mice have partial hair at 3 to 5 months, when mature they
weigh less than wild type littermates, the sebaceous glands are not always
associated with utricles, the skin is hyperkeratotic but does not ulcerate easily.
Hr
-/-
mice begin to lose hair in the anterior region, and then the alopecia
progresses in a posterior manner defined by a sharp border. Krt14creHDAC1
fl/fl

mice possess sparse abnormal hair in the anterior region and this becomes
exacerbated in the posterior region with no defining hair wave border. The timing
of alopecia and the nature of its progression differs between these mice. This
suggests that the underlying mechanism of alopecia is probably different
between these mice. Pigmentation differences suggest that the dermal cysts of
the respective mice develop differently. Hr
-/-
mice have enlarged sebaceous
glands associated with utricles and dermal cysts. Krt14creHDAC1
fl/fl
mice also
have enlarged sebaceous glands but they do not always associate with utricles
and rarely associate with dermal cysts (Figure 4). This suggests a separation of
components within the pilo-sebacous unit in Krt14creHDAC1
fl/fl
mice, which is not
observed in Hr
-/-
mice.
It is not clear whether our observed utricles and dermal cysts in the
Krt14creHDAC1
fl/fl
mice represent different stages of cyst development or if they
derive from different follicular compartments as was suggested for the hairless
phenotype (Panteleyev et al., 1998). Our molecular staining does not support the
latter hypothesis, but we are looking at times involving cyst maintenance rather

91

than formation and molecular expression may have changed in the interim. While
there are some differences in the phenotypes produced by HDAC loss of
function, similar phenotypic trends may be due to the fact that hairless binds to
and may recruit HDAC1, HDAC3 and HDAC5 to its target genes (Potter et al.,
2001; Potter, Zarach, Sisk, & Thompson, 2002).

3.4.5 Comparison with dermal cysts from conditional -catenin reduction
mice
Dermal cyst formation has also been observed in Krt14cre -catenin
fl/fl

mice (Huelsken, Vogel, Erdmann, Cotsarelis, & Birchmeier, 2001). These mice
exhibited lack of embryonic hair development in -catenin negative skin with no
BMP or SHH expression that are downstream events. Further more, in -catenin
positive skin that did develop hair, these hairs turned into cysts during the first
hair cycle. These cysts were positive for KRT10 and Involucrin in the inner most
epithelial layers, and lacked sebaceous glands. This lead the authors to surmise
that hair follicle cell fates were redirected to skin epithelial fates. In contrast,
Krt14creHDAC1
fl/fl
mice exhibit sebaceous hyperplasia, no KRT10 staining in the
dermal cysts, and an increase in Involucrin staining.
Overall, we observe cyst formation where proper morphogenesis is
required in order for more complex epithelial organs to form. This is observed in
polycystic kidney in which primary cilia are defective. Patients who suffer from

92

Birt–Hogg–Dubé syndrome develop lung, skin, and kidney cysts (Hartman et al.,
2009). Here we observe ectodermal follicular organs that degenerate into cysts
from hair follicles. Cyst development is a default mechanism of morphogenesis
gone wrong. Complex tissues and their respective organs will not form if
molecular signaling cascades are abnormal or tissue competency is
compromised.

3.4.6 Future Directions
The epigenetic process in development is now recognized as an important
biological modulator of gene expression. Krt14creHDAC1
fl/fl
mice exhibit drastic
skin epithelial phenotypes and can be a great disease model. However, the
downstream targets of HDAC1 have yet to be elucidated. Other investigators
have also looked into the roles of epigenetic process on the morphogenesis and
regeneration of skin ectodermal organs. EZH2 is a histone methyltransferase and
a part of the Polycomb Group Complex 2. EZH2 conditional knock out mice
exhibited a thinner dermal epithelium that led to reduced progeniotor populations
(Ezhkova et al., 2009; Ezhkova et al., 2011). DNMT1 is a DNA methyltransferase
that maintains DNA methylation patterns after cell replication. DNMT1 is
expressed in the basal layer epithelium and its functional loss promotes cellular
differentiation by allowing expression of differentiation genes that are normally
methylated (Sen, Reuter, Webster, Zhu, & Khavari, 2010).

93

By utilizing system biology techniques, such as microarray analysis and
ChIP sequencing, we hope to study these targets in the future. With other
epigenetic study on skin morphogenesis, we also expect the understanding of
these skin pathologies will move to a new era.

94

Chapter 4: Repair versus regeneration: Msx2 homeobox gene is required to
generate epidermal stem cells for de novo hair regeneration after wounding
4.1 Introduction
Wound healing is an important physiological process that repairs injured
tissue to restore functionality. Response to injury generally can be categorized as
repair or regeneration. In regenerative wound healing, damaged parts are
replaced with normal fully functional tissues similar to those before wounding. In
reparative wound healing, damaged parts are mended with reparative connective
(scarring) and epithelial tissues which do not rebuild the original tissue
architectures (Bullard, Longaker, & Lorenz, 2003). Mammalian fetuses or very
young animals exhibit scarless wound healing, and these animals undergo
regenerative wound healing. The ability to regenerate tissues declines in adult
animals as they age. These animals usually exhibit the repair mechanism, and
this ability, like the regenerative ability, declines as the animal ages.
In nature, amphibians exhibit the ability to regenerate their amputated
limbs or tails with well organized digital skeleton, skin, muscle, and nerves
(Brockes & Kumar, 2005; Tanaka, 2003). Since there are no stem cells stored
away, cells have to go through a de-differentiation process to generate the
blastema (Kragl et al., 2009; Yu et al., 2010). Mammals do not exhibit such
remarkable limb regeneration ability. However, the human fetus has been
observed to regenerate an amputated finger during the first trimester (Allan et al.,

95

2006). Adolescent children (1 to 11yrs) are observed to regenerate an amputated
digit tip, including the bone (Gurtner, Callaghan, & Longaker, 2007; Illingworth,
1974). These regenerative abilities quickly decline with age and are eventually
lost altogether. Biologically, while the repair process saves the lives of patients
and is an achievement, with the promise of regenerative medicine, scientists
aspire to rebuild organs for those suffering from disease. This may be achieved
by delivery of stem cells or delivery of exogenous genes that can reprogram cells
to the fate we desire.
How is the response of repair and regeneration determined? What
molecules are required for the reprogramming that produce cells capable to
regenerate? If we can decipher what molecular mechanism is required, maybe
we can reprogram these cells to once again possess the ability to regenerate.
Recently, scientists have learned that it is possible to reprogram the fate of cells
by over-expressing just a few molecules (Patel & Yang, 2010). The success of
induced pluripotential stem cells is the most dramatic example of success
(Takahashi et al., 2007). There is also success in converting exocrine pancreatic
glands in vivo into insulin secreting beta cells (Zhou, Brown, Kanarek, Rajagopal,
& Melton, 2008), and the conversion of skin fibroblasts into neurons (Vierbuchen
et al., 2010). This reprogramming was done with exogenously delivered
molecules in vivo or ex vivo, and has the potential to cause problems. It would be
greatly beneficial if we can identify the method organisms use to do their own

96

reprogramming, as seen in the regeneration of amphibian limbs in the
regenerating blastema.
Recently, a remarkable work showed that new hairs can regenerate via a
de novo mechanism in a wound bed if the right environment is provided (C. M.
Chuong, 2007; M. Ito, Yang, Andl, Cui, Kim, Millar, & Cotsarelis, 2007a). The de
novo hair regeneration only occurs in the center of the wound. Lineage studies
demonstrated that these cells are not derived from existing hair bulge stem cells.
Rather, they are derived from inter-follicular epidermis. Thus there must be some
endogenous reprogramming process to convert their fates into hair forming stem
cells.
Hair follicles play a significant role in the wound healing process. Cells
from the hair bulge, collar region, and the outer root sheath of the hair follicle
contribute to re-epithelialize the skin (M. Ito et al., 2005; Levy, Lindon, Harfe, &
Morgan, 2005). This re-epithelialization process is guided by coordinated
signaling events occurring between the dermis and epidermis. The ability of a
tissue to respond to a signal (cellular competency) and the ability of a tissue to
send a signal (inducibility) are necessary components of regeneration. What
mechanism programs a non-competent keratinocyte cell from the wound margin
to a keratinocyte cell at the wound center to be competent and respond to
inductive signals?

