Portal de Periódicos da CAPES

Mechanisms regulating epidermal stem cells

2012; Springer Nature; Volume: 31; Issue: 9 Linguagem: Inglês

10.1038/emboj.2012.67

1460-2075

Benjamin Beck, Cédric Blanpain,

Wnt/β-catenin signaling in development and cancer

Review20 March 2012free access Mechanisms regulating epidermal stem cells Benjamin Beck Benjamin Beck Interdisciplinary Research Institute (IRIBHM), Université Libre de Bruxelles (ULB), Brussels, Belgium Search for more papers by this author Cédric Blanpain Corresponding Author Cédric Blanpain Interdisciplinary Research Institute (IRIBHM), Université Libre de Bruxelles (ULB), Brussels, Belgium WELBIO, Université Libre de Bruxelles (ULB), Brussels, Belgium Search for more papers by this author Benjamin Beck Benjamin Beck Interdisciplinary Research Institute (IRIBHM), Université Libre de Bruxelles (ULB), Brussels, Belgium Search for more papers by this author Cédric Blanpain Corresponding Author Cédric Blanpain Interdisciplinary Research Institute (IRIBHM), Université Libre de Bruxelles (ULB), Brussels, Belgium WELBIO, Université Libre de Bruxelles (ULB), Brussels, Belgium Search for more papers by this author Author Information Benjamin Beck1 and Cédric Blanpain 1,2 1Interdisciplinary Research Institute (IRIBHM), Université Libre de Bruxelles (ULB), Brussels, Belgium 2WELBIO, Université Libre de Bruxelles (ULB), Brussels, Belgium *Corresponding author. Interdisciplinary Research Institute (IRIBHM), Université Libre de Bruxelles (ULB), 808, Route de Lennik, BatC, C6-130, Brussels 1070, Belgium. Tel.: +32 2 555 4190; Fax: +32 2 555 4655; E-mail: [email protected] The EMBO Journal (2012)31:2067-2075https://doi.org/10.1038/emboj.2012.67 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The skin epidermis contains different appendages such as the hair follicle and the sebaceous glands. Recent studies demonstrated that several types of stem cells (SCs) exist in different niches within the epidermis and maintain discrete epidermal compartments, but the exact contribution of each SC populations under physiological conditions is still unclear. In addition, the precise mechanisms controlling the balance between proliferation and differentiation of epidermal SC still remain elusive. Recent studies provide new insights into these important questions by showing the contribution of hair follicle SC to the sebaceous lineage and the importance of chromatin modifications and micro-RNAs (miRs) in regulating epidermal SCs renewal and differentiation. In this review, we will discuss the importance of these papers to our understanding of the mechanisms that control epidermal SC functions. Introduction The skin epidermis is a stratified epithelium that constitutes a barrier protecting the animals against microorganisms, external stress and water loss. To ensure these vital functions, the skin epidermis must replace the cells that are naturally lost and must be able to repair the tissue following injuries. The skin epidermis contains different appendages such as the hair follicle, the sweat gland and the sebaceous gland (SG), which ensure protection, camouflage and thermoregulation. Different stem cells (SCs) present in the skin epidermis regulated the maintenance and repair of the different part of the skin epidermis during physiological conditions (Blanpain et al, 2007). During embryonic development, epidermis originates from a single layer of ectodermal cells that will undergo a stratification process allowing the formation of the different layers of differentiated cells required to fulfil its barrier function (basal, spinous and granular layers). The basal layer cells express high level of basal integrins, keratin 5 (K5) and K14, while the spinous layers express K1 and K10 and the granular layers accomplish the terminal differentiation process and expressed filaggrin and loricrin. P63, a gene of the p53 family, is expressed in basal cells of all stratified epithelia and is thought to represent a master regulator of the stratification process (Blanpain and Fuchs, 2007). In the absence of p63, epidermis indeed fails to form a stratified epithelium (Mills et al, 1999; Yang et al, 1999). Different theories have been proposed to account for the critical p63 function during