| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Dermatology, Boston University School of Medicine, Boston, Massachusetts 02118, USA;
* Department of Dermatology, University Hospital Eppendorf, University of Hamburg, Hamburg, Germany;
Department of Clinical Chemistry and Pathobiochemistry, University Hospital Benjamin Franklin, Free University, Berlin, Germany;
Department of Anatomy and Cell Biology, University of Goteborg, Goteborg, Sweden; and
Deutsches Rheumaforschungzentrum, Berlin, Germany
1Correspondence: Department of Dermatology, Boston University School of Medicine, 609 Albany St., Boston, MA 02118, USA. E-mail: vladbotc{at}bu.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: hair cycle • bone morphogenetic protein • telogen • anagen • sonic hedgehog
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. This cyclic activity of the HF is
governed by the epithelial–mesenchymal interactions between HF
keratinocytes and fibroblasts of the dermal papilla. The HF transition
between distinct hair cycle stages is regulated by a tightly controlled
balance between numerous growth stimulatory and inhibitory factors
(1
2
3)
.
The HF transition from telogen to anagen is a unique process of organ
regeneration characterized by sudden activation of cell proliferation
in the proximal follicular epithelium (secondary hair germ and bulge)
containing HF stem cells (4
5
6
7)
. This leads to invasion of
the elongating HF into subcutaneous (s.c.) tissue that is accompanied
by reactivation of cell differentiation programs, followed by
construction of the hair matrix and inner root sheath (3)
,
as well as by differentiation of melanocyte precursors leading to
melanogenesis (8
9
10)
. Full restoration of the hair
fiber-producing unit is associated with formation of the epithelial
hair bulb surrounding the dermal papilla located deep in the s.c.
tissue (1
2
3)
.
Induction of cell proliferation in the germinative compartment of the
telogen HF leading to anagen development can be induced experimentally
by mechanical or chemical stimuli. For example, mechanical removal of
the hair shaft (depilation) induces anagen in mice and humans (1
, 3
, 11
, 12)
. Abnormal shedding of hair in mouse mutants with
constitutive deletion of the adhesion molecule desmoglein-3 or
proteolytic enzyme L-cathepsin also induces anagen (13
, 14)
. Deletion of STAT3 transcription factor is associated with a
long-term delay of the hair follicles in telogen (15)
.
Chemical stimulation of a variety molecular pathways—e.g.,
immunosuppressants (cyclosporin A, FK506), a potassium channel opener
(minoxidil), neuropeptides (substance P, ACTH), mast cell
secretagogues, or an estrogen receptor antagonist—results in anagen
induction (16
17
18
19
20)
. This suggests that the local balance
of hair growth stimulators and inhibitors in the proximal part of the
HF (bulge, secondary hair germ, dermal papilla) is critical for
initiation of a new hair growth wave.
Since HF morphogenesis and anagen development have many similarities
and both result in construction of the fiber-producing hair bulb, it
has long been suspected that both processes are regulated by similar
mechanisms (1
2
3
, 21
, 22)
. This concept was recently
supported by demonstrating that Sonic hedgehog (Shh) is a factor
controlling both HF development and telogen–anagen transition. Shh is
an essential epithelial signal that promotes HF morphogenesis in
embryonic skin. Constitutive deletion of Shh results in the arrest of
HF morphogenesis at the bud stage (23
, 24)
, whereas the
overexpression of Shh or its downstream effector Gli2 induces basal
cell carcinoma development (25
, 26)
. It was recently
demonstrated that increased intradermal expression of Shh stimulates HF
transition from telogen to anagen (27)
. Conversely, Shh
blockade by neutralizing antibody alters HF transition from telogen to
anagen in the postnatal skin (28)
.
It was proposed some time ago that telogen skin contains an inhibitor
of hair growth and that neutralization or inactivation of this
inhibitor might trigger HF transition from telogen to anagen in
postnatal skin (12)
. Bone morphogenetic proteins 2 and 4
(BMP2/4) inhibit induction of many ectodermal derivatives such as the
neural tube, tooth, or feather during embryogenesis
(29
30
31
32)
. We have recently shown that neutralization of
the inhibitory activity of BMP2/4 by the BMP antagonist noggin is also
essential for HF induction (33)
. Therefore, we
hypothesized that, as in HF induction during embryogenesis, BMP2/4
serve as important inhibitor of anagen initiation in postnatal skin and
that neutralization of BMP4 by its antagonist noggin is required for
the HF telogen–anagen transition.
