Cutaneous cancer stem cell maintenance is
dependent on b-catenin signalling
Ilaria Malanchi1, Hector Peinado2, Deepika Kassen1, Thomas Hussenet1, Daniel Metzger3, Pierre Chambon3,
Marcel Huber4, Daniel Hohl4, Amparo Cano2, Walter Birchmeier5& Joerg Huelsken1
Continuous turnover of epithelia is ensured by the extensive self-
renewal capacity of tissue-specific stem cells1. Similarly, epithelial
tumour maintenance relies on cancer stem cells (CSCs), which co-
opt stem cell properties2. For most tumours, the cellular origin of
these CSCs and regulatory pathways essential for sustaining stem-
ness have not been identified. In murine skin, follicular morpho-
Here we identify a population of cells in early epidermal tumours
characterized by phenotypic and functional similarities to normal
bulge skin stem cells. This population contains CSCs, which are
the only cells with tumour initiation properties. Transplants
derived from these CSCs preserve the hierarchical organization
of the primary tumour. We describe b-catenin signalling3as being
essential in sustaining the CSC phenotype. Ablation of the b-cate-
nin gene results in the loss of CSCs and complete tumour regres-
sion. In addition, we provide evidence for the involvement of
increased b-catenin signalling in malignant human squamous cell
carcinomas. Because Wnt/b-catenin signalling is not essential for
thus be targeted to eliminate CSCs4and consequently eradicate
squamous cell carcinomas.
Two populations of epithelial stem cells have been identified in
murine skin1,5. One is located in the bulge region of hair follicles and
expression6,7with concurrent exclusion of dead or non-keratinocyte
cells by using 7-amino-actinomycin D (7-AAD), CD31 and CD45
(Fig. 1a, e). Alternatively, and with identical results, we isolate bulge
stem cells as CD341yellow-fluorescent-protein-positive (YFP1)
cells from K14-cre:ROSA26-lox-stop-lox-YFP mice, which express
YFP in keratinocyte lineages only (data not shown). CD34-positive
keratinocytes in normal skin. In cutaneous tumours derived by
13-acetate (DMBA/TPA)) carcinogenesis we observe a nine-fold
increase in the CD341or CD341YFP1population (Fig. 1b, d;
n54; 17.264.8% (mean6s.d.) of keratinocyte-lineage-derived
tumour cells). CD34-expressing tumour cells are located in close
contact to stromal areas (Fig. 1g, h) and also express additional,
established markers of bulge skin stem cells8,9such as Gas6,
ephrinA4, Sox9, Runx1, T-cell-specific transcription factor 3
(TCF-3) and tenascinC (Fig. 1i–m and Supplementary Fig. 1).
Although the DMBA/TPA model generates mainly benign papillo-
matous lesions10, occasionally progression to squamous cell carcino-
mas (SCCs) is observed. Within these advanced tumours the CD341
found at the invading front, in line with recent reports identifying
CSCs as the main source of tumour invasion11,12. CD34 seems also to
be of functional importance in skin tumour formation6, although its
mode of action is currently not understood. Together these data
identify a population of cells in early and advanced epidermal skin
tumours that closely resembles that of bulge stem cells.
The in vivo tumorigenic capacity of CD341cells isolated by mag-
netic cell sorting (MACS) (Fig. 1c and Supplementary Fig. 2) was
compared with that of unsorted tumour cells by serial dilution
experiments in an orthotopic transplantation system, using tumour
cells in conjunction with primary keratinocytes and dermal fibro-
over 100-fold more potent in initiating secondary tumours than
a stable population of CD341cells that retain tumour initiation
potential, giving rise to tertiary tumours (Fig. 2b and data not
shown). In contrast, CD342cells (Supplementary Fig. 2) never pro-
duce tumours (n59; 2.53104cells per transplant), although
extensive proliferation is detected in this population, which presum-
derived from CD341cells closely resemble the architecture of the
parental tumour and maintain a small population of CD341cells
among the majority of keratin-10-expressing cells committed to dif-
while producing differentiated progeny further substantiates the
stem-cell nature of cells contained in the CD341population. The
relative abundance of this CSC-containing population and the
enrichment of tumour-initiating cells is comparable to that of other
solid tumours13,14. However, in contrast with these other systems,
orthotopic transplantations in theskin areaccompanied by aninitial
lackofnutrient supplyandastronginflammatory response—mainly
by innate immune cells—which may explain why more tumour cells
(unsorted or CD341) are required to generate secondary tumours in
provide compelling evidence for this idea by combining chemical
carcinogenesis with lineage tracing of bulge stem cells, using K15-
creERT:ROSA26-lox-stop-lox-lacZ mice as a genetic tool (Supple-
mentary Fig. 3). Taken together, these data strongly suggest that
bulge-derived CD341stem cells are a source of tumorigenesis.
