The regulatory circuits that coordinate epidermal differentiation during development are still not fully understood. Here, we report that the transcriptional regulator ID1 is enriched in mouse basal epidermal progenitor cells and find ID1 expression to be diminished upon differentiation. In utero silencing of Id1 impairs progenitor cell proliferation, leads to precocious delamination of targeted progenitor cells and enables differentiated keratinocytes to retain progenitor markers and characteristics. Transcriptional profiling suggests that ID1 acts by mediating adhesion to the basement membrane while inhibiting spinous layer differentiation. Co-immunoprecipitation reveals ID1 binding to transcriptional regulators of the class I bHLH family. We localize bHLH Tcf3, Tcf4 and Tcf12 to epidermal progenitor cells during epidermal stratification and establish TCF3 as a downstream effector of ID1-mediated epidermal proliferation. Finally, we identify crosstalk between CEBPA, a known mediator of epidermal differentiation, and Id1, and demonstrate that CEBPA antagonizes BMP-induced activation of Id1. Our work establishes ID1 as a key coordinator of epidermal development, acting to balance progenitor proliferation with differentiation and unveils how functional crosstalk between CEBPA and Id1 orchestrates epidermal lineage progression.

Progenitor cells in self-renewing tissues commonly undergo a coordinated program of differentiation to accommodate tissue function. Although many aspects of progenitor cell differentiation are known, the transcriptional programs that orchestrate epidermal differentiation during development are not well defined. Epidermal p63+ (Trp63) progenitor cells are specified from a single-layered ectoderm at embryonic day (E) 8.5 (Zhao et al., 2015). Following this initial epidermal specification, progenitors commit to stratification and a transient proliferative yet suprabasal cell population co-expressing K5 (Krt5)/K14 (Krt14) progenitor and K10 (Krt10) differentiation marker forms (Koster and Roop, 2007). At around E15, this intermediate cell population initiates terminal differentiation, and proliferation becomes restricted to the basal progenitor layer (Blanpain and Fuchs, 2006). Continued growth of the embryo and expansion of the stratified embryonic epidermis thus requires coordination of progenitor cell cycle exit, delamination and differentiation through transcriptional effectors (Bao et al., 2013; Blanpain et al., 2006; Klein et al., 2017; Mulder et al., 2012; Okuyama et al., 2004). It is likely that seemingly antagonistic transcriptional programs are interdependent (Boxer et al., 2014; Hopkin et al., 2012) and enable fine-tuning of epidermal lineage progression.

Several families of secreted signaling molecules have been implicated in skin development. Epidermal bone morphogenetic protein (BMP) signaling is complex (Blessing et al., 1996), but collective evidence suggests a role in modulating progenitor cell proliferation and promoting differentiation through activation of pSMAD1/5 transcriptional programs (Botchkarev and Sharov, 2004). The inhibitors of differentiation (ID) proteins are established BMP-sensing target genes in several cellular contexts (Genander et al., 2014; Korchynskyi and ten Dijke, 2002; Kowanetz et al., 2004). ID proteins are typically expressed in stem and progenitor cells, only to be downregulated upon differentiation (Lasorella et al., 2001), suggesting – paradoxically from an epidermal BMP signaling perspective – that ID proteins could, as the name implies, enable progenitor self-renewal by suppressing differentiation.

Expression of Id1 is associated with stemness in several cellular contexts (Jankovic et al., 2007; Liu et al., 2013; Malaguti et al., 2013; Nam and Benezra, 2009; Niola et al., 2012; Romero-Lanman et al., 2012; Ying et al., 2003; Zhang et al., 2014). In the adult skin, ID1 mediates hair follicle stem cell quiescence and is required for hair follicle progenitor cell specification (Genander et al., 2014). In contrast, forced expression of ID1 in immortalized human HaCaT organotypic cultures results in a hyperproliferative, disorganized epidermis (Rotzer et al., 2006) and ID1 is highly expressed in adult human psoriatic skin (Bjorntorp et al., 2003), suggesting that the phenotypic outcome of ID1 is context dependent but impinging on the balance between progenitor cell proliferation and differentiation.