To explore the roles for noggin and BMP4 in the control of HF transition from telogen to anagen, we analyzed the expression of BMP4, BMPR-IA, and noggin during the depilation-induced hair cycle in the postnatal C57BL/6 mice using semiquantitative RT-PCR, in situ hybridization, and immunohistology. The hair growth-modulatory activities of noggin and BMP4 were also tested in vivo. In addition, the expression of the selected molecules implicated in the control of the telogen–anagen transition was compared between noggin- and/or BMP4-treated skin and the corresponding controls. Using this experimental approach, we demonstrate here that neutralization of the inhibitory activity of BMP4 by the BMP antagonist noggin indeed initiates a new hair growth phase in postnatal skin.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
34
35
36)
were used to determine
noggin expression in postnatal spontaneously cycling HF by
histoenzymatic staining cryosections with ß-galactosidase. All mice
were fed water and murine chow ad libitum and kept under 12 h
light/dark cycles. Active hair growth (anagen) was induced in the back
skin during the telogen phase of the hair cycle by application of the
wax-rosin mixture with subsequent depilation, as described before
(12)
. HF telogen–anagen transformation was studied using
at least four mice per time point: telogen (unmanipulated skin), early
anagen (day 3 postdepilation), late anagen (days 8–12 postdepilation)
(12)
. In all experiments, the neck region of back skin was
harvested parallel to the vertebral line and was embedded, using a
special technique for obtaining longitudinal cryosections through the
HF from one defined site (37)
.
Semiquantitative RT-PCR
Semiquantitative RT-PCR analysis of noggin, BMP4, BMPR-IA, and
constitutively expressed ß-actin was performed as described
previously (38
39
40)
. Total RNA was isolated from
full-thickness back skin samples of C57BL/6 mice harvested at telogen
(unmanipulated skin), early anagen (3 days postdepilation), and late
anagen (8–12 days postdepilation) stages of the hair cycle. Skin was
homogenized in liquid nitrogen using a mortar and total RNA was
isolated using a single step guanidine thiocyanate-phenol-chloroform
method with RNAzol B (Biotech Laboratories, Inc., Houston, TX). cDNA
was synthesized by reverse transcription of 3 µg total RNA, using a
cDNA synthesis kit (Invitrogen, San Diego, CA). The following sets of
oligonucleotide primers were used: for ß-actin, 5'-TGG AAT CCT GTG
GCA TCC ATG AAA C-3' and 5'-TAA AAC GCA GCT CAG TAA CAG TCC G-3'; for
noggin, 5'-GAA GGA TCT GAA CGA GAC GC-3' and 5'-TTA CAC TCG GAA ATG ATG
G-3'; for BMP4, 5'-CTC CCA AGA ATC ATG GAC TG-3' and 5'-AAA GCA GAG CTC
TCA CTG GT-3'; for BMPR-IA, 5'-AGA AGC TAG CTG GTT TAG AG 3' and 5'-ATT
AGC TTC AAA ACT GCT CG-3'. Primers used were designed according to the
reported sequences in the GenBank databases. Amplification was
performed using Taq polymerase (Life Technologies, Inc.,
Grand Island, NY) over 34 cycles, using an automated thermal cycler
(Perkin-Elmer Corp., Norwalk, CT). Each cycle consisted of the
denaturing at 94°C (1 min), annealing at 60°C (45 s), and extension
at 72°C (45 s). PCR products were analyzed by agarose gel
electrophoresis and enzymatic digestion using standard methods.
Staining was densitometrically assessed with video scanner using Scan
Pack 2.0 (Biometra, Goettingen, Germany).
In situ hybridization and immunohistochemistry
In situ hybridization using Dig-labeled riboprobes for noggin,
BMP4, and Shh mRNAs was performed as described (23
, 33
, 41)
. Skin cryosections of 8- to 12-wk-old LacZ heterozygous
noggin knockout (±) females (n=4) were used for analyses of
noggin expression in postnatal HF by histoenzymatic
ß-galactosidase staining, as described previously
(33
34
35
36)
. Immunohistochemical detection of bone
morphogenetic protein receptor IA (BMPR-IA), Lef-1, and Ki-67 was
performed according to described protocols (38
39
40
, 42)
.
