Content uploaded by Ines Sequeira
Author content
All content in this area was uploaded by Ines Sequeira on Jan 08, 2016
Content may be subject to copyright.
Chapter 5
Hair Follicle Stem Cells
Emilie Legué, Inês Sequeira, and Jean-François Nicolas
Abstract The cyclic renewal of hair follicles offers
an exclusive opportunity to study stem cells behaviors,
that is their contributions and mode of maintenance.
The hair follicle renewal comprises a growth phase
of morphogenesis and elongation that is interrupted
by a phase of massive apoptosis followed by a rest-
ing phase. The hair follicle contains two types of stem
cells: stem cells that function during the growth phase
and ensure the continuous elongation of the hair folli-
cle and stem cells that maintain the hair follicle over
cycles. Lineage-tracing analyses demonstrated a so far
unexpected heterogeneity of contribution to the diverse
hair follicle lineages among the long-term stem cells.
This heterogeneity however does not reflect the hetero-
geneity of the stem cell markers identified to date and
the contribution of a cell likely depends on the position
adopted by its descendants. Moreover, recent studies
demonstrated that the pool of long-term stem cells was
highly dynamic, undergoing rounds of depletion and
replenishment by symmetrical division and contribu-
tion of cells from the previously called non-permanent
differentiated part of the hair follicle. In contrast, dur-
ing the elongation of the hair follicle, asymmetrically
dividing cells with stereotypical behaviors sustain hair
follicle growth. Their functioning is interrupted by the
intervention of apoptosis. Therefore, the hair follicle
utilizes two strategies to maintain itself. This illustrates
the notion that stemness is not so much the properties
of a cell but the properties of a system that can adopt
E. Legué ()
Memorial Sloan-Kettering Cancer Center, New York,
NY 10065, USA
e-mail: leguee@mskcc.org
various strategies to achieve its function, the long-term
maintenance of an organ.
Keywords Hair follicle ·Outer root sheath ·Stem
cells ·Growth phase ·Markers ·Dermal papilla
Introduction
The hair follicle (HF) is a skin appendage typical of
mammals. As the skin, it comprises an epithelial com-
ponent, the HF proper, and a mesenchymal component,
the dermal papilla. The HF is a mini-organ with a very
well described morphology. The mature HF at the peak
of its growth contains several epithelial cell types orga-
nized in concentric layers: the outer root sheath (ORS),
the companion layer, the inner root sheath (IRS) that
includes the Henle, Huxley and IRS cuticle layers, and
the hair shaft that includes the cuticle, the cortex and
the medulla (Fig. 5.1a). The ORS is in continuity with
the epidermis via the isthmus (at the level of the seba-
ceous glands) and the infundibulum (Fig. 5.1a). The
IRS cells serve as a sheath guiding the growth of the
hair shaft; they are present only below the sebaceous
glands and the arrector pili muscle. They become more
and more keratinicized towards the distal end of the HF
(the part closest to the sebaceous glands) and are shed
as the hair grows. The hair shaft is visible on the out-
side of the skin. At the base (or proximal end) of the
HF is the matrix, also called the bulb, that encases the
dermal papilla.
One of the striking properties of the HF is its capac-
ity to self-renew in a cyclic manner during the life of
the animal (Hardy, 1992; Müller-Röver et al., 2001)
35
M.A. Hayat (ed.), Stem Cells and Cancer Stem Cells, Volume 3,
DOI 10.1007/978-94-007-2415-0_5, © Springer Science+Business Media B.V. 2012
36 E. Legué et al.
Fig. 5.1 Hair follicle morphology and cycle. (a)Themature
HF organization. Longitudinal section of an anagen pelage HF
of a mT/mG (membrane-targeted td-Tomato) transgenic mouse
(Muzumdar et al., 2007). Cell membranes are outlined by
expression of the td-Tomato fluorescent protein. The HF is orga-
nized in several concentric layers of different cell types: the hair
shaft (composed of medulla, cortex and cuticle), the inner root
sheath (IRS, composed of the IRS cuticle, Huxley and Henle
layers) and the outer root sheath (ORS). Long-lived SCs are
believed to be located in the ORS under the sebaceous glands
in a region called the bulge. The matrix, at the proximal end of
the HF, is a proliferative zone that surrounds the dermal papilla
(DP) and contains precursors that divide asymmetrically. These
matrix SCs form a germinative layer abutting the dermal papilla
and are arranged in distinct proximal-distal sectors, each sector
contributing to a specific inner structure (IRS precursors in grey,
cuticle and cortex precursors in brown, and medulla precursors
in green). Cell fates are assigned according to the position of the
precursors along the proximal-distal axis. (b)TheHFcyclein
adult mouse back skin. The HF cycle comprises three phases:
catagen, telogen and anagen. During catagen, the lower two-
thirds of the HF regress by cell apoptosis and the dermal papilla
moves upwards along with the cells of the remaining HF. During
telogen the HF enters a resting phase, during which only a few
cells remain, forming two structures, the secondary hair germ
and the bulge; the club hair is the hair shaft from the previous
cycle. Anagen, the growth phase, starts after an exchange of sig-
nals between the dermal papilla and the remaining cells. The
bulge and secondary hair germ cells proliferate to form the new
matrix and ORS. SG sebaceous glands, DP dermal papilla, ORS
outer root sheath, IRS inner root sheath
(Fig. 5.1b). The HF cycle consists in three phases
during which its cellular organization and morphol-
ogy drastically change. The anagen is the growth phase
when the HF displays its typical organization in con-
centric layers. It is followed by catagen, a destruction
phase during which the HF regresses due to the apop-
tosis of most of the HF epithelial cells and the dermal
papilla moves upwards, staying in close contact with
the remnant epithelial cells. After catagen there is a
resting phase, telogen, when only a few cells remain
around the hair club in close proximity with the dermal
papilla. These remaining cells include the stem cells
(SCs) recruited for HF renewal.
The cells of the telogen HF were sometimes collec-
tively designated as the bulge. Historically, they were
termed hair germ (Chase et al., 1951). The bulge was
identified in human anagen HFs as a swelling of the
ORS at the level of insertion of the arrector pili mus-
cle, just under the sebaceous glands. In mouse vibrissa
and tail HF, the bulge is clearly visible. However, it is
not as prominent in anagen of the other HFs categories,
it can only be seen if the hair club from the previous
5 Hair Follicle Stem Cells 37
cycle is still present: the region where the old hair
club is attached to the new anagen HF is then called
bulge. Closer examination of the telogen HF revealed
that two distinct structures remain after catagen: the
bulge and the secondary hair germ. The secondary hair
germ is the most proximal population, closer to the
dermal papilla, whereas the bulge cells are around the
club hair. The distinction between these two structures
appears increasingly important as they are biochemi-
cally distinct and as their roles during the transition
from telogen to anagen HF and their involvement in the
maintenance of HF SCs may be different (see below).