The CEBP family of transcription factors (CCAAT/enhancer binding proteins) are induced upon epidermal differentiation, and couple cell cycle exit with commitment to differentiation in several cell types, including the epidermis (Loomis et al., 2007; Lopez et al., 2009; Oh and Smart, 1998; Zhu et al., 1999). In the hematopoietic lineage, CEBPs are able to redirect chromatin binding of general signaling pathway mediators, including SMAD proteins, thereby acting to define cellular identity during lineage specification and differentiation (Trompouki et al., 2011). In other systems, SMADs bind and inhibit the transcriptional activity of CEBPs (Coyle-Rink et al., 2002; Zauberman et al., 2001), suggesting that crosstalk between SMADs and CEBPs is a commonly employed mechanism in place to fine-tune gene regulation.

Here, we explored published single-cell profiling datasets (Fan et al., 2018) to identify enrichment of Id1 in E13 progenitor cells committed to differentiation. Employing an in utero lentiviral injection strategy (Beronja et al., 2010) to target in vivo epidermal progenitors in combination with transcriptional profiling of cultured epidermal progenitor cells allowed us to delineate the consequences of Id1 silencing on progenitor proliferation and differentiation during the onset of epidermal stratification. We demonstrate that targeting of ID1 leads to a thinning of the developing epidermis, impairs epidermal progenitor proliferation and results in loss of Id1-silenced progenitor cells. Furthermore, in vivo targeted cells co-express progenitor (K5) and differentiation (K10) markers and transcriptional profiling reveals upregulation of differentiation markers in vitro, indicating that ID1 acts to repress epidermal differentiation in progenitor cells.

Co-immunoprecipitation demonstrates that ID1 binds the basic helix-loop-helix (bHLH) TCF family of transcription factors, and we visualize the presence of Tcf3, Tcf4 and Tcf12 in epidermal progenitor and committed keratinocytes during epidermal stratification. We continue to establish TCF3 as a repressor of epidermal progenitor cell proliferation in vitro. Focusing on CCAAT enhancer binding protein alpha (CEBPA) as an ID1-dependent regulator of epidermal cell cycle exit and differentiation in vivo, we confirm upregulation of CEBPA after silencing of Id1 in the epidermis and demonstrate TCF3 binding to Cebpa regulatory chromatin. Finally, we unearth a new role for CEBPA in antagonizing BMP-induced Id1 promoter activation, establishing a regulatory mechanism of ID1 expression in epidermal progenitor cells likely to be relevant for other BMP-sensing genes. Collectively, these data establish ID1 as an essential orchestrator of epidermal differentiation and demonstrate functional crosstalk between CEBPA and ID1, enabling coordinated epidermal progenitor cell differentiation.

ID1 is expressed in epidermal progenitor cells during skin development

Although epidermal development has been studied in some detail (Flora and Ezhkova, 2020; Miroshnikova et al., 2019; Miyai et al., 2016; Soares and Zhou, 2018), the process of epidermal differentiation is not fully understood. We exploited a published single-cell RNA-sequencing data set (Fan et al., 2018) from E13 mouse epidermis, a developmental time point where epidermal progenitor cells have not yet initiated terminal differentiation. Re-analysis revealed two main epidermal clusters: cluster 1 was marked by the progenitor cell marker Krt15, and differentiation-associated Krtdap segregated to cluster 2 (Fig. S1A-C). Gene Ontology (GO) profiling of differentially expressed genes (DEGs) showed enrichment of biological processes linked to tissue and epithelial development in cluster 1, whereas cluster 2 DEGs were associated with epithelial differentiation (Fig. S1D), suggesting that cluster 2 epidermal cells have committed to epidermal differentiation. Interestingly, classification of protein types found in the most significant GO terms in cluster 1 and 2 revealed that epithelial differentiation is associated with DNA-binding transcriptional regulators and cytoskeletal rearrangement (Fig. S1E). Collectively, these results highlight the importance of transcriptional regulation in epidermal commitment to differentiation.