epidermal morphogenesis. p63 has been proposed to be required for the initiation of the stratification programme (Mills et al, 1999), for epidermal SC renewal (Yang et al, 1999; Senoo et al, 2007) or both processes (Truong et al, 2006). HF morphogenesis begins during embryonic development through a series of epithelial-to-mesenchymal interactions and it is complete by postnatal day 8 in mice. After reaching a critical size, the lower part of the HF degenerates in a process called catagen. The permanent portion of the HF called the bulge region enters a resting stage (telogen) before entering in a new growing stage (anagen) (Paus et al, 1999). Throughout the life of the animals, HF will alternate cycle of growth and degeneration called the hair cycle. The cyclic HF regeneration is mediated by the presence of multipotent SC located in the bulge region. The first defined characteristic of bulge SC that has been identified is their relative quiescence, which can be visualized by their ability to retain nucleotide analogues during pulse chase experiments and called for this reason labelled retaining cells or LRCs (Cotsarelis et al, 1990; Morris and Potten, 1999). Bulge SC have been purified based on their slow cycling property (Tumbar et al, 2004) or the expression of several markers including CD34/CD49f (α6 integrin) (Trempus et al, 2003; Blanpain et al, 2004), keratin 15 (Morris et al, 2004), Lgr5 (Jaks et al, 2008). While bulge SC are able to differentiate into all epidermal lineage upon transplantation into immunodeficient mice, adult bulge SC lineage tracing experiments demonstrated that bulge SCs participate to the homeostasis of all HF cells below the SGs but not contribute to SG and IFE under physiological conditions (Morris et al, 2004; Levy et al, 2005). However, during pathological conditions such as during wounding, bulge SC are rapidly mobilized and participate into the repair of the IFE (Ito et al, 2005; Jaks et al, 2008; Nowak et al, 2008). The continuous replacement of terminally differentiated cells that are shed from the skin surface is mediated by the presence of multiple small epidermal proliferation units (EPUs) scattered all along the IFE that contain one SCs or progenitor cells (Ghazizadeh and Taichman, 2005). Upon division, these SC can undergo three fates: most of the divisions are asymmetric, giving rise to one SC and one committed cell, the remaining divisions are symmetric (renewal or differentiation). Based on clonal analysis and mathematical modelling, it has been recently proposed that the homeostasis of the IFE can be maintained by the presence of only one type of progenitors that stochastically choose between these three different fates with a fixed probability (Clayton et al, 2007; Doupe et al, 2010). Recent studies demonstrated that even more progenitors and SC exist in the epidermis and contribute to the maintenance of the other epidermal compartments such as the isthmus, the junctional zone, the infundibulum (Nijhof et al, 2006; Jensen et al, 2008; Jensen et al, 2009; Raymond et al, 2010; Snippert et al, 2010). But it is not clear what is the precise contribution of each SC populations under physiological conditions and how the compartmentalization of the epidermis is orchestrated. In addition, while different signalling pathways regulating epidermal SC functions have been identified, the precise mechanisms controlling the balance between proliferation and differentiation of epidermal SC still remain elusive. Especially, the impact of chromatin modifications and micro-RNAs (miRs) expression on epidermal SCs renewal and differentiation is also still poorly understood. Recently, a flurry of papers published in the EMBO Journal provides new insights into these outstanding questions (Chikh et al, 2011; Driskell et al, 2011; Mejetta et al, 2011; Petersson et al, 2011). In this review, we discuss the importance of these papers to the understanding of the heterogeneity of skin SCs, and the role of miR and histone modifications on epidermal SC functions. Contribution of bulge SC to the sebaceous gland lineage SG is a part of the pilosebaceous unit that through holocrine differentiation of the sebocytes secretes the sebum, which maintain the waterproof nature of the fur. The important