During catagen, the HF regresses up to the region
just under the sebaceous glands. Therefore, the matrix
and the part of the HF organized in concentric layers
is often called the “non-permanent part” of the HF,
in contrast with the “permanent part” of the HF that
designates the cells around the hair club just below
the sebaceous glands, the isthmus and the infundibu-
lum. However, as discussed below, recent studies have
unveiled that cells from the non-permanent part of the
HF could contribute to the so-called permanent part
showing an interesting dynamic of maintenance of the
HF SCs.
Throughout the cycle, dermal papilla cells always
stay in close contact with some HF epithelial cells.
There is a constant exchange of signals between these
two components of the HF and this is essential for the
maintenance of the integrity of the HF, the initiation of
anagen and the growth of the HF. The dermal papilla
is therefore an essential component of the HF dynam-
ics, potentially playing a role as an organizing center
(Legué and Nicolas, 2005; Legué et al., 2010).
The HF is an excellent model to study SC behav-
iors, both their contributions and mode of growth and
renewal, thanks to its capacity to self-renew and to
regenerate its several cell types at each cycle. The tran-
sition from telogen to anagen allows for studying the
contribution of the SCs to each cell type of the HF and
test whether the HF SCs are one homogeneous popula-
tion or a more complex pool of precursors. Moreover,
anagen is not only a phase of growth but also a phase
of morphogenesis when the various cell types of the
mature HF are produced and arranged to build the
typical HF organization. The transition between the
SCs and the precursors responsible for the growth of
the HF via a phase of organogenesis during anagen
makes the HF a very amenable model to study lineage
relationships from SCs to the multiple differentiated
cell types they produce. Finally, the reiteration of the
cycles provides a unique chance to assess how the
pool of HF SCs is maintained from one cycle to the
other. The classic view of SCs is that it is a popula-
tion that is set aside, divides rarely and is maintained
over long term; in this view, the SCs are also at the
top of the lineage hierarchy giving rise to descen-
dants whose contributions are progressively restricted.
However, recent studies of HF SCs and of other sys-
tems have challenged this SC classic model and tend
to suggest that the SCs are a more dynamic pool of
cells, and that there may be a more stochastic selec-
tion of the SCs than a stereotypical progression from a
multipotent cell.
In this chapter, we discuss the behavior of HF SCs
that can be identified during the growth phase of the HF
and the SCs that ensure HF renewal over cycles. We
will first focus on the growth phase precursors at the
origin of the HF organogenesis and growth, and then
we will describe in detail the organization of the HF in
the resting phase where the cells recruited for the next
cycle are localized. Next, we will review the current
knowledge about the contributions of these cells and
present the models that are under debate. Finally, we
will discuss how the HF illustrates the concept(s) of
stemness in parallel with other models.
Hair Follicle Stem Cells During
Growth Phase
At each cycle, a new HF is produced during ana-
gen (Fig. 5.1b). This is a genuine morphogenesis: at
each anagen, the entire HF structure (proximal matrix
cells encasing the dermal papilla, differentiated cells
arranged in concentric layers) needs to be regenerated.
Moreover, the anagen HF is a dynamic structure as
it is constantly growing during anagen. This results
in the upward elongation of the hair shaft outside the
epidermis and also in the downward elongation of the
HF under the epidermis, into the dermis. Therefore, it
entails the production and the spatial arrangement of
the differentiated cells into concentric layers.
During elongation of the HF, the dividing cells are
located in the matrix and in the ORS. Moreover, using
temporal clonal analysis, it was shown that the matrix
contains the precursors for the internal layers of the HF
(IRS, cuticle, cortex and medulla) (Fig. 5.1a) (Legué
38 E. Legué et al.
and Nicolas, 2005). The precursors for the ORS are
present in the ORS itself. Furthermore, this analy-
sis revealed that the precursors in the matrix have
highly stereotyped behaviors. First, these precursors
have very restricted contributions (to the IRS, or to
the cuticle and the cortex, or to the medulla). Second,
they are arranged at different proximal positions in
the matrix along the dermal papilla according to the
layer they contribute to, such that the most proximal
precursors contribute to the outermost layer of the
internal structures (the IRS) and the most distal precur-
sors contribute to the most internal layer (the medulla)
(Fig. 5.1a). Third, they produce the differentiated cells
of the HF through a stem cell mode of growth. The
precursors abutting the dermal papilla divide through
asymmetrical division generating a new precursor
abutting the dermal papilla and a cell that will divide
once or twice (a transient amplifying cell) before gen-
erating differentiated cells. Therefore, the restricted
precursors for the internal cell types of the HF form
a germinative layer of permanent self-renewing cells
abutting the dermal papilla. The functioning of the per-
manent precursors in the germinative layer lasts as long
as anagen, it is thus restricted to the elongation phase
of the HF, and these precursors are not contributing to
the renewal of the next HF (Greco et al., 2009) (Inês
Sequeira, unpublished results).
These behaviors account for the production of the
HF differentiated epithelial cells and the elongation
of the HF. However, in the beginning of the anagen,
the new matrix and ORS have to be formed. This is
the step of morphogenesis during anagen. Early on,
the ORS and the inner layers lineages are separated.
Interestingly, a transient pool of oligopotent precur-
sors that eventually contribute cells to several inner
layer cell types is established in the beginning of ana-
gen. The precursors of this transient pool generate
cells that distribute at different levels of the germi-
native layer. However, not all the precursors transit
through this pool, some display contribution to only
one layer, probably due to the fact that their descen-
dants may contribute cells to only one position in the
germinative layer. Therefore, during the early steps
of anagen, the contributions of the SCs giving rise to
the precursors present in the germinative layer are not
stereotyped, they most likely depend on the positions
in relation to the dermal papilla that the precursors
and their descendants occupy as morphogenesis of the
matrix and ORS proceeds. So far, there is no evidence
that the positions of the descendants of early anagen
precursors are predetermined. The final positions of
the descendants of early anagen precursors depend on
the number of cells produced by a given precursor and
the level of intercalation of the descendants cells. The
higher the level of intercalation and the number of cells
produced by a precursor, the higher the diversity of
positions occupied by its descendants in the germina-
tive layer along the dermal papilla. It then determines
the probability for a given precursor to receive less or
more diverse signals from the dermal papilla that trans-
late into less or more diverse contributions. In such
a model, any given cell can contribute to any (and
potentially all) internal structure; however, their final
contribution is determined by their position. This will
be further discussed below in the section dedicated to
the contributions of the cells present in the telogen HF
to the anagen HF.