has been implicated in a variety of human tumours3,20. We therefore
examined the involvement of this pathway in chemical-derived skin
1E´cole Polytechnique Fe ´de ´rale de Lausanne (EPFL)/ISREC (Swiss Institute for Experimental Cancer Research) and National Center of Competence in Research (NCCR) ‘Molecular
Oncology’, Chemin des Boveresses 155, 1066 Epalinges, Switzerland.2Departamento de Bioquı ´mica, Instituto de Investigaciones Biome ´dicas ‘Alberto Sols’, CSIC-UAM, Arturo
Duperier 4, 28029 Madrid, Spain.3Institut de Ge ´ne ´tique et de Biologie Mole ´culaire et Cellulaire (IGBMC), BP 10142, CU de Strasbourg, 67404 Illkirch, France.4Laboratory of
Cutaneous Biology, Dermatology CHUV and FBM UNIL, Avenue de Beaumont 29, 1011 Lausanne, Switzerland.5Max Delbru ¨ck Centrum, Robert-Roessle-Strasse 10, 13122 Berlin,
Vol 452|3 April 2008|doi:10.1038/nature06835
tumours and assessed b-catenin signalling activity in vivo by using a
reporter mouse strain with lacZ knock-in at the conductin/axin2
locus, which is a well-established general target gene of Wnt signal-
observe strong signalling activity in tumours (Fig. 3b), in particular
in the region where CD341cells localize. Similarly, nuclear b-cate-
nin, a hallmark of active signalling, is preferentially found in CD341
CSCs (Fig. 3c; 21.768.4% (mean6s.d.) of CD341cells versus
3.3%61.2% (mean6s.d.) of CD342cells; see ref. 12). This signal-
ling activity prompted us to perform a dedicated tumour regression
experiment by inducing b-catenin gene ablation in established
Fig. 3g). b-Catenin deletion results in complete regression, whereas
the growth of control tumours is unaffected by the activation of
creERT2. Regression is characterized byextensive terminal differenti-
ation and the complete absence of mitotic figures within five to six
weeks (Fig. 3d). Similarly, spontaneous regression was observed in a
skin tumour model (TG.AC)22expressing activated H-ras (data not
shown). In addition, deletion of b-catenin in the skin also prevents
tumour initiation by chemical carcinogenesis or H-ras (Supple-
mentary Figs 4 and 5). The first phenotype we detect in b-catenin-
deficient tumours two weeks after deletion is a marked loss of the
CD341population of CSCs as analysed by FACS and bromodeox-
yuridine (BrdU) long-term label-retaining assays (Fig. 3e, f; n57;
0 200 400 600 8001,0000 200 400 600 8001,0000 200 400 600 800 1,000
0 200 400 600 800 1,000
Figure 1 | Skin cancers contain cells with bulge stem cell phenotype.
a–d, Density plots of cells with stem cell phenotype defined as CD341and
viable (7AAD2) as well as CD312CD452or K14-cre:R26–YFP1(n54
each): adult mouse skin (a), murine papilloma (b) and murine SCC
(d). c, Purity of CD341tumour cells after enrichment by MACS.
e, f, Immunofluorescence of CD34 in normal murine skin (e) and tumour
(f). DAPI, 4,6-diamidino-2-phenylindole. g, CD341cells (red) in murine
magnification with small groups of invading tumour cells. h–m, expression
of bulge stem cell markers in papillomas as detected by RNA in situ
hybridization: CD34 (h), Gas6 (i), ephrinA4 (j), SOX9 (k), Runx (l) and
TCF-3 (m). Scale bars, 50mm.