Rabbit antisera for Lef-1 and BMPR-1A were generated, as described
(43
, 44)
. Rabbit antiserum against murine Ki-67 was
obtained from Dianova (Hamburg, Germany). Secondary goat anti-rabbit
TRITC-conjugated immunoglobulin G (IgG) was obtained from Jackson
Immuno-Research (West Grove, PA). In all immunofluorescence procedures,
nuclei were counterstained by TO-PRO-3 (45)
. Multicolor
confocal microscope (Zeiss, Thornwood, NY) and digital image analysis
system (Pixera, Los Gatos, CA) were used for analysis and preparation
of images.
Pharmacological manipulations in vivo
Noggin protein was isolated from the supernatants of
noggin-producing CHO B3.A4 cells as described previously
(29)
; 200 µl of Affi-gel blue beads (Bio-Rad, Hercules,
CA; 100 µm in diameter) were soaked with 200 µl of mouse normal
serum (control) or 200 µl of 10 µg/ml noggin (33)
.
Beads were then injected into the back skin of mice, with all HF in the
resting stage (n=7 for the control group and n=7
for the group treated with noggin), as identified by their pink back
skin color (33)
. Skin was harvested on days 5–18 after
implantation.
Recombinant human BMP4 was expressed in Escherichia coli with a carboxyl-terminal His Tag using a modified pQE Vector (Qiagen, Chatsworth, CA) and purified under denaturing conditions on Ni-NTA Agarose (Qiagen). BMP4 was dimerized as described (N. Cerletti et al., European Patent Application 0433 225 A1; June 19, 1991) and the protein concentration was determined by SDS-PAGE; 100 µl of BMP4 in concentration 1 µg/ml was injected intradermally on days 0, 1, 3, and 5 postdepilation (n=5 for the experimental group and n=4 for the control group). Skin was harvested on day 12 postdepilation (i.e., 7 days after the last injection), when all depilated control HF had reached late anagen.
Quantitative histomorphometry
In the skin samples treated by BMP4 or noggin, the percentage of
HF at defined hair cycle stages (telogen, anagen II, anagen VI) was
assessed and defined on the basis of accepted morphological criteria
(16
, 46)
. To identify the defined substages of hair cycle
as precisely as possible, histochemical detection of endogenous
alkaline phosphatase activity was used, as this highlights the dermal
papilla as a useful morphological marker for staging HF anagen
development (47)
. For every in situ hybridization or
immunoreactivity pattern, > 150 hair follicles derived from three
animals per hair cycle stage were examined. In pharmacological
experiments, at least 50–60 longitudinal HF sections in 50–60
microscopic fields derived from 4–5 BMP4-treated animals were analyzed
and compared with those of 50–60 HF for the corresponding control
group. All sections were analyzed at 100 or 200x microscopic
magnification; means and SE were calculated from pooled
data. Differences were judged as significant if the P value
was <0.05, as determined by the independent Student’s t
test for unpaired samples.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
By semiquantitative RT-PCR, high steady-state levels of BMP4 and
BMPR-IA were detected in telogen skin, whereas very low levels of
noggin transcripts were found (Fig. 1
A–C). Hair follicle transition from telogen to late anagen
was associated with significant up-regulation (P<0.05) of
steady-state levels of noggin transcripts (Fig. 1B
).
However, steady-state levels of BMP4 and BMPR-IA mRNAs in anagen skin
remained as high as in telogen and were not changed significantly (Fig. 1A
, 1C
).
|
BMP4 transcripts were found in the dermal papilla by in situ
hybridization in the secondary hair germ of telogen HF (Fig. 2
). Dermal cells also showed BMP4 mRNA expression in telogen skin (Fig. 2A
). Noggin message was absent in telogen HF (Fig. 2B
). However, weak expression of noggin mRNA was visible in
few dermal cells (Fig. 2B
), which is consistent with the
semiquantitative RT-PCR results (Fig. 1B
). In heterozygous
noggin knockout (±) mice, the lack of noggin in telogen HF
was further supported by the absence of ß-galactosidase activity (the
gene product partially replaced noggin in these animals;
Fig. 2C
). In contrast to noggin, strong BMPR-IA
immunoreactivity was seen in the secondary hair germ, sebaceous gland,
and arrector pili muscle (Fig. 2D
, 2E
). The
expression patterns for BMP4, BMPR-IA, and noggin described above were
seen in all hair follicles examined.