In addition to known transcriptional regulators such as Mafb, Hes1 and Klf4 (Blanpain et al., 2006; Lopez-Pajares et al., 2015; Segre et al., 1999), we identified Id1 to be enriched in cluster 2 (P=0.007, expression fold-change 0.4, likelihood-ratio test, FindAllMarkers in Seurat package), although expressed also in cluster 1 (Fig. 1A,B; Fig. S1A,B,F; Table S1). Localization of ID1 protein in the developing epidermis revealed prominent ID1 expression at E13.5 in most epidermal progenitor cells, in line with the E13 single-cell RNA-sequencing expression data. At E14.5, ID1 immunoreactivity is less uniform and, starting from E15.5, epidermal ID1 expression becomes enriched in basal progenitor cells and is gradually diminished as progenitors delaminate from the basal layer and commit to differentiation (Fig. 1C; Fig. S1G). Analogous to the ID1 in vivo expression pattern, cultured primary epidermal progenitors downregulate Id1 mRNA and corresponding protein when asked to differentiate in vitro (Fig. 1D,E). Interestingly, ID1 protein levels are sustained during the first 24 h of differentiation, indicating a role of ID1 in the transition from basal epidermal progenitor to committed epidermal keratinocyte.

Fig. 1.

ID1 is expressed in epidermal progenitor cells during skin development. (A) Feature plot displaying expression of Id1 in E13 epidermis. (B) Id1 expression at E13 epidermis show enrichment of Id1 in cluster 2 (blue) compared with 1 (red). (C) ID1 protein is abundantly expressed in the E13.5 and E14.5 epidermis. At E18.5, ID1 immunoreactivity is enriched in basal progenitors. (D) Id1 mRNA expression decreases with differentiation in vitro. One-way ANOVA comparing 24, 48 and 72 h with 0 h. (E) ID1 levels are high in progenitor cells and relatively sustained during initiation of in vitro differentiation (24 h of Ca2+ treatment). K10 is used as a positive differentiation control. (F) Western blot visualizing ID1 protein levels after shId1 or shScr targeting of epidermal progenitors at 0 and 24 h of differentiation. (G) Expression of Id2 and Id3 is not altered upon shId1 targeting compared with shScr, displayed as counts per million (CPM) from RNA-sequencing data (n=3 in both groups). (H) Reactome enrichment analysis of genes differentially expressed (>2-fold change) in shId1 compared with shScr progenitors. (I,J) Relative expression of selected genes based on reactome enrichment analysis, normalized to expression in shScr samples. Dashed line in C,D indicates the boundary between the basal layer of the epidermis and dermis. Data are represented as mean±s.e.m. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 using multiple unpaired t-test (G,I,J) or ANOVA (D). n=3 in quantifications (H-J). n.s, not significant. Scale bar: 50 μm.

Fig. 1.

ID1 is expressed in epidermal progenitor cells during skin development. (A) Feature plot displaying expression of Id1 in E13 epidermis. (B) Id1 expression at E13 epidermis show enrichment of Id1 in cluster 2 (blue) compared with 1 (red). (C) ID1 protein is abundantly expressed in the E13.5 and E14.5 epidermis. At E18.5, ID1 immunoreactivity is enriched in basal progenitors. (D) Id1 mRNA expression decreases with differentiation in vitro. One-way ANOVA comparing 24, 48 and 72 h with 0 h. (E) ID1 levels are high in progenitor cells and relatively sustained during initiation of in vitro differentiation (24 h of Ca2+ treatment). K10 is used as a positive differentiation control. (F) Western blot visualizing ID1 protein levels after shId1 or shScr targeting of epidermal progenitors at 0 and 24 h of differentiation. (G) Expression of Id2 and Id3 is not altered upon shId1 targeting compared with shScr, displayed as counts per million (CPM) from RNA-sequencing data (n=3 in both groups). (H) Reactome enrichment analysis of genes differentially expressed (>2-fold change) in shId1 compared with shScr progenitors. (I,J) Relative expression of selected genes based on reactome enrichment analysis, normalized to expression in shScr samples. Dashed line in C,D indicates the boundary between the basal layer of the epidermis and dermis. Data are represented as mean±s.e.m. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 using multiple unpaired t-test (G,I,J) or ANOVA (D). n=3 in quantifications (H-J). n.s, not significant. Scale bar: 50 μm.