The cells at the origin of the anagen HF and the
cells sustaining the elongation of the HF can therefore
be considered as SCs functioning during anagen. Their
functioning is limited to one anagen by the intervention
of catagen apoptosis. These anagen SCs are different
from the SCs present in telogen that are maintaining
the HF over the cycles.
Organization of Hair Follicle Long-Term
Stem Cells
Anagen lasts for 2–2.5 weeks in mouse, followed by
the phase of destruction – catagen – during which
most of the anagen HF cells die by apoptosis. The
infundibulum and the isthmus are not affected by apop-
tosis in catagen. A few bulge cells die (Ito et al., 2004)
but most of them are spared. The hair germ is not
present during anagen, this structure becomes visible
at the end of catagen (Ito et al., 2004). Over the last
decade, new biochemical markers have been identified
that are expressed in specific populations of cells in the
telogen HF.
The infundibulum and isthmus cells are in
continuity with interfollicular epithelial cells and they
express the same markers, such as K14 and K5.
However, subdomains can be identified by expression
of specific markers. The most distal domain of the
infundibulum expresses Sca1. These cells do not con-
tribute to the renewal of the HF (Jensen et al., 2008),
5 Hair Follicle Stem Cells 39
therefore they mark the limit of the HF compartment
and is not considered to be a HF SC marker. Lrig1
defines a junctional zone (that includes the lower part
of the infundibulum and the upper part of the isth-
mus) (Jensen et al., 2009), and the cells of the upper
part of the isthmus specifically express the marker
MTS24 (Nijhof et al., 2006) (Fig. 5.2b). In the upper-
most isthmus, there is an additional population that
expresses Blimp1 (Horsley et al., 2006). These cells
are located at the opening of the sebaceous gland canal.
Below the upper isthmus defined by the expression
of MTS24 and Lrig1, and above the bulge there is
a population of cells that expresses specifically Lgr6
(Snippert et al., 2010b), defining the central isthmus.
There is a limited overlap between the Lgr6 cluster
and the MTS24 and Lrig1 population (Snippert et al.,
2010b).
The secondary hair germ and the bulge are two
anatomically distinct structures. The secondary hair
germ cells specifically express P-cadherin. However,
the bulge and the secondary hair germ share some
markers. They both express Keratin 15 (K15) (Liu
et al., 2003; Lyle et al., 1998; Morris et al., 2004)
(Fig. 5.2b). K15 expression does not overlap with
Lgr6 in the isthmus allowing for a clear differenti-
ation between bulge/secondary hair germ region and
the isthmus region. The bulge and the secondary hair
germ also express CD34, however, only the most dis-
tal cells of the secondary hair germ seem to express
CD34 indicating a heterogeneity within the secondary
hair germ population. Lgr5 is another marker shared
by bulge cells and secondary hair germ cells; however,
only the most proximal cells of the bulge are posi-
tive for Lgr5 (Jaks et al., 2008). Therefore, within the
K15 population, three subsets of cells can be distin-
guished based on marker expression: the CD34 only
cells, the CD34/Lgr5 cells, the Lgr5 only cells. The
CD34/Lgr5 population encompasses both the bulge
and secondary hair germ, and may be indicative of a
transition between these two populations.
K6+
Lgr5
CD34
Lgr6
K15
Blimp1
Lrig1
Sca1
Lrig1
MTS24
Lgr6
Blimp1
CD34+ and
α6-integrin+
Lgr5+
Pcad+
L
gr5
C
D
34
L
gr6
K
1
5
Bli
m
p1
L
r
ig
1
Contribution of the HF stem cell subpopulations
DP
DP
SG
secondary
hair germ
bulge
dermis epidermis
20µm K15+
Lineage tracing Reconstitution assays
Unipotent sebaceous gland
progenitors
Sebaceous glands
Interfollicular epidermis,
infundibulum, sebaceous
glands and isthmus
All skin lineages
Interfollicular epidermis,
isthmus, sebaceous glands,
and occasionally HF
All skin lineages
All skin lineages All skin lineages
nd. All skin lineages
Isthmus and HF All skin lineages
bulge
secondary
hair germ
SG
ABC
(Horsley et al., 2006)
(Jensen et al., 2009)
(Snippert et al., 2010)
(Morris et al., 2004)
(Blanplain et al., 2004)
(Jaks et al., 2008)
infundibulum
isthmus
Fig. 5.2 The long-lived hair follicle stem cells. (a) The telo-
gen HF organization. Longitudinal confocal optical section of a
telogen pelage HF of a mT/mG transgenic mouse (Muzumdar
et al., 2007). Cell membranes are outlined by expression of
the td-Tomato fluorescent protein. (b) HF SC markers and their
individual localization in telogen. Each SC subpopulation is
illustrated by their distinct gene/protein-expression: P-cadherin
(Pcad) is expressed in the secondary hair germ (red circle ),
CD34 and α6-integrin (high or low expression) in the outer layer
of the bulge (violet circle), Lgr5 is expressed in the secondary
hair germ and lower bulge cells (light green), K15 is expressed
both in the bulge and secondary hair germ (orange stripes), and
K6 subpopulation is restricted to the innermost layer of the bulge
which is also CD34–(grey) and they do not contribute to HF
renewal. Located more distally are the Lgr6 subpopulation in the
central isthmus (dark green), the MTS24 and Lrig1 in the upper
isthmus and lower infidibulum (light blue) and, restricted to the
sebaceous gland duct entrance, the Blimp1 subpopulation (red
stripes). Sca1 is expressed in the cells of the upper infundibu-
lum (dark blue). The Sca1 cells never contribute to HF renewal
and thus delineate the distal limit of the HF compartment. SG
sebaceous glands, DP dermal papilla. (c) Contributions of the
HF stem cell subpopulations to the HF and epidermis, according
to lineage tracing studies and to reconstitution assays
40 E. Legué et al.
Within the CD34-positive population, two sub-
populations have been identified: a basal popula-
tion expresses high levels of α6-integrin whereas
a suprabasal population expresses low levels of
α6-integrin (Blanpain et al., 2004).
K15, Lgr5, CD34 as well as Lgr6, MTS24 and Lrig1
are often referred to as HF SC markers (Horsley et al.,
2006; Jaks et al., 2008; Jensen et al., 2009; Lyle et al.,
1998; Morris et al., 2004; Nijhof et al., 2006; Snippert
et al., 2010b; Trempus et al., 2003). Several types of
experiments were used to identify the cells expressing
these markers as potential HF SCs.