|
Early steps of the HF telogen–anagen transition (anagen I-III, 3 days
postdepilation) were associated with down-regulation of BMP4
transcripts in the dermal papilla and secondary hair germ (Fig. 2F
). However, many dermal cells showed BMP4 mRNA expression
in early anagen skin (Fig. 2F
). In early anagen HF, noggin
mRNA was seen not only in the dermal papilla, but also in the secondary
hair germ and distal outer root sheath (Fig. 2G
). The
appearance of ß-galactosidase activity in the secondary hair germ and
dermal papilla of early anagen HF of LacZ heterozygous noggin knockout
(±) mice (Fig. 2H
) confirmed the latter results. This
suggested that in postnatal skin, noggin expression is not restricted
to follicular mesenchyme and can also be seen in the epithelium.
However, BMPR-IA immunoreactivity was undetectable in the secondary
hair germ (Fig. 2I
), previously reported to have numerous
proliferating cells during early anagen (48)
. Instead,
BMPR-IA was seen in those HF compartments that show a very low
proliferative activity during anagen (dermal papilla, distal outer root
sheath; Fig. 2I
). These data do not contradict the overall
constant total expression of mRNA transcripts observed by RT-PCR in
whole skin homogenates.
During late anagen (days 8–12 postdepilation), BMP4 transcripts were
seen in the proximal hair matrix, dermal papilla, and outer and inner
root sheaths of fully developed anagen HF (Fig. 2K
).
Prominent expression of noggin mRNA was observed in the entire cycling
portion of the anagen HF (hair matrix, dermal papilla, proximal outer
and inner root sheaths; Fig. 2L
). This is consistent with
our RT-PCR data demonstrating a significant increase of the noggin
message in late anagen skin compared with telogen (Fig. 1B
).
In addition, many mesenchymal cells in the perifollicular dermis and
subcutis around late anagen HF also showed noggin mRNA expression (Fig. 2L
). BMPR-IA immunoreactivity was seen in the hair matrix
and in proximal outer and inner root sheaths of late anagen HF (Fig. 2M
, 2N
). Therefore, hair cycle-associated
expression patterns of BMP4, noggin and BMPR-IA suggested involvement
of BMP signaling in the control of epithelial–mesenchymal interactions
in the HF during its transition from telogen to anagen.
Noggin induces hair growth phase in postnatal skin in vivo
To correlate the phenomenological data described above (Fig. 1
,
Fig. 2
) with the putative functional effects of noggin on the HF
telogen–anagen transition, agarose beads soaked by noggin or by normal
mouse serum as a control were injected into C57BL/6J telogen skin. The
beads provided long-term release of the noggin in close proximity to
telogen HF, and skin was studied 5–18 days after implantation.
Apparent skin color in mice is strikingly dependent on the hair cycle
stage and varies from white-pink in telogen to gray-black in anagen,
reflecting the stringent coupling of follicular melanogenesis to anagen
(46
, 49
, 50)
. Therefore, anagen development was monitored
by assessing the dynamics of skin color changes in a previously shaven
telogen skin, as well as by the emergence of new pigmented hair shafts
through the epidermis.
As shown in Fig. 3A
, noggin-treated (but not control) animals displayed anagen induction in
those areas where beads had been implanted with a change of skin color
to gray-black and in the emergence of new pigmented hair from the skin
surface. Histological analysis confirmed that noggin-induced anagen
development occurred both in tylotrich and non-tylotrich HF. In the
immediate vicinity of the noggin-soaked beads, only anagen VI
HF-producing pigmented hair was seen (Fig. 3C
). In contrast,
in skin areas 2–3 mm distant from the implanted beads, earlier anagen
stages (anagen II-IV HF) were also seen (not shown). No anagen
development was found in skin areas located at a distance of more than
3 mm from the noggin-soaked beads (not shown); no anagen development at
all was observed in the control animals, which displayed only telogen
HF (Fig. 3A
, 3B
).
|
BMP4 induces selective arrest of hair growth phase development in
the secondary (non-tylotrich) hair follicles
To test whether BMP4 inhibits anagen development in postnatal
skin, the effects of BMP4 administration on depilation-induced HF
telogen–anagen transition were assessed. BMP4 or vehicle control was
administered intracutaneously on days 0, 1, 3, and 5 after depilation,
and skin was examined 7 days after the last administration, i.e., on
day 12 postdepilation. As shown in Fig. 4
A, control mice displayed new fur, indicating that all HF
reached anagen VI (Fig. 4B
). However, BMP4-treated animals
showed only a few anagen spots and the skin remained white-pink in
color, indicating arrest of anagen development (Fig. 4A
).