In addition to Id1, Id2 and Id3 are also expressed in the developing epidermis. Expression analysis of Id2 and Id3 using E13 single-cell RNA-sequencing fails to reveal enrichment in committed progenitors (cluster 2) (Fig. S1H,I; Table S1). Analysis of Id gene co-expression shows that more than half (59%) of E13 epidermal progenitors express Id1 but not Id2 or Id3, whereas 36% of progenitor cells co-express all three Id genes. Only a small percentage (5%) of progenitors at E13 are Id gene negative (Fig. S1J). These data suggest that the family of ID proteins likely have both redundant and non-redundant functions in epidermal development. Based on the statistically significant enrichment of Id1, but not Id2 or Id3, in cluster 2 (Table S1), we focus on delineating the role of ID1 in the developing epidermis.

ID1 is associated with epidermal lineage progression

To address the function of ID1 in epidermal progenitor cells, we transduced primary epidermal progenitor cells with either control scrambled shRNA (shScr) or an shRNA-targeting Id1 (shId1) (Fig. 1F). In line with the distinct expression profiles of Id1 compared with Id2 and Id3 (Fig. S1H,I), silencing of Id1 in cultured epidermal progenitors did not lead to a compensatory upregulation of Id2 and Id3 mRNA (Fig. 1G). Reactome pathway analysis including DEGs upregulated in shId1 compared with shScr-transduced primary epidermal progenitors revealed biological processes associated with formation of the cornified envelope and keratinization, indicative of epidermal differentiation (reactome enrichment P=7*10−12 and 9*10−9, respectively). In contrast, downregulated DEGs were linked to extracellular matrix (ECM) modulation and laminin signaling (reactome enrichment P=2.8*10−5, P=1.4*10−4 and P=8.7*10−5) (Fig. 1H-J; Fig. S1K), suggesting that Id1 influences gene programs altering the composition of the ECM and/or adhesion to the basement membrane. All in all, the in vivo ID1 expression pattern, in combination with transcriptional profiling, associates ID1 with epidermal lineage progression.

ID1 counteracts epidermal progenitor delamination

To assess the function of ID1 in vivo, we exploited the method of ultrasound-guided in utero injections of high titer lentiviral particles (Beronja et al., 2010) (Fig. 2A). Id1 was either silenced in wild-type mice using an shRNA-targeting Id1, or depleted by injection of a CRE-expressing lentivirus (LV-CRE) into a conditional Id1 transgenic (Id1fl/fl) mouse line at E9.5 (Nam and Benezra, 2009). Delivery of shId1 (as judged by H2BGFP reporter expression) successfully reduced in vivo ID1 protein expression at E14.5 when compared with embryos targeted with shScr, in line with in vitro knockdown efficiency (Fig. 2B,C). Similarly, LV-CRE injection resulted in efficient loss of ID1 immunoreactivity in Id1fl/fl-targeted epidermal progenitors at E14.5, whereas ID1 protein was sustained in Id1+/fl-targeted skin (Fig. S2A), indicating that injection of either shId1 or LV-CRE efficiently ablated epidermal ID1 expression in vivo.

Fig. 2.

ID1 counteracts epidermal progenitor delamination. (A) Schematic of in utero lentiviral injections. (B) E14.5 ID1 immunoreactivity is reduced in H2BGFP-positive cells upon injection of lentiviral shId1, but not control shScr. (C) Id1 mRNA expression is reduced in epidermal progenitors targeted in vitro with shId1 compared with shScr. (D) The percentage of transduced H2BGFP-positive epidermal progenitors is reduced in shId1-targeted epidermis compared with shScr at E14.5. (E) Quantification of percentage of H2BGFP-positive targeted K14-expressing basal progenitors and H2BGFP-positive K14 negative suprabasal keratinocytes indicate selective loss of shId1-targeted progenitor cells. (F) Total number of epidermal nuclei per 150 μm is significantly reduced in shId1 compared with shScr-targeted epidermis at E16.5. (G) Quantification of K14-positive basal progenitor layer thickness in shId1 and shScr epidermis at E16.5. (H) Representative images of K14 thickness at E16.5 in shId1 and shScr epidermis. (I) K10-positive spinous layer thickens with epidermal development. Silencing of Id1 impairs spinous layer development. (J) Representative images of K10 distribution at E16.5 in shId1 and shScr epidermis. (K,L) Terminal differentiation is temporally and spatially normal in shId1 compared with shScr as judged by expression of differentiation marker TGM1 and involucrin at E15.5. (M) Differentiation markers expression in shScr and shId1 cells. Statistical analysis is performed between shScr and shId1 at 0 or 96 h. Dashed line in B,H,J-L indicates the boundary between the basal layer of the epidermis and dermis. Data are represented as mean±s.d. *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test. n=3-8 in quantifications (D-F,I,M), n=2-3 in G. Scale bars: 50 μm.

