Deletion of the epidermis derived laminin γ1 chain leads to defects
in the regulation of late hair morphogenesis
Anja Fleger-Weckmann1*, Yasemin Üstün1*, Jennifer Kloepper2, Ralf Paus3, Wilhelm
Bloch4, Zu-Lin Chen5, Jeannine Wegner6,7, Lydia Sorokin6,7, Lutz Langbein8, Beate
Eckes1, Paola Zigrino1, Thomas Krieg1,9,10 and Roswitha Nischt1
1Dermatology,
University of Cologne, Germany
2Dermatology,
University of Lübeck, Germany
3Dermatology,
University of Manchester, UK
4German
Sport University Cologne, Germany
5Neurobiology
and Genetics, Rockefeller University, New York, USA
6Physiological
Chemistry and Pathobiochemistry, University of Muenster, Germany
7Cells-in-Motion
8German
9Center
Cluster of Excellence, University of Muenster, Germany
Cancer Research Center, Heidelberg, Germany
for Molecular Medicine Cologne (CMMC), University of Cologne, Germany
10Cologne
Excellence Cluster on Cellular Stress Responses in Aging-Associated
Diseases (CECAD), University of Cologne, Germany
* These authors contributed equally
Running title
Epidermal laminin γ1 regulates late hair morphogenesis
Key words
Skin, basement membrane, laminin-511, laminin-211, hair follicle
Address for correspondence:
Professor Thomas Krieg, MD
Dermatology, University of Cologne
Kerpener Str. 62
50931 Cologne, Germany
phone +49-221-478 82245
Email [email protected]
1
Abstract
Laminins are the most abundant non-collagenous basement membrane (BM)
components, composed of an α, β and γ chain. The laminin γ1 chain, encoded by
LAMC1, is the most abundant γ chain. The main laminin isoforms in the dermoepidermal junction (DEJ) are laminin-332, laminin-511 and laminin-211, the latter
being restricted to the lower part of hair follicles (HFs). Complete deletion of LAMC1
results in lethality around embryonic day 5.5. To study the function of laminin γ1
containing isoforms in skin development and maturation after birth, we generated
mice lacking LAMC1 expression in basal keratinocytes (LAMC1EKO) using the keratin
14 (K14) Cre/loxP system. This deletion resulted in loss of keratinocyte derived
laminin-511 and in deposition of fibroblast derived laminin-211 throughout the whole
DEJ. The DEJ in areas between hemidesmosomes was thickened, whereas
hemidesmosome morphology was normal. Most strikingly, LAMC1EKO mice showed
delayed HF morphogenesis accompanied by reduced proliferation of hair matrix cells
and impaired differentiation of hair shafts (HS). However, this deletion did not
interfere with early HF development, since placode numbers and embryonic hair
germ formation were not affected. Microarray analysis of skin revealed down
regulation of mainly different hair keratins. This is due to reduced expression of
transcription factors such as HoxC13, FoxN1, FoxQ1 and Msx2, known to regulate
expression of hair keratins. While the role of laminin-511 in signaling during early hair
germ formation and elongation phase has been described, we here demonstrate that
epidermal laminin-511 is also a key regulator for later hair development and HS
differentiation.
2
Abbreviations: basement membrane (BM), bone morphogenetic protein (BMP)
dermo-epidermal junction (DEJ), embryonic day (E), Forkhead box N1 (FOXN1), hair
follicle (HF), hair shaft (HS), Homeobox C13 (HOXC13), immunofluorescence (IF),
inner root sheath (IRS), keratin 5 (K5), keratin 14 (K14), keratin associated proteins
(KAPs), Lymphoid enhancing factor (Lef1), Homeobox msh-like 2 (MSX2),
polymerase chain reaction (PCR), postnatal day (P), reverse transcriptase (RT),
Sonic hedgehog (Shh), transient amplifying (TA), transcription factor (TF)
3
Introduction
BMs are highly structured extracellular matrices that form a dynamic interface
separating
and
connecting
simultaneously
epidermal
with
mesenchymal
compartments. Its components are regulators of cellular processes essential for skin
integrity as illustrated by several acquired and inherited diseases [1], [2]. The major
components of the DEJ are members of the laminin, nidogen, collagen IV and the
heparan sulfate proteoglycan (perlecan and agrin) families [3], [4].
Laminins are the most abundant non-collagenous structural components of BMs and
play important roles in tissue morphogenesis and homeostasis by regulating cell
survival, differentiation, migration and adhesion [5], [6]. Furthermore, they link the BM
to epithelial transmembrane receptors, primarily β1 and β4 integrins [7]. Laminins are
heterotrimeric proteins composed of α, β, and γ subunits [8]. Currently, 5 α-, 3 β- and
3 γ-subunits have been identified in mice and humans forming the 18 known laminin
heterotrimers [5], [9]. The different isoforms exhibit tissue and developmental
specificity [10], [11], [12], reflecting their diverse functions in vivo.
Laminins in the skin have been implicated in the regulation of HF development,
driven by a series of epithelial-mesenchymal crosstalk [13]. HF development is
initiated by the induction of epithelial hair placodes by the first message of the
underlying mesenchyme [14] and proceeds with the down growth of hair germs into
the dermis to form HFs [15]. Induced by the epithelial message derived from
elongating HFs, dermal papilla cells condensate and become surrounded by the hair
bulb. Signals from dermal papilla cells stimulate proliferation of adjacent hair matrix
cells of the HF. During HF differentiation, matrix cells produce progenitors that
differentiate and give rise to HS and inner root sheath of the HFs [16], [17], [18], [13].
As the cells move distally into the different HF compartments, they start to
4
differentiate, visible by the expression of characteristic marker proteins (keratins).
Differentiation of matrix cells is regulated by a number of factors, such as
Wnt/Lymphoid enhancing factor 1 (Lef1) [19], [20], fibroblast growth factor 7 (Fgf7)
[21] and members of the bone morphogenetic protein (BMP) family [22], [23]. These
are important in HF proliferation and differentiation during hair morphogenesis.
However, the role of extracellular matrix proteins, which provide critical anchorage to
the cells, in this process is not elucidated.
The laminin γ1 chain is the most ubiquitously expressed γ subunit, present in 10 of
the 18 known laminin isoforms [5]. Previous studies showed that laminin γ1 is a
prerequisite for the formation of an embryonic BM, since LAMC1 complete knock out
mice die at embryonic day (E) 5.5 [24]. Expression of laminin-111 and -511 is already
detected at the morula stage in mouse development [25]. At the DEJ, laminins are
synthesized by both dermal [26], [27] and epidermal cells [28]. The main secreted
laminin variants at the DEJ are laminin-332 and laminin-511. Depending on the
developmental stage, laminin-211 is also present in the lower part of the HFs [29],
[9]. During HF elongation, the laminin composition of the BM undergoes changes:
while laminin-511 expression pattern is not altered, laminin-332 [30] and laminin-111
[31] are downregulated. Previous studies have shown the importance of laminin-511
in early embryonic hair germ elongation. The loss of laminin-511 in transgenic mice
resulted in a reduced follicular proliferation, thus in a reduced number of hair germs
after regression of the HFs [25]. Further known laminin γ1 chain containing isoforms
in the skin are laminin-311, and -321, however, their presence in murine skin is
discussed controversially [9].