BMP4-treated skin showed selective arrest of depilation-induced anagen
development in all non-tylotrich (secondary) HF (Fig. 4C
, 4D
).
|
Non-tylotrich HF represent 
90% of follicles in mouse fur and are
identified by the smaller size of the proximal hair bulb, shorter and
thinner hair, and one sebaceous gland (51)
. After BMP4
treatment, all non-tylotrich HF were arrested in telogen-anagen I,
whereas all non-tylotrich HF in the control skin reached anagen VI
(Fig. 4C
). In contrast, BMP4 did not affect anagen
development in the tylotrich (primary) HF. Tylotrich HF represent
5–10% in mouse fur and are characterized by two sebaceous glands,
larger volume of the proximal hair bulb, and longer and thicker hair
(51)
. On day 12 after depilation and 7 days after last
administration of BMP4, all tylotrich HF, similar to the control,
reached anagen VI (Fig. 4C
, 4D
).
Furthermore, tylotrich HF showed a high number of Ki-67-positive cells
in the hair matrix (Fig. 4E
), a characteristic feature of
the late anagen HF (2
, 3)
. However, non-tylotrich HF had
Ki-67-positive cells only in the distal outer root sheath and did not
show proliferating cells in the secondary hair germ (Fig. 4E
). This suggests that BMP4 selectively inhibits cell
proliferation in the germinative compartment of the non-tylotrich HF,
leading to the arrest of anagen development in this HF type.
Noggin increases Shh expression in the HF whereas BMP4
down-regulates Shh
To further explore mechanisms that may involve the reciprocal
modulation of anagen development by noggin and BMP4, the expression of
markers implicated in the control of both HF induction and hair growth
phase initiation was determined in skin treated with noggin, BMP4, or
corresponding vehicle controls. Because administration of Shh cDNA in
an adenovirus vector induces new hair growth in postnatal telogen skin
(27)
, Shh mRNA expression was examined. Lef-1 as a
downstream effector of Wnt/ß-catenin pathway was also studied because
of its expression in the bulge of the telogen HF (52)
and
because Lef-1 is down-regulated by BMP4 and up-regulated by noggin in
embryonic and postnatal HF (33
, 36)
.
The expression of these markers was compared between skin samples in
which anagen development had been induced by depilation (control
telogen/anagen I and anagen VI, days 1 and 12 postdepilation,
respectively), the samples treated with BMP4 that show selective arrest
of telogen–anagen transition and anagen development in the
non-tylotrich HF and apparently normal anagen VI tylotrich HF, and
samples treated with noggin-soaked agarose beads that display
exclusively anagen VI HF (Fig. 5
).
|
Expression of Shh mRNA was slightly above background in the secondary
hair germ and distal outer root sheath of the control telogen/anagen I
HF (Fig. 5A
). After BMP4 treatment, Shh expression was not
seen in the non-tylotrich HF at telogen/anagen I stage (Fig. 5C
). In the control anagen VI HF, Shh transcripts were seen
in a unilateral cluster of matrix cells (Fig. 5B
),
displaying the characteristic expression pattern described previously
(53)
. After BMP4 treatment, expression of Shh mRNA was
strongly down-regulated in the matrix of tylotrich anagen VI HF (Fig. 5C
). In turn, the noggin-treated non-tylotrich HF showed
ectopic bilateral expression of Shh transcript throughout the entire HF
matrix (Fig. 5D
).