Fig. 2.

ID1 counteracts epidermal progenitor delamination. (A) Schematic of in utero lentiviral injections. (B) E14.5 ID1 immunoreactivity is reduced in H2BGFP-positive cells upon injection of lentiviral shId1, but not control shScr. (C) Id1 mRNA expression is reduced in epidermal progenitors targeted in vitro with shId1 compared with shScr. (D) The percentage of transduced H2BGFP-positive epidermal progenitors is reduced in shId1-targeted epidermis compared with shScr at E14.5. (E) Quantification of percentage of H2BGFP-positive targeted K14-expressing basal progenitors and H2BGFP-positive K14 negative suprabasal keratinocytes indicate selective loss of shId1-targeted progenitor cells. (F) Total number of epidermal nuclei per 150 μm is significantly reduced in shId1 compared with shScr-targeted epidermis at E16.5. (G) Quantification of K14-positive basal progenitor layer thickness in shId1 and shScr epidermis at E16.5. (H) Representative images of K14 thickness at E16.5 in shId1 and shScr epidermis. (I) K10-positive spinous layer thickens with epidermal development. Silencing of Id1 impairs spinous layer development. (J) Representative images of K10 distribution at E16.5 in shId1 and shScr epidermis. (K,L) Terminal differentiation is temporally and spatially normal in shId1 compared with shScr as judged by expression of differentiation marker TGM1 and involucrin at E15.5. (M) Differentiation markers expression in shScr and shId1 cells. Statistical analysis is performed between shScr and shId1 at 0 or 96 h. Dashed line in B,H,J-L indicates the boundary between the basal layer of the epidermis and dermis. Data are represented as mean±s.d. *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test. n=3-8 in quantifications (D-F,I,M), n=2-3 in G. Scale bars: 50 μm.

Even though the transduction efficiency of shId1 and shScr were comparable when cultured epidermal progenitors were infected (Fig. S2B), the percentage of H2BGFP reporter-expressing progenitors was reduced in E14.5 shId1 epidermis compared with shScr (Fig. 2D; Fig. S2C). At E16.5, shId1 embryos displayed significantly fewer H2BGFP/K14-positive progenitor cells compared with embryos targeted with shScr, whereas the number of H2BGFP-positive, K14-negative suprabasal cells remained high (Fig. 2E; Fig. S2D). At E18.5, shId1-targeted basal progenitor cells were scarce and remaining epidermal H2BGFP expression was localized to K14-negative suprabasal keratinocytes or developing hair follicles (Fig. S2E). Analogously, we found a reduction of the number of Id1-silenced progenitor cells at E16.5 in Id1fl/fl embryos compared with E14.5 (Fig. S2F-I), indicating that progenitors lacking ID1 are selectively lost during epidermal development.

To assess whether loss of Id1-targeted progenitors is compensated by untargeted progenitors, we quantified the total number of cells in shId1 epidermis at E14.5 and E16.5. We found a reduction in the epidermal cell number at E16.5, but not E14.5, in shId1 epidermis compared with shScr-targeted embryos (Fig. 2F; Fig. S2J). We then analyzed apoptosis to see whether cell death could explain the selective loss of Id1-depleted progenitors, but did not detect increased CC3-immunoreactivity in shId1 compared with shScr-targeted E14.5 epidermis (Fig. S2K,L). Collectively, these results suggest that ID1 counteracts precocious epidermal progenitor cell delamination, independent of apoptosis.