To study the function of laminin γ1 chain containing isoforms in skin development and
maturation in later stages, and to avoid early embryonic lethality, we generated mice
5
that do not express LAMC1 in basal keratinocytes from early developmental stages
onwards (referred to as LAMC1EKO) using the K14 Cre/loxP system [32].
This report demonstrates that epidermal derived laminin γ1 chain is a key regulator
for late hair morphogenesis and differentiation. We showed that the deletion of
epidermal laminin γ1 resulted in the loss of keratinocyte derived laminin-511 and in
ectopic deposition of fibroblast derived laminin-211 in the whole DEJ. These changes
in laminin composition led to ultrastructural defects of the dermo-epidermal BM zone
between hemidesmosomes and in delayed hair growth, accompanied by reduced
proliferation of hair matrix cells and defects in hair differentiation. Furthermore, our
studies suggest an important role of the BMP-MSX2-HOXC13-FOXN1 signaling axis
in regulation of the processes required for HF differentiation and proliferation.
6
Results
Depletion of keratinocyte specific laminin γ1 from the dermo-epidermal BM
The main laminins in the murine dermo-epidermal BM are laminin-332 and the
laminin γ1 chain containing isoform laminin-511 and, in the lower part of the HF,
laminin-211.
To analyze the functional impact of the γ1 chain containing laminins in late hair
development, homozygous floxed laminin γ1 mice [33] were crossed with a
transgenic mouse strain expressing Cre-recombinase under the control of the human
K14 promoter [32]. This results in the mouse strain LAMC1EKO which lacks laminin γ1
chain expression and secretion by basal keratinocytes. The human K14 promoter
becomes active at around E14.5, and activity in postnatal mice is restricted to the
mitotically active basal keratinocytes, the outer root sheath of the HFs and the oral
epithelium [34]. After birth, homozygous LAMC1EKO mice were easily distinguished
from their control littermates by syndactyly of the hind limbs (Fig. 1a), as well as by
growth retardation, reduced body size and delayed appearance of the hair coat. Of
151 mice analyzed, hind limb syndactyly was observed in 98.7% of mice carrying the
Cre-transgene and only 1.3% of mice displayed partial hind limb syndactyly. Delayed
hair growth was observed in 100% of LAMC1EKO mice. Macroscopically, the
observed hair phenotype did not reveal a high variability in severity, and it correlated
with syndactyly and reduced body growth.
To analyze the influence of epidermal laminin γ1 chain depletion on the development
of the skin at the molecular level, we performed immunofluorescence (IF) stainings
for different laminin chains on skin sections of LAMC1EKO and control mice. Laminin
composition of the DEJ was analyzed at early and late time points of hair
morphogenesis (E18.5, P1 and P10) before entry into first catagen phase at P16. At
7
all time points investigated, skin sections of LAMC1EKO mice revealed reduced
deposition of the laminin γ1 chain in the DEJ, but not a complete loss, when
compared to control skin, indicating that laminin 1 in the DEJ was produced by other
cell types, such as fibroblasts. Staining with antibodies directed against different
laminin α chains revealed absence of laminin α5 chain in the DEJ of LAMC1EKO skin,
while staining of BMs of capillaries, arterioles and nerves was still detected. Laminin
α2 deposition was restricted to the lower part of the HFs in control skin, however,
mutant skin sections revealed prominent continuous laminin α2 staining of the
dermo-epidermal BM (Fig. 1b). Our findings confirm that laminin-511 at the DEJ is
synthesized by keratinocytes, while fibroblasts produce the laminin γ1 chain
containing laminin-211 as previously reported [27], [35]. This result is consistent with
data obtained by in situ hybridization (Fig. S1a) and with reverse transcriptase (RT)polymerase chain reaction (PCR) using RNA isolated from fibroblasts or
keratinocytes from control or LAMC1EKO skin (Fig. S1b). By contrast, the localization
of the laminin α3 and α4 chains and deposition of laminin-332 was not changed in
LAMC1EKO mice (data not shown). Since binding of the laminin γ1 chain to nidogen-1
and -2 [36] mediates the connection of laminins to the collagen IV network and to
other proteins like perlecan, we also investigated the distribution of these major BM
molecules. However, IF stainings did not reveal changes in the localization of these
proteins at the DEJ of LAMC1EKO skin (data not shown).
In agreement with our staining results, Western blot analysis of serum-free
supernatants of primary control and mutant keratinocytes showed loss of both laminin
γ1 and α5 chains in the supernatants of mutant keratinocytes, indicating that laminin
α5 chain is not secreted in the absence of laminin γ1 (Fig. 1c). This finding is in
concordance with the concept that laminin assembly into trimeric molecules takes
place intracellularly and that only laminin trimers are secreted [37]. By Western blot
8
analysis using keratinocyte lysates, we could indeed confirm that laminin α5 chain,
although not secreted by keratinocytes of LAMC1EKO mice, was still present in these
cells, whereas no intracellular laminin γ1 was detected. This was different in
fibroblasts, which secreted the laminin γ1 chain normally, as expected from previous
studies [27], [35] (Fig. 1c).
Ultrastructural analysis of skin sections at P1 revealed that a discontinuous dermoepidermal BM in areas between hemidesmosomes of LAMC1EKO skin. Partially
preserved lamina densa between the hemidesmosomes appeared diffuse with less
compact extracellular matrix deposits when compared to control littermates.
However, hemidesmosome formation per se was not affected in the mutants.
Furthermore, skin of P9 mutant mice revealed alterations of the epidermal
ultrastructure with less compact packing of basal keratinocytes and enlarged
intercellular spaces. Basal keratinocytes showed apoptotic features such as pyknotic
nuclei reflecting chromatin condensation (Fig. 1d).
HF morphogenesis is impaired after depletion of laminin γ1 chain expression in
basal keratinocytes
Apart from syndactyly of the hind limbs, the most striking finding in LAMC1EKO mice
was the complete absence of a hair coat at P7 to P10 when control mice were fully
covered with pigmented hair. Mutant mice developed a sparser hair coat that covered
the mice by P18, and pigmentation of HSs was not observed before P15. To
elucidate the effect of the laminin γ1 chain containing laminins on hair development,
we characterized HF morphology and HF developmental stages in control and
LAMC1EKO skin sections at P1 and P10.
9
Changes in HF morphology were assessed by hematoxylin and eosin stainings of
longitudinal back skin sections from mice at P1 and P10 (Fig. 2a). At P1, mutant skin
morphology did not reveal any major differences compared to control animals (Fig.