The downstream effector of Wnt/ß-catenin signaling, the transcription
factor Lef-1, was seen in the dermal papilla and secondary hair germ of
the control telogen/anagen I HF (Fig. 5E
), as well as in the
precortical zone and hair matrix of the control anagen VI HF (Fig. 5F
). Also, weak Lef-1 immunoreactivity was seen in the
dermal papilla of these HF (Fig. 5F
). Compared with the
control, no alterations in Lef-1 immunoreactivity were seen in the
non-tylotrich telogen/anagen I HF treated by BMP4 (not shown). However,
the dermal papilla of the tylotrich anagen VI HF after BMP4 treatment
displayed an increase in Lef-1 expression (Fig. 5G
). Neither
BMP4 nor noggin affected Lef-1 immunoreactivity in the precortical zone
or hair matrix of anagen VI HF (Fig. 5G
, 5H
). This
suggests that Lef-1 as a transcriptional effector in the Wnt pathway is
not involved in the telogen–anagen transition and that modulation of
anagen development by BMP4 and noggin in vivo is mediated, at least in
part, by Shh.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 3)
. Here we demonstrate that, in striking similarity
to HF morphogenesis (33)
, initiation of the active growth
stage (anagen) in the resting (telogen) HF requires modulation of
BMP4/BMPR-IA signaling by noggin.
In the telogen HF, BMP4 is produced by both dermal papilla fibroblasts
and secondary germ keratinocytes and might interact with BMPR-IA, which
is selectively expressed in the secondary germ (Fig. 2)
, thus
preventing the onset of anagen development. Secondary hair germ
keratinocytes in murine telogen hair follicles also express Smad1, a
signal transduction effector for BMPR-IA (D. Lange et al., unpublished
results). This suggests that the BMPR-IA expressed in telogen HF is
functional. We further show that activation of hair growth phase in the
postnatal telogen HF is associated with up-regulation of noggin in the
HF epithelium and mesenchyme (Fig. 2)
. Since noggin binds the BMP4
molecule with an affinity 10- to 15-fold higher than does BMPR-IA
(54)
, the locally produced noggin might prevent BMP4
interaction with BMPR-IA expressed in the secondary HF germ of the
telogen HF.
Indeed, we show that noggin induces active hair growth in postnatal
telogen skin in vivo (Fig. 3)
. In contrast, BMP4 plays important
inhibitory roles during the HF telogen–anagen transition. BMP4 and
BMPR-IA are down-regulated in the germinative compartment of the early
anagen HF (Fig. 2)
, which is characterized by a high rate of
keratinocyte proliferation (48)
. Furthermore, BMP4
administration blocks both the depilation-induced anagen development
and keratinocyte proliferation in the secondary hair germ of the
non-tylotrich (secondary) HF (Fig. 4)
. These results fully support a
previous report that postnatal telogen skin contains endogenous
inhibitor (s) of anagen development or so-called chalone(s)
(12)
and identify BMP4 as one of these inhibitors.
However, BMP4 administration does not affect anagen development in the
tylotrich (primary) HF (Fig. 4)
, which make up 5–10% of all HF in
mouse dorsal skin and are characterized by a large hair bulb, long
straight hair, and two sebaceous glands (51
, 55)
. This is
not surprising, because it was recently shown that the induction of
tylotrich and non-tylotrich HF in embryonic skin requires differential
molecular pathways. Whereas induction of the tylotrich HF is strikingly
dependent on signaling through the newly identified TNF receptor
homologue Edar, the induction of the non-tylotrich HF is Edar
independent and most likely occurs via activation of the
Wnt/ß-catenin/Lef-1 pathway (56)
. This would be
consistent with a recent observation that ß-catenin acts downstream
of Eda/Edar and upstream of BMP during placode formation
(57)
.
In turn, the Wnt pathway represents an important target for BMP
regulation, because excess BMP signaling in noggin knockout mice
results in down-regulation of the Lef-1 transcription factor and noggin
overexpression leads to the ectopic Lef-1 expression in the HF
(33
, 36)
. We have recently shown that constitutive
deletion of the BMP antagonist noggin selectively affects induction of
the non-tylotrich (secondary) HF in embryonic skin (58
; V.
A. Botchkarev et al., unpublished results). This supports a concept
that Edar and Wnt/Lef-1 signaling pathways at certain stages of
development are regulated independently of each other (56
, 59)
.
However, the molecular basis of the differential response of tylotrich
vs. non-tylotrich HF to BMP4 stimulation remains to be dissected. We
found no differences in the expression of BMP4, BMPR-IA, noggin,
ß-catenin, Lef-1, and Shh between tylotrich and non-tylotrich HF in
the postnatal skin. This is consistent with data obtained from
embryonic skin where tylotrich (primary) and non-tylotrich (secondary)
HF were characterized by the essentially similar expression patterns of
BMP4, BMPR-IA, and Shh (33
, 41
, 56
, 60)
. It remains to be
elucidated whether Edar and its ligand ectodysplasin are expressed in
postnatal HF and whether signaling through Edar induces anagen
development in tylotrich HF.