Loss of ID1 results in epidermal thinning without affecting terminal differentiation

To assess whether the reduction of epidermal cell number at E16.5 corresponded to loss of progenitor or differentiated cell layers, we first quantified the thickness of the entire epidermis as well as the basal K14-positive progenitor layer and found both to be diminished when comparing shId1 with shScr embryos (Fig. 2G,H; Fig. S2M). As our in vitro transcriptional profiling implicated ID1 in epidermal differentiation (Fig. 1I,J), we characterized the dynamic formation of the K10-expressing spinous layer between E14.5 and E16.5. Whereas the thickness of the spinous layer increased in shScr-targeted epidermis during development, it failed to do so efficiently in shId1-infected epidermis (Fig. 2I), and at E16.5 there was a significant reduction in the thickness of the spinous layer in shId1 epidermis compared with shScr-targeted epidermis (Fig. 2I,J). Owing to the rapid loss and few remaining CRE-targeted cells in Id1fl/fl epidermis (Fig. S2H), we failed to detect any reduction in either the E16.5 spinous layer or the overall epidermal thickness (Fig. S2N-P), suggesting that the expanding epidermis is able to compensate for the mosaic genetic depletion of ID1.

In vivo ID1 expression pattern is consistent with a role for ID1 in commitment to differentiation rather than terminal differentiation. To assess whether the absence of ID1 induced precocious terminal differentiation in vivo, we analyzed E15.5 epidermis, the earliest time point at which we could confidently detect markers of terminal differentiation. Although we found a general thinning of the spinous and granular differentiated cell layers marked by TGM1 and involucrin in shId1 epidermis compared with shScr-transduced skin (Fig. 2K,L), TGM1 and involucrin were correctly expressed, temporally as well as spatially. Cultured shId1 and shScr progenitor cells both upregulated spinous and granular differentiation markers when asked to terminally differentiate for 96 h in vitro (Fig. 2M), although the upregulation was less prominent when Id1 was silenced. In line with our transcriptional profiling (Fig. 1I,J), differentiation markers [with the exception of K1 (Krt1) and Tgm5] were enriched in shId1 compared with shScr progenitor cells before differentiation. Taken together, these data suggest that progenitor cells lacking ID1 do not initiate terminal differentiation precociously.

Progenitor cells devoid of ID1 co-express basal and differentiation markers

During epidermal development, basal progenitor cell crowding induces delamination and subsequent differentiation (Miroshnikova et al., 2018). Delaminating epidermal progenitors couple induction of spinous fate markers with suppression of basal gene programs (Blanpain et al., 2006). Revisiting our transcriptomics analysis (Fig. 1H), we identified upregulation of spinous markers in shId1-targeted epidermal progenitors in vitro (Fig. 3A), suggesting that the basal-to-spinous transition is affected in the absence of ID1. Analysis of basal (K5) and spinous (K10) marker co-expression in E16.5 epidermis revealed enrichment of K5/K10 double-positive epidermal progenitors in shId1 compared with shScr (Fig. 3B,C). Analogously, the percentage of K5/K10 double-positive progenitors was significantly increased in LV-CRE-targeted Id1fl/fl compared with Id1+/fl epidermis (Fig. 3D; Fig. S3A), indicating that loss of ID1 leads to an accumulation of cells co-expressing markers normally representative of distinct cell states.

Fig. 3.

Progenitor cells devoid of ID1 co-express basal and differentiation markers. (A) Spinous marker expression after silencing of Id1 in cultured epidermal progenitor cells. (B,C) In vivo silencing of Id1 increases the incidence of K5/K10 double-positive cells at E16.5 compared with shScr-targeted epidermis. Arrowheads indicate K5/K10 double-positive cells. Dashed line indicates the boundary between the basal layer of the epidermis and dermis. (D) The percentage of K5/K10 double-positive cells is significantly increased in Id1fl/fl compared with Id1+/fl epidermis when ID1 is ablated using a CRE lentivirus. (E) Cultured epidermal progenitors were targeted with shId1 or shScr and asked to differentiate for 24 h before collection and transcriptional profiling. (F) Number of differentially expressed genes (DEGs) in shId1 compared to shScr targeted cells at 24 h of differentiation. (G) Overlap between differentially expressed genes at 0 and 24 h of differentiation. (H) Genes associated with a basal progenitor cell state are reduced in shId1 compared with shScr-targeted epidermal progenitor cells (n=3). (I) Basal genes are sustained in shId1-targeted progenitors upon differentiation. (J,K) EdU incorporation upon differentiation of shId1- and shScr-infected progenitors (n≥12 for each sample). (L,M) In vivo silencing of Id1 leads to an increased number of suprabasal cells co-expressing EdU and K10. Arrowheads indicate EdU/K10 double-positive cells. Dashed line indicates the boundary between the basal layer of the epidermis and dermis. Data are represented as mean±s.e.m. *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test. n=3-6 in quantifications (A,C,D,H,I,M). n.s, not significant. Scale bars: 50 μm (B,L); 100 μm (K).