2a). However, at P10, HF orientation, length and structure were affected in the
mutant skin (Fig. 2a). While HFs of control mice were uniformly arranged at an angle
of 30° to the epidermal surface, the orientation of HFs in mutant skin was
disorganized, showing varying angles of HFs between 40° and 90°. Furthermore,
mutant mice lacked HSs with the typically striped pigmentation pattern and HFs that
were pigmented revealed a uniform light-brown color. HFs in mutant skin were much
shorter and did not grow deeply into the subcutis, rather they were curved and
thickened in some areas when compared to HFs in control skin (Fig. 2a). At the
ultrastructural level, the follicular BM of LAMC1EKO mice at P3 revealed thickened BM
areas compared to control mice (Fig. 2b).
Morphologically, HF development proceeds through 8 stages, divided into 3 phases:
the induction phase (stage 0 and 1), the organogenesis phase (stage 3-5) and the
cyto-differentiation phase (stage 6-8) [18], [13]. Deletion of laminin γ1 chain timely
correlates with placode formation (E13.5). The HF stages at P1 and P10 were
compared in control and mutant mice. At P1, in control animals 40% of the back skin
HFs were in stage 6. In contrast, only 8% of HFs in mutant mice were at this stage.
At P10, almost all HFs were in stage 8 in controls, whereas in mutant mice only half
of the HFs reached this stage (Fig. 2c), indicating a delay in hair growth of
approximately 8 days compared to control littermates.
To determine if the delay in HF morphogenesis results from reduced proliferation of
hair matrix cells that give rise to the growing HS, the number of proliferating hair
matrix cells below Auber’s line, i.e. the widest diameter of the dermal papilla [38], was
10
determined by Ki67 staining (Fig. 2d). Indeed, numbers of proliferating matrix cells
were significantly reduced (p<0.005) in LAMC1EKO as compared to control HFs (Fig.
2e).
In the mouse, HF morphogenesis starts at E13.5 with formation of the placode and
proceeds with its progression to hair germ to hair peg. Embryos at E16.5 exhibit all
three HF precursors, which were assessed in the skin of LAMC1EKO and control
embryos by performing IF co-staining for P-cadherin and K14 (Fig. 3a, b). The total
number of HF precursors at the distinct stages was determined and revealed no
difference between control and LAMC1EKO mice (Fig. 3a). Quantification of
embryonic hair placode numbers at day 16.5 (Fig. 3c) and mature HF numbers in 8week old adult mice (Fig. 3d) revealed comparable numbers in control and
LAMC1EKO skin. Together, these data indicate that K14-Cre mediated epidermis
specific deletion of the laminin γ1 chain does not significantly interfere with HF
induction.
Differential gene expression after keratinocyte specific laminin γ1 depletion
To identify genes whose expression were affected by depletion of LAMC1 expression
in keratinocytes, the mRNA expression profile of the epidermis and dermis from
LAMC1EKO split skin was analyzed at E18.5 and P6. At P6, HFs have already invaded
deeply into the dermis and subcutis, thus, the dermal compartment of split skin can
contain parts of HFs resulting from incomplete separation of the two compartments.
Analyzing split skin from 2 male control and mutant mice at P6 revealed differential
expression of 596 genes in the dermis and 373 genes in the epidermis (p<0.05;
>±1.5-fold) of LAMC1EKO mice compared to age-matched control samples. Among
the genes with altered expression in the dermis, 209 genes were specifically
11
downregulated and 348 were upregulated. In the epidermis of LAMC1EKO mice,
expression of 77 genes was specifically upregulated, while expression of 257 genes
was specifically downregulated (Fig. S2). Most of the upregulated genes were
involved in immunological processes and were mainly detected in the dermis. The
most strongly downregulated genes in the LAMC1EKO skin included genes encoding
hair keratins or keratin associated proteins (KAPs) (listed in Fig. 4a). The analysis
further revealed downregulation of transcription factors known to directly or indirectly
regulate expression of hair keratins found in the cortex, medulla and cuticle layers of
the HS (listed in Fig. 5a). Altered expression of these genes was confirmed by
analysis of the respective proteins (see below).
The expression profile of split skin (into dermis and epidermis) at E18.5 failed to
reveal major differences between LAMC1EKO and control skin. In total, ten genderunspecific genes exhibited significant changes in expression. Among these, we
identified a group of upregulated immune-associated genes (data not shown).
Differentiation of the HS is impaired in LAMC1EKO HFs
Rapidly dividing keratinocytes in the hair matrix form seven concentrically arranged
layers of HFs [39]. Since we had identified reduced proliferation of hair matrix cells in
the LAMC1EKO mutants (Fig. 2d, e) we asked whether this defect might interfere with
HF differentiation. We therefore performed IF stainings of skin sections using hair
keratin antibodies specific for different HS compartments as well as for HF-specific
epithelial keratins [40], [41]. Immunostainings of P6 skin sections did not reveal
differentiation defects of the companion layer (K75) and the IRS (K28) of LAMC1EKO
mice. By contrast, the development of the HS, which itself comprises medulla (K81),
matrix and/or cortex (K33, K35, K81, K85, K86) and cuticle (K35, K82, K85), was
12
affected, as shown by the altered expression patterns of the HS keratins in the
mutant mice (Fig. 4b). This result agrees with our array analysis, which revealed
downregulation of these keratins at the transcriptional level (Fig. 4a).
In addition to keratins, microarray analysis of LAMC1EKO split skin has also revealed
downregulation of the transcription factors Homeobox C13 (HOXC13), Forkhead box
N1 (FOXN1), Forkhead box Q1 (FOXQ1) and Homeobox msh-like 2 (MSX2), all
known to directly or indirectly regulate expression of hair keratins found in the cortex,
medulla and cuticle layers of the HS [42], [43], [44] (Fig. 5a). Downregulation of the
transcription factors Msx2, FoxN1 and HoxC13 was confirmed at the protein level by
IF and/or by immunoblot analysis. IF analysis showed reduced FoxN1 protein in the
HS and IRS of mutant HFs at P6 and P10. Further, staining for HoxC13 revealed a
reduced amount, as well as an impaired localization of the protein in the companion
layer and matrix region of HFs of LAMC1EKO animals (Fig. 5b). Reduced FoxN1,
HoxC13 and Msx2 expression was further confirmed by immunoblot analysis using
skin lysates of mutant and control mice (P6 and P10) (Fig. 5c). In addition to these
transcription factors, the regulatory proteins Lef1 and BMP2, known to be important
for HF proliferation and differentiation [19], [20], [22], [23] were also downregulated
(Fig. 5a). As a consequence of reduced BMP signaling activity in LAMC1EKO skin, we
detected reduced pSmad1/5 expression at P6 and at P10 (Fig. 5d). A working model
for the effect of the identified regulatory molecules on HF morphogenesis is shown in
Fig. 6.
13
Discussion
This study is the first to describe the role of the epidermal laminin γ1 chain, the most
abundant laminin γ chain, as an important regulator for late hair morphogenesis and
differentiation. As previously reported, the complete deletion of laminin γ1 results in
embryonic lethality of mice preventing skin and hair analysis after birth [24]. Here, we
generated mice with LAMC1 deficiency in basal and outer root sheath keratinocytes
(LAMC1EKO) induced during late embryogenic development, using the K14 Cre/loxP
system.