97

Msx proteins are members of the muscle segment homeobox gene family
and encode transcription factors. Msx2 has been shown to be involved in digit,
and craniofacial development (Jabs et al., 1993; Liu et al., 1994). In newborn
mice, Msx1 is important for digital regeneration. Msx2 is shown to be induced
and important for axolotl limb regeneration (Carlson, Bryant, & Gardiner, 1998).
In the skin, Msx2 null mice exhibit a delayed hair regeneration response after hair
plucking (Ma, Liu, Wu, Plikus, Jiang, Bi, Liu, Muller-Rover, Peters, Sundberg,
Maxson, Maas, & Chuong, 2003a). Recently, we showed that these mice showed
acceleration in wound closure following small full thickness wounding, implying a
faster repair response. These results imply a role for Msx2 in modulating repair
versus regenerative responses. Thus, we utilized the large wound procedure (M.
Ito, Yang, Andl, Cui, Kim, Millar, & Cotsarelis, 2007a) to analyze the de novo hair
regeneration process in Msx2
-/-
mice.

4.2 Materials & Methods
4.2.1 Generation of transgenic mice
Animal utilization and care were approved by the Institutional Animal Care
and Utilization Committee (IACUC) of the University of Southern California.
Msx2-functionally deficient mice (Msx2tm1Rilm/Mmcd), or Msx2
-/-
, were
generated by inserting the neo cassette into the Nde1 site, 5’ to the Msx2
homeobox exon 2 region, as described previously (Satokata, Ma, Ohshima, Bei,

98

Woo, Nishizawa, Maeda, Takano, Uchiyama, Heaney, Peters, Tang, Maxson, &
Maas, 2000a). Krt14Noggin transgenic mice were generated by microinjection of
human Krt14 promoter chicken Noggin-poly A inserts into the male pronucleus of
fertilized eggs of C57BL/6J x CBA/J mice, followed by reimplantation of injected
eggs into pseudopregnant C57BL/6J x CBA/J females as previously described
(M. Plikus et al., 2004). Msx2 promoter-LacZ reporter mice were described
previously (Brugger et al., 2004). Krt14creROSAMsx2tm mice were generated. In
brief, Tg(KRT14-cre) /B6.129S4-Gt(ROSA)26Sortm1Sor/J mice were mated with
Msx2tm mice to generate Krt14creROSAMsx2tm.

4.2.2 Large wounds
All animal protocols were performed according to the USC IACUC review
board. Adult (3mth, 6mth, 9mth, and 1yr old) mice were anesthetized and surgery
was performed to remove a 1.5cm x 1.5cm piece of dorsal skin. Buprenex was
injected sub-cutaneously and Ketophen was added to the drinking water as an
ad libum analgesic. These mice were sacrificed at various time points and the
wounds harvested. To assay for new hair formation, wounds were examined
carefully every two days. Hair filaments in the wound center can become visible
around PWD19. To observe earlier hair germs, wound epithelium has to be
removed and stained to visualize developing hair germs.

99

4.2.3 Separation of epidermis and dermis
The separation of epithelium from the dermis was performed as
mentioned in Ito 2007. Briefly, full thickness wounds were incubated in 20mm
EDTA in Phosphate Buffered Saline (PBS) at 37
o
C overnight. The next day the
epithelium was gently peeled away from the dermis with fine watchmaker’s
forceps.

4.2.4 Histological preparations
The wound tissues were fixed in 4% PFA and dehydrated in a graded
alcohol series. The tissue was cleared in Xylene and embedded in paraffin wax.
Seven micron sections were cut on a microtome. H&E sections were done
according to an accepted protocol. Whole mount tissues were fixed in 4% PFA
and then stored at 4
o
C in PBS with NaAzide.

4.2.5 Immunohistochemical and histochemical procedures
Whole mount and section IHC was performed as previously mentioned.
Briefly, tissues were permeabilized with methanol and blocked with 3% H2O2 for
30 min, and then blocked with mouse serum for 1hr. The primary antibody was
added and incubated over night at 4
o
C. The tissue was washed with TBST and
the secondary antibody was added for 1hr at room temperature. The tissue was
washed with TBST and if utilized, a tertiary antibody was added for 1 hr at room

100

temperature. The tissue was washed and color was developed using the AEC kit
or fluorescence was visualized. The fluorescent secondary was Alexa 488. K17
antibody is from Lab Vision (MS-489-SO).

4.2.6 Alkaline Phosphatase Staining
To detect hair bud germs, alkaline phosphatase staining was performed
as mentioned in Ito 2007. Briefly, full thickness wounds were excised and
epidermis separated from the dermis. The dermis was fixed in acetone at 4oC
and then incubated in NBT/BCIP to exhibit alkaline phosphatase staining.

4.2.7 -galactosidase staining
Tissues expressing the LacZ gene were incubated in Xgal buffer and color
was developed as previously mentioned (Yeh 2009). Briefly, wound epithelium
was separated from wound dermis using 20mM EDTA and incubated overnight
at 4oC in X-Gal staining solution (2mM MgCl2, 0.01% sodium deoxycholate,
0.02% NP-40, 0.1% 4-chloro-5-bromo- 3-indolyl beta-galactosidase [X-Gal]
[Research Organics, Cleveland, OH], and 5mM potassium ferrocyanide in
phosphate-buffered saline [PBS]).

101

4.2.8 Statistics
Photographs are representative samples of at least 5 replicates. Hair
follicle counts are the average of at least 5 samples with the standard error. The
wound healing assay has high inter-sample variability, yet clear differences can
be seen in de novo hair follicle number between the control and Msx2
-/-

specimens.

4.3 Results
4.3.1 In large wounds, Msx2
-/-
mice show a failure of de novo hair
regeneration
The role of Msx2 in large wounds (M. Ito, Yang, Andl, Cui, Kim, Millar, &
Cotsarelis, 2007a) was studied by creating full thickness 1.5cm x 1.5cm wounds
in wild type C57BL/6 and transgenic mice (Figure 4-1). Earlier, when we
performed 3mm wounds (small wounds), wounds in Msx2
-/-
mice closed faster
(Yeh et al., 2009). Paradoxically, in the large wounds, Msx2
-/-
mice healed slower
than wild type (C57BL/6) mice. Msx2
-/-
mice had not completely re-epithelialized
by PWD (post wound day) 19 when wild type mice had completed this process
(Figure 4-1A). This resulted in a delayed wound re-epithelialization of
approximately one week. Full thickness dermal wounds were collected at
different time points (Figure 4-1A) and de novo regenerated hair follicles were
counted at PWD 19 (Figure 4-1B). C57BL/6 wild type mice exhibited an average

102

follicle count of 65 (+/-14), n = 5 follicles while Msx2
-/-
mice exhibited no follicles
(n = 5). Alkaline phosphatase staining was utilized to detect early growth of hair
follicles in the center of the wound. Wild type mice exhibited early de novo hair
bud growth while Msx2
-/-
mice demonstrated none (Figure 4-1C). Cytokeratin 17
(K17) immunostaining of wound epithelium was utilized to detect hair follicles.
Wild type mice exhibited K17 staining of new hair follicles in the center of the
wound at PWD 22 while Msx2
-/-
mice did not, even by PWD 41 (Figure 4-1D). Ito
(M. Ito, Yang, Andl, Cui, Kim, Millar, & Cotsarelis, 2007a) demonstrated that
regenerative wound healing took place in the center of the wound and not at the
wound margin. Our data suggest that Msx2 is required for de novo hair
regeneration in the center of the wound.

103

Figure 4-1 Altering Msx2 Levels Inhibits wound closure and de novo Hair
regeneration.
Msx2
-/-
exhibited a delayed wound closure rate compared to wild type (A). Msx2
-/-

exhibited less hair regeneration than wild type (B). Newly regenerating hair
follicles in skin stained for alkaline phosphatase activity (C) and KRT17 protein
(D). New follicles only formed in the center of the wound. Dashed yellow line
marks wound border. Green box marks wound center. Red box marks wound
margin. Green dots mark new hair follicles.