0 2 4 6 8 10
101 102 103 104
101 102 103 104
0 200 400 600800 1,000
101 102 103 104
5 × 102
5 × 103
Transplanted tumour cells
5 × 104
5 × 105
Frequency of tumour formation (%)
Figure 2 | Cancer stem cells efficiently initiate secondary tumours that
recapitulate the organization of the primary tumour. a, Diagram
summarizing the frequency of tumour formation in orthotopic tumour
transplants using unsorted cells (filled diamonds) or CD341cells (open
circles) in varying amounts. The n value for each point is shown.
b, Abundance of CSCs (CD3417AAD2CD312CD452) in secondary
tumours derived from orthotopic transplantations of CD341cells.
c–e, Assessment of the distribution of proliferating tumour cells by
immunofluorescence analysis for CD34 (red) and Ki67 (green) (c) and by
short-term BrdU incorporation in CD342cells (d) or CD341cells
(e). f, g, Hierarchical organization of primary tumours (f) is maintained in
secondary tumours (g) as determined by immunofluorescence analysis of
CD34 (red) and keratin10 (K10, green). Scale bars, 50mm.
NATURE|Vol 452|3 April 2008
wild-type, 17.264.8% (mean6s.d.); b-cateninD/D, 3.062.0%;
Supplementary Fig. 6). This loss of CD341cells precedes any
alterations in tumour cell proliferation, apoptosis or senescence,
which remain unaltered at this time point, whereas proliferation is
affected only laterwhencellsdifferentiate (Supplementary Figs 7and
8, and data not shown). b-Catenin-deficient tumour cells devoid of
the CD341population lose their ability to initiate secondary
tumours, even if transplanted in large numbers (up to 106; n511).
In the reverse experiment we find that the tamoxifen-induced
expression of an active, non-degradable mutant of b-catenin is
sufficient to expand (2.5-fold) the pool of bulge stem cells within
five days (Supplementary Fig. 9; n58; wild-type, 1.860.2%
(mean6s.d.); b-cateninfloxEx3/1:K14-creERT2, 4.661.1%). We con-
clude that b-catenin has an essential function in the maintenance of
Similarly to murine tumours, human skin SCCs show constitutive
activation of b-catenin signalling as revealed by nuclear localization
tumours, Fig. 4a) than previously reported23–25. Likewise, expression
of the general Wnt/b-catenin target gene conductin is often detected
than would be predicted from a small population of CSCs. This
emphasizes that although b-catenin signalling is required for tumor-
igenesis (see also below) it is not strictly a marker of CSCs. Human
skin stem-cell markers have not been described and CD34 is not
expressed in human skin or SCCs, which precluded a CSC analysis
of human tumours. We therefore tested the importance of b-catenin
signalling in human SCC cell lines. We find that short hairpin RNA
(shRNA)-mediated knockdown of b-catenin strongly decreases
tumour growth of human SCC13 cells in xenografts (Fig. 4c and
Supplementary Fig. 11). Residual tumour areas are always positive
for b-catenin, indicating selection for cells that had escaped knock-
down (Supplementary Fig. 11). The main function of b-catenin in
SCCs seems to be lymphoid-enhancer binding factor/T-cell-specific
transcription factor (LEF/TCF)-regulated transcription, because
expression of a dominant-negative mutant of LEF-1 (DNLEF-1)
decreased tumour formation in human HN-30 SCC xenografts and
in murine orthotopic tumour transplants (Supplementary Fig. 12
and data not shown). This is supported by increased expression of
the absence of b-catenin (Supplementary Fig. 13).
Using transplantation experiments in pre-cancerous murine and
advanced human SCCs we show that b-catenin signalling is essential
for tumorigenesis. In the murine system, this signalling specifically
sustains CSCs, which are necessary for long-term tumour growth.
essential for skin tumorigenesis, but not adult skin homeostasis26. As
CSCs are expected to be responsible for tumour recurrence4, strat-
egies targeting CSCs may lead to more effective therapies. We now
provide evidence that differential requirements for b-catenin signal-
ling in cancer versus normal stem cells exist in skin, which could be
exploited for future therapy.
Primary tumour cells, keratinocytes and dermal cells were prepared and trans-
Tumour growth (%)
0 200 400 600 800 1,0000 200 400 600 800 1,000
Figure 3 | b-Catenin signalling is essential to maintain skin tumorigenesis.
a, b, b-Catenin signalling activity as detected by lacZ expression from the
(b). c, Nuclear, dephosphorylated b-catenin is enriched in CD341cells
(right) as detected by immunofluorescence analysis of cytospins after
magnetic sorting. DAPI, 4,6-diamidino-2-phenylindole. d, b-Catenin-
negative tumours undergo terminal differentiation after six weeks as
characterizedbyabundantkeratinization. e, f,AbundanceofCD341cellsin
control tumours (e) and b-catenin-deficient tumours (f) two weeks after
induction of deletion. g, Tumour regression after tamoxifen-induced
K14-creERT2:b-catenin1/D; mutant (open triangles), K14-creERT2:b-
cateninD/D). Tumour numbers at 8 weeks were set to 100%. Error bars show
s.d. Insets: immunohistochemistry (IHC) to detect efficiency of b-catenin
gene ablation. Scale bars, 50mm (a, b), 400mm (d) and 100mm (g).
