University of Southern California Dissertations and Theses

Implications of constitutively active bone morphogenetic protein (BMP) signaling in skin morphogenesis and skin postnatal homeostasis

(USC Thesis Other)

Implications of constitutively active bone morphogenetic protein (BMP) signaling in skin morphogenesis and skin postnatal homeostasis

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content IMPLICATIONS OF CONSTITUTIVELY ACTIVE BONE MORPHOGENETIC
PROTEIN (BMP) SIGNALING IN SKIN MORPHOGENESIS AND SKIN
POSTNATAL HOMEOSTASIS

by

Andrew W. Hennigan

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(EXPERIMENTAL AND MOLECULAR PATHOLOGY)

December 2011

Copyright 2011 Andrew W. Hennigan
ii

Acknowledgements

Foremost among the many people that deserve acknowledgement is my mentor Dr.
Krzysztof Kobielak. Always willing to discuss the research or ways to approach to
problems, Kris provided the support for not only this project but also all my endeavors
the last two years. For this I am so grateful.
My committee: Dr. Randy Widelitz, Dr. Cheng-Ming Choung, and Dr. Agnieszka
Kobielak. Their insight and feedback over the last two years was invaluable in shaping
and refining my research.
All my wonderful colleagues who taught me, helped me, and suportd me: Dr. Eve
Kandyba, Yvonne Leung, Christine Cao, Dr. Hanwei Zhang, Marie Rippen, and Leila
Sheikholeslam

iii

Table of Contents

Acknowledgements ii
List of Figures iv
Abstract v
Methods 1
Introduction 5
Results 18
Chapter 1: Strong Phenotype 18
Chapter 2: Medium Phenotype 24
Chapter 3: Progressive Phenotype 25
Discussion 35
Future Directions 41
Bibliography 43

iv

List of Figures

Figure 1: Canonical Bone Morphogenetic Protein Signaling Pathway 6
Figure 2: Epidermal Morphogenesis: Asymmetric and Symmetric Division 7
Figure 3: Hair Follicle Morphogeneis is a Complex Panoply of Singals 9
Figure 4: Hair Follicle Morphogenesis Transitions to Homeostasis 11
Figure 5: BMP Inhibitor Noggin Plays Role in Morphogenesis and Homeostasis 14
Figure 6: The Role of BMP in Hair Cycling and Homeostasis 15
Figure 7: Generation of K14Alk3 Q233D-HA mouse 20

Figure 8: K14Alk3 Strong and Medium Phenotypes: Impaired Hair Follicle
Morphogenesis 23

Figure 9: K14Alk4 Progressive Phenotype 26
Figure 10: K14Alk3 Progressive Phenotype: Loss of Hair but Maintenance of
Differentiation 29

Figure 11: Increase of the a6 high/CD34 high Hair Follicle Stem Cell Population in
K14Alk3 Progressive Phenotype 32

Figure 12: In Vitro Analysis of a6 high/CD34 high Hair Follicle Stem Cell Population 34
Figure 13: Model of the Multiple Roles Constitutively Active BMP Signaling Plays in
Morphogenesis and Homeostasis of Epidermis and Epidermal Appendages 40

v

Abstract
The skin and its array of epidermal appendages with their complex 3-dimensional
structure provide an excellent model in which to study morphogenesis, pattern formation,
and homeostasis. The biochemical signals that govern these biological processes continue
to be unraveled. Bone Morphogenetic Proteins (BMP) play an integral role in both
morphogenesis and homeostasis of the epidermis, however the precise details of this role
remains a source of inquiry.
To further study the role of BMP signaling in the epidermis and its appendages both in
morphogenesis and homeostasis we have generated transgenic mice harboring a
constitutively active form of the BMPR1A receptor under a Keratin 14 (K14) promoter.
This will drive expression of the transgene in the basal layer of the epidermis as well as
the Outer Root Sheath (ORS) of the hair follicle.
At strong levels of transgene expression hair follicle morphogenesis is markedly
impaired. Epidermal morphogenesis appeared normal, though the balance of
differentiation and proliferation was perturbed. Interestingly, in the animals which
expressed low levels of the K14Alk3 transgene, there was a progressive loss hair with
age. This loss of hair follicle homeostasis implicates BMP as not only a crucial player in
morphogenesis, but important in finely balancing proliferation and quiescence in the hair
follicle stem cells.
1

Methods
Model Generation
Transgenic mice for BMP gain-of-function studies, were engineered using CD1 mice.
Mice were generated to express a constitutively active form of human BMPR receptor 1A
(hALK3Q233D-HA, under the control of a human keratin 14 promoter (Vassar,
Rosenberg et al. 1989). For the transgene construct, a constitutively active form of human
BMPR receptor 1A was fused on a C-terminal part to human influenza hemagglutinin
epitope tag (HA). The K14-BG (beta-globin bovine polyA) expression cassette was used
for construction of K14– hALK3Q233D-HA. (Gat, DasGupta et al. 1998) The HA-tagged
hALK3Q233D-HA cDNA was cut and subcloned from original construct provided by Dr.
Rosenblum's Lab by using a Not restriction enzyme fragment (~3.4kb) ligated into the
Not (sticky ends)-cut K14-BG cassette (Hu, Piscione et al. 2003).
The linear construct of hALK3Q233D-HA, driven by keratin 14 (K14) was prepared by
SacI and SphI and after purification on low melting 1% agarose gel was used to generate
transgenic animals. Offspring of the founding mice yielded litters in the expected
Mendelian ratios as confirmed by both phenotype and genotyping.
Mice with strong expression of the Alk3Q233D-HA were significantly smaller in size
compared to their control littermates and died in the first couple days after birth, while
less severe or medium phenotype animals failed to survive beyond 4 weeks. Finally
transgenic animals with the weak and viable phenotype, these were similar in size to their
control littermates and allow to establish the transgenic line. These transgenic animals
2

were fertile and able to reproduce, but lost their hair progressively after approximately 3-
4 months.
PCR
Mice were screened for transgene genomic integration by polymerase chain reaction
(PCR) using primers which span the human Keratin 14 promoter and Alk3 transgene.
Forward: 5'-CCTACAGCTCCTGGGCAACGTG-3' and Reverse: 5'-
CCCAGTGCCATGAAGCATACT3'. Tail genomic DNA was extracted by lysis using
Protinase K (10mg/mL) at 56 degrees C. PCR products were amplified in separate
reactions using three-stage PCR program: 94 C for 4 minutes, 94 C for 30 seconds, 57 C
for 45 seconds, 72 C for 60 seconds (35 cycles), 72 C for 5 minutes.
Western Blot
6% SDS gel was prepared and samples were applied with loading buffer after 10 mins at
100º C. Samples were run at 90V and transferred to membrane at 75V for 1.5 hours at
room temperature. Membrane was blocked with 5% milk in TBS for 30 mins. After 1h at
room temperature (RT), primary antibody HA (rat, 1:500; Roche) was added for
overnight (O/N) incubation in a cold room at 4
o
C. Following washingin TBS-T,
secondary antibody, HRP anti Rat (1:5000, Sigma) was applied for 1h at room
temperature. The blot was then visualized by SuperSignal West Dura Chemiluminescent
Substrate (Thermo).
IHC
Tissue samples were embedded in OCT, frozen and sectioned at 12 um. The sections
were fixed in 4% paraformaldhyde for 10 minutes and subject to immunohistochemistry.
3