K15, Lgr5 and Lgr6 expressing cells have been
genetically fate-mapped and their progeny contributed
to the anagen HF (Jaks et al., 2008; Morris et al.,
2004; Snippert et al., 2010b). Moreover, K15 and Lgr5
expressing cells were tested for colony-forming abil-
ity and they both efficiently formed colonies that could
be successfully passaged in vitro. Furthermore, both
the K15 and Lgr5 cells were able to reconstitute HFs
and all skin lineages in skin reconstitution assays.
Expression-profiling of K15 and Lgr5 cells showed
that other putative SC markers were expressed in these
populations: for instance K15 cells express CD34 and
β1-integrin, Lgr5 cells express CD34 and K15.
The two CD34 sub-populations, high α6-integrin
and low α6-integrin expressing cells, have different
expression profiles, however, they both form colonies
in vitro and reconstitute all skin lineages in reconstitu-
tion assays, identifying CD34 cells as a whole as HF
SCs (Blanpain et al., 2004). Moreover, the CD34 pop-
ulation contains label-retaining cells (LRCs), that are
cells that infrequently divide which is thought to be a
hallmark of SCs (Tumbar et al., 2004). Sox9 has also
been identified in the bulge, overlapping the CD34+
cell population. However, it is also expressed in the
ORS cells during anagen (Vidal et al., 2005).
The MTS24 population also contains LRCs:
MTS24 cells can successfully reconstitute skin and
their expression profile is similar to the CD34 popula-
tion expression profile (Nijhof et al., 2006). However,
their position in the telogen HF makes them an unlikely
HF SC population. Lrig1 overlaps in the isthmus with
MTS24, and as the MTS24 population, the Lrig1 cells
can reconstitute all skin lineages but lineage-tracing
analysis of Lrig1 cells showed that they contributed
to the junctional zone, the sebaceous glands, the
infundibulum and the interfollicular epidermis, and not
to the HF (Jensen et al., 2009).
Interestingly, even though LRCs have been a way
to identify SCs – this was how the bulge was primar-
ily identify as the HF SC reservoir (Cotsarelis et al.,
1990) – some bona fide HF SC populations do not con-
tain LRCs. Lgr5 and Lgr6 cells which both contribute
cells to the anagen HF are not LRCs. This challenges
a stem cell dogma and indicates that HF SCs are more
dynamic than previously thought. This will be further
discussed below.
Hair Follicle Stem Cells Have
Heterogeneous Contributions
As presented above, the cells of the telogen HF that
contribute to the next anagen HF can be divided into
several and sometimes overlapping populations. An
obvious question is whether these populations have
or not similar contributions to the various HF cell
types. To address this question, it would be neces-
sary to fate-map each of these populations to assess
their contributions. In some cases, it had been done but
these studies failed to reveal any restricted contribu-
tions of a population identified by a specific marker.
Genetic fate-mapping of the K15, Lgr5 and Lgr6 pop-
ulations showed that these cells contributed to all HF
lineages (Jaks et al., 2008; Morris et al., 2004; Snippert
et al., 2010b). Reconstitution assays showed that K15,
CD34, Lgr5, MTS24 and Lrig1 cells were also able
to contribute to all HF structures but this assay might
not be well-suited for assessing in vivo contributions
(Blanpain et al., 2004; Jaks et al., 2008; Jensen et al.,
2009; Morris et al., 2004; Nijhof et al., 2006).
Does this mean that each cell in the telogen HF
can contribute to all anagen HF structures? The fate-
mapping analyses and the reconstitution assays tested
the potential of the cells at the population level,
however, is each cell within a specific population mul-
tipotent? And more broadly, is each stem cell present
in telogen multipotent?
Grafting experiment of cells generated in vitro by
a single cell isolated from the upper part of the HF
showed that although most of the grafted cells were
contributing to all HF lineages, in some cases the
grafted cells contributions were restricted to a few or
one cell type (Claudinot et al., 2005). This suggested
that some HF SCs might have restricted contributions.
Clonal fate-mapping studies unequivocally showed
5 Hair Follicle Stem Cells 41
that this was the case. The labeling of a unique pre-
cursor gave rise to labeled cells contributing to a few
or even a single HF structure (Jaks et al., 2008; Legué
et al., 2010; Zhang et al., 2009). Surprisingly, the con-
tributions of the SCs are highly heterogeneous, some
precursors contribute to only one cell type whereas
some are multipotent and all the intermediate com-
binations are represented. This shows that the pool
of HF SCs is highly heterogeneous in terms of the
final contributions of its precursors. It is unlikely that
the heterogeneity of the HF SCs is determined by the
expression of a specific marker since the fate-mapping
of at least some populations did not reveal a specific
contribution. It is probable that a clonal analysis within
a specific population of HF SC would show hetero-
geneity of the contributions of the precursors within
this population, as did the clonal analyses of all the
HF SCs. However, maybe the proportions of each cat-
egory of precursors would change when comparing a
population to another. Alternatively, one marker might
not be sufficient to define a restricted population, inter-
sectional fate-mapping of precursors expressing two
markers may then reveal the existence of populations
with distinct contributions.
But again, as for the precursors present at the begin-
ning of anagen, the telogen cells recruited to initiate
the anagen might potentially be able to contribute to
all HF structures, and the more restricted final contri-
butions observed in some cases would be due to the
fact that the descendants of some of the telogen HF
SCs occupy less diverse positions and receive a more
homogeneous set of signals.
Recently, additional markers for the upper parts
of the telogen HF have been identified. The Lgr6
population is the more proximal of these markers
(Fig. 5.2b). It contributes cells to the anagen HF
during the first cycle of renewal, however, it seems
that this population loses its capacity to contribute
to the anagen HF after a few cycles (Snippert et al.,
2010b). Fate-mapping analysis showed that this pop-
ulation contributes mainly to the sebaceous glands
and the inter-follicular epidermis. Above is a popula-
tion that expresses Lrig1 and MTS24 (Jensen et al.,
2009; Nijhof et al., 2006). The Lrig1 cells contribute
to the isthmus, the sebaceous glands, the infundibulum
and the inter-follicular epidermis (Jensen et al., 2009).