The main laminin γ1 containing isoforms in the DEJ are laminin-511 [9] and laminin211, the latter found in the lower part of the HFs. Our analysis showed that
LAMC1EKO mice lack laminin α5 chain deposition at the DEJ, confirming that laminin511 is deposited by keratinocytes [35] and moreover that laminin molecules are
secreted only following heterotrimerization, while assembly takes place intracellularly.
Furthermore, laminin α2, normally restricted to the lower part of the HFs was
ectopically deposited throughout the whole BM of LAMC1EKO skin, confirming
fibroblasts as the source of laminin-211 expression and secretion [27] (see also Fig.
1c). Consistent with these findings, laminin α2 has already been identified as a
product of mesenchymal origin in previous studies [27], [9].
Since the laminin γ1 chain was proposed to play an essential role in the integration of
other major BM molecules such as collagen IV, perlecan and nidogens, their
distribution could have been affected by the inactivation of LAMC1. However, these
molecules were normally distributed in LAMC1EKO skin, implicating that fibroblast
derived laminin γ1 chain is sufficient for their integration into the DEJ.
The phenotypic changes in LAMC1EKO mice, however, demonstrate that the
ectopically deposited, fibroblast derived laminin-211 is not fully able to functionally
14
compensate for the lack of laminin-511. The described phenotypes can be caused
either by loss of laminin-511 or by ectopic deposition of laminin-211. In the follicular
BM, particularly in the area surrounding the DP, laminin-211 is present during HF
morphogenesis in both, control and mutant skin. Thus, its expression should not alter
the HF phenotype of LAMC1EKO mice, but the loss of laminin-511 is causal for the
observed defects. However, there is an altered distribution of laminin-211 in mutant
HFs, so it may be the local balance of these laminins that is necessary for proper HF
development. By contrast, ultrastructural defects of interfollicular epidermis and the
interfollicular DEJ can be attributed to the combined loss of laminin-511 and the
ectopic presence of laminin-211, the latter being absent from control interfolliclular
epidermis.
Laminins mediate outside-in signaling (intracellularly) through transmembrane
receptors such as integrins [45]. Laminin-211 and -511 are recognized by partially
overlapping sets of cell surface receptors including integrins α6β1 and α7β1 [46],
[47], [5]. Dystroglycan is also expressed by epidermal cells [48] and binds with high
affinity to laminin-211 [49], but weakly, if at all, to laminin-511 [50]. An abundant
receptor for laminin-511 in the skin is integrin α3β1, which, however, does not bind to
laminin-211 [51], [52]. Therefore, the ultrastructure of the lamina densa was markedly
abnormal in LAMC1EKO mice, possibly due to the ectopic deposition of laminin-211,
which is not able to bind to integrin α3β1. This results in rounding-off of basal
keratinocytes, likely due to a partial loss of cell polarity. However, hemidesmosomal
adhesion of basal keratinocytes to the underlying BM was not disturbed,
demonstrating that the interaction of laminin-332 with integrin α6β4 is sufficient to
prevent epidermal detachment and blistering.
15
Of note, LAMC1EKO mice exhibited delayed HF morphogenesis and a defect in HF
differentiation. Findings of previous studies on mice with ablation of laminin α5 [25],
integrin β1 [53], [54] and integrin α3 [55] support our hypothesis that late hair
morphogenesis is affected by the lack of interaction between laminin-511 and integrin
α3β1, which has a direct role in proliferation of hair matrix cells. For laminin α5
deficient mice it was shown that embryos have reduced numbers of hair germs [25].
However, we found that hair placode formation was not affected in our mutant mice
due to the later onset of the K14 promoter and induction of LAMC1 deletion at around
E14.5 (data not shown) [34]. Mice with an epidermis specific deletion of laminin α5
induced by keratin 5 (K5)-Cre show a delay of two days in the appearance of hair
after birth. In these mice, HF morphology was not disturbed, but HF density was
slightly reduced (L. Sorokin, personal communication). This differed from our results
and might be explained by different time of onset of K5 and K14 promoters.
Interestingly, both studies found that loss of laminin-511 at the DEJ was associated
with an upregulation of laminin α2 in this region. In addition, laminin α4 was
upregulated upon K5-Cre mediated ablation of LAMA5, which was not seen in our
LAMC1EKO mice. In skin, transplantation experiments, where laminin-511 deficient
skin was transplanted onto full-thickness WT skin, resulted in a failure of hair germ
elongation leading to regression of HFs [25]. This regression of HFs, however, was
not observed in LAMC1EKO skin. Similar to our findings with LAMC1EKO skin,
embryonic laminin α5 deficient mice showed disrupted lamina densa and reduced
proliferation of hair matrix cells. Topically application of laminin-511 in the transplant
experiments, however, rescued these defects and also restored HF development
[25]. Laminin-511 in signaling during early hair germ elongation phase was implicated
by the reduced expression of the molecules sonic hedgehog (Shh) and Gli1 [25]. Our
gene expression analysis did not reveal any reduction in molecules of the Shh/Gli1
16
signaling pathway at E18.5 and P6. Resembling the findings in our LAMC1EKO mice,
K5-mediated deletion of β1 integrin also lead to markedly reduced proliferation of
matrix keratinocytes and to a disrupted lamina densa [53]. More severe, β1 integrin
deletion was responsible for impaired invagination of HFs and eventual hair loss [53],
[54]. Integrin α3 deficient animals also displayed changes in BM ultrastructure, but
their HF phenotype was less severely disturbed than that observed in the absence of
integrin β1 and laminin α5 as the animals showed progressive hair loss only as adults
[56], [55]. Interestingly, the deletion of integrin β1 in the dermis did not affect HF
morphogenesis, suggesting that laminin-511 interacts with β1 integrins at the surface
of epithelial rather than on dermal papilla cells [57].
Our results suggest that loss of laminin-511 due to epidermal depletion of laminin γ1
causes abnormal expression of different hair keratins during late HF morphogenesis,
resulting in disruption of normal hair formation. We show that this is associated with
changes in the BMP-MSX2-HOXC13-FOXN1 signaling axis, which seems to be
crucially implicated in regulating HS differentiation and matrix cell proliferation.
Microarray analyses demonstrated that gene regulation in the skin of LAMC1EKO mice
at E18.5 was not markedly altered, probably due to a long half-life of laminin γ1
protein in the skin. At P6, however, many genes encoding hair keratins and keratinassociated proteins (KAPs) were downregulated. This was accompanied by a
downregulation of the transcription factors HoxC13, FoxN1 and FoxQ1, which are all
known to directly or indirectly regulate expression of hair keratins found in the cortex,
medulla and cuticle layers of the HS [42], [43], [44]. The transcription factors HoxC13
[42] and FoxN1 [58], [59] have been shown to be essential for proper HS
differentiation as the corresponding mutants completely lack external hair. It has
been described that HoxC13 and FoxN1 might function in a common pathway of HF
differentiation, where FoxN1 acts downstream of HoxC13 in a regulatory signaling
17
cascade regulating expression of terminal differentiation markers [60]. HoxC13 was
also described to target FoxQ1 expression during regulation of HF differentiation [61].