Number of tumours
Number of tumours50
P < 0.0008
Figure 4 | Functional importance of b-catenin signalling in human skin
SCCs. a, b, Quantification of nuclear b-catenin (a, n595) and conductin
(b, n550) expression in human skin SCCs. a, Tumours displaying nuclear
of positive cells. c, shRNA-mediated knockdown of b-catenin in human
SCC13 cells strongly decreases tumour growth (tumour volume in mm3) in
xenografts (P,0.0008; Student’s t-test, n55; horizontal bars indicate the
NATURE|Vol 452|3 April 2008
MACS (Miltenyi) was used to isolate cell populations enriched or depleted for
The skin-specific ablation of b-catenin and the K14-cre, b-cateninfloxEx3, K14-
creERT2, conductinlacZ, ROSA26-lox-stop-lox-YFP and ROSA26-lox-stop-lox-lacZ
alleles have been described16,21,27–29. The K15-creERT2transgene used 5kilobases
daily for five days. For tumour initiation, mice were treated once with 100mg of
DMBA followed by 12.5mg of TPA twice per week. Mice expressing the v-H-ras
transgene (TG.AC)22, were treated with 5mg of TPA twice per week only.
Xenografts of human SCC cells were performed in Nude mice by using
growth-factor-reduced Matrigel with 23106HN30 or SCC13 cells stably trans-
duced by lentiviruses expressing dominant-negative DNLEF-1 (NM_010703;
nucleotides 1086–2183) or shRNAs directed against b-catenin30or GFP,
Full Methods and any associated references are available in the online version of
the paper at www.nature.com/nature.
Received 19 December 2007; accepted 8 February 2008.
1.Ito, M.et al. Stem cells in thehair follicle bulge contribute to wound repair but not
to homeostasis of the epidermis. Nature Med. 11, 1351–1354 (2005).
Al-Hajj, M., Becker, M. W., Wicha, M., Weissman, I. & Clarke, M. F. Therapeutic
implications of cancer stem cells. Curr. Opin. Genet. Dev. 14, 43–47 (2004).
Reya, T. & Clevers, H. Wnt signalling in stem cells and cancer. Nature 434,
Reya, T., Morrison, S. J., Clarke, M. F. & Weissman, I. L. Stem cells, cancer, and
cancer stem cells. Nature 414, 105–111 (2001).
Mackenzie, I. C. Retroviral transduction of murine epidermal stem cells
demonstrates clonal units of epidermal structure. J. Invest. Dermatol. 109,
tumor development in mice. Cancer Res. 67, 4173–4181 (2007).
Blanpain, C., Lowry, W. E., Geoghegan, A., Polak, L. & Fuchs, E. Self-renewal,
multipotency, and the existence of two cell populations within an epithelial stem
cell niche. Cell 118, 635–648 (2004).
Tumbar, T. et al. Defining the epithelial stem cell niche in skin. Science 303,
Liu, Y., Lyle, S., Yang, Z. & Cotsarelis, G. Keratin 15 promoter targets putative
epithelial stem cells in the hair follicle bulge. J. Invest. Dermatol. 121, 963–968
10. Quintanilla, M., Brown, K., Ramsden, M. & Balmain, A. Carcinogen-specific
mutation and amplification of Ha-ras during mouse skin carcinogenesis. Nature
322, 78–80 (1986).
11. Hermann, P. C. et al. Distinct populations of cancer stem cells determine tumor
growth and metastatic activity in human pancreatic cancer. Cell Stem Cell 1,
12. Gaspar, C. et al. Intracellular b-catenin accumulation underlies cancer stemness
and metastatic behaviour in an Apc mouse model for mammary tumorigenesis.
13. Al-Hajj, M., Wicha, M. S., Benito-Hernandez, A., Morrison, S. J. & Clarke, M. F.
Prospective identification of tumorigenic breast cancer cells. Proc. Natl Acad. Sci.