cellular turnover within the SG is thought to be fuelled by the presence of SCs. Until recently, very little was known about the cellular and molecular mechanisms that control SG morphogenesis and homeostasis. Lineage tracing performed during epidermal development demonstrate that SG originates from bipotent embryonic progenitors common for both HF and SG lineages that expressed Shh and Sox9 at one time of their specification (Levy et al, 2005; Nowak et al, 2008). Lineage tracing of adult bulge SC was performed using different inducible CRE expressed under the regulatory region of K15, Lgr5 and K19, which are expressed specifically in bulge SC during telogen (Morris et al, 2004; Ito et al, 2005; Levy et al, 2005; Youssef et al, 2010). K19CREER and Lgr5CREER are knockin mice, in which the CREER is inserted into the endogenous locus of these genes (Barker et al, 2007; Means et al, 2008), whereas K15CREPR are transgenic mice made by random integration of fragment of the K15 promoter (Morris et al, 2004). In mice expressing one of these inducible CRE together with a reporter gene (e.g., Rosa/YFP or Rosa/LacZ), tamoxifen (TAM) or RU486 administration induces the activity of the inducible CRE, allowing the excision of a stop cassette leading to the reporter gene expression in bulge SC and all their future progeny. These experiments showed that only rarely were SG cells labelled using these different bulge specific inducible CRE, suggesting that the SG is mainly maintained independently of bulge SC during physiological condition. The transcriptional repressor Blimp1 is expressed during morphogenesis in the upper HF adjacent to the emergence of the newly formed SGs. Blimp1 expressing cells encompass progenitors committed to SG lineage as demonstrated by lineage tracing using Blimp1CRE mice. The persistence of Blimp1 expressing cells in adult HF suggested that the SG could be maintained by unipotent progenitors during adult homeostasis. In the absence of Blimp1, increased bulge SC proliferation is observed, possibly through a derepression of c-Myc mediated by Blimp1 (Horsley et al, 2006). Two independent groups have shown that cells located in the upper follicle called the upper isthmus (UI), expressing high level of MTS24 or low level of Sca1, are highly clonogenic in vitro and multipotent in vivo, suggesting that this region of the HF might contain a subset of SCs that sustain SG homeostasis (Nijhof et al, 2006; Jensen et al, 2008). Another population of cells located at the junction zone (JZ) between UI and IFE, expressing Lrig1 and enriched for Blimp1 transcript, contain multipotent SC as demonstrated by transplantation assays (Jensen et al, 2009). Finally, a recent study using Lgr6CREER, which is expressed in the UI, demonstrated the presence of unipotent SG SCs responsible for the long-term labelling of the SG and the surrounding UI HF (Snippert et al, 2010). It is still unclear whether SGs are only maintained by Lgr6+ SCs, a substet of which expressed MTS24 and possibly Blimp1, or that several population of SCs, including HF SCs, are mobilized and contribute to the SG turnover during adult life. In a recent study, Petersson et al (2011) used new bulge SC lineage tracing to show that a subpopulation of bulge SCs may contribute to the homeostasis of SG. Using new K15CreERT2 transgenic mice crossed with the Rosa26–YFP or Rosa–LacZ reporter mice, the authors performed lineage tracing of bulge SCs. TAM administration labels cells preferentially in the bulge region (90% of LacZ+ cells) while the remaining labelled cells are already localized in the SG and its duct. Clonal analysis showed that a subset of YFP+ cells apparently coming from the bulge migrates to the UI and the SG 5 days later, as suggested by the increased number of the SG labelled. These K15CREER-labelled cells then locally self-renew and continue to contribute to SGs homeostasis up to 6 months after TAM administration. K15CREER-labelled cells also renew and expand overtime within the bulge region, consistent with the notion that some of the bulge SCs undergo symmetric self-renewing division within their niche, as previously suggested (Waghmare et al, 2008; Zhang et al, 2009, 2010). The absence of contiguous trail of YFP+ cells between the bulge