104

Figure 4-1 Continued:

105

The lack of new hair formation is not simply due to a delay in the
maturation of wound epidermis. Up to 90 days, there is still no new hair formation
in Msx2
-/-
mice. Furthermore, de novo hair regeneration was reported to be
reduced in aging mice (M. Ito, Yang, Andl, Cui, Kim, Millar, & Cotsarelis, 2007a).
However, we did not detect differences in the de novo hair regeneration between
Msx2
-/-
mice aged 3 - 4 months, 6 - 7 months, or 9 - 10 months.
Thus while Msx2 null mice showed accelerated wound repair when small
wounds were produced (Yeh et al., 2009), these mice showed a reduced ability
to regenerate hairs. Earlier we used in situ hybridization and Msx2 promoter
LacZ reporter mice to show that Msx2 is transiently induced in the wound margin
of the small wounds (Yeh et al., 2009). We wondered how Msx2 is expressed in
the large regenerative wounds?

4.3.2 Msx2 is expressed in the wound
The Msx2 is expressed in two phases in the large regenerative wounds:
peripherally along the wound margin at early times, followed by punctate
expression in the central wound bed at later times.
Msx2 promoter, LacZ reporter mice were utilized in the large wounds to
visualize the expression pattern of Msx2. Transgenic mice with 1.9kb Msx2
promoter linked to the -galactosidase was used (Brugger et al., 2004). This was

106

shown to be consistent in the small, 3mm skin wounds (Yeh et al., 2009). Full
thickness wounds were collected daily from PWD 13 - 18. Msx2 expression
became apparent at the wound edge approximately 12 hours after wounding.
This Msx2 expression was in the epithelia in the periphery during the early phase
of wound re-epithelialization (arbitrarily defined as PWD14 here) (Figure 4-2A).
At this stage, Msx2 expression was not observed in the center of the wound
epithelium on the dorsal side. Nor was Msx2 expression observed on the ventral
side of the wound epithelium where the epithelium contacts the dermal
mesenchyme (Figure 4-2C). In the late phase of wound re-epithelialization
(arbitrarily defined as after PWD 15 here), the Msx2 pattern changed and
became punctate throughout the wound at approximately PWD16. It maintains
this pattern at PWD17 - 19 (Figure 4-2B and D, and data not shown). We also
observed weaker Msx2 expression in the dermis of the central wound bed (data
not shown).

107

Figure 4-2 Msx2 promoter reporter mice suggested Msx2 is expressed at
the wound margin in early stage regeneration and punctate in late stage
regeneration.
Six representative samples during early stage de novo hair regeneration
demonstrate the Msx2 LacZ expression pattern (A). A time course study
demonstrates the Msx2 LacZ expression pattern changes to become punctate
during late stage de novo hair regeneration. A ventral view of PWD13 to 15
wound epithelial demonstrating the Msx2 LacZ expression pattern at the wound
margin. Histology of a late stage de novo wound exhibiting Msx2 LacZ
expression demonstrated a punctate pattern (D). (Dashed line marks wound
border. Green box marks wound center. Red box marks wound margin)

108

Figure 4-2 Continued:

109

The early phase of Msx2 expression is also observed in the small
reparative wounds, but the late phase of Msx2 was only observed in the large
regenerative wounds. Therefore, we think that the late phase of Msx2 expression
has more to do with de novo hair formation. The expression of Msx2 in the
dermis also prompted us to consider whether the epidermal or dermal Msx2
expression is more important to hair regeneration.

4.3.3 Msx2 expression in the epidermis is required for de novo hair
regeneration
In the large wounds, new hair germs are not derived from existing hair
bulge stem cells, but from adjacent suprabasal epidermal cells (M. Ito, Yang,
Andl, Cui, Kim, Millar, & Cotsarelis, 2007a). Therefore, the wound epithelium
probably generates new epidermal stem cells through some reprogramming
process. To determine if Msx2 expression was essential in the skin epidermis for
de novo hair regeneration, Tg(KRT14-cre) /B6.129S4-Gt(ROSA)26Sortm1Sor/J
mice were mated with Msx2
tm
mice to generate Krt14creROSAMsx2
tm
mice (n=6)
(Figure 4-3A). These mice exhibited a very similar skin and hair phenotype to
Msx2
-/-
mice including cyclic alopecia (Figure 4-3A). LacZ expression analysis
demonstrated cre activity was restricted to the basal epithelium of the skin organ
(Figure 4-3B). Large full thickness wounds were performed on these mice. The
Krt14creROSAMsx2
tm
mice also exhibited slower healing, they had not

110

completely re-epithelialized by PWD19 when wild type mice had completed this
process (Figure 4-1A). This resulted in a delayed wound re-epitheliazation of
approximately one week similar to the Msx2
-/-
mice (Figure 4-1A). The
Krt14creROSAMsx2
tm
mice exhibited no de novo hair regeneration (Figure 4-3C).
Therefore, in the large wounds the presence of Msx2 in the epidermis is essential
for de novo hair regeneration.

111

Figure 4-3 Epithelial specific reduction of Msx2 activity results in poor
tissue regeneration and suppression of de novo hair regeneration.
(A) Krt14creROSAMsx2
tm
mouse phenotype is very similar to Msx2
-/-
. (B) LacZ
staining pattern demonstrated epithelial specific expression. (C)
Krt14creROSAMsx2
tm
exhibited delayed wound closure when compared to
wildtype mice (Figure 4-1A) and did not exhibit de novo hair regeneration. (D)
H&E and skin differentiation gene expression patterns exhibit relatively normal
histology.

112

Figure 4-3 Continued:

113

4.3.4 The Mad/Smad factor binding site in the Msx2 promoter is important
for Msx2 expression during wounding
To study which domains of the Msx2 promoter are important for the
induction of Msx2 expression during the wound response, we used different LacZ
reporter mice. Three different promoter constructs were utilized: 1.9kb promoter,
partial length 560bp promoter, and a minimal 52bp promoter (Figure 4-4A). The
minimal 52bp promoter contains two evolutionary conserved protein factor
binding sites: binding site of Mad/Smad-related BMP signal transducers and a
consensus site for Antennapedia superclass homeodomain proteins (Brugger
2004). Mutations were introduced into either the binding site of Mad/Smad
factors or the binding site of homeodomain proteins. This resulted in a minimal
52bp promoter-LacZ reporter smad binding site mutant, and a minimal 52bp
promoter-LacZ reporter homeodomain protein binding site mutant (Figure 4-4A).
These mutant Msx2 promoter-LacZ reporter mice were utilized in the large
wound assay to determine which domain modulated the Msx2 expression
pattern. The Msx2 expression patterns observed between the three promoter
constructs were very similar. The 1.9kb and 560bp promoter constructs exhibited
a similar expression pattern and level. However, the 52bp promoter construct
exhibited a similar expression pattern but a lower level of expression. When
either domain of the 52bp promoter construct is mutated, Msx2 expression in the
wounds disappeared (Figure 4-4B). Thus, the binding sites of Mad/Smad-related

114

BMP signal transducers and for Antennapedia superclass homeodomain proteins
coordinately induce Msx2 expression during de novo hair regeneration.

115

Figure 4-4 The Mad/Smad factor and Homeodomain protein binding sites
regulate Msx2 expression during de novo hair regeneration.
(A) Three Msx2 promoter-LacZ reporter contructs were utlilized to determine
Msx2 promoter activity during de novo wound healing. (B) The Mad/Smad factor
binding site or the homeodomain protein binding site was mutated to generate
transgenic reporter mice. (C) Mice with the Msx2 52bp mutated Mad/Smad
binding site or a mutated Homeobox protein binding site driving LacZ expression
were subjected to full thickness wounds. Wounds were harvested and stained for
LacZ. When the Mad/Smad factor binding site or the homeodomain protein
binding site was mutated Msx2LacZ expression was suppressed during de novo
hair regeneration.