USA 100, 3983–3988 (2003).
14. O’Brien, C. A., Pollett, A., Gallinger, S. & Dick, J. E. A human colon cancer cell
capable of initiating tumour growth in immunodeficient mice. Nature 445,
15. Morris, R. J. Keratinocyte stem cells: targets for cutaneous carcinogens. J. Clin.
Invest. 106, 3–8 (2000).
16. Huelsken, J., Vogel, R., Erdmann, B., Cotsarelis, G. & Birchmeier, W. b-Catenin
controls hair follicle morphogenesis and stem cell differentiation in the skin. Cell
105, 533–545 (2001).
17. Merrill, B. J., Gat, U., DasGupta, R. & Fuchs, E. Tcf3 and Lef1 regulate lineage
initiation of hair follicle development. Dev. Cell 2, 643–653 (2002).
of DNLef1 in mouse epidermis results in differentiation of hair follicles into
squamous epidermal cysts and formation of skin tumours. Development 129,
20. Chan, E. F., Gat, U., McNiff, J. M. & Fuchs, E. A common human skin tumour is
caused by activating mutations in b-catenin. Nature Genet. 21, 410–413 (1999).
21. Lustig, B. et al. Negative feedback loop of Wnt signaling through upregulation of
conductin/axin2 in colorectal and liver tumors. Mol. Cell. Biol. 22, 1184–1193
22. Leder, A., Kuo, A., Cardiff, R. D., Sinn, E. & Leder, P. v-Ha-ras transgene abrogates
the initiation step in mouse skin tumorigenesis: effects of phorbol esters and
retinoic acid. Proc. Natl Acad. Sci. USA 87, 9178–9182 (1990).
23. Brasanac, D., Boricic,I., Todorovic, V.,Tomanovic, N. &Radojevic, S.Cyclin Aand
cell carcinoma of the skin. Br. J. Dermatol. 153, 1166–1175 (2005).
24. Doglioni,C.etal.Alterationsofb-cateninpathwayinnon-melanoma skintumors:
loss of a-ABC nuclear reactivity correlates with the presence of b-catenin gene
mutation. Am. J. Pathol. 163, 2277–2287 (2003).
25. Papadavid, E., Pignatelli, M., Zakynthinos, S., Krausz, T. & Chu, A. C. Abnormal
immunoreactivity of the E-cadherin/catenin (a-, b-, and c-) complex in
premalignant and malignant non-melanocytic skin tumours. J. Pathol. 196,
26. Oskarsson, T. et al. Skin epidermis lacking the c-Myc gene is resistant to Ras-
driven tumorigenesis but can reacquire sensitivity upon additional loss of the
p21Cip1gene. Genes Dev. 20, 2024–2029 (2006).
27. Harada, N. et al. Intestinal polyposis in mice with a dominant stable mutation of
the beta-catenin gene. EMBO J. 18, 5931–5942 (1999).
28. Li, M. et al. Skin abnormalities generated by temporally controlled RXRa
mutations in mouse epidermis. Nature 407, 633–636 (2000).
29. Srinivas, S. et al. Cre reporter strains produced by targeted insertion of EYFP and
ECFP into the ROSA26 locus. BMC Dev. Biol. 1, 4 (2001).
30. van de Wetering, M. et al. Specific inhibition of gene expression using a stably
integrated, inducible small-interfering-RNA vector. EMBO Rep. 4, 609–615
Supplementary Information is linked to the online version of the paper at
support and advice with FACS analysis; P. Dotto for suggestions during
establishment of the transplantation model; D. Trono for lentiviruses; and
I. Stamenkovich for counsel in skin pathologies. I.M., D.K. and J.H. were supported
in part by the Swiss League against Cancer, the SNF and the Swiss NCCR in
Molecular Oncology, and H.P. was supported by the Spanish Association for
Cancer Fight (AECC). Work in Madrid was supported in part by the Spanish
Ministry of Education and Science to A.C.
Author Contributions H.P., D.K. and T.H. performed research and analysed data;
D.H. analysed data; A.C. designed research; D.M., P.C., M.H., D.H. and W.B.
contributed vital reagents; I.M. and J.H. designed and performed research,
analysed data, and wrote the paper.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. Correspondence and requests for materials should be
addressed to J.H. (firstname.lastname@example.org).