The sections were then permealized in 0.1% Triton X-100 in PBS for 10 minutes. They
were blocked in 5% HI-NGS, 1% BSA in PBS-T for 1 hour at room temperature. The
primary antibody was incubated overnight at 4 C 1% BSA in PBS. Secondary antibodies
were incubated for 1 hour at room temperature in PBS. The slides were stained with
DAPI (1:2000) and closed with 80% glycerol. When appropriate the MOM kit (Vector
Laboratories) was used.
Antibodies used: CD34 (eBioscince, 1:100), CD104 (BD Pharmingen, 1:300), K1 (gift
Ab, Fuchs E. Lab, NY 1:200), K5 (gift Ab, Fuchs E. Lab, NY 1:200), K15 (Thermo,
1:100), pSMAD1/5/8 (Cell Sginaling, 1:50), E-Cadherin (gift Ab, Fuchs E. Lab, NY
1:100), HA rat, 1:50; Roche)
Histology
Frozen tissue samples were sectioned at 12 µm and fixed for 10 minutes with 4%
paraformaldehyde in PBS. Standard Hematoxylin and Eosin (H&E) was performed for
basic histological analysis.
Frozen sections (12µm) were used for Oil red Staining. After air drying, samples were
fixed in 10% formalin. Following fixation, samples were treated with 100% propylene
glycol and then stained with Oil Red Staining solution (0.5g Oil Red O in 100 mL
propylene glycol) for 10minutes at 60C. After differentiation in 85% propylene glycol,
the sections were counterstained with Hematoxylin for 30 seconds. After washing the
sections were mounted in 80% glycerol.
Frozen sections (12um) were used for Alkaline Phosphatase (AP) staining. Following
fixation in 4% paraformaldehyde and wash, the sections were stained with Alkaline
4

Phosphatase stain solution (Takara) for 1 hour at 37 C. The sections were then
counterstained with 1% methyl green for 10 minutes and mounted in 80% glycerol.
FACS analysis
Each experiment consisted of scraping the back skin of an adult K14Alk3 mouse with a
scalpel removing the fat and fascia. The back skins were then placed dermis down in
0.25% trypsin (Gibco) overnight at 4 C. The next day the epidermis was scraped off the
dermis , neutralized with 0.05 mMCa media, and strained through 70 µm and 40 µm
pores (BD Biosciences). The cells were then resuspended in 1% Chelex in PBS and
stained with primary antibodies coupled to fluorophores.
The primary antibodies for FACS analysis were anti-α6 integrin (CD49f, BD
Pharminigen (1:200) coupled to PE and anti-CD34 (1:50) coupled to Alexa-700
(eBiosciences). Cells were gated for viablity and single events and then sorted based on
α6-integrin and CD34 expression (using FACSARIA).
Cell Culture:
FACS-isolated a6 high/CD34 high cells were plated on mitomycin-treated (8µg/mL) 3T3
fibroblasts in E-media supplemented with 0.3 mM Ca on 10cm cell culture plates
(Gernier). These cells were cultured in 5% CO
2
and 37º until confluence at which point
they were trypsinized using 0.05% trypsin (Gibco). Cells were then re-plated in treated
3T3 feeder. At 3 passages, cells were trypsinized and plated without feeder. The media
was correspondingly changed to 0.05mM low Ca
2+
media.

5

Introduction
The skin is one of the largest organs of the body and plays a crucial in creating and
maintaining an internal milieu in which an organism can carry out complex biochemical
reactions necessary for survival. Placed at the boundary of our body and the
environment, the skin is constantly bombarded by heat, cold, UV, chemical and
mechanical insult. The stratified epithelium that composes skin is dynamic; constantly
replacing dead and damaged cells in order to maintain its protective barrier function.
Additionally, the skin consists of a number of specialized appendages like sweat glands,
sebaceous glands, and hair follicles that supplement its function of maintaining
homeostasis. The skin and each of these appendages must contain a population of
pluripotent cells that allow regeneration through life and maintain a functional
epithelium. Furthermore, the great diversity of these appendages in nature: skin, scale,
nails, feather, and hair underlie the complexity of skins pattern formation and
morphogenesis. There are a multitude of signals and pathways which give rise to the
diversity of epidermal morphogenesis. Many aspects of these pathways remain to be
uncovered including those of Bone Morphogenetic Protein (BMP), which is studied here.

BMP signaling is the largest member of the transforming growth factor (TGF)-β
superfamily and encompasses a number of separate ligands. Though first isolated from
bone, BMP proteins play multifunctional roles in development and homeostasis. The
BMP family consists of 20 proteins synthesized as pro-peptides(Botchkarev 2003;
Botchkarev and Sharov 2004). Upon cleavage, the monomers hetro- or homodimerize
6

and are secreted into the extracellular matrix (ECM). Once biologically active, BMP
proteins interact with a family of BMP receptors which complex with the ligands. A type
I receptor (BMPR1A or BMPR1B) hetrodimerizes with a type II receptor (BMPR-II) and
subsequently the type II receptor phophorylates the type I. In the canonical BMP
pathway, the type I receptor phosphorylates a SMAD protein, like SMAD 1, 5, or 8.
These activated, phospho-SMADs proteins interact with a Co-SMAD (common partner
SMAD) protein, then after complexing, this heterodimers, translocate to the nucleus.
There the heterocomplex(es) interact with genes regulatory elements in promoter
or enhancer regions to activate or repress target genes. Additionally, there is a group of
secreted BMP inhibitors which serve to modulate BMP signaling. These proteins include
Noggin, Chordin, and Follistatin (Botchkarev 2003).

Figure 1. Canonical Bone Morphogenetic Protein Signaling Pathway
7

Epidermal and Follicular Morphogenesis
During gastrulation the skin arises from the ectoderm as a function of BMP signaling.
BMP4 strongly induces an epidermal fate while inhibiting a neuronal fate in the ectoderm
(Wilson, Lagna et al. 1997). The now single-layer of multipotent epidermal cells persists
until E12.5 at which point signals from the mesenchyme induce stratification. As
stratification progresses in the embryonic epidermis, the basal layer of keratinocytes
increasingly divides asymmetrically. This new suprabasal layer begins to express Keratin
1(K1) and forms the first layer of stratification (Lechler and Fuchs 2005). In conjunction
with the asymmetric divisions mediated by spindle orientation, Notch signaling promotes
the basal to suprabasal switch in differentiation (Blanpain, Lowry et al. 2006; Blanpain,
Horsley et al. 2007; Williams, Beronja et al. 2011).

Figure 2. Epidermal Morphogenesis: Asymmetric and Symmetric Division: During
epidermal morphogenesis basal keratinocytes first divide symmetrically, parallel to the
basement membrane. When the epidermis begins stratification and differentiation, this
division becomes asymmetric. (Adapted from Lechler 2005, Fuchs 2009)

8

The hair follicle is a dynamic and complex structure that goes through cyclic phases of
growth (anagen), degeneration (catagen), and resting (telogen) throughout life. It is the
only such appendage of the body to behave in such a way physiologically. More so, these
cycles are well established and characterized (Muller-Rover, Handjiski et al. 2001). This
fact combined with it location just below the surface of the skin and its complex 3-
demensional structure makes the hair follicle an extremely informative model to study
tissue morphogenesis and homeostasis.
Like other epithelial appendages the morphogenesis of the hair follicle involves the
complex interaction of both the epithelium and mesenchyme. During initiation of hair
follicle development the single layer of multipotent epithelia progenitor cells grow
downward (invaginate) into the dermis. This down growth requires intricate cellular
reorganization involving changes in cell-cell junctions and transcriptional program (Rhee,
Polak et al. 2006). These changes occur as a result of two different signals: Wnt
signaling to stabilize β-catenin and Noggin to inhibit BMP signaling the synergy of
which produces the transcription factor Lef1. Upon receiving the two signals, the
stabilized β-catenin interacts with Lef1 to downregulate E-cadherin expression and thus
altering cell-cell adhesion at the placode formation (Jamora, DasGupta et al. 2003).