116

Figure 4-4 Continued:

117

4.3.5 Krt14noggin mice exhibited a mildly enhanced de novo hair
regeneration response.
To test whether BMP dependent regulation of Msx2 expression is indeed
important, we examined de novo hair regeneration with large wounds in
Krt14noggin mice. K14noggin mice have reduced BMP pathway activity at the
dermal-epidermal junction (M. Plikus et al., 2004). Large wounds on the dorsal
skin of Krt14noggin mice showed a similar wound closure rate as that seen in
wild type mice (Figure 4-5A). de novo hair regeneration in the center of the
wounds showed a slight increase in Krt14noggin (avg = 75 hairs (+/-12), n = 5)
over wild type (avg = 65 hairs +/-14, n = 5) mice (Figure 4-5B). Alkaline
phosphatase staining of wounds at PWD 22 showed that Krt14noggin mice
exhibited de novo hair regeneration in the center of the wound bed but not at the
margin, in a similar pattern as seen in wild type mice (Figure 5C). Thus reducing
BMP in the basal layer does not affect de novo hair regeneration to a large
extent. (Figure 4-5D) Immunostaining for KRT17 nicely highlights developing hair
follicles present in the center of the wounds in both Krt14noggin and control
mouse skin.

118

Figure 4-5 Inhibiting BMP in the basal layer does not affect de novo hair
regeneration. (A) Krt14noggin mice exhibited a similar wound closure rate
compared to wild type. (B) Krt14noggin mice exhibited comparable hair
regeneration as wild type. (C and D) Newly regenerating hair follicles in skin
stained for alkaline phosphatase activity (C) and KRT17 protein (D). (C and D)
New follicles only formed in the center of the wound. (Dashed yellow line marks
wound border. Green box marks wound center. Red box marks wound margin.
Green dots mark new hair follicles)

119

Figure 4-5 Continued:

120

4.4 Discussion
4.4.1 Wound healing verus tissue regeneration
The ability of adult mammals to regenerate following injury has
diminished. While axolotls can regenerate their limbs and tails, the regenerating
ability in adult mammals are very limited. Paradoxically, recent work shows that
large wounds in the skin lead to regeneration as shown by the de novo formation
of new hair follicles (M. Ito, Yang, Andl, Cui, Kim, Millar, & Cotsarelis, 2007a).
Since new hairs only occur in the center of the big wound, it implies that the
periphery of the wound undergoes the repair response, while the center of the
wound, topologically away from hypothetical repair signals produced by the
wound margin, are able to mount a regeneration response. Therefore repair and
regeneration could be in competition. To regenerate functional skin ectodermal
organs after wounding, differentiated cells have to be reprogrammed to become
competent to respond to inducing signals. These competent cells then can enter
morphogenesis and be organized into tissues. Any missing integral part of this
reprogramming leads to a failure of regeneration and tips the balance towards
repair. It would be very useful to learn the molecular pathways involved in
modulating the repair versus regeneration response.

We now can use small and large wounds to analyze repair and
regenerative wounds. Both are full thickness wounds inflicted on the dorsal skin

121

of the mice. Small wounds are defined here as smaller than 3mm. A large wound
is defined as 1.5cm X 1.5cm. After initial closure and wound contraction, they
should be still about 10 mm in diameter in order for hair regeneration to occur.
Small wounds are characterized by a repair type of response. The wound beds
are filled with granulation tissues, followed by re-epithelialization. Eventually, the
healed wound shows fibroblasts with an extracellular matrix, and a flat multi-layer
epidermis (Yeh et al., 2009). In the large wound, peripheral regions show a
similar repair type of response. In the center of the wound bed, morphogenesis is
able to take place and is similar to those observed in development (M. Ito, Yang,
Andl, Cui, Kim, Millar, & Cotsarelis, 2007a; Millar, 2002). Eventually, hair follicles
form in the central part of the wound. With these two models we can analyze the
different molecular components utilized by keratinocytes in the repair and
regenerative regions, and begin to study the molecular mechanisms involved.

Using Msx2
-/-
mice, we showed that their small wounds close faster than
the control (Yeh et al., 2009), implying a stronger repair response in the absence
of Msx2. In this work, we perform large wounds on Msx2
-/-
mice. We showed
genetic evidence that Msx2 is required for the successful de novo formation of
hair follicles. Analysis of the Msx2 promoter and expression showed that a mere
52 bp segment is sufficient to drive expression of Msx2 in the wound. Further,
using Krt14cre to specifically knock out Msx2 in the epidermis, we show that

122

epidermal Msx2 is essential for generating stem cells required for the formation
of new hair germs. Further analyses of the Msx2 promoter showed that both the
MAD/SMAD factor binding site and a homeobox protein binding site are required
to mediate the induction of Msx2 genes at wound sites. Since BMP signaling is
mediated by SMAD sites, we further tested if a reduction of BMP in Krt14noggin
mice will change the regenerative response. The results show only a modest
increase. This suggests that suppression of BMP or induction of a BMP
antagonist is not the only inducers for the expression of Msx2.

4.4.2 What events are downstream to Msx2?
Msx2 has been shown to be an important modulator in growth and
differentiation (Alappat, Zhang, & Chen, 2003). In the skin, we have shown that
adhesion molecules NCAM, integrin 1, and DCC in the epithelium are affected
by over-expression of Msx2 (Jiang et al., 1999). Msx2
-/-
mice can still form hairs
in embryonic development, but affect hair regeneration in two ways. One is that
they show delay of regeneration after hair plucking. The second is that Msx2
affects hair shaft differentiation (Ma, Liu, Wu, Plikus, Jiang, Bi, Liu, Muller-Rover,
Peters, Sundberg, Maxson, Maas, & Chuong, 2003b). Fox n1, Ha3, and Lef1 are
shown to be downstream to Msx2 in hair shaft differentiation, and these defects
may lead to the subsequent hair loss specifically at the onset of catagen, leading
to the dramatic cyclic alopecia phenotype. Furthermore, functionally null Msx2

123

mice exhibit skull ossification defects that resemble parietal formina (PFM)
(Satokata, Ma, Ohshima, Bei, Woo, Nishizawa, Maeda, Takano, Uchiyama,
Heaney, Peters, Tang, Maxson, & Maas, 2000b). Msx2 modulates bone
formation and mineralization (Liu et al., 1996; Satokata, Ma, Ohshima, Bei, Woo,
Nishizawa, Maeda, Takano, Uchiyama, Heaney, Peters, Tang, Maxson, & Maas,
2000b) by binding to Runx2, an inducer of osteocalcin, and inhibiting Runx2
activity (Shirakabe, Terasawa, Miyama, Shibuya, & Nishida, 2001). Similarly,
Msx2 and Twist participate in osteoblast proliferation and differentiation (Ishii et
al., 2003).
It is possible that the Msx family of transcription factors is modulating
chromatin reprogramming. Yoshizawa et al (Yoshizawa et al., 2004) showed that
Msx2 recruited HDAC1 to repress transcriptional activity and Lee et al (H. Lee,
Habas, & Abate-Shen, 2004) showed that Msx1 binds to histone protein 1b (H1b)
to inhibit transcription and myogenesis. The interaction of Msx proteins with
epigenetic modulators leads to the possibility of chromatin remodeling. This
chromatin remodeling is very important in determining a cellular state and is the
difference between a stem cell and a differentiated cell. The clear requirement of
Msx2 in de novo hair regeneration provides clues for future investigation on its
molecular mechanism.

124

4.4.3 What factors may induce Msx2 expression in wound sites?
The Msx2 promoter-LacZ reporter data demonstrate the Mad/Smad factor
binding and homeobox protein binding sites are required for Msx2 expression
during de novo hair regeneration. This suggests that a bone morphogenic protein
(BMP) and a homeobox protein might be forming a complex that is initiating
Msx2 expression. Msx2 in turn modulates the de novo process possibly by
reprogramming the cell to a more competent state. Interestingly, when Noggin is
over expressed in the basal layer of the epidermis as in Krt14noggin mice (Figure
4-5), the de novo process is unaffected and in contrast mildly enhanced. It could
be possible that Msx2 is regulated by a different member from the TGF-
superfamily during de novo hair regeneration. This might explain how the
Mad/Smad factor binding site is required for Msx2 expression and why
Krt14noggin mouse develop more hair. The BMP pathway Smads are 1, 5 and 8.
TGF- and activin pathway Smads are 2 and 3. Smad4 is common to all
pathways of the TGF- superfamily. Additional work will be required to decipher
the link from wound cytokine response to the reprogramming of epidermal stem
cells competent to generate new hairs.