Interestingly, there is a partial overlap of the regulated genes in HOXC13 deficient
mice [60] and our LAMC1EKO mice. Ten out of 19 significantly regulated keratins in
the skin of LAMC1EKO mice were also downregulated in skin of HoxC13 mutants:
K26, K31, K32, K33a, K33b, K34, K35, K72, K84 and K86 [60]. Consistent with these
findings, MSX2, which is known to regulate the expression of FOXN1 and also of hair
keratins by a parallel signaling pathway [62], was also downregulated in LAMC1EKO
skin. Bone morphogenetic proteins (BMPs) act upstream of HoxC13, FoxN1 and
Msx2, and regulate postnatal differentiation and proliferation of HFs. They are
expressed in proliferating hair matrix cells and differentiating hair precursor cells.
Expression of the BMP inhibitor Noggin, under the control of the Msx2 promoter lead
to downregulation of hair keratins as well as of FoxN1 and HoxC13 expression. Along
with the findings from LAMC1EKO mice, differentiation of the HS, but not of the IRS
was disturbed in Msx2-Noggin transgenic mice [23]. In LAMC1EKO skin, we also
detected reduced phospho-Smad1/5 (pSmad1/5) expression in the skin at P6 and
P10, demonstrating that BMP signaling is altered in the absence of laminin γ1.
It has been shown that activated BMP signaling keeps HF stem cells in a quiescent
state. However, when BMP signaling is turned off, stem cells begin to proliferate and
produce transient amplifying (TA) progenitors that express Shh. Subsequently, BMP
signaling promotes TA cells to specialize and differentiate into the HS or the IRS
[22]. Thus, BMP inhibitor Noggin expressed by the dermal papilla sustains
proliferation and specification of TA cells within the hair bulb [63], [64]. The model in
Fig. 6 summarizes all the signaling pathways that we found altered in LAMC1EKO hair
follicles and which contribute to HS differentiation. During hair differentiation, BMP
signaling induces a signaling cascade in which Smad1/5 proteins become
18
phosphorylated. Epithelial BMP, through phosphorylated Smad1/5 induces Msx2,
which in turn activates HoxC13. HoxC13 was shown to directly regulate the
expression of FoxN1 and of HS keratins, either directly by binding to keratin
promotors or via FoxN1 [60]. HoxC13 also targets FoxQ1 expression and thereby
regulates HF differentiation [61]. Msx2 can also regulate FoxN1 not only via HoxC13,
but also directly.
Lef1, which mediates nuclear responses to Wnt signaling, was also suggested to play
an activating role in hair maturation. It cooperates with BMPs to activate HS
differentiation and keratin expression [23]. Lef1 mutant mice show disturbed HF
morphogenesis with downregulation of K31 [20], which, interestingly, is also
significantly downregulated in the skin of LAMC1EKO mice (Fig. 4a). These similarities
support the hypothesis that downregulation of Lef1 contributes to the phenotype of
LAMC1EKO mice.
All these regulatory proteins involved in HS maturation, however, are downregulated
in the absence of laminin-511 and altered localization of laminin-211, thereby
resulting in a reduced proliferation of matrix cells and a disturbed expression of
differentiation specific HS keratins (Fig. 6).
Taken together, our data suggest that there is a direct link between laminin-511mediated signaling and the BMP-MSX2-HOXC13-FOXN1 signaling axis, which
controls the regulation of HS differentiation and proliferation of hair matrix cells.
19
Experimental Procedures
Generation and genotyping of mice with keratinocyte specific deletion of the
laminin γ1 chain (LAMC1EKO)
To generate the K14-Cre/LAMC1 knockout mouse strain, homozygous floxed laminin
γ1 mice (LAMC1flox) [33] were crossed with male K14-Cre transgenic mice [32]. Both
mouse strains were back-crossed for at least 8 generations onto the C57BL6/J
background. To obtain knockout mice that carry homozygous LAMC1flox alleles and
the Cre-recombinase transgene (LAMC1EKO mice), homozygous LAMC1flox mice
were crossed with heterozygous LAMC1flox mice carrying the K14-Cre transgene.
Littermates that were used as control animals in all experiments were either
homozygous LAMC1flox or heterozygous for the LAMC1flox allele and the Cre
transgene. Genotyping was performed by PCR using genomic DNA from mouse tail
biopsies. Wild-type, floxed or deleted alleles of LAMC1 were assessed using the
following
primer
sets:
for
detection
of
wildtype
allele
5’-
CTCAGAGCTGGCTTCTCACAT-3’ and 5’-CATTTCCCCACAAGTGGTTCTT-3’, for
detection of floxed and deleted allele 5’-CCTACATTTTGAATGCAAGGATTGG-3’ and
5’-GATTTTCAAAGAAGCAGAGTGTG-3’. Animals were housed in specific-pathogenfree facilities and all animal experiments were performed in compliance with German
Regulations for Welfare of Laboratory Animals and were approved by the
Regierungspräsidium Köln Germany (NRW authorization 8.87-51.05.20.13.004).
Histological analysis and indirect IF staining of skin sections
For histology, back skin samples were fixed in 4% (w/v) paraformaldehydephosphate-buffered saline on ice for 45 min, embedded in paraffin wax and sectioned
(30-60nm thickness). Paraffin sections were subjected to trypsin digestion or citric
acid-based antigen retrieval. Alternatively, skin tissue was snap-frozen in optimal20
cutting-temperature compound (OCT Tissue Tec). Frozen sections were fixed with
ice-cold 100% ethanol or acetone. To block unspecific binging, skin sections were
incubated for one hour with 10% (v/v) donkey or goat serum in phosphate-buffered
saline, then incubated with the primary antibody, either for one hour at room
temperature or overnight at 4°C. Bound antibodies were detected with Alexa Fluor
coupled secondary antibodies. For visualization of nuclei, sections were stained with
DAPI. IF stainings were documented with the Olympus IX71 microscope. Information
on antibodies used is provided in the supplementary methods.
Hematoxylin and eosin staining of frozen skin sections was performed, according to
standard protocols. Light microscopy was performed with the Leica DM 4000 D
microscope.
Immunoblot analysis
Proteins of cell or total skin lysates from control and LAMC1EKO mice at P1 or
precipitates of cell culture supernatants (50µg) were resolved in 4-12% SDSpolyacrylamide gradient gels (Invitrogen) under reducing conditions and transferred
to nitrocellulose membranes (GE Healthcare). The membranes were blocked in 5%
low-fat milk powder in TRIS-buffered saline and incubated with the respective primary
antibodies, followed by incubation with appropriate secondary antibodies conjugated
to horseradish peroxidase. Reactive bands were visualized using ECL detection
reagent (GE Healthcare) on X-ray films. Equal protein loading was assessed by
incubating the membranes with antibodies directed against GAPDH or by Ponceau S
staining. Detailed information on antibodies used is provided in the supplementary
methods.