9

Figure 3. Hair Follicle Morphogeneis is a Complex Panoply of Singals: Hair follicle
morphogenesis involves a complex repertoire of signals derived from both the epidermis
and mesenchyme. These signals include β-catenin,WNTs, BMP, Noggin, Sonic
Hedgehog(Shh), and others. (adapted from Millar, SE 2002; Jamora, C 200)
Previous work had demonstrated the requirement of Wnt signaling for hair follicle
morphogenesis. When Dickkopf-related protein 1 (Dkk1), a potent inhibitor of Wnt
signaling, was ectopically expressed in the skin under K14 promotion there was a failure
of placode formation However this loss of Wnt signaling does not affect the
differentiation of the epidermis (Andl, Reddy et al. 2002; Millar 2002). Indeed when β-
catenin, a downstream target of Wnt signaling, is expressed in the epidermis in a
truncated, stabilized form which mimics physiological activation, the skin undergoes de
novo hair follicle morphogenesis (Gat, DasGupta et al. 1998).
10

Botchkarev and colleagues had demonstrated that Noggin stimulated hair follicle
induction and originated in the mesenchyme, specifically the dermal papilla (DP), a
signaling center of highly specialized fibroblasts. Noggin inhibits BMP4 and this
interaction not only influences hair follicle induction but also the proliferation and
differentiation of the epidermal keratinocytes (Botchkarev, Botchkareva et al. 1999).
Indeed subsequent work showed that Noggin is in fact necessary for the induction of
secondary (Nontylotrich) hair follicles. Deletion of Noggin was associated with increased
levels of BMP2 and BMP4 as well as a downregulation of genes which specify epidermal
cell fates including β-catenin, Lef1, and Shh (Chiang, Swan et al. 1999). Additionally an
early marker of dermal papilla, p75NTR, was lost (Botchkarev, Botchkareva et al. 2001).
Not only is Noggin, and thus BMP inhibition, required for proper morphogenesis of the
hair follicle, but in postnatal skin the switch from resting to growth (telogen to anagen)
requires the downregulation or BMP signaling and is mediated by increased levels of
Noggin. (Botchkarev, Botchkareva et al. 2001).
11

Figure 4. BMP Inhibitor Noggin Plays Role in Morphogenesis and Homeostasis:
BMP signaling is tightly and dynamically regulated. One mechanism is the BMP
inhibitor Noggin which functions both during hair follicle placode formation as well as
the switch from telogen to anagen in the mature follicle.
One of the receptors for BMP ligands which are inhibited by Noggin is BMPR1A, which
is expressed in the developing hair follicle placode (Botchkarev 1999). Using an in
vitro/in vivo hybrid assay, the ablation of BMPR1A in the dermal papilla destroyed its
ability induce hair formation, underlying the complexity of the epithelial-mesenchymal
interaction (Rendl 2008). Additionally, when β-catenin was removed in the postnatal DP
using a Cre mediated ablation, the hair follicles lost proliferative potential, entered
catagen pre-maturely, and failed to regenerate. This demonstrates the mesechymal-
epidermal interaction present not only in development but in homeostasis (Enshell-
Seijffers, Lindon et al. 2010).
12

The conditional ablation of this receptor using a K14 Cre/BMPR1A flox/flox system
supports the role by which BMP inhibition is required for placode induction as
morphogenesis is initiated. However, the continued absence of BMP signaling results in
gross abnormalities in hair follicle morphogenesis and BMPR1A is essential for that
morphogenesis. BMPR1A knock-out (KO) follicles displaying lack of AE13 and AE15
expression which mostly results in a loss of organization of all differentiation layers of
hair follicle, but still presence of progenitors cells of matrix with proliferative ability.
Interestingly these conditional ablation studies also demonstrated a role for BMP
signaling in priming the hair follicle to respond to WNT signaling. (Kobielak, Pasolli et
al. 2003)
Hair Follicle Homeostasis
Once the hair follicle undergoes proper morphogenesis, it proceeds through cycles of
anagen (growth), catagen (degeneration ), and telogen (resting) (Muller-Rover, Handjiski
et al. 2001). In order for the hair follicle to maintain this process of regeneration
throughout life it must contain a population of slow-cycling adult stem cells in a
protective niche (Cotsarelis, Sun et al. 1990) (Morris and Potten 1999; Tumbar, Guasch
et al. 2004). These stem cells not only participate in the regeneration of the hair follicle
during normal skin homeostasis, but also are able to respond and regenerate the epidermis
when wounded (Ito, Liu et al. 2005). These cells must not only be able to proliferate and
differentiate when necessary, but also self-renew so as not to be depleted throughout life.
In the hair follicle there are two different populations within this niche of hair follicle
stem cells based on their attachment to the basement membrane. However these two
13

population display similar ability to regenerate hair and thus it seems the niche itself
plays an important role in maintaining stemness(Blanpain, Lowry et al. 2004).
Additionally there are a number of marker of this niche/bulge are including Keratin 15
(K15), CD34, and α6-integrin (Trempus, Morris et al. 2003; Morris, Liu et al. 2004;
Ohyama, Terunuma et al. 2006).
Morphogenesis and growth continues until P14 at which point catagen begins and the
lower two-thirds of the hair follicle degenerates, bringing the DP in contact with bulge
reservoir of the hair follicle stem cell. During telogen the bulge of the hair follicle is
quiescent. When again entering anagen, the hfSC become activated to form the transit
amplifying progenitors and matrix from which cells proliferate and differentiate to form
the nascent hair shaft(HS) and inner root sheath(IRS).

14

Figure 5. Hair Follicle Morphogenesis Transitions to Postnatal Homeostasis:
Folowing morphogenesis, the hair follicle enters a homestatic cycle which involes
periods of growth, destruction, and growth which takes place for the remainder of the
follicles life.

These phases of the hair cycle are complexly and tightly regulated by a large number of
molecules and pathways. Many of the same pathways implicated in the morphogenesis of
the hair follicle are implicated in the hair cycle. β-catenin activity has been demonstrated
to be required for the cyclic activation of the hfSCs residing in the quiescent bulge (Gat,
DasGupta et al. 1998). Ablation of β-catenin after morphogenesis using a K14 Cre-lox
results in the complete loss of hair flowing the first hair cycle (Huelsken, Vogel et al.
2001). While transient activation of stabilized β-catenin also using a K14 Cre-lox system
induces resting hair follicles to enter the anagen phase of the hair cycle (Van Mater,
Kolligs et al. 2003; Lo Celso, Prowse et al. 2004; Lowry, Blanpain et al. 2005). BMP
15

signals also originate from the dermis, providing macroenvironmental cues for the
maintenance of hair follicle homeostasis. BMP from the dermis serves to prime the hair
follicle during telogen so that it may respond appropriately to WNT/β-catenin signals and
enter anagen. This divides telogen into a refractory and competent phase and shows the
crucial role of BMP in the postnatal hair cycle as well as morphogenesis. (Plikus, Mayer
et al. 2008)

Figure 6. The Role of BMP in Hair Cycling and Homeostasis:
BMP plays an important role in the continued homeostasis and cycling of the hair
follicle. It remains "ON" during the resting telogen phase, is switched off during the
telogen to anagen transition, and is subsequently back "ON" during hair differentiation.

When BMPR1A is conditionally ablated following morphogenesis using a K14 Cre-lox
system otherwise quiescent hfSCs are activated to proliferate, expanding the niche and
16

losing its slow cycling characteristic. Interestingly, the SCs are not lost, instead they form
downgrowths which are long-lived and express some SC makers like Sox4, Lhx2, and
Shh but do not terminally differentiate. Indeed the ablation of BMPR1A in the SCs
elevates levels of Lef1 and β-catenin, and surprisingly the loss of BMPR1A stabilizes
nuclear β-catenin through a PTEN/AKT mediated pathway (Kobielak, Stokes et al.
2007).
Given the complex morphogenesis of the skin and hair follicle, and the scope of BMP
signaling pathways ligands and receptors we sought to further elucidate the role of BMP
signaling. Previous over-expression on BMP ligand BMP6 under a Keratin 10 (K10)
promoter located in the suprabasal layer of the skin suggested the BMP role in regulation
of the balance between proliferation and differentiation in the epidermis. Strong
expression of the transgene repressed proliferation while weak or mosaic expression
resulted in epidermal thickening (Blessing, Schirmacher et al. 1996). However this work
relied on overexpression of the secreted ligand in a non-physiologic manner, since
expression of BMP6 is downregulated after birth. Here we over-express a constitutively
active form of the BMPR1A receptor under the promotion of Keratin 14 (K14). This
system had a few advantages since no ligand was secreted extracellularly and the
overactivation of BMP signaling was driven in basal layer of epidermis during
morphogenesis and in ORS of hair follicle during hair cycle, thus it recapitulate
endogenous expression of BMPR1A in the skin. We would then ideally remove some
weaknesses in previous studies in which ectopic expression of proteins drove the
acquisition of a phenotype. This system would allow us to look at the skin
17

morphogenesis as well as skin postnatal homeostasis, and explain discrepancies of BMP
signaling in skin. Furthermore, our model would allow us to look at the effect BMP
signaling has on the balance of proliferation and differentiation in the epidermis.
Thus we sought to answer several related questions within the context of our model
system. What effect, if any, did this constitutive activation of the BMPR1A have on the
morphogenesis of epidermis and hair follicle. K14 and its partner Keratin 5 (K5) are
expressed in the first layer of the epidermis, the basal, and as such this transgene would
turn on at a relatively early developmental time point. This would provide insight into
both the epidermis itself and the hair follicle which invaginates from the epidermis.
Secondly, the transgene would provide insight into the effect overexpression has on the
maintenance of the dynamic structure of the epidermis and its appendages like the hair
follicle. Specifically within the maintenance of homeostasis, the effect on the stem cells
of the hair follicle for which BMP is an important signaling component.