Some cells within the Lrig population express Blimp1,
defining a population that specifically contributes to
the sebaceous glands (Horsley et al., 2006). This does
not exclude that some cells Lrig+Blimp1–may con-
tribute to the sebaceous glands, but Blimp1 is a marker
of a unipotent precursors population exclusively con-
tributing to the sebaceous gland. Therefore, there
might be a certain level of correspondence between the
populations defined by markers and their contributions
in the upper part of the telogen HF. If the contribu-
tion to the anagen HF is considered as one fate, then
there is a gradient of contributions: the Lgr5 population
contributes to the HF and the isthmus, the Lgr6 popula-
tion contributes marginally to the HF and mainly to the
isthmus, the sebaceous glands, the infundibulum and
the interfollicular epidermis (Snippert et al., 2010b),
and the Lrig1/MTS24 cells contribute to the isthmus,
the sebaceous glands, the infundibulum and the inter-
follicular epidermis (Jensen et al., 2009) (Fig. 5.2c).
However, when put out of their physiological con-
text (in reconstitution assays for instance) or during
wound-healing, the contributions of these populations
are broader, the change of context maybe bringing
them in contact with signals that are not reaching them
under physiological conditions.
Models for Hair Follicle Renewal
The recent advances in the understanding of HF SCs
markers and contributions lead to increasingly detailed
models of the cell dynamics during HF renewal.
In the early 1990s, Sun et al. (1991) proposed the
bulge activation model, in which the slow cycling
SCs in the bulge would respond to activating signals
coming from the dermal papilla and initiate the mor-
phogenesis and the growth of a new anagen HF (Sun
et al., 1991) (Fig. 5.3a). At the time, the bulge had
just been identified as the potential location for SCs.
The bulge was shown to contain label-retaining cells
(LRCs) (Cotsarelis et al., 1990), a slow cell cycle being
considered a hallmark of long-lived SCs. In this model,
bulge cells are slow-cycling cells that can divide upon
activation by dermal papilla signals. They provide the
cells necessary to generate the matrix, whose cells
divide actively during anagen but transiently, since the
matrix is lost during catagen. The upward movement
of the dermal papilla during catagen is necessary to
ensure close contact between bulge cells and the der-
mal papilla, and thus the activation of the bulge cell
in the telogen HF (Panteleyev et al., 1999). Although
42 E. Legué et al.
Fig. 5.3 Models for hair follicle renewal (a) The bulge acti-
vation hypothesis, proposed by Cotsarelis et al. (1990). The
bulge region is the location of the HF SCs (LRCs). At the onset
of the anagen, signals from the dermal papilla cells (orange
arrow) activate the bulge cells to proliferate and form the new
HF. (b) The hair follicle predetermination hypothesis, proposed
by Panteleyev et al. (2001). In anagen I, dermal papilla sig-
nals (orange arrow 1) first activate the secondary hair germ
cells. The proliferation of hair germ cells induces the prolifer-
ation the bulge SCs (orange arrow 2). In anagen III, the active
downward growth of bulge-derived cells results in the forma-
tion of the ORS (white arrows), and the upward proliferation
of hair germ cells (white arrows) results in formation of the
ascending compartment of the internal HF layers (hair shaft and
IRS). Recent studies confirmed this hypothesis and proposed a
two-step mechanism for HF stem cells activation (Greco et al.,
2009). (c) Recycling stem cells models from Jaks et al. (2008)
and Hsu et al. (2011). The ORS cells in anagen contribute to the
secondary hair germ and to the bulge. Lgr5+cells labeled in the
ORS during anagen (left panels, green) survive catagen apopto-
sis and their progeny contribute to the secondary hair germ and
to the mid and lower bulge (Jaks et al., 2008). Hsu et al. (2011)
(right panels) propose that ORS cells that divided only a few
times (ORShigh) are located more distally and contribute to the
bulge (blue), ORS cells that divided a few more times (ORSmid )
contribute to the secondary hair germ (red) and finally proximal
ORS cells that actively divided during anagen (ORSlow) con-
tribute to the innermost layer of the bulge expressing K6 (grey)
(Hsu et al., 2011). DP dermal papilla, ORS outer root sheath
the signals that trigger anagen initiation are still not
known, this model remains a strong basis on which
newer models were proposed. However, this model
does not address the roles of the bulge versus the
secondary hair germ, nor does it address how the
pool of SCs is maintained over several cycles or the
contributions of the SCs to the various cell types of
the HF.
Based on an extensive review of the HF literature
up to 2001, Panteleyev et al. (2001) proposed the hair
follicle predetermination hypothesis. In this model, the
secondary hair germ and the bulge have distinct roles.
The secondary hair germ, located closest to the dermal
papilla, is responsible for the production of the cells
that form the matrix and consequently the internal
layers of the HF, while the bulge cells generate the
ORS (Panteleyev et al., 2001) (Fig. 5.3b). In mature
anagen, some bulge-derived cells have the ability to
migrate downwards through the ORS (Oshima et al.,
2001) and reach the matrix as shown by the isolation
of clonogenic cells from different portions of the ORS
and the fate-mapping of bulge-grafted cells. The pre-
determination model hypothesizes that bulge-derived
cells reaching the matrix form the lateral disc in late
anagen. During catagen, the lateral disc cells survive
the massive apoptosis and form the secondary hair
germ. Compared to the bulge activation hypothesis, in
the predetermination model, the cells of the secondary
hair germ are receiving the activation signals from the
dermal papilla instead of the bulge cells. Bulge cells
become activated subsequently by the secondary hair
germ cells (Panteleyev et al., 2001). In this model, each
cell population has a stereotypical or predetermined
fate with the secondary hair germ cells giving rise to
the internal layers and the bulge generating the ORS at
anagen initiation; during middle-late anagen, the bulge
5 Hair Follicle Stem Cells 43
cells migrating along the ORS form the lateral disc
and during catagen the lateral disc resists apoptosis
and forms the secondary hair germ of the subsequent
telogen.
As hypothesized by the predetermination model,
the secondary hair germ cells are indeed the first to
respond to dermal papilla signals and they prolifer-
ate first during anagen (Greco et al., 2009). This was
termed the “two-step mechanism for stem cell acti-
vation”. It is therefore tempting to propose that the
proximity of the secondary hair germ cells with the
dermal papilla is at the origin of the differential tim-
ings of activation between the bulge and the secondary
hair germ cells. Moreover, it was shown that telogen
HF individual cells could have restricted contributions
with some cells generating descendants exclusively in
the ORS and some exclusively in the internal layers.
Again, it is tempting to postulate that the proximity
to dermal papilla signals might influence the fate and
lineage decisions during early anagen (Legué et al.,
2010): secondary hair germ cells would preferentially
contribute to the internal layers of the HF while more
distal cells, i.e. bulge cells, would preferentially con-
tribute to the ORS. However, within the telogen HF,
the segregation between the ORS lineage and internal
lineages was not complete whereas the predetermi-
nation model implies a strict segregation of the two
lineages. Instead a more stochastic, as opposed to pre-
determined, choice of fate might happen: as anagen
proceeds and cells arrange spatially in relation to the
dermal papilla, descendants of the telogen cells have
more chances to contribute to the internal structures if
they are close to the dermal papilla or more chances to
contribute to the ORS if they stay away from it.