Electron microscopy
21
Briefly, specimens were fixed in 2% glutaraldehyde/2% (w/v) paraformaldehyde (in
0.1M cacodylate buffer pH 7.4), followed by incubation in 2% OsO4/0.1M cacodylate
buffer, pH 7.4. After washing, the specimens were stained in 1% uranyl acetate,
dehydrated through a graded series of ethanol, and embedded in araldite resin.
Semithin sections (1μm) and ultrathin sections (70nm) were cut with a diamond knife
using an ultramicrotome (Reichert, Bensheim, Germany).
Isolation and cultivation of primary keratinocytes
Keratinocytes were isolated between postnatal day 1 and 2. The epidermis was
removed en bloc after Dispase treatment overnight (5mg/ml in serum-free DMEM) at
4°C. The epidermis was then incubated in a Trypsin substitute TrypLETM Select
(Gibco) for 40 min at room temperature. Afterwards, epidermal cells were dissolved
in serum-free DMEM and resuspended in low Ca2+ FAD medium. Low Ca2+ FAD
medium contains 10% (v/v) fetal calf serum (FCS) treated with Chelex 100 resin
(BioRad, Munich, Germany), adenine (1.8x10–10M), hydrocortisone (0.5µg/ml), insulin
(5g/ml), cholera toxin (10–10M), EGF (10ng/ml), streptomycin/penicillin (100U/ml) and
glutamine (0,1mg/ml). 106 cells were plated onto 6cm collagen type I-coated dishes
(BioCoat, BD) together with Mitomycin C-treated 3T3 fibroblasts. Information on
antibodies used is provided in the supplementary methods.
Statistical analysis
Statistical analyses of the data was performed using Mann-Whitney-Test. Differences
with P<0.05 were considered to be statistically significant (* p<0.05, ** p<0.005).
22
Acknowledgements
We thank Marion Reibetanz for expert technical assistance, and Dr. Ulrike Mayer
(University of Norwich) for stimulating discussion. We also thank Dr. Sidney
Strickland (Rockefeller University, New York, USA) for kindly providing the floxed
laminin γ1 strain. This work was supported by Deutsche Forschungsgemeinschaft
through SFB 829 (to WB, PZ, TK, BE, RN) at the University of Cologne and SFB
1009 (to LS) at the University of Muenster.
23
References
1. Erickson AC, Couchman JR (2000) Still more complexity in mammalian basement
membranes. J Histochem Cytochem 48: 1291-1306.
2. Hamill KJ, Kligys K, Hopkinson SB, Jones JC (2009) Laminin deposition in the
extracellular matrix: a complex picture emerges. J Cell Sci 122: 4409-4417.
3. Yurchenco PD (2011) Basement membranes: cell scaffoldings and signaling platforms.
Cold Spring Harb Perspect Biol 3.
4. Timpl R, Brown JC (1996) Supramolecular assembly of basement membranes. Bioessays
18: 123-132.
5. Durbeej M (2010) Laminins. Cell Tissue Res 339: 259-268.
6. Miner JH, Yurchenco PD (2004) Laminin functions in tissue morphogenesis. Annu Rev
Cell Dev Biol 20: 255-284.
7. Hynes RO (1992) Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69:
11-25.
8. Timpl R (1996) Macromolecular organization of basement membranes. Curr Opin Cell Biol
8: 618-624.
9. Aumailley M, Rousselle P (1999) Laminins of the dermo-epidermal junction. Matrix Biol 18:
19-28.
10. Sanes JR, Engvall E, Butkowski R, Hunter DD (1990) Molecular heterogeneity of basal
laminae: isoforms of laminin and collagen IV at the neuromuscular junction and
elsewhere. J Cell Biol 111: 1685-1699.
11. Paulsson M (1992) Basement membrane proteins: structure, assembly, and cellular
interactions. Crit Rev Biochem Mol Biol 27: 93-127.
12. Yurchenco PD, Wadsworth WG (2004) Assembly and tissue functions of early embryonic
laminins and netrins. Curr Opin Cell Biol 16: 572-579.
13. Schmidt-Ullrich R, Paus R (2005) Molecular principles of hair follicle induction and
morphogenesis. Bioessays 27: 247-261.
14. Andl T, Reddy ST, Gaddapara T, Millar SE (2002) WNT signals are required for the
initiation of hair follicle development. Dev Cell 2: 643-653.
15. Schneider MR, Schmidt-Ullrich R, Paus R (2009) The hair follicle as a dynamic
miniorgan. Curr Biol 19: R132-142.
16. Oro AE, Scott MP (1998) Splitting hairs: dissecting roles of signaling systems in
epidermal development. Cell 95: 575-578.
17. Millar SE (2002) Molecular mechanisms regulating hair follicle development. J Invest
Dermatol 118: 216-225.
18. Paus R, Muller-Rover S, Van Der Veen C, Maurer M, Eichmuller S, et al. (1999) A
comprehensive guide for the recognition and classification of distinct stages of hair
follicle morphogenesis. J Invest Dermatol 113: 523-532.
19. Zhou P, Byrne C, Jacobs J, Fuchs E (1995) Lymphoid enhancer factor 1 directs hair
follicle patterning and epithelial cell fate. Genes Dev 9: 700-713.
20. DasGupta R, Fuchs E (1999) Multiple roles for activated LEF/TCF transcription
complexes during hair follicle development and differentiation. Development 126:
4557-4568.
21. Rosenquist TA, Martin GR (1996) Fibroblast growth factor signalling in the hair growth
cycle: expression of the fibroblast growth factor receptor and ligand genes in the
murine hair follicle. Dev Dyn 205: 379-386.
22. Wilson N, Hynd PI, Powell BC (1999) The role of BMP-2 and BMP-4 in follicle initiation
and the murine hair cycle. Exp Dermatol 8: 367-368.
23. Kulessa H, Turk G, Hogan BL (2000) Inhibition of Bmp signaling affects growth and
differentiation in the anagen hair follicle. EMBO J 19: 6664-6674.
24. Smyth N, Vatansever HS, Murray P, Meyer M, Frie C, et al. (1999) Absence of basement
membranes after targeting the LAMC1 gene results in embryonic lethality due to
failure of endoderm differentiation. J Cell Biol 144: 151-160.
24
25. Li J, Tzu J, Chen Y, Zhang YP, Nguyen NT, et al. (2003) Laminin-10 is crucial for hair
morphogenesis. EMBO J 22: 2400-2410.
26. Woodley DT, Bachmann PM, O'Keefe EJ (1988) Laminin inhibits human keratinocyte
migration. J Cell Physiol 136: 140-146.
27. Schuler F, Sorokin LM (1995) Expression of laminin isoforms in mouse myogenic cells in
vitro and in vivo. J Cell Sci 108 ( Pt 12): 3795-3805.
28. Stanley JR, Hawley-Nelson P, Yaar M, Martin GR, Katz SI (1982) Laminin and bullous
pemphigoid antigen are distinct basement membrane proteins synthesized by
epidermal cells. J Invest Dermatol 78: 456-459.