18

Results
Chapter 1: Strong Phenotype Results
Generation of K14Alk3 Mouse
A C-terminal portion of a constitutively active form of human BMPR receptor 1A fused
to human influenza hemagglutinin epitope tag (HA), hALK3Q233D-HA, driven by
keratin 14 (K14) was used to generate transgenic animals (Fig. 7A). K14 is expressed in
the basal layer of the epidermis in the skin and Outer Root Sheath (ORS) of hair follicle,
as well as in basal layer of other stratified epitheliums including the oral mucosa,
esophagus, trachea, bladder, and cervix (Vassar, Rosenberg et al. 1989). As visualized
by IHC, transfection of wild-type kertatinocytes with the K14Alk3Q233D-HA construct
induced strong expression of the HA tag which correlated with activation of phospho-
SMAD1, 5, 8 proteins, downstream effectors of the canonical BMP signaling pathway,
(Fig. 7B). In addition, the transfected kertatinocytes expressed the correct size
(molecular weight) of the active form of human BMP receptor 1A (K14Alk3Q233D-HA)
as checked by western blot using a HA-antibody (Fig. 7C). ). In vivo, transgenic animals
displayed strong and uniform expression of the transgene properly located in the basal
layer of the epidermis (Fig. 7D). These results indicate the transgene is both functional
and appropriately expressed within the basal layer of the epidermis.
Offspring of the founding mice yielded litters in the expected Mendelian ratios as
confirmed by both phenotype and genotyping (Fig. 7G and Fig. 7E, respectively). The
K14Alk3Q233D-HA mice exhibited a variable level of transgene expression, which
resulted in three different phenotypes (Fig.7F). Mice with strong expression of the
19

Alk3Q233D-HA were significantly smaller in size compared to their control littermates
and died in the first couple days after birth (Fig. 7G, P7). This early lethality was likely
due to defect in the esophagus or oral epithelium, as the transgenic pups had no milk in
their stomachs. Additionally there was a less severe or medium phenotype. These animals
failed to survive beyond 4 weeks, once again probably due to a defect in the oral
epithelium (Fig. 7G, P21). Finally there was a weak and not yet viable phenotype, where
transgenic animals were similar in size to their control littermates. These transgenic
animals were fertile and able to reproduce, but lost their hair progressively after
approximately 3-4 months (Fig. 9A).

20

Figure 7. Generation of K14Alk3 Q233D-HA mouse: A. K14Alk3 Q233D-HA
construct used in the generation of the transgenic mouse. PCR primers noted in brackets.
B. Western blot of keratinocytes transfected with K14Alk3 Q233D-HA construct and
positive for HA tag. C. Immunohistochemistry (IHC) for HA tag of K14Alk3 Q233D-HA
construct in transfected keratinocytes. D. IHC for HA and E-cadherin in the epidermis of
transgenic animal E. Genotyping of K14Alk3 mice. F. Western blot demonstrating
different levels of transgene expression G. Phenotype of K14Alk3 Q233D-HA mice at
P7 and P21
21

Strong K14Alk3 Q233D-HA Phenotype
The strongest K14Alk3 Q233D-HA animals died in the first couple days after birth. They
were smaller and exhibited a more rigid and less elastic skin when compared them to
their control littermates (Fig 7G, P7). Since the early lethality of these strong transgenic
animal impaired further analysis of the phenotype, to address long-term consequences
how active BMP signaling affect skin morphogenesis a skin graft experiment was
performed. P0 lethal head and back skin along with control was grafted onto an
immunocompromised mouse. Thirty days post graft, the control skin exhibited normal
and strong hair growth while the transgenic graft revealed almost no visible hair growth
(Fig. 8A). These grafts were harvested for further histological analysis, which confirmed
the normal skin and hair follicles morphogenesis of the control graft whereas, the
histology of the transgenic graft showed significant abnormalities in the grafted
transgenic skin. Morphogenesis was severely impaired with a few normal hair follicles
and instead mass-like down growths of the epidermis were observed. Additionally the
epidermis of the transgenic graft exhibited hyperplasia (Fig. 8B). As compared to the
control’s graft, the transgenic one had 71% less hair follicles (Fig. 8C).
To further look into the morphogenesis of the skin and effect on the mesenchymal dermal
papilla (DP), alkaline phosphate staining (AP) was performed and showed just a few
normal DPs in transgenic animals. Normal staining was visible around the a few anagen
hair follicles, but interestingly a AP staining was expanded in the upper dermis just below
the epidermal-dermal junction especially in the area were no invaginations were visible in
22

the transgenic graft (Fig. 8D). Since hair follicle morphogenesis in the transgenic graft
was severely impaired, it was important to further examine if another epidermal lineages,
like sebaceous glands are formed. Oil red staining showed existence of sebaceous
glands in the Tg skin graft but it was limited only to mass-like downgrowths of the
epidermis, indicating that impaired hair follicle morphogenesis also affected the amount
of sebaceous glands and their lineage formations (Fig. 8E). Furthermore,
immunohistochemistry showed proper differentiation of the epidermal layers as indicated
by keratin 1 (K1), a maker of spinous layer, and Loricrin , a marker of the granular layer
of epidermis (Fig. 8F).
These results from the strong phenotype mice show that constitutive BMPR1A signaling
markedly reduces normal hair follicle morphogenesis. This is likely a developmental
event, as the skin exhibited severely impaired hair follicle invagination which correlated
with lack of dermal papillae condensation and occasionally visible mass-like down
growths of the epidermis. Additionally, the strong K14Alk3 Q233D-HA animals showed
no defect in the morphogenesis or differentiation of the epidermis. While the transgenic
epidermis looks hyperplastic, given the graft possible effects from the wound healing
process it is unclear whether this observation is directly or indirectly related the
constitutive BMP signaling.
23

Figure 8. K14Alk3 Strong and Medium Phenotypes: Impaired Hair Follicle
Morphogenesis: Constitutive active BMP signaling affects hair follicle morphogenesis.
A. Skin grafts from P0 transgenic and control animals. The transgenic exhibited a
complete lack of hair growth. B. H&E staining of skin grafts showing epidermal
hyperplasia and mass-like downgrowths of the epidermis. C. Graph showing significant
loss of hair formation in transgenic graft. D. Alkaline Phosphatase (AP) staining
depicting the derma papilla (DP) present but wrapping around the downgrowths in the
transgenic as well as expanding in the upper dermis just below the epidermal-dermal
junction. E. Oil red staining showed formation of sebaceous glands in transgenic graft but
it was only limited to the mass-like downgrowths of the epidermis. F. Loricrin and
Keratin 5 immunohistochemistry demonstrating proper differentiation of epidermal layers
in the transgenic graft. G. P8 lethal transgenic animal with littermate control. H. H&E
staining of P8 back skin. Note the shorter and more sparsely spaced hair follicles in the
transgenic. I. Graph demonstrating a small but significant decrease in hair follicles in
the transgenic. J. Loricrin/K5 and K1/K5 staining which shows proper epidermal
differentiation in both samples.