125

Chapter 5: Conclusions and Perspectives
5.1 Conclusion
We conclude that a mechanism of hierarchical gene expression modulates
tissue competency. Perturbing histone acetylation led to the inhibition of feather
bud formation and perturbing histone methylation led to the disruption of feather
bud patterning (Figure 2-7). The disruption of HDAC1 activity leads to abnormal
skin morphogenesis (Figure 3-4). Finally, we showed genetic evidence that Msx2
is required for the successful de novo formation of hair follicles (Figure 4-1).
Epigenetic enzymes controlling histone acetylation (HDACs) and methylation
(HMTs), signaling molecules (Bmp and SHH), and transcription factors (Msx2)
controlling gene expression coordinately modulate tissue competence. This
competence can be modulated by perturbing this hierarchy at any of the levels.

5.2 Perpective
The goal is to understand how tissue morphogenesis occurs in hopes of
recapitulating the process in the clinical setting. The desire is to be able to
increase the competence of a tissue in order to tilt the balance from wound
healing to tissue regeneration. The problem is a lack of understanding the basis
of tissue competency. Since every normal somatic cell has identical genetic
content, then epigenetic mechanisms may play a role in tissue competence.

126

The number of studies addressing the role of epigenetic mechanisms in
tissue morphogenesis has increased dramatically over the past 5 years. Novel
transgenic mouse models utilizing conditional modulation of epigenetic genes
permits the study of previously embryonic lethal processes. There are significant
studies that have addressed the role of histone methylation in skin morhogenesis
(Ezhkova et al., 2009; Ezhkova et al., 2011). We have begun addressing the role
of histone acetylation by studying HDACs. The next logical progression to study
the role of histone acetylation in skin morphogenesis is histone
acetyltransferases (HATs). CBP and p300 have been show to be involved in
epithelial proliferation and differentiation in vitro (Teo, Ma, Nguyen, Lam, & Kahn,
2005). However, perturbing the activity of these HATs in skin specific roles during
morphogenesis in vivo has the potential to discover more knowledge of tissue
competence. Furthermore, histone phosphorylation and ubiquination remains to
be addressed in this theatre. All of these processes need to be studied to further
understand the basis of hierarchical gene modulation of tissue competence.

5.3 Future Directions
The paucity of known molecules involved in skin morphogenesis that are
modulated by perturbing specific genes is significant. Downstream genetic
identification is of upmost importance. This needs to be addressed in our studies
here. One methodology that shows promise in the ability to detect these

127

downstream pathways is bioinformatics. By utilizing chromatin
immunoprecipitation (ChIP) assays with DNA chip sequencing assays (ChiP
seq), investigators are increasing the knowledge base of downstream pathways.
This is a powerful tool that aids in the discovery of hierarchical gene expression
mechanisms. However, this technology is cost prohibitive to many investigators
and the analysis of such data is not standardized. Although these issues can be
inhibitory, as technologies advances, these studies will become more available.
After perturbing histone acetylation that leads to the inhibition of feather
bud formation, and perturbing histone methylation that leads to the disruption of
feather bud patterning, we must discover the downstream genes that are
responsible. The reduction of HDAC1 activity leads to abnormal skin
morphogenesis through downstream genes that are responsible for the abnormal
skin appendage phenotypes. The functionally null Msx2 mice exhibited no de
novo formation of hair follicles. Which specific genes are immediately
downstream of Msx2 that modulate the competence of the wound epithelium?
These questions, along with others, must be answered in order to understand
tissue competence.

128

Bibliography
Ahmad, W., Zlotogorski, A., Panteleyev, A. A., Lam, H., Ahmad, M., Faiyaz ul
Haque, M., et al. (1999). Genomic organization of the human hairless gene
(HR) and identification of a mutation underlying congenital atrichia in an arab
palestinian family. Genomics, 56(2), 141-148. doi:10.1006/geno.1998.5699
Alappat, S., Zhang, Z. Y., & Chen, Y. P. (2003). Msx homeobox gene family and
craniofacial development. Cell Research, 13(6), 429-442.
doi:10.1038/sj.cr.7290185
Alibardi, L. (2003). Adaptation to the land: The skin of reptiles in comparison to
that of amphibians and endotherm amniotes. Journal of Experimental
Zoology.Part B, Molecular and Developmental Evolution, 298(1), 12-41.
doi:10.1002/jez.b.24
Allan, C. H., Fleckman, P., Fernandes, R. J., Hager, B., James, J., Wisecarver,
Z., et al. (2006). Tissue response and Msx1 expression after human fetal digit
tip amputation in vitro. Wound Repair and Regeneration : Official Publication
of the Wound Healing Society [and] the European Tissue Repair Society,
14(4), 398-404. doi:10.1111/j.1743-6109.2006.00139.x
Alonso, L., & Fuchs, E. (2003). Stem cells in the skin: Waste not, wnt not. Genes
& Development, 17(10), 1189-1200. doi:10.1101/gad.1086903
Andl, T., Reddy, S. T., Gaddapara, T., & Millar, S. E. (2002). WNT signals are
required for the initiation of hair follicle development. Developmental Cell,
2(5), 643-653.
Balighi, K., Lajevardi, V., Moeineddin, F., Jelani, M., Tamizifar, B., Nikoo, A., et
al. (2009). A novel deletion mutation in the human hairless (HR) gene in an
iranian family with atrichia and papular lesions. Clinical and Experimental
Dermatology, 34(7), e498-500. doi:10.1111/j.1365-2230.2009.03578.x
Brockes, J. P., & Kumar, A. (2005). Appendage regeneration in adult vertebrates
and implications for regenerative medicine. Science (New York, N.Y.),
310(5756), 1919-1923.

129

Brugger, S. M., Merrill, A. E., Torres-Vazquez, J., Wu, N., Ting, M. C., Cho, J. Y.,
et al. (2004). A phylogenetically conserved cis-regulatory module in the Msx2
promoter is sufficient for BMP-dependent transcription in murine and
drosophila embryos. Development (Cambridge, England), 131(20), 5153-
5165. doi:10.1242/dev.01390
Brunmeir, R., Lagger, S., & Seiser, C. (2009). Histone deacetylase
HDAC1/HDAC2-controlled embryonic development and cell differentiation.
The International Journal of Developmental Biology, 53(2-3), 275-289.
doi:10.1387/ijdb.082649rb
Bullard, K. M., Longaker, M. T., & Lorenz, H. P. (2003). Fetal wound healing:
Current biology. World Journal of Surgery, 27(1), 54-61. doi:10.1007/s00268-
002-6737-2
Candi, E., Rufini, A., Terrinoni, A., Dinsdale, D., Ranalli, M., Paradisi, A., et al.
(2006). Differential roles of p63 isoforms in epidermal development: Selective
genetic complementation in p63 null mice. Cell Death and Differentiation,
13(6), 1037-1047. doi:10.1038/sj.cdd.4401926
Carlson, M. R., Bryant, S. V., & Gardiner, D. M. (1998). Expression of msx-2
during development, regeneration, and wound healing in axolotl limbs. The
Journal of Experimental Zoology, 282(6), 715-723.
Chen, G., Fernandez, J., Mische, S., & Courey, A. J. (1999). A functional
interaction between the histone deacetylase Rpd3 and the corepressor
groucho in drosophila development. Genes & Development, 13(17), 2218-
2230.
Chuong, C. -. (1998). Molecular basis of epithelial appendage morphogenesis.
Austin, TX: Landes Bioscience.
Chuong, C. M. (2007). Regenerative biology: New hair from healing wounds.
Nature, 447(7142), 265-266.
Chuong, C. M., & Edelman, G. M. (1985). Expression of cell-adhesion molecules
in embryonic induction. I. morphogenesis of nestling feathers. The Journal of
Cell Biology, 101(3), 1009-1026.