Another postulate of the predetermination model is
that cells from the lateral disc survive catagen to form
the secondary hair germ. However, this was invali-
dated by following the fate of the cells of the lateral
disc. These cells express Sonic Hedgehog (Shh) among
other markers. Genetic marking of Shh-expressing
cells in anagen and their fate-mapping in telogen
showed that they did not survive catagen (Greco et al.,
2009). Therefore the origin of the secondary hair germ
is not the Shh expressing cells of the lateral disc.
The origin of the secondary hair germ and its
relationship to bulge cells is still a matter of debate.
According to Greco et al. (2009) and Ito et al.
(2004) the secondary hair germ forms at the transi-
tion between catagen and telogen, bulge cells are at the
origin of the secondary hair germ (Greco et al., 2009;
Ito et al., 2004) and the number of cells in the hair germ
(determined by P-cadherin expression) stays constant
during telogen (Greco et al., 2009). However, Zhang
et al. (2009) propose that the secondary hair germ
increases its cell number at the transition between telo-
gen and anagen by translocation of cells from the bulge
into the secondary hair germ (Zhang et al., 2009). All
these studies suggest that the secondary hair germ cells
originate from the bulge, but they lack definitive evi-
dence such as lineage tracing analysis especially to
confirm the formation of the secondary hair germ from
bulge cells at the end of catagen.
Interestingly, Zhang et al. (2009) showed that bulge
cells were supplying cells to the secondary hair germ
without dividing thus depleting the pool of bulge cells
at the end of telogen. They moreover showed that sym-
metrical divisions of the bulge cells during anagen
replenish the bulge without contribution to the rest of
the anagen HF (Zhang et al., 2009). This was the first
evidence for the mode of maintenance (or self-renewal)
of the bulge population at the single cell level and it
was challenging the view that bulge cells were a per-
manent population whose cells divided asymmetrically
to generate a new bulge (stem) cell and a transient
amplifying cell.
An even more dynamic view of the bulge and sec-
ondary hair germ populations emerged recently from
growing evidence for contribution of cells from the
so-called non-permanent anagen HF to the secondary
hair germ and the bulge. The first indication that ana-
gen cells escaped catagen apoptosis came from the
fate mapping of Lgr5 expressing cells in anagen (Jaks
et al., 2008). Lgr5 cells are located in the ORS, which
is an actively cycling population during anagen, how-
ever, they (or their progeny) contribute to both the
hair germ and the bulge in the subsequent telogen HF
(Fig. 5.3c). As the predetermination model, this model
suggests that cells from the non-permanent part of the
HF escape apoptosis and contribute cells to the telogen
HF. However, contrasting with the predetermination
model, these cells are not located in the matrix but in
the ORS and they contribute not only to the secondary
hair germ but also to the bulge. This further indicates
that the pool of bulge cells is not permanent in the
sense that new cells are added from the non-permanent
part at each cycle.
Finally, following the fate of cells in relation to the
number of times they divide, Hsu et al. (2011)were
44 E. Legué et al.
able to show that cells that contribute to the anagen
ORS and divide slowly could survive catagen and con-
tribute to the bulge and the secondary hair germ of the
next telogen HF, implying that a new bulge and sec-
ondary hair germ are formed after each catagen. They
show that the bulge is a composite between “old” bulge
cells and new added cells from the ORS and that the
potential of ORS cells to contribute to the bulge dimin-
ishes with the number of divisions they undergo. The
cells of the more distal part of the ORS that divide the
least are more likely to contribute to the bulge; the cells
located in an intermediate part of the ORS that divide a
few more times preferentially contribute to the forma-
tion of the secondary hair germ after catagen; finally,
the cells of the more proximal part of the ORS are the
cells that divided most and they do not contribute to
the bulge nor the secondary hair germ. However, some
of them escape apoptosis and contribute to the inner-
most layer of cells around the club hair that express
K6 but not CD34 and are not involved in generating
cells during the next anagen (Fig. 5.3c). These K6 cells
may instead constitute a niche and provide signals such
as FGF18 and BMP6 that maintain the HF in telogen
(Hsu et al., 2011). Contribution of the ORS cells to the
telogen bulge or hair germ was also suggested by the
HFs lacking Sox9. In Sox9 mutants, the first HFs were
formed but lacked ORS and they never formed a proper
bulge after regression (Nowak et al., 2008; Vidal et al.,
2005).
The latter studies (Hsu et al., 2011; Jaks et al., 2008)
suggest that the bulge population is more dynamic than
previously thought and that the terms “permanent” and
“non-permanent” parts of the HF might be inappro-
priate since the so-called permanent part of the HF is
in fact changing from one cycle to the other and that
the non-permanent part supplies cells whose progeny
contributes to the next cycle and potentially to several
cycles.
Stemness: Set of Properties of a Cell
or of a System?
Historically, the study of the renewal of the cells
of the hematopoietic system has lead to a domi-
nant model for tissue homeostasis (Gilbert, 1997).
Following this model an undifferentiated, multipotent,
slow cycling cell, the hematopoietic SC is at the top
of a cellular hierarchy with, down, more restricted
but still multipotent lymphoid and myeloid SCs, then
transient amplifying cells and finally differentiated
post-mitotic cells. Furthermore, the model considers
that all hematopoietic SCs are equivalent, that their lin-
eages are stereotyped and that the transitions of cell
states are irreversible. The notion that a niche (the
osteoblastic niche of the bone narrow) is actively main-
taining the hematopoietic SCs in a quiescent state was
a crucial addition to this model (Schofield, 1978)as
it explained the maintenance of an hematopoietic SC
compartment for life.
This model and the associated concepts have been
very influential and useful to help deciphering the
renewal of other systems that then were studied
(Weissman et al., 2001). This applied to the HF sys-
tem as evident from the above section on the models.
Recently however, as described in this chapter, the
detailed analysis of critical biochemical markers and
the use of increasingly sophisticated techniques for lin-
eage tracing have revealed an unexpected complexity
of the strategy of renewal in the HF system and has
lead to proposals that challenge several of the elements
of the initial hematopoietic SC model.