and the SGs suggests that the K15-derived SG progenitors arising from the bulge are not bipotent at the clonal level, and leave the bulge niche once committed to the SG lineage. The apparent contribution of bulge SC to the SG homeostasis is not limited to a particular stage of the hair cycle, although the flux of bulge SC to the SG seems to increase during the activation stage of bulge SC that occurs during the initiation of HF regeneration. The migration of bulge SC to the SG has been substantiated by two different experimental approaches. The level of expression of SG progenitor markers (Lgr6, Lrig1 and Blimp1) in YFP-labelled cells increases with time after TAM administration, consistent with a progressive enrichment of SG progenitors within the progeny of K15CREER-derived cells. The authors also developed a new method to perform ex vivo live cell imaging of whole mount preparation of the skin epidermis in order to visualize the migration of bulge cells in real time. This elegant approach allowed the authors to follow the fate of individual bulge SC during 22 h. As expected, most of the labelled bulge SC remained quiescent during the observation but a fraction of those cells divide and migrate locally, downward but also upward towards the UI and SG (Figure 1). However, the whole mount preparation could not be imaged long enough to visualize the whole process of bulge SC migration to the SG and their subsequent contribution to the sebocyte differentiation. Figure 1.Bulge SC contribution to the sebaceous lineage. (A) Scheme highlighting the different epidermal SC populations found in the skin epidermis: the IFE SC (shown in blue), the bulge SC (Bu in red), the isthmus SC (Is in green) and the junctional SC (JZ in yellow). (B) Lineage tracing experiments have shown that IFE is maintained independently of the HF by the presence of number small unit of proliferation. (C) Lineage tracing of bulge SC using K15CREPR, Lgr5CREER and K19CREER demonstrate their ability to contribute to the homeostasis of the hair follicle but not the SG or the IFE. (D) Lgr6 lineage tracing show the long-term renewal of Lgr6+ isthmus cells and their contribution to the sebaceous lineage. (E) The recent study of Petersson and colleagues using a new K15CreER shows the potential contribution of bulge SCs to the SG homeostasis. Download figure Download PowerPoint It has been previously demonstrated that overexpression of a mutant of Lef1 that cannot bind β-catenin (K14ΔNLef1) leads to development of ectopic SGs, suggesting that β-catenin/Lef1 signalling regulates SG differentiation (Merrill et al, 2001; Niemann et al, 2002, 2007). To determine the cellular origin of the de novo SGs in K14ΔNLef1, the authors performed lineage tracing experiments using K15CreERT2/Rosa–YFP in K14ΔNLef1 mice. The presence of YFP-labelled cells in de novo SG above and below the bulge demonstrates that these lesions arise from the bulge SC that progressively expressed markers of the SG progenitors (Lgr6, Lrig1, Blimp1 and Plet1) ectopically. Interestingly, while de novo SG arise from cells below and above the bulge region in K14ΔNLef1, de novo SGs arise only in their normal niche when ΔNLef1 is preferentially expressed in the bulge region (K15ΔNLef1 transgenic mice). These new results are very interesting and intriguing. Indeed, while it has been previously reported that bulge SC can contribute to the SG lineage using K15CrePR and Lgr5CreER lineage tracing, but the contribution of the SG was reported to be limited to only rare follicles (Morris et al, 2004; Jaks et al, 2008). One possibility is that these new CREER mice target a subpopulation of the upper bulge more prone to give rise to the SG lineages than the previously described bulge specific inducible CRE (Morris et al, 2004; Levy et al, 2005; Jaks et al, 2008; Youssef et al, 2010) or that these new K15CREER also targeted a rare subpopulation of SG progenitors known to participate to both isthmus and SG maintenance such as Lgr6+ cells (Snippert et al, 2010), Lrig1+ cells (Jensen et al, 2009) or MTS24+ cells (Nijhof et al, 2006). The long-term labelling of SG by resident progenitors observed using Lgr6CREER (Snippert et al, 2010) or K15CreER in this study (Petersson et al, 2011), demonstrates that the