Fig. 3.

Progenitor cells devoid of ID1 co-express basal and differentiation markers. (A) Spinous marker expression after silencing of Id1 in cultured epidermal progenitor cells. (B,C) In vivo silencing of Id1 increases the incidence of K5/K10 double-positive cells at E16.5 compared with shScr-targeted epidermis. Arrowheads indicate K5/K10 double-positive cells. Dashed line indicates the boundary between the basal layer of the epidermis and dermis. (D) The percentage of K5/K10 double-positive cells is significantly increased in Id1fl/fl compared with Id1+/fl epidermis when ID1 is ablated using a CRE lentivirus. (E) Cultured epidermal progenitors were targeted with shId1 or shScr and asked to differentiate for 24 h before collection and transcriptional profiling. (F) Number of differentially expressed genes (DEGs) in shId1 compared to shScr targeted cells at 24 h of differentiation. (G) Overlap between differentially expressed genes at 0 and 24 h of differentiation. (H) Genes associated with a basal progenitor cell state are reduced in shId1 compared with shScr-targeted epidermal progenitor cells (n=3). (I) Basal genes are sustained in shId1-targeted progenitors upon differentiation. (J,K) EdU incorporation upon differentiation of shId1- and shScr-infected progenitors (n≥12 for each sample). (L,M) In vivo silencing of Id1 leads to an increased number of suprabasal cells co-expressing EdU and K10. Arrowheads indicate EdU/K10 double-positive cells. Dashed line indicates the boundary between the basal layer of the epidermis and dermis. Data are represented as mean±s.e.m. *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test. n=3-6 in quantifications (A,C,D,H,I,M). n.s, not significant. Scale bars: 50 μm (B,L); 100 μm (K).

ID1 co-ordinates the basal-to-suprabasal progenitor state transition

Interestingly, most cells with double progenitor/differentiation marker expression were not in contact with the basement membrane (Fig. 3B), suggesting that ID1-silenced progenitors failed to downregulate progenitor markers during delamination. To this aim, we sequenced shId1- and shScr-targeted epidermal progenitors after 24 h of in vitro differentiation (Fig. 3E,F). Reactome analysis confirmed the involvement of ID1 in coupling inhibition of differentiation to modulation of ECM (Fig. S3B). Interestingly, 35% of the DEGs found in shId1-targeted epidermal progenitors were significantly altered upon differentiation of shId1 progenitors (Fig. 3G), suggesting that aspects of ID1 function are conserved in basal progenitors and differentiating keratinocytes.

Focusing on markers associated with the basal state revealed that silencing of Id1 in progenitor cells resulted in reduced expression of genes linked to basal membrane anchoring and cell cycle progression (Fig. 3H), potentially coupling loss of ID1 to progenitor cell delamination and cell cycle exit. Paradoxically however, comparing transcriptional profiles of shId1 and shScr progenitors after 24 h of differentiation revealed that shId1-targeted progenitors sustained expression of basal markers normally lost during differentiation (Fig. 3I). These data suggest that, although epidermal progenitors downregulate basal markers upon ID1 silencing, their response to in vitro differentiation cues is impaired.

