Skip to main content

Regulation of epidermal stem cells by a molecular clock during tissue homeostasis and cancer

Final Report Summary - EPISTEMCLOCK (Regulation of epidermal stem cells by a molecular clock during tissue homeostasis and cancer)

The epidermis needs to renew constantly in adults to maintain its function. This process is called homeostasis and relies on a population of epidermal stem cells (EpSCs) that self-renew and can undergo terminal differentiation (Blanpain 2009). Failures in their strict regulation can lead to premature ageing or to the development of tumours (Owens 2003). Little is still known about the molecular pathways relevant for the transition between the inactive and active states of EpSCs within the stem cell niche. The aim of our work was then to understand the molecular mechanisms that control the behaviour of normal adult stem cells during tissue homeostasis and how their deregulation contributes to carcinogenesis. Preceding data in our laboratory showed that molecular clock machinery is expressed differentially in active EpSCs compared with inactive EpSCs.

The molecular clock combines autoregulatory cycles of transcriptional regulation, post-translational modifications, cellular sublocalisation and proteosomal degradation (Levi 2007). The critical regulator is composed of a heterodimer of Clock / Bmal that binds to E-boxes at the promoters of the Period (Per), Cryptochrome (Cry) and Dec genes to induce their transcription. Functional oscillations of these genes orquestrate physiological functions including metabolic activity, wakefulness/sleep, hormone segregation, and cerebral activity, among others. We wanted to understand the correlation of the molecular clock with the skin correct maintenance.

The two main objectives of the project were to understand:

(1) the role of the molecular clock on the behaviour of the EpSCs during epidermal homeostasis;
(2) whether there is a role for the molecular clock of cancer stem cells in skin Squamous cell carcinoma.

To approach the first objective, we decided to monitor the clock activity of the EpSCs using a Per1::Venus reporter mouse model described in Cheng et al. In this mouse model, the fluorescent protein Venus is expressed recapitulating the endogenous Per1 expression. It was then possible to separate by fluorescence-activated cell sorting (FACS) the Venus positive and negative populations inside the stem cells of the back skin of Per1::Venus mice (the skin stem cells express the membrane markers alpha6 and CD34).

We set up a collaboration with the Bioinformatics Unit at CRG and we obtained the global transcriptome of both populations. In the arrays results we found that Venus positive cells had higher levels of expression of Wnt-signalling genes such as Fzd2 and Dkk3, as well as inhibitors of the Tgf pathway such as Smad7 or Smurf. Wnt and Tgfb pathways are well known EpSC regulators. This result suggests that the EpSCs have a different predisposition to respond to dormancy or activation stimuli depending on the state of their molecular clock, what constitutes then a fine-tuning mechanism for the correct skin homeostasis.

In order to understand if deregulation of the clock in EpSCs could predispose tissues to carcinoma formation, we decided to study what is the state of the molecular clock in carcinoma cancer stem cells compared with normal EpSCs. We proposed to perform immunohistochemistry of about 150 tumours that had been collected in the Hospital del Mar during the last two years. Commercial antibodies have been tested to this end, but we could not find specific staining for Bmal, Dec2, Per2 or Per1.

To overpass this issue I designed peptides for each of these proteins in order to immunise rabbits and obtain in-house antibodies. However, the antibodies produced were not good enough to perform neither western blot analysis nor immunofluorescence. This fact has retarded considerably the overall performance of the project.

To overpass this, we decided to study the state of the clock in isolated single cancer stem cells from carcinomas. The idea was to create 'humanised' carcinomas transplating SCC cells, in the back skin of mice. We needed also to engineer a fluorescent clock reporter in order to know the state of the molecular clock at each time point at the single cell level. Mixing the SCC cells with keratinocytes and cancer associated fibroblasts, we were able to set up the right conditions to give rise to tumours that resemble the SCC seen in humans.

However, due to different issues, we were not able to create a good fluorescent clock reporter to infect the cells with, so we decided to move to mouse models to approach the above question.