24

Chapter 2: Medium Phenotype Results
Medium K14Alk3 Q233D-HA Phenotype
The medium transgenic phenotype is smaller in size and usually survived until 3-4 weeks
of age, which usually allowed for analysis of skin morphogenesis in these transgenic. A
hair phenotype manifests on the top of the head progresses down the back of these
animals. There is a significant decrease as well as delay in amount of hair follicles growth
in these animals. The number of whiskers was also decreased (data not shown). The skin
of the transgenic animal contained significant differences from that of the control. The
head, where phenotype is first evident and strongest, had a very thin epidermal layers and
thick stratum corneum. Skin on the back exhibited similar reduced amount of hair follicle
formed, but instead of a thin epidermis visible on the head area there was evidence of
hyperplasia. Furthermore, the skin of these medium transgenic animals was rigid and
lacked of elasticity when handled.
Analysis of the P8 lethal transgenic animal and a control littermate showed a less severe
but consistent, medium phenotype (Fig. 8G). Histologically, the transgenic hair follicles
appeared shorted and more spread out (Fig. 8H). Indeed the transgenic demonstrated a
10% decrease in number of hair follicles in the back skin (Fig. 8I). Additionally the
transgenic animals possessed a smaller layer of dermal adipose tissue compared to the
control (Fig. 8H). Consistent with the graft, epidermal differentiation of the back skin
remained normal as visualized by Loricrin and K1 (Fig. 8J). However, looking the head
where the phenotype is strongest, we observed different results. The head region was
covered by a very thin epidermis with spare and arrested follicular
25

invagination/morphogenesis. When stained for markers of differentiation, K1 and K5, the
thin epidermis exhibited premature differentiation (Fig. 8K, P18, lower pannel).
The K14Alk3 Q233D-HA medium phenotype consistently exhibited a decrease in hair
morphogenesis, although less when compare to strong phenotype, thus it suggests that
BMP signaling work in dose dependent manner. Likewise the back skin failed to show
any effect by the constitutively active BMP signaling on epidermal differentiation.
However, on the head where the phenotype is most obvious and strongest the epidermis
showed premature differentiation with overlap of markers for the basal (K5) and
suprabasal layers (K1).
Chapter 3: Weak Phenotype Results
Weak Progressive K14Alk3 Q233D-HA Phenotype
While the disruption of hair follicle morphogenesis in the animals most strongly
expressing the transgene provides a fascinating phenotype in which to study the role of
BMP signaling in morphogenesis, more interestingly there is also a weak progressive
phenotype. These animals manifest a bald phenotype on their head similar to the lethal/
strong and medium phenotypes, but are otherwise comparable in size to control
littermates, viable, and fertile. Additionally hair follicle morphology, epidermal
differentiation, and follicular differentiation appear normal at the beginning of life.
However, at around 4 months these transgenic animals begin to progressively lose their
hair until the animals are almost completely bald usually after they're past one year in age
(Fig. 9A). Since the transgene is expressed postnatally in ORS of hair follicle, part of
which is occupied by hair follicle stem cell in the bulge region, this progressive loss of
26

hair in the presence of the constitutively active BMP signaling raised interesting question
about the homeostasis of the hair follicle stem cells (hfSC).

Figure 9. K14Alk4 Progressive Phenotype: Progressive loss of hair in weak phenotype
of K14Alk3 Q233D-HA with age. A. Progressive loss of hair by transgenic animals over
the course of months. B. Transgenic animals showed lack of hair follicles growth at P140
after back skin hairs were shaved at previous telogen at P90 when compare to control
mice C. Graphical comparison of transgenic and control hair cycles. D. H&E staining of
transgenic and control samples at various time points of the hair cycle. Note at P140 the
transgenic animals appear lack of hair follicles anagen activation still at telogen stage
while the control already is in full anagen.

27

Progressive Loss of Hair
With the dynamic nature of the hair follicle stem cells and their niches during the hair
cycle, it was important to elucidate the precise timing at which the transgenic animals
began to differ from the wild-type controls. Through the first postnatal hair cycle both
transgenic and control animals remained the same as evidenced through both histology
and hair shaving – regrowth experiment (data not shown) This was also true during the
second postnatal hair cycle where the transgenic and control animals exhibited no
difference (Fig. 9B, P60). It was not until after the third postnatal hair follicle telogen
that the transgenic began to differ from that of the control as evidenced by shaving –
regrowth experiment (Fig. 9C). At P140 the transgenic hair follicles still remained in
telogen whereas at that time point the control was already in full anagen (Fig. 9D).
Next it was important to investigate what differences there might be between the
transgenic and control at the time points older than 1 yr at which the transgenic had
become predominantly bald. This phenotype raised interesting questions about the
maintenance and homeostasis of the hfSCs in these animals. Was the long-term BMP
signaling resulting in a loss of hfSCs or conversely was the transgene forcing the hfSCs
to remain increasingly quiescent? After 1 year of age in transgenic animal skin showed
changes in morphology as demonstrated by H&E staining (Fig. 10A). The epidermis is
thickened, exhibiting hyperplasia while the hair follicle remained in a quasi quiescent
telogen-like stage (Fig. 10A-C). Since almost all clubs hairs were already missing it
suggests accelerated exogen phase at that time point, (Fig. 10A-C as compare to controls
lower pannels). Alkaline phosphatase staining showed the presence of the arector pili
28

muscle and sometimes the DP structures at the bottom of telogen-like hair follicle. Other
times the DP looked improperly localized wrapping around a structure of telogen -like
hair follicle (Fig. 10B). Interestingly, in transgenic animals oil red staining still revealed
the presence of sebaceous glands attached to telogen-like hair follicles and sebum-like
substance in the hair canal of the some of the follicles (Fig. 10C). The makers of
epidermal differentiation, Loricrin and K1, were both expressed appropriately in their
respective layers (Fig. 10D).

29

Figure 10. K14Alk3 Progressive Phenotype: Loss of Hair but Maintenance of
Differentiation: Characterization of progressive K14Alk3 phenotype. A. H&E staining
of one year old bald transgenic mouse. Note the epidermal hyperplasia, loss of hair shaft
along with hair club – accelerated exogen like phase, and hair follicle remained in a
quiescent telogen-like stage . B. AP staining showed the presence DP structures and the
arector pili muscle still attached to structure of telogen-like hair follicle, however
sometimes DP was improperly localized around that structures. C. Oil read staining
showing a the remainder of sebaceous glands in the transgenic animal. Note the sebum in
the hair canal and the decrease in dermal adipose tissues. D. Lor/B4 and K1/B4 staining
demonstrating the maintenance of proper epidermal differentiation in the transgenic.

30

Next it was important to determine if the hfSCs are still maintained in the skin of
transgenic animals with progressive phenotype. The answer to this question would help
to reveal the potential molecular process implicit in the loss of hair. To achieve this, well-
established markers of the stem cell population: α6 integrin and CD34 (Trempus, Morris
et al. 2003) were used for fluorescent activated cell sorting (FACS) of the hfSCs
populations Representative transgenic and control animals of the same age were sorted
with histology confirming representative phenotype (Fig. 11A and 11B). Very
surprisingly, and across both several experiments and several different time points, the α6
high/CD34 high population in the back skin of the transgenic animals was elevated
compared to age-matched controls (Fig 11C). At 15 months the transgenic animal's α6
high/CD34 high population was 6.7% while the control's was 3.1%. Additionally this
increase was significant, ranging from 100% to a 300% increase of the α6 high/CD34
high population (Fig 11D). Conversely, the α6 low/CD34 high suprabasal fraction was
lowered, though less significantly in the transgenic animals, ranging from 20-50% less. In
seeking to confirm this counterintuitive result, IHC for CD34 and K15, a marker of the
stem cell bulge, were performed on these samples. Indeed there is evidence that the
transgenic animals maintain a subset of cells that are positive for CD34 and K15,
however this population of cells can also be seen in the epidermis (Fig. 11E and 11F).
CD34 is expressed by other cell types including endothelial progenitor cells and
hematopoietic derived cells or CD34 could be expressed differently as a result of the
introduction of the transgene. Given this fact and the counterintuitive results of the FACS
experiments, we thought it is important to confirm these results by crossing the K14Alk3
31

Q233D-HA mice on the background of K15GFP reporter mice (Fig. 11G) The
expression of GFP driven by K15 promoter in the bulge of the hair follicle would allow
to mark live hair follicle stem cells as the hair loss phenotype progressed (Morris, Liu et
al. 2004).