130

Chuong, C. M., Widelitz, R. B., Ting-Berreth, S., & Jiang, T. X. (1996). Early
events during avian skin appendage regeneration: Dependence on epithelial-
mesenchymal interaction and order of molecular reappearance. The Journal
of Investigative Dermatology, 107(4), 639-646.
Dhouailly, D. (1975). Formation of cutaceous appendages in dermo-epidermal
recombinaitons between reptiles, birds and mammals. Wilhelm Roux’ Arch.
Entwicklungsmech Org., 177, 323-340.
Driskell, R. R., Giangreco, A., Jensen, K. B., Mulder, K. W., & Watt, F. M. (2009).
Sox2-positive dermal papilla cells specify hair follicle type in mammalian
epidermis. Development (Cambridge, England), 136(16), 2815-2823.
doi:10.1242/dev.038620
Ezhkova, E., Lien, W. H., Stokes, N., Pasolli, H. A., Silva, J. M., & Fuchs, E.
(2011). EZH1 and EZH2 cogovern histone H3K27 trimethylation and are
essential for hair follicle homeostasis and wound repair. Genes &
Development, 25(5), 485-498. doi:10.1101/gad.2019811
Ezhkova, E., Pasolli, H. A., Parker, J. S., Stokes, N., Su, I. H., Hannon, G., et al.
(2009). Ezh2 orchestrates gene expression for the stepwise differentiation of
tissue-specific stem cells. Cell, 136(6), 1122-1135.
doi:10.1016/j.cell.2008.12.043
Fischle, W., Wang, Y., & Allis, C. D. (2003). Histone and chromatin cross-talk.
Current Opinion in Cell Biology, 15(2), 172-183.
Fuchs, E., & Horsley, V. (2008). More than one way to skin. Genes &
Development, 22(8), 976-985. doi:10.1101/gad.1645908
Greco, V., Chen, T., Rendl, M., Schober, M., Pasolli, H. A., Stokes, N., et al.
(2009). A two-step mechanism for stem cell activation during hair
regeneration. Cell Stem Cell, 4(2), 155-169. doi:10.1016/j.stem.2008.12.009
Guo, X., & Wang, X. F. (2009). Signaling cross-talk between TGF-beta/BMP and
other pathways. Cell Research, 19(1), 71-88. doi:10.1038/cr.2008.302
Gurtner, G. C., Callaghan, M. J., & Longaker, M. T. (2007). Progress and
potential for regenerative medicine. Annual Review of Medicine, 58, 299-312.
doi:10.1146/annurev.med.58.082405.095329

131

Haberland, M., Montgomery, R. L., & Olson, E. N. (2009). The many roles of
histone deacetylases in development and physiology: Implications for disease
and therapy. Nature Reviews.Genetics, 10(1), 32-42. doi:10.1038/nrg2485
Han, M., Yang, X., Farrington, J. E., & Muneoka, K. (2003). Digit regeneration is
regulated by Msx1 and BMP4 in fetal mice. Development (Cambridge,
England), 130(21), 5123-5132. doi:10.1242/dev.00710
Hartman, T. R., Nicolas, E., Klein-Szanto, A., Al-Saleem, T., Cash, T. P., Simon,
M. C., et al. (2009). The role of the birt-hogg-dube protein in mTOR activation
and renal tumorigenesis. Oncogene, 28(13), 1594-1604.
doi:10.1038/onc.2009.14
Hosokawa, R., Deng, X., Takamori, K., Xu, X., Urata, M., Bringas, P.,Jr, et al.
(2009). Epithelial-specific requirement of FGFR2 signaling during tooth and
palate development. Journal of Experimental Zoology.Part B, Molecular and
Developmental Evolution, 312B(4), 343-350. doi:10.1002/jez.b.21274
Huelsken, J., Vogel, R., Erdmann, B., Cotsarelis, G., & Birchmeier, W. (2001).
Beta-catenin controls hair follicle morphogenesis and stem cell differentiation
in the skin. Cell, 105(4), 533-545.
Ikegami, K., Ohgane, J., Tanaka, S., Yagi, S., & Shiota, K. (2009). Interplay
between DNA methylation, histone modification and chromatin remodeling in
stem cells and during development. The International Journal of
Developmental Biology, 53(2-3), 203-214. doi:10.1387/ijdb.082741ki
Illingworth, C. M. (1974). Trapped fingers and amputated finger tips in children.
Journal of Pediatric Surgery, 9, 853.
Ishii, M., Merrill, A. E., Chan, Y. S., Gitelman, I., Rice, D. P., Sucov, H. M., et al.
(2003). Msx2 and twist cooperatively control the development of the neural
crest-derived skeletogenic mesenchyme of the murine skull vault.
Development (Cambridge, England), 130(24), 6131-6142.
doi:10.1242/dev.00793
Ito, M., Liu, Y., Yang, Z., Nguyen, J., Liang, F., Morris, R. J., et al. (2005). Stem
cells in the hair follicle bulge contribute to wound repair but not to
homeostasis of the epidermis. Nature Medicine, 11(12), 1351-1354.

132

Ito, M., Yang, Z., Andl, T., Cui, C., Kim, N., Millar, S. E., et al. (2007a). Wnt-
dependent de novo hair follicle regeneration in adult mouse skin after
wounding. Nature, 447(7142), 316-320.
Ito, M., Yang, Z., Andl, T., Cui, C., Kim, N., Millar, S. E., et al. (2007b). Wnt-
dependent de novo hair follicle regeneration in adult mouse skin after
wounding. Nature, 447(7142), 316-320.
Ito, Y., Sarkar, P., Mi, Q., Wu, N., Bringas, P.,Jr, Liu, Y., et al. (2001).
Overexpression of Smad2 reveals its concerted action with Smad4 in
regulating TGF-beta-mediated epidermal homeostasis. Developmental
Biology, 236(1), 181-194. doi:10.1006/dbio.2001.0332
Jabs, E. W., Muller, U., Li, X., Ma, L., Luo, W., Haworth, I. S., et al. (1993). A
mutation in the homeodomain of the human MSX2 gene in a family affected
with autosomal dominant craniosynostosis. Cell, 75(3), 443-450.
Jiang, T. X., Liu, Y. H., Widelitz, R. B., Kundu, R. K., Maxson, R. E., & Chuong,
C. M. (1999). Epidermal dysplasia and abnormal hair follicles in transgenic
mice overexpressing homeobox gene MSX-2. The Journal of Investigative
Dermatology, 113(2), 230-237.
Jiang, T. X., Stott, S., Widelitz, R. B., & Chuong, C. M. (1998). Current methods
in the study of avian skin appendages. In C. M. Chuong (Ed.), Molecular
basis of epithelial appendage morphogenesis (pp. 359-408). Austin, TX:
Landes Bioscience.
Jones, P. H., Simons, B. D., & Watt, F. M. (2007). Sic transit gloria: Farewell to
the epidermal transit amplifying cell? Cell Stem Cell, 1(4), 371-381.
Jung, H. S., Francis-West, P. H., Widelitz, R. B., Jiang, T. X., Ting-Berreth, S.,
Tickle, C., et al. (1998). Local inhibitory action of BMPs and their relationships
with activators in feather formation: Implications for periodic patterning.
Developmental Biology, 196(1), 11-23.
Jung, H. S., Oropeza, V., & Thesleff, I. (1999). Shh, bmp-2, bmp-4 and fgf-8 are
associated with initiation and patterning of mouse tongue papillae.
Mechanisms of Development, 81(1-2), 179-182.
Kragl, M., Knapp, D., Nacu, E., Khattak, S., Maden, M., Epperlein, H. H., et al.
(2009). Cells keep a memory of their tissue origin during axolotl limb
regeneration. Nature, 460(7251), 60-65. doi:10.1038/nature08152

133

Kubicek, S., Schotta, G., Lachner, M., Sengupta, R., Kohlmaier, A., Perez-
Burgos, L., et al. (2006). The role of histone modifications in epigenetic
transitions during normal and perturbed development. Ernst Schering
Research Foundation Workshop, (57)(57), 1-27.
Lagger, G., O'Carroll, D., Rembold, M., Khier, H., Tischler, J., Weitzer, G., et al.
(2002). Essential function of histone deacetylase 1 in proliferation control and
CDK inhibitor repression. The EMBO Journal, 21(11), 2672-2681.
doi:10.1093/emboj/21.11.2672
Le Beyec, J., Xu, R., Lee, S. Y., Nelson, C. M., Rizki, A., Alcaraz, J., et al.
(2007). Cell shape regulates global histone acetylation in human mammary
epithelial cells. Experimental Cell Research, 313(14), 3066-3075.
doi:10.1016/j.yexcr.2007.04.022
LeBoeuf, M., Terrell, A., Trivedi, S., Sinha, S., Epstein, J. A., Olson, E. N., et al.
(2010). Hdac1 and Hdac2 act redundantly to control p63 and p53 functions in
epidermal progenitor cells. Developmental Cell, 19(6), 807-818.
doi:10.1016/j.devcel.2010.10.015
Lee, E. R., Murdoch, F. E., & Fritsch, M. K. (2007). High histone acetylation and
decreased polycomb repressive complex 2 member levels regulate gene
specific transcriptional changes during early embryonic stem cell
differentiation induced by retinoic acid. Stem Cells (Dayton, Ohio), 25(9),
2191-2199. doi:10.1634/stemcells.2007-0203
Lee, H., Habas, R., & Abate-Shen, C. (2004). MSX1 cooperates with histone H1b
for inhibition of transcription and myogenesis. Science (New York, N.Y.),
304(5677), 1675-1678. doi:10.1126/science.1098096
Levy, V., Lindon, C., Harfe, B. D., & Morgan, B. A. (2005). Distinct stem cell
populations regenerate the follicle and interfollicular epidermis.
Developmental Cell, 9(6), 855-861.
Li, L., Zheng, P., & Dean, J. (2010). Maternal control of early mouse
development. Development (Cambridge, England), 137(6), 859-870.
doi:10.1242/dev.039487
Liu, Y. H., Ma, L., Kundu, R., Ignelzi, M., Sangiorgi, F., Wu, L., et al. (1996).
Function of the Msx2 gene in the morphogenesis of the skull. Annals of the
New York Academy of Sciences, 785, 48-58.