SG contain resident progenitor that locally self-renew and that not all SGs are constantly renewed through a continuous flux of bulge SC. It will be important to determine what mechanisms regulate the asymmetric cell fate decision of bulge SCs during the specification of SG progenitors and what regulate the flux of bulge SC towards the SG. Regulation of p63 expression by an autoregulatory loop involving miRs p63, a member of the p53 family of protooncogenes, has been identified as a key regulator of stratified epithelium development. Mice lacking p63 are born alive but lack all major stratified epithelia including the epidermis (Mills et al, 1999; Yang et al, 1999). p63 has several isoforms depending on the transcription site and alternative splicing (Yang et al, 1998). ΔNp63 is the major isoform express in the epidermis and is expressed mainly in the basal layer of the epidermis (Koster et al, 2004; Laurikkala et al, 2006). Subsequent studies suggest that p63 can regulate renewal and differentiation of the epidermal progenitors (Truong et al, 2006; Senoo et al, 2007) but the mechanisms regulating p63 expression in epidermal cells remain elusive. miRNAs provide an additional layer of complexity to the transcriptional regulatory switch of p63 during epidermal differentiation. miR-203 is upregulated during epidermal stratification suprabasally at E15.5 and was shown to directly represses p63 expression in the suprabasal differentiated cells to promote differentiation and stratification of the epidermis (Lena et al, 2008; Yi et al, 2008). In a recent study published in the EMBO Journal, Chikh et al (2011) showed that iASPP regulates p63 expression in the epidermis through the regulation of miRNA expression. Apoptosis stimulating proteins of p53 (ASSP) are a family of proteins that regulates p53-mediated apoptosis. ASPP1, ASPP2 potentiate the pro-apoptotic function of p53, while the third member iASPP inhibits this function (Bergamaschi et al, 2003). iASPP is mostly expressed in epithelial cells from several tissues including skin (Herron et al, 2005) and is frequently mutated in human cancers (Bergamaschi et al, 2003). Structural studies predicted that iASPP preferentially binds to p63, suggesting that iAPSS could be involved in the regulation of p63 expression in the skin (Robinson et al, 2008). In this new study, the authors showed that iASPP is expressed in the early developing mouse epidermis and colocalizes with p63 in the nucleus of basal epidermal cells at E15.5 and thereafter. The same pattern of iASPP expression is found in human epidermis. Stimulation of human keratinocyte differentiation decreases both ΔNp63 and iASPP expression, suggesting that iASPP might be involved in p63 downregulation. The authors performed ChIP experiments to demonstrate that p63 directly binds to iASPP promoter in vivo. Overexpression of the two p63 isoforms stimulates iASPP expression, while their knockdown by siRNA downregulates iASPP, thus functionally demonstrating that p63 regulates iASPP in vitro. Reciprocally, downregulation of iASPP decreases expression of TAp63 and ΔNp63 at protein level, demonstrating that iASPP and p63 form a positive autoregulatory feedback loop. Intriguingly, iASPP knockdown does not regulate p63 mRNA expression, suggesting that iASPP may regulate the expression of p63 protein through a posttranscriptional or a postranslational mechanism. By profiling miR expression after iASPP KD, the authors found that miR-574-3p and miR-720, two miRs predicted to target p63, were upregulated following iASPP silencing. Antagomirs against these miRs restored the expression of p63 in the context of iASPP knockdown, suggesting that iASPP represses miR-574-3p and miR-720 expression, which in turn promote p63 expression. To determine the role of iASPP in keratinocytes, the authors used transcriptional profiling by microarray of iASPP-depleted cells. This experiment showed that iASPP is involved in the regulation of several members of the desmosomal complexes, tight junction and gap junction components, and a decrease in β1 integrin expression. Cells depleted for iASPP acquire a phenotype of differentiated cells with reduced proliferation. Human skin equivalent using organotypic culture of iASPP-depleted keratinocytes