32

Figure 11. Increase of the a6 high/CD34 high Hair Follicle Stem Cell Population in
K14Alk3 Progressive Phenotype: Isolation of hair follicle stem cells from transgenic
K14Alk3 Q233D-HA animals. A. Phenotype of transgenic and control animals used in
FACS experiment at 15 months B. Morphology of back skin transgenic and control
animals stained by H&E C. FACS analysis of back skin from transgenic and control
littermates. P9 gate represent a6 low/CD34 high suprabasal fraction. P8 gate represent
a6 high/CD34 high basal hfSC population. Note the increase of a6/CD34 population in
the transgenic as compared to control. D. Graph showing the percent difference between
transgenic and control a6 high/CD34 high hfSC populations over different FACS
experiments and time points. E. K15/K5 staining showing K15 positive cells remaining
in the transgenic animals. F. CD34/K5 staining displaying an absence of CD34 positive
cells in transgenic samples. G. Schematic of K15GFP reporter mouse crossed K14Alk3
animal and showing phenotype at 8 months.

33

During the FACS experiments, the α6high/CD34high fraction was collected and cultured
(Fig 12A). These cells were able to be passaged multiple times in vitro, further
supporting that the bald transgenic animals might maintain a population of hfSCs. While
the collected α6hi/CD34hi cells of the control and transgenic animals were both able to
be passaged extensively. While in the presence of feeder and 0.03mM Ca
2+
media
transgenic and control cells exhibited little difference. However, once passaged to
0.05mM Ca
2+
media in the absence of 3T3 feeder the transgenic cells tended to be longer
and more spread morphologically compared to controls (Fig. 12B). These transgenic cells
were also more slowly growing then controls and tended to less readily form colonies
despite similar attachment in low calcium media (Fig. 12C)
While low expression of the constitutively active BMPR1A transgene had no discernable
effect on homeostasis of the hair follicle at least for the first 4 months after the birth.
Over time, these animals developed loss of hair phenotype, likely due to the increasing
quiescence of the hair follicle stem cells. Interestingly and surprisingly, when flow
cytometry was used to sort cells from the back skins of these animals, there was a larger
population of a6high/CD34high positive cells as compare to hairy control littermates at
the same age.

34

Figure 12. In Vitro Analysis of a6 high/Cd34 high Hair Follicle Stem Cell
Population: In vitro analysis of FACS isolated hair follicle stem cells from K14Alk3
Q233D-HA animals. A. hfSCs were sorted by FACS using antibodies against CD34 and
a6 integrin. P8 gate represents a6 high/CD34 high hfSC population. P9 P9 gate
represent a6 low/CD34 high suprabasal fraction. B. Morphology of transgenic and
control cells. Note the longer, more spindle shape of the transgenic. C. Growth assay
using low Ca media (0.05 mM Ca) without feeder and showing lower growth for the
transgenic cells despite similar levels of attachment.

35

Discussion
The balance of BMP signaling has been demonstrated to be exquisitely tuned not only in
the morphogenesis of both the epidermis and the hair follicle, but also during postnatal
skin homeostasis including the cycling of the hair follicle. Here, the expression of a
constitutively active form of BMPR1A under a K14 promoter shows how changes in the
fine balance of BMP signaling can lead to both developmental and long-term effects in
the skin and skin appendages.
BMPR1A and Epidermal Morphogenesis:
As previous studies have illustrated, differing levels of BMP transgene expression can
result in different skin phenotypes. When BMP6 was ectopically expressed under the
promotion of K10 (in suprabasal layer), proliferation of epidermis was repressed at high
levels of transgene expression but enhanced at moderate levels. Additionally,
differentiation was unaffected in the strong expressers while the moderate expressers
exhibited K14 staining through all layers of stratification and Keratin 6 (K6) staining in
suprabasal layer (Blessing, Schirmacher et al. 1996). Our study displays both similarities
as well as differences to this previous work. We observed at the highest and medium
levels of expression on the head, the constitutively active BMP signaling results in a very
thin epidermis which suggests that proliferation is decreased. However, this thin head
skin diverges from the previous work and exhibits premature epidermal differentiation.
Conversely at lower levels of expression, the progressive phenotype displayed a similar
hyperplasic epidermis but no defect in differentiation. One of explanation of these
differences between both systems would be that ectopic expression of BMP6 ligand in
36

the suprabasal layer (BMP6 is not express in adult skin) compared to our expression of a
constitutively active form the BMPR1A receptor (in its “physiologic” location in the
basal layer) was not that strong, and not activate as efficiently canonical BMP pathway in
the proliferative, basal layer of epidermis. Additional reason would be that although
BMP6 ligand can interact with BMP receptor 1A, however at lower binding affinity
(Botchkarev 2003; Botchkarev and Sharov 2004). The strength of this epidermal
premature differentiation phenotype, in fact is very consistent between our three
different phenotypes (strong, medium and weak BMP expressers) but only on the head
area whereas the rest of back skin is less or not affected or demonstrate relatively normal
differentiation of epidermis. Potentially, several different reasons could be considered to
explain the regional disparities in this model: one of them would be that the transgene,
under the promotion of K14, is turned at an earlier developmental time point in the head
area; another would be that the microenvironment of head dermis is more sensitive to any
BMP signaling imbalance and that the ratio of BMP activators to inhibitors varies in
different regions. Therefore the head region would be more permissive to the effects of
the constitutively active BMP signaling. The epidermis in this location is very thin and
there is overlap between K1 and K5 makers of the suprabasal and basal layers
respectively. This precocious epidermal stratification further emphasis the role BMP
signaling in balancing proliferation and differentiation within the basal layer of the
epidermis. However, back skin areas from medium and low phenotypes transgenic
animals (as well as skin graft from strong transgenic phenotype) exhibited relatively
normal differentiation, but sometimes with epidermal hyperplasia. This discrepancy in
37

phenotype observed in the same animal in the different area (head vs. back skin)
illustrates that BMP signaling in the skin is not a simple on/off switch but rather a
continuously balanced gradient where different strength in BMP signaling (high, medium
and low) could modulate different biological outcome, the nature of which must be
further investigated.
BMPR1A-CA and Hair Follicle Morphogenesis:
BMP has long been known to play an important and complex role in hair follicle
morphogenesis. BMP inhibition via Noggin is required for hair placode formation as
shown by Noggin ablation, and the ablation of the BMPR1A receptor results in the
failure of progenitor cells to terminally differentiate into IRS and hair shaft (Kulessa,
Turk et al. 2000; Kobielak, Pasolli et al. 2003; Andl, Ahn et al. 2004). In both our
strongest phenotype and medium phenotype hair follicle morphogenesis is affected at
levels which correspond to transgene expression in the skin. The strongest transgenic
animals where postnatal skin where transplanted on immunodeficient host mouse exhibit
no hair follicle formation (Fig 8) and limited mass-like downgrowths of the epidermis,
whereas the medium phenotype shows reduced overall numbers of hair follicles. These
results are consistent and suggest that constitutively active BMP signaling (if strong
enough) results in inhibition of early placodes formation during morphogenesis. This
remains consistent with earlier studies regarding the role of BMP in hair morphogenesis
and placode formation. In an in vitro organ culture system, addition of BMP4 soaked
beads to embryonic back skin resulted in a loss of hair placode formation and this was
recapitulated in Noggin
-/-
animals (Botchkarev, Botchkareva et al. 1999). This support the
38