In animals with a synchronized pattern of HF cycling, such as mice and
rats, anagen develops spontaneously as a wave, which is propagated from
anagen skin areas to neighboring telogen skin (3
, 11
, 61)
.
It is now well appreciated that actively growing HF secrete growth
factors into the skin and induce substantial remodeling of skin
architecture, innervation, and microvasculature (38
, 39
, 62
, 63)
. Therefore, it was logical to expect that anagen HF secretes
factor(s) capable of inducing anagen phase in neighboring telogen HF.
Shh is one such candidate anagen inducer, since temporarily
overexpressed Shh induces anagen in telogen mouse skin and
Shh-neutralizing antibody inhibits anagen development (27
, 28)
. However, as we show here, Shh is produced by the unilateral
cluster of hair matrix keratinocytes only in the anagen VI HF (Fig. 5)
.
This renders it rather unlikely that Shh, secreted by late anagen HF,
significantly influences the closely located telogen HF to also enter
into anagen.
Our data suggest that the BMP antagonist noggin is at least one other
candidate signaling molecule for anagen wave propagation. Noggin is
also expressed in the late anagen HF and is a potent anagen inducer in
vivo (Fig. 2
, Fig. 3
). In contrast to Shh, noggin is abundantly
expressed not only in the dermal papilla, but also in the epithelium of
the cyclic portion of late anagen HF, as well as in numerous connective
tissue cells surrounding HF (Fig. 2)
. Noggin expression in follicular
epithelium may in turn be controlled by BMP4, known to positively
regulate noggin in other models (64
, 65)
via its
interaction with BMPR-IA (Fig. 2)
. This invites the hypothesis that a
gradient of noggin, secreted by anagen HF and perifollicular
mesenchymal cells, neutralizes the anagen inhibitory activity of BMPs
and initiates transition of the telogen HF to anagen (Fig. 6
).
|
Most important, our data suggest that the anagen-inducing effect of
noggin is mediated at least in part by activation of Shh signaling in
the HF. We show that Shh is up-regulated in the HF after noggin
treatment, whereas BMP4 down-regulates Shh (Fig. 5)
. This suggests a
model in which noggin neutralizing BMPs represent important upstream
effectors of Shh during activation of hair growth in non-tylotrich
telogen HF (Fig. 6)
. Although the expression patterns of Shh receptors
Patched-1/2 during noggin-induced HF telogen–anagen transition remain
to be elucidated, this model is consistent with recently published
observations that excess BMP4 antagonizes Shh signaling during tooth
and limb development (66
, 67)
. However, the possible
interactions of BMP signaling with other pathways implicated in anagen
induction (e.g., estrogen receptor signaling or the STAT3 pathway;
15
, 19
, 68
) remain to be carefully dissected.
Taken together, our data suggest that the molecular mechanisms that
direct HF induction in embryonic skin and initiate anagen during
postnatal HF cycling show striking similarities and are more highly
conserved than previously appreciated. Our study encourages exploration
of BMP antagonists for anagen manipulation in human skin affected by
hair growth disorders (androgenetic alopecia, alopecia areata, telogen
effluvium), all of which are characterized by elongation of telogen and
shortening of anagen (2
, 22)
.
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 27, 2001.
Revision received June 11, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Y. Fessing, T. Y. Sharova, A. A. Sharov, R. Atoyan, and V. A. Botchkarev Involvement of the Edar Signaling in the Control of Hair Follicle Involution (Catagen) Am. J. Pathol., December 1, 2006; 169(6): 2075 - 2084. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Zhang, X. C. He, W.-G. Tong, T. Johnson, L. M. Wiedemann, Y. Mishina, J. Q. Feng, and L. Li Bone Morphogenetic Protein Signaling Inhibits Hair Follicle Anagen Induction by Restricting Epithelial Stem/Progenitor Cell Activation and Expansion Stem Cells, December 1, 2006; 24(12): 2826 - 2839. [Abstract] [Full Text] [PDF]
|
|