NATURE|Vol 452|3 April 2008
METHODS Download full-text
Cell preparation, culture and transplantation.Isolation of cells and transplan-
tations were performed as described31. In brief, skins of newborn mice were
collected and incubated in 0.25% trypsin in PBS for 16h. Epidermis was sepa-
rated from the dermal compartment, chopped and incubated for 1h at 21uC in
low-calcium medium (LCM). Undissolved tissue clumps were removed by fil-
tration and keratinocytes were plated on collagen (Vitrogen100 collagen)-
treated dishes. The dermal compartment was chopped and incubated for
20min in 0.35% collagenase solution (Sigma) and diluted fivefold with Hanks
balanced salt solution (HBSS), and undigested tissue was removed by filtration.
The dermal cell suspension was centrifuged at 72g to obtain two fractions: pellet
(P1) and supernatant (S1): P1 was resuspended and centrifuged at 10g, and S1
was centrifuged at 220g. The supernatant of P1 and the pellet of S1 were pooled,
centrifuged at 72g and washed three times in HBSS. Fibroblasts were resus-
pended in LCM and plated. Primary adult mouse keratinocytes were prepared
as described above except for an initial removal of hair with a depilatory cream
(Cre `me De ´pilatoire; Veet). Cells were grown at 5% CO2and 34uC in LCM
(2gl21KCl, 1gl21MgSO4?7H2O, 34gl21NaCl, 11gl21NaHCO3, 0.7gl21
NaH2PO4-H2O, 1gl21glucose, phenol red (Sigma), non-essential and essential
amino acid mix (Invitrogen), vitamins (Invitrogen), antibiotic/antimycotic
(Invitrogen), epidermal growth factor (100ngml21), glutamine (Invitrogen),
45mM CaCl2, 4% fetal calf serum (FCS; Chelex100 treated)).
Tumours were harvested from mice, chopped with a razor blade and incu-
bated in collagenase solution (0.35% in HBSS) for 30–45min. The suspension
was filtered, then washed three times with PBS; cells were resuspended in LCM.
Lentiviral infections were performed for 24h in the presence of Polybrene. Cells
were washed three times with PBS and allowed to recover for one or two days
Orthotropic transplantations were performed on the back of Nude mice by
ness excision wounds. Mixtures of freshly prepared tumour cells (103to 106),
primary keratinocytes (106) and dermal cells (83105) were injected into these
Studies of human tumour samples were approved by the ethical commission
of Lausanne University, and patients’ informed consent was obtained. Mouse
experiments were performed in accordance with Swiss guidelines and approved
by the Veterinarian Office of Vaud, Switzerland.
Immunodetection, FACS and in situ hybridization. FACS staining was per-
formed with (3–5)3106cells in 400ml of PBS supplemented with 3%FCS and
CD31 and CD45 (Alexa488-streptavidin) antibodies (Pharmingen and
eBioscience). Antibodies directed against the following antigens were used for
immunodetection on frozen or paraffin tissue sections: CD34 (Pharmingen),
Ki67 (Novocastra), BrdU (Sigma), keratin10 (Covance), b-catenin (BD
Transduction Laboratories, Millipore and ref. 16), active caspase-3 (Cell
Signaling) and tenascinC (G. Orend and R. Chiquet-Ehrismann). Nuclear
b-catenin was detected by using a method described by the Clevers group32. In
situ hybridizations were performed as described previously16with the following
probes: CD34 (NM_133654, nucleotides
(NM_015732, nt395–1039), Conductin (NM_004655, nt1092–1872), Gas6
(NM_019521, nt310–1509), ephrinA4 (NM_007910, nt295–762), Sox9
(NM_011448, nt10–2530), Runx1 (NM_001111022.1, nt56–1819) and TCF3
LTR and BrdU assays. For analysis of long-term label-retaining (LTR) cells, 10-
day-old mice were injected with BrdU (50mg per kg body weight) every 12h,
with a total of five injections. After 30 days, mice were killed and LTR cells were
detected by immunohistochemistry. To achieve short-term labelling of cells
undergoing DNA synthesis, adult mice received one intraperitoneal injection
of BrdU at 100mg per kg body weight 2h before being killed.
31. Lichti, U. et al. In vivo regulation of murine hair growth: insights from grafting
defined cell populations onto nude mice. J. Invest. Dermatol. 101, 124–129 (1993).
32. Batlle, E. et al. b-catenin and TCF mediate cell positioning in the intestinal
epithelium by controlling the expression of EphB/ephrinB. Cell 111, 251–263