are thicker, less proliferative and present premature differentiation characterized by the expression of K1, involucrin and loricrin-positive cells within the basal compartment. This phenotype was partially rescued by specific antagomirs against miR-574-3p and miR-720, suggesting that iASPP promotes self-renewal and inhibits differentiation by repressing miR-574-3p and miR-720 expression (Figure 2). Figure 2.iASPP regulates keratinocyte renewal and differentiation by regulating miR expression. Under physiological conditions, p63 regulates iASPP expression, which in turn, represses miR-574-3p and miR-720. Upon iASPP knockdown in human keratinocytes, these miRs are upregulated and inhibit p63 expression, leading to defect in keratinocyte adhesion, renewal and differentiation (Chikh et al, 2011). Download figure Download PowerPoint This study shows that p63, a major regulator of epidermal morphogenesis and differentiation, is indirectly regulated by iASPP through the modulation of miR expression. While miR-574-3p and miR-720 regulate p63 expression, iASPP KD does not exactly phenocopy p63 KD (Carroll et al, 2006; Truong et al, 2006). Indeed, while iASPP-depleted keratinocytes show a decrease of β1 integrin and an increase of α3 integrin expression without modifications of β4 integrin; p63 knockdown induces a strong decrease in the expression of β1, β4 and α3 integrins at mRNA level (Carroll et al, 2006). In addition, iASPP KD impairs cell proliferation similarly as p63 KD, but also induces a premature differentiation of skin equivalent, which is inhibited upon p63 silencing (Truong et al, 2006), suggesting that iASPP may regulate other signalling pathways in addition of p63. Future studies will be required to clarify the discrepancy between iASPP and p63 KD, and determine what could be the other targets regulated by miR-574-3p and miR-720. Role of the polycomb repressor complex in the regulation of epidermal SCs The polycomb group proteins (PcG) constitute a chromatin remodelling complex referred to as Polycomb Repressor Complexes (PRC). PRC2 is composed of Ezh2 (or Ezh1), Eed, suz12 and RBAP46/48 (also known as Rbbp7/4) (Margueron and Reinberg, 2011). This complex is recruited to chromatin and Ezh2 mediates the trimethylation of histone H3 on Lysine 27 (H3K27me3). This histone mark induces recruitment of PRC1, which participates to repress adjacent locus. PRC1 complexes are more variable and are composed of Ring1a/b together with Bmi1, Cbx and PH proteins that mediate the monoubiquitination of H2A at Lys119, which induces gene silencing (Simon and Kingston, 2009). Deletion of PRC2 complexes subunit leads to the upregulation of lineage commitment genes normally silenced in embryonic SCs that lead to early embryonic lethalilty (O'Carroll et al, 2001; Voncken et al, 2003). In the skin, loss of Ezh2 impaired proliferation and induced the premature differentiation of the epidermis during embryonic development but this phenotype disappeared postnatally (Ezhkova et al, 2009). Combined deletion of Ezh1 and Ezh2 leads to hair loss and hyperproliferation of IFE cells (Ezhkova et al, 2011). In HF lineages, Ezh1/Ezh2 cKO induces upregulation of Ink4b/Ink4a/Arf and subsequent overexpression of p16/p19 (Ezhkova et al, 2011). Since Ink4b/Ink4a/Arf shRNA restore HF cells self-renewal in vitro, these results suggest that Ezh1/Ezh2 regulate HF SCs proliferation mostly by repressing Ink4b/Ink4a/Arf locus, similarly to Bmi1, another PRC2 subunit (Jacobs et al, 1999; Cordisco et al, 2010). In addition, Jmjd3 histone demethylase, which is involved in H3K27me3 demethylation, has been shown to be upregulated following keratinocyte differentiation and its overexpression in basal epidermal cells induces premature differentiation. This effect is associated to the derepression of differentiation-related genes such as krt1 and S100A8 and ectopic expression of krt1, krt10, filaggrin and involucrin. Inhibition of Jmjd3 demethylase activity by point mutation blocked its promoting effect on differentiation, suggesting that jmjd3 stimulates epidermal differentiation by the demethylation of H3 at several loci (Sen et al, 2008). Jarid2, a jumonji/JmjC domain-containing protein, has been shown to help the recruitment of PRC2 to the