The idea was to have skin tumours in mice, expressing a fluorescent protein to signal the state of the clock at the single cell level. To this end, we crossed K5-SOS mice (which develop spontaneous papillomas on the skin, Sibila et al, 2000) with the Per1::Venus mice explained above. The markers for EpSCs alpha6 and cd34 have been also described to label the cancer stem cell population inside carcinomas (Malanchi et al.). So we could compare the percentage of active-green cells in normal tissue compared to tumours. The result was that the number of stem cells (alpha6bright / cd34+) that expressed Venus under the control of the Per1 promoter decreased in the tumours compared with normal skin. This is in line with Fu et al. that showed that mice where Per2 is inactive are more predisposed to develop tumours compared with wild type animals and with Gery et al. that showed that Per1 levels are down in human lung and breast cancers compared with normal tissue.

Moreover, we were able to FAC sort the two cancer stem cell populations, positive or not to clock-Venus fluorescence and we compared their global transcriptomes. Detailed analysis of the arrays shows that Venus positive cells present overexpression of genes related to keratinisation such as filaggrin, hornerin, loricrin, Tprg, involucrin and cornifelin. As well as downregulation of genes associated to less differentiated cels such as Keratin15.

Thus, this result suggests that the Venus positive cells representing around the 5 % of the cancer stem cell in the skin tumours (defined for the high expression of alpha6 and the expression of cd34) corresponds to the more differentiated cells inside that population. To understand if deregulation of the clock predisposes the tissue to form carcinomas, we crossed Bmal conditional KO mice (BmalloxP / loxP crossed with Keratin14-CrePgr) with K5-SOS mice. Preliminary results show that Bmal-/- mice still develop tumours althought there is a delay in the tumour formation and progression.

Thus, we have demonstrated that the molecular clock machinery of EpSCs can be present in different states of activation. This means that these cells have a different predisposition to respond to external stimuli of activation, assuring the correct homeostasis of the skin.

Moreover, we have indications showing that the clock machinery is downregulated in carcinomas and that deregulation of the fine-tuning of the clock machinery in skin stem cells can lead to a predisposition to develop skin tumours. This could have a great socio-economic impact since Squamous cell carcinomas is one of the most diagnosed type of solid tumour in industrialised countries. Understanding why carcinoma cells abnormally respond to homeostatic cues could help in the identification of new diagnosis markers important for the detection of the disease at early stages and could rise the possibility to follow novel anti-tumour strategies.

Moreover, from the technical point of view, we have set up an assay in order to develop human SCCs in the back skin of mice. This assay can be used to study the effect of anticancer drugs in a relatively quick and cheap way.


(1) Blanpain C, Fuchs E. Epidermal homeostasis: a balancing act of stem cells in the skin. Nat Rev Mol Cell Biol 10:207 (2009)
(2) Owens DM, Watt FM. Contribution of stem cells and differentiated cells to epidermal tumours. Nat Rev Cancer 3:444 (2003)
(3) Levi F, Schibler U. Circadian rhythms: mechanisms and therapeutic implications. Annu Rev Pharmacol Toxicol 47:593 (2007)
(4) Sibilia M. et al. The EGF receptor provides an essential survival signal for SOS-dependent skin tumor development. Cell. 21;102(2):211-20 (2000)
(5) Cheng HY. et al. Segregation of expression of mPeriod gene homologs in neurons and glia: possible divergent roles of mPeriod1 and mPeriod2 in the brain.. Hum Mol Genet. Aug 15;18(16):3110-24 (2009)
(6) Malanchi I. et al. Cutaneous cancer stem cell maintenance is dependent on beta-catenin signalling. Nature (2008)
(7) Fu L. et al. The circadian gene Period2 plays an important role in tumor suppression and DNA damage response in vivo. Cell. (2002) Oct 4;111(1):41-50. Erratum in: Cell 2002 Dec 27;111(7):1055
(8) Gery S et al. The circadian gene per1 plays an important role in cell growth and DNA damage control in human cancer cells. Mol Cell. May 5;22(3):375-82. (2006)