134

Liu, Y. H., Ma, L., Wu, L. Y., Luo, W., Kundu, R., Sangiorgi, F., et al. (1994).
Regulation of the Msx2 homeobox gene during mouse embryogenesis: A
transgene with 439 bp of 5' flanking sequence is expressed exclusively in the
apical ectodermal ridge of the developing limb. Mechanisms of Development,
48(3), 187-197.
Longworth, M. S., Wilson, R., & Laimins, L. A. (2005). HPV31 E7 facilitates
replication by activating E2F2 transcription through its interaction with
HDACs. The EMBO Journal, 24(10), 1821-1830.
doi:10.1038/sj.emboj.7600651
Ma, L., Golden, S., Wu, L., & Maxson, R. (1996). The molecular basis of boston-
type craniosynostosis: The Pro148-->His mutation in the N-terminal arm of
the MSX2 homeodomain stabilizes DNA binding without altering nucleotide
sequence preferences. Human Molecular Genetics, 5(12), 1915-1920.
Ma, L., Liu, J., Wu, T., Plikus, M., Jiang, T. X., Bi, Q., et al. (2003a). 'Cyclic
alopecia' in Msx2 mutants: Defects in hair cycling and hair shaft
differentiation. Development (Cambridge, England), 130(2), 379-389.
Ma, L., Liu, J., Wu, T., Plikus, M., Jiang, T. X., Bi, Q., et al. (2003b). 'Cyclic
alopecia' in Msx2 mutants: Defects in hair cycling and hair shaft
differentiation. Development (Cambridge, England), 130(2), 379-389.
Michailidis, E., Theos, A., Zlotogorski, A., Martinez-Mir, A., & Christiano, A. M.
(2007). Atrichia with papular lesions resulting from novel compound
heterozygous mutations in the human hairless gene. Pediatric Dermatology,
24(5), E79-82. doi:10.1111/j.1525-1470.2007.00448.x
Millar, S. E. (2002). Molecular mechanisms regulating hair follicle development.
The Journal of Investigative Dermatology, 118(2), 216-225.
Mills, A. A., Zheng, B., Wang, X. J., Vogel, H., Roop, D. R., & Bradley, A. (1999).
P63 is a P53 homologue required for limb and epidermal morphogenesis.
Nature, 398(6729), 708-713. doi:10.1038/19531
Montgomery, R. L., Davis, C. A., Potthoff, M. J., Haberland, M., Fielitz, J., Qi, X.,
et al. (2007). Histone deacetylases 1 and 2 redundantly regulate cardiac
morphogenesis, growth, and contractility. Genes & Development, 21(14),
1790-1802. doi:10.1101/gad.1563807

135

Muneoka, K., Allan, C. H., Yang, X., Lee, J., & Han, M. (2008). Mammalian
regeneration and regenerative medicine. Birth Defects Research.Part C,
Embryo Today : Reviews, 84(4), 265-280. doi:10.1002/bdrc.20137
Muneoka, K., Han, M., & Gardiner, D. M. (2008). Regrowing human limbs.
Scientific American, 298(4), 56-63.
Nowak, J. A., Polak, L., Pasolli, H. A., & Fuchs, E. (2008). Hair follicle stem cells
are specified and function in early skin morphogenesis. Cell Stem Cell, 3(1),
33-43.
Olson, L. E., Tollkuhn, J., Scafoglio, C., Krones, A., Zhang, J., Ohgi, K. A., et al.
(2006). Homeodomain-mediated beta-catenin-dependent switching events
dictate cell-lineage determination. Cell, 125(3), 593-605.
doi:10.1016/j.cell.2006.02.046
Panteleyev, A. A., Ahmad, W., Malashenko, A. M., Ignatieva, E. L., Paus, R.,
Sundberg, J. P., et al. (1998). Molecular basis for the rhino yurlovo (hr(rhY))
phenotype: Severe skin abnormalities and female reproductive defects
associated with an insertion in the hairless gene. Experimental Dermatology,
7(5), 281-288.
Panteleyev, A. A., Paus, R., Ahmad, W., Sundberg, J. P., & Christiano, A. M.
(1998). Molecular and functional aspects of the hairless (hr) gene in
laboratory rodents and humans. Experimental Dermatology, 7(5), 249-267.
Panteleyev, A. A., van der Veen, C., Rosenbach, T., Muller-Rover, S., Sokolov,
V. E., & Paus, R. (1998). Towards defining the pathogenesis of the hairless
phenotype. The Journal of Investigative Dermatology, 110(6), 902-907.
doi:10.1046/j.1523-1747.1998.00219.x
Patel, M., & Yang, S. (2010). Advances in reprogramming somatic cells to
induced pluripotent stem cells. Stem Cell Reviews, 6(3), 367-380.
doi:10.1007/s12015-010-9123-8
Plikus, M., Wang, W. P., Liu, J., Wang, X., Jiang, T. X., & Chuong, C. M. (2004).
Morpho-regulation of ectodermal organs: Integument pathology and
phenotypic variations in K14-noggin engineered mice through modulation of
bone morphogenic protein pathway. The American Journal of Pathology,
164(3), 1099-1114.

136

Plikus, M. V., Widelitz, R. B., Maxson, R., & Chuong, C. M. (2009). Analyses of
regenerative wave patterns in adult hair follicle populations reveal macro-
environmental regulation of stem cell activity. The International Journal of
Developmental Biology, 53(5-6), 857-868. doi:10.1387/ijdb.072564mp
Pott, P. (1775). Chirurgical observations relative to the cataract, the polypus of
the nose, the cancer of the scrotum, the different kinds of ruptures and the
mortification of the toes and feet. Hawes: London, , 1-208.
Potter, G. B., Beaudoin, G. M.,3rd, DeRenzo, C. L., Zarach, J. M., Chen, S. H., &
Thompson, C. C. (2001). The hairless gene mutated in congenital hair loss
disorders encodes a novel nuclear receptor corepressor. Genes &
Development, 15(20), 2687-2701. doi:10.1101/gad.916701
Potter, G. B., Zarach, J. M., Sisk, J. M., & Thompson, C. C. (2002). The thyroid
hormone-regulated corepressor hairless associates with histone deacetylases
in neonatal rat brain. Molecular Endocrinology (Baltimore, Md.), 16(11), 2547-
2560.
Satokata, I., Ma, L., Ohshima, H., Bei, M., Woo, I., Nishizawa, K., et al. (2000a).
Msx2 deficiency in mice causes pleiotropic defects in bone growth and
ectodermal organ formation. Nature Genetics, 24(4), 391-395.
Satokata, I., Ma, L., Ohshima, H., Bei, M., Woo, I., Nishizawa, K., et al. (2000b).
Msx2 deficiency in mice causes pleiotropic defects in bone growth and
ectodermal organ formation. Nature Genetics, 24(4), 391-395.
doi:10.1038/74231
Schmidt-Ullrich, R., Tobin, D. J., Lenhard, D., Schneider, P., Paus, R., &
Scheidereit, C. (2006). NF-kappaB transmits eda A1/EdaR signalling to
activate shh and cyclin D1 expression, and controls post-initiation hair
placode down growth. Development (Cambridge, England), 133(6), 1045-
1057. doi:10.1242/dev.02278
Schmittwolf, C., Kirchhof, N., Jauch, A., Durr, M., Harder, F., Zenke, M., et al.
(2005). In vivo haematopoietic activity is induced in neurosphere cells by
chromatin-modifying agents. The EMBO Journal, 24(3), 554-566.
doi:10.1038/sj.emboj.7600546
Sen, G. L., Reuter, J. A., Webster, D. E., Zhu, L., & Khavari, P. A. (2010).
DNMT1 maintains progenitor function in self-renewing somatic tissue. Nature,
463(7280), 563-567. doi:10.1038/nature08683