BMP gradient hypothesis that different level of active BMP signaling correlate to
different phenotypes in hair follicle pattern formation. This is consistent with the low
expressing phenotype which exhibited normal morphogenesis and hair follicle number
(Fig 9B and 9D).
One of the interesting results from characterizing the strongest phenotype of skin graft
was the accumulation of AP staining, a marker for the dermal papilla and other dermal
component (including arrector pili muscle) of the hair follicle, in the upper dermis at the
epidermal-dermal junction. During hair follicle morphogenesis the reciprocal interaction
between epidermis and dermis is really important for initiation or placode formation.
Given that the transgene is expressed in the epidermal components only at basal layer, it
most likely indirectly affects expansion the AP staining and suggests that dermal papilla
condensation was impaired at the very early time point as well. Therefore, the increased
BMP signaling presumably inhibits hair placode formation, in both direct and indirect
manners by affecting a reciprocal interaction between epidermis and surrounding dermal
fibroblast and thus impaired the DP condensation. To better understand this mechanism
further investigation must be proceed in the future.
BMPR1A, Skin, and Hair Follicle Homeostasis:
Since both the strong and medium phenotype, were early lethal, it was thus hard to
address long term consequences of the constitutively active BMPR1A on the homeostasis
of the skin and hair follicle. However the low, progressive phenotype lives to old age.
While they exhibit no difference from controls early in life (except a head area as
discussed previously), as they age the homeostasis of the epidermis is affected. The
39

hyperplasia and loss of elasticity of these animals point to a role for BMP signaling in
maintaining epidermal homeostasis throughout life. Our observation is that transgene
expression remains consistent through life and this raises an interesting question
regarding the mechanism of this phenotype. One possible speculation would be changes
of the micro or macroenvironmental BMP levels in skin with aging and thus tilting the
balance of homeostasis towards more quiescent stage? Although this remains intriguing,
and merits further study.
Even more dramatic is progressive loss of hair follicle on the low phenotype animals later
in life. This phenotype suggested the progressive effect of BMP signaling on homeostasis
of hair follicles, presumably on their stem cells. A previous works have investigated this
balance of signaling in the hair follicle and ablation or overexpression of the BMPR1A
receptor has resulted in defects of the hair follicle as a result of perturbing the hfSCs
(Kobielak, Stokes et al. 2007; Plikus, Mayer et al. 2008). Combined with previous
knowledge of the role BMPR1A plays in the maintenance of hfSC quiescence and our
own in vitro work, the constitutively active BMP signaling would seem to increase the
quiescence of the hfSCs.
Finally, the other interesting question would be the formation of sebaceous glands in the
strong phenotype. The sebaceous gland is attached to the hair follicle and maintains a
unipotenital population of cell that express Blimp1 and are the progenitors for the other
cell types of the sebaceous gland. However, when this population is perturbed, hfSCs
from the bulge have been shown able to maintain sebaceous gland homeostasis.(Horsley,
O'Carroll et al. 2006; Fuchs and Horsley 2008). When invagination occurs, but the hair
40

follicle does not form properly, leaving an epidermal downgrowth there appears no effect
on the fate decision of sebaceous progenitors to form the sebaceous gland. In the
progressive phenotype this formation and maintenance of the sebaceous gland also
appears unaffected by the constitutively active BMP signaling despite the loss of hair
follicle homeostasis.

Figure 13. Conclusions on Skin and Hair Follicle Morphogenesis and Homeostasis:
Scheme depicting diverse effects of constitutively active BMP signaling in the model
described here. At high levels of transgene expression, skin and hair follicle
morphogenesis is disrupted, but even low levels of BMP signaling will disrupt long-term
homeostasis of the epidermis and particularly the hair follicle.

41

Future Directions
While our work provides some increased insight into the role BMP signaling plays it the
morphogenesis of the skin and hair follicle as well as their long-term homeostasis, there
are a number of interesting facets to the story which have yet to be learned. The question
of the hair follicle stem cell fate as a consequence of the active BMP signaling remains
open. Some data points to their continued existence despite the complete loss of hair in
these animals; other data is ambiguous. The potential of the K15GFPK14Alk3 mice are
just being tapped to help answer these questions. Furthermore, if these stem cells remain
present in some form, rescue experiments would help elucidate the question from an
additional perspective.
This system also provides the opportunity for several alternate rescue experiments.
While a pilot experiment using the activator of differentiation TPA were performed, this
and other topical treatments are one approach. Beyond topical approaches, beads
harboring recombinant activator proteins such as those from the WNT family could be
used in an attempt to prod the progressive phenotype hair follicles out of quiescence.
Finally crossing the K14Alk3 mouse with a transgenic harboring a K14 promoted
activator is a third avenue.
Additionally, the paucity of dermal adipose is a consistent phenotype of K14Alk3 mice at
all levels of expression and phenotype. This raises interesting questions about this
relationship between dermal BMP and BMP signaling in the epidermis.(Plikus, Mayer et
al. 2008) While the constitutively active BMPR1A studied as the confounding situation
arising from overexpression of a secreted ligand, the lack of dermal adipose remains
42

unexplained. Whether this is a direct or indirect effect provides an intriguing line of
experiments. Furthermore, given the progression of phenotype with age in the low-
expressing animals and the fact transgene expression remains consistent through, it
remains intriguing to investigate whether the macroenvironmental levels of BMP change
with age. Could this explain the progressive phenotype: that the finely tuned balance of
BMP is altered naturally with age and in conjunction with the transgene shifts this
balance.

43

Bibliography

Andl, T., K. Ahn, et al. (2004). "Epithelial Bmpr1a regulates differentiation and
proliferation in postnatal hair follicles and is essential for tooth development."
Development 131(10): 2257-2268.

Andl, T., S. T. Reddy, et al. (2002). "WNT signals are required for the initiation of hair
follicle development." Dev Cell 2(5): 643-653.

Blanpain, C., V. Horsley, et al. (2007). "Epithelial stem cells: turning over new leaves."
Cell 128(3): 445-458.

Blanpain, C., W. E. Lowry, et al. (2004). "Self-renewal, multipotency, and the existence
of two cell populations within an epithelial stem cell niche." Cell 118(5): 635-
648.

Blanpain, C., W. E. Lowry, et al. (2006). "Canonical notch signaling functions as a
commitment switch in the epidermal lineage." Genes Dev 20(21): 3022-3035.

Blessing, M., P. Schirmacher, et al. (1996). "Overexpression of bone morphogenetic
protein-6 (BMP-6) in the epidermis of transgenic mice: inhibition or stimulation
of proliferation depending on the pattern of transgene expression and formation of
psoriatic lesions." J Cell Biol 135(1): 227-239.

Botchkarev, V. A. (2003). "Bone morphogenetic proteins and their antagonists in skin
and hair follicle biology." J Invest Dermatol 120(1): 36-47.

Botchkarev, V. A., N. V. Botchkareva, et al. (2001). "Noggin is required for induction of
the hair follicle growth phase in postnatal skin." Faseb J 15(12): 2205-2214.

Botchkarev, V. A., N. V. Botchkareva, et al. (1999). "Noggin is a mesenchymally derived
stimulator of hair-follicle induction." Nat Cell Biol 1(3): 158-164.

Botchkarev, V. A. and A. A. Sharov (2004). "BMP signaling in the control of skin
development and hair follicle growth." Differentiation 72(9-10): 512-526.

Chiang, C., R. Z. Swan, et al. (1999). "Essential role for Sonic hedgehog during hair
follicle morphogenesis." Dev Biol 205(1): 1-9.

Cotsarelis, G., T. T. Sun, et al. (1990). "Label-retaining cells reside in the bulge area of
pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin
carcinogenesis." Cell 61(7): 1329-1337.
44

Enshell-Seijffers, D., C. Lindon, et al. (2010). "Beta-catenin activity in the dermal papilla
of the hair follicle regulates pigment-type switching." Proc Natl Acad Sci U S A
107(50): 21564-21569.

Fuchs, E. and V. Horsley (2008). "More than one way to skin." Genes Dev 22(8): 976-
985.