promoter of repressed genes in ESC (Peng et al, 2009; Shen et al, 2009; Landeira et al, 2010). However, conflicting results have been published concerning the role of Jarid2 in regulating the methyltransferase activity of PRC2 (Herz and Shilatifard, 2010; Landeira and Fisher, 2011). Little is known about the mechanisms that contribute to the recruitment of PRC2 complex to the promoter of repressed genes in adult tissue. A recent study by Mejetta et al (2011) published in EMBO Journal showed that Jarid2 regulates epidermal progenitor proliferation and differentiation in the mouse skin epidermis. Moreover, Jarid2 also recruits PRC1 to the promoters of lineage-specific genes in embryonic SCs (Landeira et al, 2010). Conditional deletion of Jarid2 in epidermal cells during embryonic development induced no initial defect of proliferation and differentiation. However, after birth, Jarid2 null epidermis is characterized by an increased number of differentiated cells as well as a decrease in the proliferation of basal epidermal progenitors. Jarid2 null keratinocytes present fewer proliferative colonies than control keratincoytes and present characteristics of terminally differentiated keratinocytes in vitro. While HF morphogenesis is apparently unaffected in the absence of Jarid2, there is a delay in the initiation of HF regeneration, despite the similar number of adult HF bulge SC. HFs present a delay in anagen re-entry, but they eventually resume hair cycle and regenerate new hairs. In addition, Jarid2 null HFs are less responsive to TPA-induced bulge SC proliferation, and their epidermis presented increased differentiation upon TPA treatment. The authors then showed by co-immunoprecipitation that Jarid2 interacts with PRC2 components Ezh2 and Suz12 in mouse keratinocytes, as it does in ESC. By performing ChIP experiments, they showed that Jarid2 localized to the promoter of several epidermal and non-epidermal target genes of the PRC2 complex previously shown to be occupied by Ezh2 in the skin epidermis (Ezhkova et al, 2009). In the absence of Jarid2, there is a small but significant decrease in the occupancy of these target genes by Suz12 as well as a decrease in the level of H3 trimethylation at these loci. These results are consistent with the notion that Jarid2 helps the recruitment of PRC2 and positively regulates the methyl transferase activity of the PRC2 complex at these loci. Accordingly, these PRC2 target genes including genes involved in keratinocyte differentiation, as well as p16, a gene that negatively regulates cell-cycle progression were upregulated in the absence of Jarid2 in newborn skin epidermis. In contrast, in the adult epidermis, the absence of Jarid2 has no effect on the expression of differentiated genes but still present a decrease in the recruitment of Suz12 at the Ink4a promoter, leading to an increase of p16 expression (Figure 3). Figure 3.Jarid2 regulates epidermal SC proliferation and differentiation. Jarid2 modulates the recruitment of the PRC2 complex and the trimethylation of lysine 27 Histone H3 (H3K27Me3). This epigenetic mark represses the Arf4 locus, leading to repression of p16/p19. In Jarid2 cKO mice (K14Cre∷Jarid2fl/fl), trimethylation of H3K27 is decreased and subsequently, p16 is upregulated leading to a delay in bulge SC activation and increased IFE differentiation. This effect could be related, at least partially, by a decrease in the recruitment of Ezh2, which that is involved in repressing of p16 and epidermal differentiation complex encompassing Filaggrin (Flg), Loricrin (Lor), Involucrin (Ivl) and late cornified envelope (Lce) genes in an AP1-dependent manner (Ezhkova et al, 2009). In the absence of Jarid2, cultured keratinocytes present a decrease in their clonogenic potential, suggesting that Jarid2 controls epidermal SC renewal. Download figure Download PowerPoint These results highlight the complexity of the regulation of epidermal SCs by histone modification. While Jarid2 and Ezh2 null epidermis present some similarities (e.g., increase of p16), there are clear differences in the timing and the extent of their phenotypes (Ezhkova et al, 2009; Mejetta et al, 2011). These apparent discrepancies between the two models suggest that Jarid2 and Ezh2 functions have onl