137

Shao, Y., Lu, J., Cheng, C., Cui, L., Zhang, G., & Huang, B. (2007). Reversible
histone acetylation involved in transcriptional regulation of WT1 gene. Acta
Biochimica Et Biophysica Sinica, 39(12), 931-938.
Shirakabe, K., Terasawa, K., Miyama, K., Shibuya, H., & Nishida, E. (2001).
Regulation of the activity of the transcription factor Runx2 by two homeobox
proteins, Msx2 and Dlx5. Genes to Cells : Devoted to Molecular & Cellular
Mechanisms, 6(10), 851-856.
Stelnicki, E. J., Komuves, L. G., Holmes, D., Clavin, W., Harrison, M. R., Adzick,
N. S., et al. (1997). The human homeobox genes MSX-1, MSX-2, and MOX-1
are differentially expressed in the dermis and epidermis in fetal and adult skin.
Differentiation; Research in Biological Diversity, 62(1), 33-41.
doi:10.1046/j.1432-0436.1997.6210033.x
Stenn, K. S., & Paus, R. (2001). Controls of hair follicle cycling. Physiological
Reviews, 81(1), 449-494.
Takahashi, K., Tanabe, K., Ohnuki, M., Narita, M., Ichisaka, T., Tomoda, K., et al.
(2007). Induction of pluripotent stem cells from adult human fibroblasts by
defined factors. Cell, 131(5), 861-872.
Takahashi, K., & Yamanaka, S. (2006). Induction of pluripotent stem cells from
mouse embryonic and adult fibroblast cultures by defined factors. Cell,
126(4), 663-676.
Tanaka, E. M. (2003). Cell differentiation and cell fate during urodele tail and limb
regeneration. Current Opinion in Genetics & Development, 13(5), 497-501.
Teo, J. L., Ma, H., Nguyen, C., Lam, C., & Kahn, M. (2005). Specific inhibition of
CBP/beta-catenin interaction rescues defects in neuronal differentiation
caused by a presenilin-1 mutation. Proceedings of the National Academy of
Sciences of the United States of America, 102(34), 12171-12176.
doi:10.1073/pnas.0504600102
Tou, L., Liu, Q., & Shivdasani, R. A. (2004). Regulation of mammalian epithelial
differentiation and intestine development by class I histone deacetylases.
Molecular and Cellular Biology, 24(8), 3132-3139.
Vierbuchen, T., Ostermeier, A., Pang, Z. P., Kokubu, Y., Sudhof, T. C., & Wernig,
M. (2010). Direct conversion of fibroblasts to functional neurons by defined
factors. Nature, doi:10.1038/nature08797

138

Wang, X., & Seed, B. (2003). A PCR primer bank for quantitative gene
expression analysis. Nucleic Acids Research, 31(24), e154.
Widelitz, R. B., Jiang, T. X., Noveen, A., Chen, C. W., & Chuong, C. M. (1996).
FGF induces new feather buds from developing avian skin. The Journal of
Investigative Dermatology, 107(6), 797-803.
Wu, M. Y., & Hill, C. S. (2009). Tgf-beta superfamily signaling in embryonic
development and homeostasis. Developmental Cell, 16(3), 329-343.
doi:10.1016/j.devcel.2009.02.012
Yeh, J., Green, L. M., Jiang, T. X., Plikus, M., Huang, E., Chang, R. N., et al.
(2009). Accelerated closure of skin wounds in mice deficient in the homeobox
gene Msx2. Wound Repair and Regeneration : Official Publication of the
Wound Healing Society [and] the European Tissue Repair Society, 17(5),
639-648. doi:10.1111/j.1524-475X.2009.00535.x
Yokoyama, H. (2008). Initiation of limb regeneration: The critical steps for
regenerative capacity. Development, Growth & Differentiation, 50(1), 13-22.
doi:10.1111/j.1440-169X.2007.00973.x
Yoshizawa, T., Takizawa, F., Iizawa, F., Ishibashi, O., Kawashima, H., Matsuda,
A., et al. (2004). Homeobox protein MSX2 acts as a molecular defense
mechanism for preventing ossification in ligament fibroblasts. Molecular and
Cellular Biology, 24(8), 3460-3472.
Yu, L., Han, M., Yan, M., Lee, E. C., Lee, J., & Muneoka, K. (2010). BMP
signaling induces digit regeneration in neonatal mice. Development
(Cambridge, England), 137(4), 551-559. doi:10.1242/dev.042424
Zhang, J. S., Wang, L., Huang, H., Nelson, M., & Smith, D. I. (2001). Keratin 23
(K23), a novel acidic keratin, is highly induced by histone deacetylase
inhibitors during differentiation of pancreatic cancer cells. Genes,
Chromosomes & Cancer, 30(2), 123-135.
Zhou, Q., Brown, J., Kanarek, A., Rajagopal, J., & Melton, D. A. (2008). In vivo
reprogramming of adult pancreatic exocrine cells to beta-cells. Nature,
455(7213), 627-632.

Abstract (if available)

Abstract The ability to engineer a tissue to replace diseased organs is a goal that when achieved will provide an incredible asset to clinicians. In order to engineer a tissue, one must understand how the specific tissue develops, maintains homeostasis, and regenerates. To gain insight into this process, we utilized the embryonic chicken skin organ (CSO) model to study the effect of histone acetylation and methylation on feather bud development. In the CSO assay, Chaetocin, a histone methyltransferase inhibitor, acts upstream of FGF and BMP to promote feather bud growth. MS275, a histone deacetylase inhibitor, acts upstream of FGF and BMP to inhibit feather bud growth. Next, we employed a tissue specific transgenic mouse model to investigate the effect of HDAC1 on mouse skin morphogenesis. Immunostaining of wild type C57BL/6 skin showed HDAC1 is predominantly located in the epidermal basal layer, sebaceous glands, inner and outer root sheaths, and matrix of pelage and vibrissae hair follicles. K14creHDACᶠˡ/ᶠˡ mice exhibited a reduction in size and weight, hyperkeratosis, alopecia, dermal cysts, hyperpigmentation, sebaceous hyperplasia and nail pathologies including

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Molecular aspects of skin early morphogenesis

PDF

Exploring the molecular and cellular underpinnings of organ polarization using feather as the model system

PDF

Canonical and non-canonical Wnt signaling in the patterning of multipotent stem cells during feather development

PDF

The role of insulin-like growth factors in size regulation during embryonic feather development

PDF

Tools to study the epicardium's response during cardiac regeneration

PDF

Macroenvironmental regulation of hair follicle stem cells

PDF

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

Asset Metadata

Creator Hughes, Michael Warren (author)

Core Title Epigenetic and genetic reprogramming during embryonic chicken feather bud morphogenesis, hair morphogenesis, and de novo hair regeneration

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Pathobiology

Publication Date 04/26/2012

Defense Date 02/19/2012

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

Tag competence,de novo,embryonic,epigenetic,feather bud,hair,histone,morphogenesis,OAI-PMH Harvest,Regeneration,reprogramming,stem cell

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Chuong, Cheng-Ming (committee chair), Maxson, Robert E., Jr. (committee member), Rice, Judd C. (committee member), Widelitz, Randall B. (committee member)

Creator Email mwhughes@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-14427

Unique identifier UC11289254

Identifier usctheses-c3-14427 (legacy record id)

Legacy Identifier etd-HughesMich-661.pdf

Dmrecord 14427

Document Type Dissertation

Rights Hughes, Michael Warren

Type texts

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

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

Repository Name University of Southern California Digital Library

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