29. Rousselle P, Lunstrum GP, Keene DR, Burgeson RE (1991) Kalinin: an epitheliumspecific basement membrane adhesion molecule that is a component of anchoring
filaments. J Cell Biol 114: 567-576.
30. Nanba D, Hieda Y, Nakanishi Y (2000) Remodeling of desmosomal and
hemidesmosomal adhesion systems during early morphogenesis of mouse pelage
hair follicles. J Invest Dermatol 114: 171-177.
31. Hayashi K, Mochizuki M, Nomizu M, Uchinuma E, Yamashina S, et al. (2002) Inhibition of
hair follicle growth by a laminin-1 G-domain peptide, RKRLQVQLSIRT, in an organ
culture of isolated vibrissa rudiment. J Invest Dermatol 118: 712-718.
32. Hafner M, Wenk J, Nenci A, Pasparakis M, Scharffetter-Kochanek K, et al. (2004) Keratin
14 Cre transgenic mice authenticate keratin 14 as an oocyte-expressed protein.
Genesis 38: 176-181.
33. Chen ZL, Strickland S (2003) Laminin gamma1 is critical for Schwann cell differentiation,
axon myelination, and regeneration in the peripheral nerve. J Cell Biol 163: 889-899.
34. Vasioukhin V, Degenstein L, Wise B, Fuchs E (1999) The magical touch: genome
targeting in epidermal stem cells induced by tamoxifen application to mouse skin.
Proc Natl Acad Sci U S A 96: 8551-8556.
35. Sorokin LM, Pausch F, Frieser M, Kroger S, Ohage E, et al. (1997) Developmental
regulation of the laminin alpha5 chain suggests a role in epithelial and endothelial cell
maturation. Dev Biol 189: 285-300.
36. Kohfeldt E, Sasaki T, Gohring W, Timpl R (1998) Nidogen-2: a new basement membrane
protein with diverse binding properties. J Mol Biol 282: 99-109.
37. Yurchenco PD, Quan Y, Colognato H, Mathus T, Harrison D, et al. (1997) The alpha
chain of laminin-1 is independently secreted and drives secretion of its beta- and
gamma-chain partners. Proc Natl Acad Sci U S A 94: 10189-10194.
38. Auber L (1952) VII.—The Anatomy of Follicles Producing Wool-Fibres, with special
reference to Keratinization. Earth and Environmental Science Transactions of the
Royal Society of Edinburgh 62: 191-254.
39. Legue E, Nicolas JF (2005) Hair follicle renewal: organization of stem cells in the matrix
and the role of stereotyped lineages and behaviors. Development 132: 4143-4154.
40. Langbein L, Schweizer J (2005) Keratins of the human hair follicle. Int Rev Cytol 243: 178.
41. Schweizer J, Langbein L, Rogers MA, Winter H (2007) Hair follicle-specific keratins and
their diseases. Exp Cell Res 313: 2010-2020.
42. Godwin AR, Capecchi MR (1998) Hoxc13 mutant mice lack external hair. Genes Dev 12:
11-20.
43. Hong HK, Noveroske JK, Headon DJ, Liu T, Sy MS, et al. (2001) The winged
helix/forkhead transcription factor Foxq1 regulates differentiation of hair in satin mice.
Genesis 29: 163-171.
44. Tkatchenko AV, Visconti RP, Shang L, Papenbrock T, Pruett ND, et al. (2001)
Overexpression of Hoxc13 in differentiating keratinocytes results in downregulation of
a novel hair keratin gene cluster and alopecia. Development 128: 1547-1558.
45. Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110: 673687.
46. Schober S, Mielenz D, Echtermeyer F, Hapke S, Poschl E, et al. (2000) The role of
extracellular and cytoplasmic splice domains of alpha7-integrin in cell adhesion and
migration on laminins. Exp Cell Res 255: 303-313.
25
47. von der Mark H, Williams I, Wendler O, Sorokin L, von der Mark K, et al. (2002)
Alternative splice variants of alpha 7 beta 1 integrin selectively recognize different
laminin isoforms. J Biol Chem 277: 6012-6016.
48. Herzog C, Has C, Franzke CW, Echtermeyer FG, Schlotzer-Schrehardt U, et al. (2004)
Dystroglycan in skin and cutaneous cells: beta-subunit is shed from the cell surface. J
Invest Dermatol 122: 1372-1380.
49. Talts JF, Andac Z, Gohring W, Brancaccio A, Timpl R (1999) Binding of the G domains of
laminin alpha1 and alpha2 chains and perlecan to heparin, sulfatides, alphadystroglycan and several extracellular matrix proteins. EMBO J 18: 863-870.
50. Yu H, Talts JF (2003) Beta1 integrin and alpha-dystroglycan binding sites are localized to
different laminin-G-domain-like (LG) modules within the laminin alpha5 chain G
domain. Biochem J 371: 289-299.
51. Kikkawa Y, Sanzen N, Sekiguchi K (1998) Isolation and characterization of laminin-10/11
secreted by human lung carcinoma cells. laminin-10/11 mediates cell adhesion
through integrin alpha3 beta1. J Biol Chem 273: 15854-15859.
52. Nishiuchi R, Takagi J, Hayashi M, Ido H, Yagi Y, et al. (2006) Ligand-binding specificities
of laminin-binding integrins: a comprehensive survey of laminin-integrin interactions
using recombinant alpha3beta1, alpha6beta1, alpha7beta1 and alpha6beta4
integrins. Matrix Biol 25: 189-197.
53. Brakebusch C, Grose R, Quondamatteo F, Ramirez A, Jorcano JL, et al. (2000) Skin and
hair follicle integrity is crucially dependent on β1 integrin expression on keratinocytes.
The EMBO Journal 19: 3990-4003.
54. Raghavan S, Bauer C, Mundschau G, Li Q, Fuchs E (2000) Conditional ablation of beta1
integrin in skin. Severe defects in epidermal proliferation, basement membrane
formation, and hair follicle invagination. J Cell Biol 150: 1149-1160.
55. Conti FJ, Rudling RJ, Robson A, Hodivala-Dilke KM (2003) alpha3beta1-integrin
regulates hair follicle but not interfollicular morphogenesis in adult epidermis. J Cell
Sci 116: 2737-2747.
56. Margadant C, Raymond K, Kreft M, Sachs N, Janssen H, et al. (2009) Integrin
alpha3beta1 inhibits directional migration and wound re-epithelialization in the skin. J
Cell Sci 122: 278-288.
57. DeRouen MC, Zhen H, Tan SH, Williams S, Marinkovich MP, et al. Laminin-511 and
integrin beta-1 in hair follicle development and basal cell carcinoma formation. BMC
Dev Biol 10: 112.
58. Flanagan SP (1966) 'Nude', a new hairless gene with pleiotropic effects in the mouse.
Genet Res 8: 295-309.