A first paradox in the HF concerns the cells in the
germinative layer of the matrix (Legué and Nicolas,
2005). Obviously, these cells exhibit the most impor-
tant canonical properties of SCs. They are undifferenti-
ated; they are at the top of a hierarchy: they self-renew
and divide asymmetrically, and they produce transient
amplifying cells whose daughter cells ultimately dif-
ferentiate; their state is probably actively maintained
at long term by signals from the dermal papilla that
therefore can be considered as their niche. However,
at the end of the anagen, they all undergo apoptosis
induced by signals extrinsic to the matrix germinative
layer and, as a consequence, they do not participate
to the next cycles. Formally and operationally, there-
fore, they cannot be considered (and generally are
not) as long-term SCs that are maintained for life.
Nevertheless, this conclusion may be misleading. It
could well be that their intrinsic character is that of a
true stem cell and their apoptosis being only circum-
stantial, brought about by the initiation of catagen. This
hypothesis could be tested by analyzing their long-
term maintenance when abolishing apoptosis. If the
matrix germinative layer cells are true long-term SCs,
this illustrates an unusual situation where the true stem
character of a cell is masked by an extrinsic operation.
5 Hair Follicle Stem Cells 45
This hypothesis further raises an intriguing possibil-
ity that, in other systems, intrinsically true SCs may
exist in places ignored simply because some extrinsic
signals interfere with their long-term maintenance.
A second paradox, maybe even more profound, con-
cerns the long-term SCs. As reported in this chapter,
it was found that these cells have a typical epithelial
morphology and express keratins such as K14 as well
as other markers that indicate that they are close to the
cells of the basal layer of the epidermis and to the ORS
cells, in contradiction with the concept of SC undif-
ferentiation. Also, we have discussed that the concept
of homogeneity of the SCs did not resist the careful
analysis of the bulge cells for a number of molecular
markers and cellular behavior (Blanpain et al., 2004;
Hsu et al., 2011; Jaks et al., 2008; Legué et al., 2010;
Zhang et al., 2009). Indeed, the telogen HF contains
several distinct cell populations of which at least three
were definitively shown by lineage tracing analysis to
contribute to all cell types of the renewed HF (Jaks
et al., 2008; Morris et al., 2004; Snippert et al., 2010b).
Furthermore, as several of these cell populations do
not overlap with the LRCs category, it was suggested
that HF SCs actively cycle. Finally, it was shown that
cells in the bulge divide symmetrically and generate
after division two daughter cells that acquire similar
fates suggesting that HF SCs are not maintained by
asymmetrical division but by rounds of depletion and
replenishment (Zhang et al., 2009). These new infor-
mation suggest that the SCs that renew the HF exhibit
a high degree of molecular and cellular heterogeneity
and do not exhibit the believed properties of true long-
lived SCs. At first, this indeed suggests that most of
the concepts derived from the hematopoietic system do
not apply to the HF system. However before discussing
how to resolve this contradiction, it is worth mention-
ing that the possibility remains that in the future, the
complexity observed will be resolved by the discovery
of a so far ignored hierarchy between the different tel-
ogen cell populations with at the top a true quiescent
SCs that divide asymmetrically. The definitive test of
this possibility will require establishing the complete
hierarchical relationships of the bulge cells using strict
single (molecular and clonal) cell analyses.
If there are no true SCs in the telogen HF in
the sense of the hematopoietic SC, we have to face
the idea that none of the so-called canonical stem
cell properties is really necessary for the HF renewal
cells. However, the difficulties raised by this puzzling
situation can be resolved if one takes into account that
in fact what matters for the organism is not to have an
undifferentiated quiescent SC that divides asymmetri-
cally but to have a renewing process that fulfills the
functions necessary for differentiation and long-term
maintenance. The renewing cell can be differentiated
if thereafter the renewing process provides the appro-
priate signals to change its characteristics and allow
the emergence of the appropriate cell type. Similarly,
the long-term maintenance could be obtained in some
cases with non-quiescent cells that frequently divide
symmetrically if the process controls the size of the
pool of stem cells and in other cases with a cell-
based asymmetric division of quiescent cells. The idea
of a renewing process implies very clearly that it
includes not only the SCs but also non-SCs, extra-
cellular matrix, signaling systems, that is the elements
of what was recognized as the niche. It is this niche-SC
system that has the renewing properties not the SCs
alone. This is illustrated by the intestinal crypt SCs,
where all cells expressing the Lgr5 marker can be a
SC. The Lgr5+cells divide symmetrically and some
of their descendants, through neutral competition, can
occupy a position close to the cell type that provides
niche environment. Therefore, the selection of a cell as
a SC is stochastic based on the probability that a given
cell occupies a position close to the niche (Snippert
et al., 2010a). To this process of renewing common to
all systems (hematopoietic, HF, intestine) correspond
different strategies designed in relation with the multi-
ple constraints of each system. In other words, the idea
of a unique renewing process used again and again at
different places and in different lineages in the organ-
ism and combining the same basic cellular operations
must probably be changed for the idea of several dis-
tinct opportunistic renewing mechanisms using largely
different basic cellular operations.
In sum, the study of HF renewal leads to a major
reevaluation of the concept of stemness. Originally and
for a long time considered as the property of a cell that
concentrates all functions, it seems now more appro-
priate to consider that it is the property of a system.
We propose that, in the case of the HF, the system is
even more complicated than a simple one to one corre-
spondence of a cell and its niche. There might actually
be two sets of stem cells whose functioning is sepa-
rated in time, one involved in the growth of the HF
during anagen and the other in the maintenance of HF
through cycles. It would involve on the one hand, for
46 E. Legué et al.
long term renewal through cycles, the cells of the bulge
and secondary hair germ and the telogen niche, and on
the other hand during anagen the matrix germinative
layer and the dermal papilla with its spatial dynamic.
Acknowledgments This work was supported by French ANR
n◦10-01 (DEV-PROCESS) to IS and JFN.
References
Blanpain C, Lowry WE, Geoghegan A, Polak L, Fuchs E
(2004) Self-renewal, multipotency, and the existence of two
cell populations within an epithelial stem cell niche. Cell
118:635–648
Chase HB, Rauch R, Smith VW (1951) Critical stages of hair
development and pigmentation in the mouse. Physiol Zool
24:1–8
Claudinot S, Nicolas J-F, Oshima H, Rochat A, Barrandon
Y (2005) Long-term renewal of hair follicles from clono-
genic multipotent stem cells. Proc Natl Acad Sci USA 102:
14677–14682
Cotsarelis G, Sun TT, Lavker RM (1990) Label-retaining cells
reside in the bulge area of pilosebaceous unit: implications
for follicular stem cells, hair cycle, and skin carcinogenesis.
Cell 61:1329–1337
Gilbert SF (1997) Developmental biology. Sinauer Associates,
Sunderland, MA, 01375-0407 U.S.A.