Gat, U., R. DasGupta, et al. (1998). "De Novo hair follicle morphogenesis and hair
tumors in mice expressing a truncated beta-catenin in skin." Cell 95(5): 605-614.

Horsley, V., D. O'Carroll, et al. (2006). "Blimp1 defines a progenitor population that
governs cellular input to the sebaceous gland." Cell 126(3): 597-609.

Hu, M. C., T. D. Piscione, et al. (2003). "Elevated SMAD1/beta-catenin molecular
complexes and renal medullary cystic dysplasia in ALK3 transgenic mice."
Development 130(12): 2753-2766.

Huelsken, J., R. Vogel, et al. (2001). "beta-Catenin controls hair follicle morphogenesis
and stem cell differentiation in the skin." Cell 105(4): 533-545.

Ito, M., Y. Liu, et al. (2005). "Stem cells in the hair follicle bulge contribute to wound
repair but not to homeostasis of the epidermis." Nat Med 11(12): 1351-1354.

Jamora, C., R. DasGupta, et al. (2003). "Links between signal transduction, transcription
and adhesion in epithelial bud development." Nature 422(6929): 317-322.

Kobielak, K., H. A. Pasolli, et al. (2003). "Defining BMP functions in the hair follicle by
conditional ablation of BMP receptor IA." J Cell Biol 163(3): 609-623.

Kobielak, K., N. Stokes, et al. (2007). "Loss of a quiescent niche but not follicle stem
cells in the absence of bone morphogenetic protein signaling." Proc Natl Acad Sci
U S A 104(24): 10063-10068.

Kulessa, H., G. Turk, et al. (2000). "Inhibition of Bmp signaling affects growth and
differentiation in the anagen hair follicle." Embo J 19(24): 6664-6674.

Lo Celso, C., D. M. Prowse, et al. (2004). "Transient activation of beta-catenin signalling
in adult mouse epidermis is sufficient to induce new hair follicles but continuous
activation is required to maintain hair follicle tumours." Development 131(8):
1787-1799.

Lowry, W. E., C. Blanpain, et al. (2005). "Defining the impact of beta-catenin/Tcf
transactivation on epithelial stem cells." Genes Dev 19(13): 1596-1611.
45

Millar, S. E. (2002). "Molecular mechanisms regulating hair follicle development." J
Invest Dermatol 118(2): 216-225.

Morris, R. J., Y. Liu, et al. (2004). "Capturing and profiling adult hair follicle stem cells."
Nat Biotechnol 22(4): 411-417.

Morris, R. J. and C. S. Potten (1999). "Highly persistent label-retaining cells in the hair
follicles of mice and their fate following induction of anagen." J Invest Dermatol
112(4): 470-475.

Muller-Rover, S., B. Handjiski, et al. (2001). "A comprehensive guide for the accurate
classification of murine hair follicles in distinct hair cycle stages." J Invest
Dermatol 117(1): 3-15.

Ohyama, M., A. Terunuma, et al. (2006). "Characterization and isolation of stem cell-
enriched human hair follicle bulge cells." J Clin Invest 116(1): 249-260.

Plikus, M. V., J. A. Mayer, et al. (2008). "Cyclic dermal BMP signalling regulates stem
cell activation during hair regeneration." Nature 451(7176): 340-344.

Rhee, H., L. Polak, et al. (2006). "Lhx2 maintains stem cell character in hair follicles."
Science 312(5782): 1946-1949.

Trempus, C. S., R. J. Morris, et al. (2003). "Enrichment for living murine keratinocytes
from the hair follicle bulge with the cell surface marker CD34." J Invest Dermatol
120(4): 501-511.

Tumbar, T., G. Guasch, et al. (2004). "Defining the epithelial stem cell niche in skin."
Science 303(5656): 359-363.

Van Mater, D., F. T. Kolligs, et al. (2003). "Transient activation of beta -catenin
signaling in cutaneous keratinocytes is sufficient to trigger the active growth
phase of the hair cycle in mice." Genes Dev 17(10): 1219-1224.

Vassar, R., M. Rosenberg, et al. (1989). "Tissue-specific and differentiation-specific
expression of a human K14 keratin gene in transgenic mice." Proc Natl Acad Sci
U S A 86(5): 1563-1567.

Williams, S. E., S. Beronja, et al. (2011). "Asymmetric cell divisions promote Notch-
dependent epidermal differentiation." Nature 470(7334): 353-358.

Wilson, P. A., G. Lagna, et al. (1997). "Concentration-dependent patterning of the
Xenopus ectoderm by BMP4 and its signal transducer Smad1." Development
124(16): 3177-3184.

Abstract (if available)

Abstract The skin and its array of epidermal appendages with their complex 3-dimensional structure provide an excellent model in which to study morphogenesis, pattern formation, and homeostasis. The biochemical signals that govern these biological processes continue to be unraveled. Bone Morphogenetic Proteins (BMP) play an integral role in both morphogenesis and homeostasis of the epidermis, however the precise details of this role remains a source of inquiry. ❧ To further study the role of BMP signaling in the epidermis and its appendages both in morphogenesis and homeostasis we have generated transgenic mice harboring a constitutively active form of the BMPR1A receptor under a Keratin 14 (K14) promoter. This will drive expression of the transgene in the basal layer of the epidermis as well as the Outer Root Sheath (ORS) of the hair follicle. ❧ At strong levels of transgene expression hair follicle morphogenesis is markedly impaired. Epidermal morphogenesis appeared normal, though the balance of differentiation and proliferation was perturbed. Interestingly, in the animals which expressed low levels of the K14Alk3 transgene, there was a progressive loss hair with age. This loss of hair follicle homeostasis implicates BMP as not only a crucial player in morphogenesis, but important in finely balancing proliferation and quiescence in the hair follicle stem cells.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Constitutively active canonical BMP signaling in the apical ectodermal ridge disrupts dorsal-ventral and anterior-posterior axes resulting in impaired development of the distal (autopod) limb domain

PDF

Morphoregulation and cycling of ectodermal organs

PDF

Characterizing the role of Dnmt1 and its downstream factors in progressive alopecia

PDF

Bimodal effects of bone morphogenetic proteins in prostate cancer

PDF

Molecular aspects of skin early morphogenesis

PDF

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

PDF

Study of the role of bone morphogenetic proteins in prostate cancer progression

PDF

Study of bone morphogenetic protein-2 and stromal cell derived factor-1 in prostate cancer

PDF

Characterization of new stem/progenitor cells in skin appendages

PDF

Macroenvironmental regulation of hair follicle stem cells

PDF

Skin appendage growth control by hormones and morphogens

PDF

Bone marrow-derived mesenchymal stromal cells in the tumor microenvironment of neuroblastoma

PDF

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

PDF

Molecular mechanism of transforming growth factor-beta signaling in skin wound healing

PDF

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

PDF

Role of STAT3 phosphorylation in mouse embryonic stem cell self-renewal and differentiation

PDF

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

PDF

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

PDF

Transcriptional and epigenetic mechanisms underlying sensory hair cell differentiation and regeneration

PDF

Roles of Hox B genes in the development of chicken skin

Asset Metadata

Creator Hennigan, Andrew (w.)

Core Title Implications of constitutively active bone morphogenetic protein (BMP) signaling in skin morphogenesis and skin postnatal homeostasis

School Keck School of Medicine

Degree Master of Science

Degree Program Experimental and Molecular Pathology

Publication Date 12/13/2011

Defense Date 12/13/2011

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

Tag BMP,Bone Morphogenetic Protein,hair follicle,K14,OAI-PMH Harvest,skin,stem cell

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Kobielak, Krzysztof (committee chair), Choung, Cheng-Ming (committee member), Kobielak, Agnieszka (committee member), Widelitz, Randall B. (committee member)

Creator Email awhennigan@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-215726

Unique identifier UC11293987

Identifier usctheses-c3-215726 (legacy record id)

Legacy Identifier etd-HenniganAn-456-0.pdf

Dmrecord 215726

Document Type Thesis

Rights Hennigan, Andrew, W.

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