59. Mecklenburg L, Nakamura M, Sundberg JP, Paus R (2001) The nude mouse skin
phenotype: the role of Foxn1 in hair follicle development and cycling. Exp Mol Pathol
71: 171-178.
60. Potter CS, Pruett ND, Kern MJ, Baybo MA, Godwin AR, et al. (2011) The nude mutant
gene Foxn1 is a HOXC13 regulatory target during hair follicle and nail differentiation.
J Invest Dermatol 131: 828-837.
61. Potter CS, Peterson RL, Barth JL, Pruett ND, Jacobs DF, et al. (2006) Evidence that the
satin hair mutant gene Foxq1 is among multiple and functionally diverse regulatory
targets for Hoxc13 during hair follicle differentiation. J Biol Chem 281: 29245-29255.
62. Cai J, Lee J, Kopan R, Ma L (2009) Genetic interplays between Msx2 and Foxn1 are
required for Notch1 expression and hair shaft differentiation. Dev Biol 326: 420-430.
63. Hsu YC, Li L, Fuchs E Emerging interactions between skin stem cells and their niches.
Nat Med 20: 847-856.
64. Genander M, Cook PJ, Ramskold D, Keyes BE, Mertz AF, et al. BMP signaling and its
pSMAD1/5 target genes differentially regulate hair follicle stem cell lineages. Cell
Stem Cell 15: 619-633.
26
Figure legends
Figure 1. Loss of laminin-511 deposition due to depletion of laminin γ1 in basal
keratinocytes leads to changes in laminin composition and to ultrastructural
defects of the DEJ. (a) LAMC1EKO mice show retarded HF morphogenesis, delayed
pigmentation and syndactyly at the hind limbs, when compared with control
littermates (P1, P3, P6, P10). Representative pictures are shown. (b) Representative
IF stainings of back skin sections of control and LAMC1EKO mice (E18.5, P1, P10)
using antibodies against the laminin γ1, α5 and α2 chains. The DEJ of LAMC1EKO
mice shows reduced deposition of laminin γ1 (arrow), loss of laminin α5 (arrowhead)
and ectopic deposition of α2 (asterisk). Scale bar: E18.5: 25µm; P1, P10: 50µm; n=58. (c) Immunoblot analysis of keratinocyte lysates (50µg protein) and protein
precipitates from 0.5ml serum-free supernatants of keratinocytes and fibroblasts
isolated from control or LAMC1EKO skin. Primary cells were cultivated from four mice
per genotype at P1. Keratinocytes and fibroblasts were starved for 48h and 24h,
respectively. Ponceau S staining or GAPDH were used as loading control. (d)
Ultrastructure of control and mutant skin sections (P1 and P9). Asterisk: less tightly
packed basal keratinocytes; arrow: areas with interrupted lamina densa between
hemidesmosomes. Insets: three-fold higher magnification of the BM zone. E:
epidermis; D: dermis. Scale bar: 1µm (P1), and 5µm (P9).
Figure 2. HF morphogenesis is delayed in mice lacking epidermal laminin γ1
expression. (a) H&E stainings of paraffin-embedded back skin sections of control
and mutant mice at P1 and P10. Arrow: curved and inappropriately angled HFs;
arrowhead: barely formed HS fragments; asterisks: thickening of the epidermis
around HF regions. Scale bar: 50µm. (b) Ultrastructure of the follicular BM
surrounding the dermal papilla in HFs of control and mutant mice (P3). Arrow:
thickening of the BM; arrowhead: intact areas of the BM. Scale bar: 1µm. (c)
27
Categorization of HFs at P1 and P10 according to characteristics of HF
morphogenesis stages 0-8 as described (below). Quantification of total number (%)
of HFs for each developmental stage; means±SEM; P1: n=5-6; P10: n=5. (d)
Immunohistochemical staining of paraffin sections for Ki67 positive hair matrix cells
(brown) in control and LAMC1EKO skin (P10). Scale bar: 25µm. (e) Quantification of
proliferating matrix cells below Auber’s line (red line) as a percent to the total number
of matrix cells; means±SEM; n=6.
Figure 3. Early embryonic HF development is not disturbed in skin of
LAMC1EKO mice. (a) Categorization of embryonic HFs at day 16.5. Counting of total
numbers (%) of HF precursors in placode, peg or germ stages; means±SD; n=4. (b)
IF stainings of back skin cryosections from control and mutant mice at E16.5. Dotted
line: polarized area for P-cadherin expression, marks anterior side of HFs. Scale bar:
20µm (one bar for all). (c) Quantification of placode numbers in E16.5 mice after IF
staining for P-cadherin. 6-9 microscopic fields per animal were analyzed; means±SD,
n=4. (d) Quantification of HF numbers which end in the epidermis in 8-week old mice
under bright field microscopy after H&E staining. 4-5 microscopic fields per animal
were analyzed; means±SD, n=5-6.
Figure 4. Altered differentiation of the HS in LAMC1EKO mice. (a) Partial list of
significantly downregulated HS keratins in the epidermis and dermis of male
LAMC1EKO skin at P6 compared to control male mice as obtained by microarray
analysis (>1.5 fold, p<0.05); n=2. Results were validated by IF; n=3 (b) IF stainings of
frozen sections of control and LAMC1EKO skin (P6) using different hair keratin
antibodies (green). Development of the companion layer (K75) and the IRS (K28) of
control and LAMC1EKO HFs were comparable. In LAMC1EKO mice, differentiation of
hair cuticle (K82, K85), cortex (K33, K35, K85, K86) and medulla (K81) of the HS
28
were abnormal compared to control mice. Co-staining with K14 antibody (red). Nuclei
(DAPI; blue). Scale bar: 20µm (one bar for all).
Figure 5. Downregulated expression and impaired localization of transcription
factors regulating differentiation of hair specific keratin genes. (a) Summarized
list of significantly down-regulated transcription factors and regulatory proteins at P6
in the epidermis and dermis of LAMC1EKO skin compared to control mice identified by
RNA microarray analysis. (b) IF stainings of paraffin sections of skin from control and
mutant mice (P1, P6 and P10) using antibodies against FoxN1 and HoxC13. Nuclei
(DAPI; blue). Insets: two-fold higher magnification. Scale bar: 20µm; n=4. (c)
Immunoblot analyses of FoxN1, HoxC13, Msx2 and (d) phosphorylated Smad1 and 5 (pSmad1/5) protein in LAMC1EKO mice compared to control animals (50µg total skin
lysates; P6, P10). GAPDH: loading control; n=4.
Figure 6. Schematic model summarizing the activity of proteins in HF
morphogenesis that are downregulated in LAMC1EKO skin. Expression of HS
keratins during HF differentiation is mainly regulated by signaling molecules of the
BMP-MSX2-HOXC13-FOXN1 signaling axis. The loss of epidermal laminin-511
results in downregulation of this BMP signaling axis, which in turn results in reduced
expression of the HS keratins depicted in the model. Red arrows: downregulated in
LAMC1EKO skin. See text for additional discussion. DP: dermal papilla, ORS: outer
root sheath, SCs: stem cells.
29