Greco V, Chen T, Rendl M, Schober M, Pasolli HA, Stokes N,
Dela Cruz-Racelis J, Fuchs E (2009) A two-step mechanism
for stem cell activation during hair regeneration. Cell Stem
Cell 4:155–169
Hardy M (1992) The secret life of the hair follicle. Trends Genet
8:55–61
Horsley V, O’Carroll D, Tooze R, Ohinata Y, Saitou M,
Obukhanych T, Nussenzweig M, Tarakhovsky A, Fuchs E
(2006) Blimp1 defines a progenitor population that governs
cellular input to the sebaceous gland. Cell 126:597–609
Hsu Y-C, Pasolli HA, Fuchs E (2011) Dynamics between stem
cells, niche, and progeny in the hair follicle. Cell 144:92–105
Ito M, Kizawa K, Hamada K, Cotsarelis G (2004) Hair folli-
cle stem cells in the lower bulge form the secondary germ, a
biochemically distinct but functionally equivalent progenitor
cell population, at the termination of catagen. Differentiation
72:548–557
Jaks V, Barker N, Kasper M, Van Es JH, Snippert HJ, Clevers H,
Toftgard R (2008) Lgr5 marks cycling, yet long-lived, hair
follicle stem cells. Nat Genet 40:1291–1299
Jensen UB, Yan X, Triel C, Woo SH, Christensen R, Owens DM
(2008) A distinct population of clonogenic and multipotent
murine follicular keratinocytes residing in the upper isthmus.
J Cell Sci 121:609–617
Jensen KB, Collins CA, Nascimento E, Tan DW, Frye M, Itami
S, Watt FM (2009) Lrig1 expression defines a distinct mul-
tipotent stem cell population in mammalian epidermis. Cell
Stem Cell 4:427–439
Legué E, Nicolas J-F (2005) Hair follicle renewal: organization
of stem cells in the matrix and the role of stereotyped lineages
and behaviors. Development 132:4143–4154
Legué E, Sequeira I, Nicolas J-F (2010) Hair follicle renewal:
authentic morphogenesis that depends on a complex progres-
sion of stem cell lineages. Development 137:569–577
Liu Y, Lyle S, Yang Z, Cotsarelis G (2003) Keratin 15 promoter
targets putative epithelial stem cells in the hair follicle bulge.
J Invest Dermatol 121:963–968
Lyle S, Christofidou-Solomidou M, Liu Y, Elder D, Albelda S,
Cotsarelis G (1998) The C8/144B monoclonal antibody rec-
ognizes cytokeratin 15 and defines the location of human hair
follicle stem cells. J Cell Sci 111:3179–3188
MorrisRJ,LiuY,MarlesL,YangZ,TrempusCS,LiS,Lin
JS, Sawicki JA, Cotsarelis G (2004) Capturing and profiling
adult hair follicle stem cells. Nat Biotechnol 22:411–417
Müller-Röver S, Handjiski B, van der Veen C, Eichmüller S,
Foitzik K, McKay IA, Stenn KS, Paus R (2001) A com-
prehensive guide for the accurate classification of murine
hair follicles in distinct hair cycle stages. J Invest Dermatol
117:3–15
Muzumdar MD, Tasic B, Miyamichi K, Li L, Luo L (2007)
A global double-fluorescent Cre reporter mouse. Genesis
45:593–605
Nijhof JG, Braun KM, Giangreco A, van Pelt C, Kawamoto H,
Boyd RL, Willemze R, Mullenders LH, Watt FM, de Gruijl
FR et al (2006) The cell-surface marker MTS24 identifies a
novel population of follicular keratinocytes with characteris-
tics of progenitor cells. Development 133:3027–3037
Nowak JA, Polak L, Pasolli HA, Fuchs E (2008) Hair fol-
licle stem cells are specified and function in early skin
morphogenesis. Cell Stem Cell 3:33–43
Oshima H, Rochat A, Kedzia C, Kobayashi K, Barrandon Y
(2001) Morphogenesis and renewal of hair follicles from
adult multipotent stem cells. Cell 104:233–245
Panteleyev A, Botchkareva N, Sundberg J, Christiano AM, Paus
R (1999) The role of the hairless (hr) gene in the regu-
lation of hair follicle catagen transformation. Am J Pathol
155:159–171
Panteleyev AA, Jahoda CAB, Christiano AM (2001) Hair folli-
cle predetermination. J Cell Sci 114:3419–3431
Schofield R (1978) The relationship between the spleen colony-
forming cell and the haemopoietic stem cell. Blood cells
4:7–25
Snippert HJ, Van der Flier LG, Sato T, van Es JH, van den Born
M, Kroon-Veenboer C, Barker N, Klein AM, Rheenen Jv,
Simons BD et al (2010a) Intestinal crypt homeostasis results
from neutral competition between symmetrically dividing
Lgr5 stem cells. Cell 143:134–144
Snippert HJ, Haegebarth A, Kasper M, Jaks V, Van Es JH, Barker
N, van de Wetering M, van den Born M, Begthel H, Vries
RG et al (2010b) Lgr6 marks stem cells in the hair folli-
cle that generate all cell lineages of the skin. Science 327:
1385–1389
Sun TT, Cotsarelis G, Lavker RM (1991) Hair follicular stem
cells: the bulge-activation hypothesis. J Invest Dermatol
96:77S–78S
Trempus CS, Morris RJ, Bortner CD, Cotsarelis G, Faircloth RS,
Reece JM, Tennant RW (2003) Enrichment for living murine
keratinocytes from the hair follicle bulge with the cell surface
marker CD34. J Invest Dermatol 120:501–511
Tumbar T, Guasch G, Greco V, Blanpain C, Lowry WE, Rendl
M, Fuchs E (2004) Defining the epithelial stem cell niche in
skin. Science 303:359–363
5 Hair Follicle Stem Cells 47
Vidal VPI, Chaboissier M-C, Lützkendorf S, Cotsarelis G, Mill
P, Hui C-C, Ortonne N, Ortonne J-P, Schedl A (2005)
Sox9 is essential for outer root sheath differentiation and
the formation of the hair stem cell compartment. Curr Biol
15:1340–1351
Weissman IL, Anderson DJ, Gage F (2001) Stem and pro-
genitor cells: origins, phenotypes, lineage commitments,
and transdifferentiations. Annu Rev Cell Dev Biol 17:
387–403
Zhang YV, Cheong J, Ciapurin N, Mcdermitt DJ, Tumbar T
(2009) Distinct self-renewal and differentiation phases in the
niche of infrequently dividing hair follicle stem cells. Cell
Stem Cell 5:267–278