To address whether silencing of Id1 not only affects expression of a cohort of basal genes, but also affects basal cell characteristics during lineage progression, we pulsed in vitro differentiated epidermal progenitors with EdU. As expected, control shScr progenitors rapidly exited the cell cycle and were largely EdU-negative at 24 h of differentiation; however, a significant fraction of differentiated progenitors targeted with shId1 still incorporated EdU after 48 h of differentiation (Fig. 3J,K). In addition, we observed an enrichment of suprabasal EdU/K10 double-positive progenitors in shId1-silenced epidermis normally absent in shScr-targeted epidermis at E16.5 (Fig. 3L,M). Our work suggests that ID1 impinges on the basal-to-suprabasal transition, acting to co-ordinate and fine-tune progenitor as well as differentiation gene programs throughout epidermal lineage transition.

Epidermal progenitor proliferation is positively regulated by ID1

Tissue development requires regulation of mechanisms balancing progenitor cell proliferation and differentiation. As downregulation of ID1 leads to diminished expression of basal progenitor cell markers (Fig. 3H), including the cell cycle regulator Ccnd1, we analyzed epidermal progenitor proliferation after manipulation of Id1. In vivo targeting of Id1 reduced progenitor cell proliferation using either an shRNA or LV-CRE strategy (Fig. 4A,B; Fig. S4A), and cultured primary shId1 epidermal progenitor cells incorporated significantly less EdU than shScr progenitors (Fig. 4C,D). In contrast, overexpression of ID1 in primary epidermal progenitor cells increased EdU incorporation in vitro (Fig. 4E,F; Fig. S4B,C), suggesting that ID1 positively regulates progenitor renewal during epidermal development.

Fig. 4.

Epidermal progenitor proliferation is positively regulated by ID1. (A) Percentage of in vivo K5/EdU-positive proliferating progenitors are reduced upon shId1 targeting compared with shScr-targeted epidermis at E16.5. (B) LV-CRE-mediated Id1 ablation reduces proliferation in K5-positive epidermal progenitors in Id1fl/fl compared with Id1+/fl epidermis. (C,D) Silencing of Id1 in cultured epidermal progenitor cells leads to reduced EdU incorporation. (E,F) Overexpression of ID1 in cultured epidermal progenitors leads to increased EdU incorporation. Data are represented as mean±s.e.m. *P<0.05, ***P<0.001 using multiple unpaired t-test. n=3 in A,B. Scale bars: 25 μm.

Fig. 4.

Epidermal progenitor proliferation is positively regulated by ID1. (A) Percentage of in vivo K5/EdU-positive proliferating progenitors are reduced upon shId1 targeting compared with shScr-targeted epidermis at E16.5. (B) LV-CRE-mediated Id1 ablation reduces proliferation in K5-positive epidermal progenitors in Id1fl/fl compared with Id1+/fl epidermis. (C,D) Silencing of Id1 in cultured epidermal progenitor cells leads to reduced EdU incorporation. (E,F) Overexpression of ID1 in cultured epidermal progenitors leads to increased EdU incorporation. Data are represented as mean±s.e.m. *P<0.05, ***P<0.001 using multiple unpaired t-test. n=3 in A,B. Scale bars: 25 μm.

Identification of ID1 gene signatures

Our data positions ID1 at the intersection of proliferation and differentiation in epidermal progenitor cells. Reasoning that silencing of Id1 negatively affects proliferation and leads to an induction of differentiation markers, forced ID1 expression would, in addition to driving cell cycle progression, shift gene expression profiles towards a basal state. To this end, we profiled epidermal progenitor cells overexpressing ID1, as well as ID1-overexpressing progenitors asked to differentiate for 24 h (Fig. 5A; Fig. S5A). Focusing on downregulated DEGs (>2-fold change) at 24 h of differentiation, GO analysis revealed significantly enriched biological processes such as epidermal progenitor cell differentiation and regulation of signaling (Fig. 5B), suggesting that differentiation is hampered when ID1 is overexpressed. Expression levels of spinous markers elevated in shId1 progenitors were found to be reduced in ID1-overexpressing progenitors compared with uninduced progenitors, albeit from a low starting level (Fig. 5C). After 24 h of differentiation, induction of spinous gene expression in ID1-overexpressing cells was prominent, although hampered, when compared with uninduced differentiated cells (Fig. S5B). In contrast to cells silenced for ID1, we did not find alterations in basal gene expression (Fig. S5C,D). These data suggest that sustained ID1 expression during differentiation acts to slow down, rather than inhibit, epidermal differentiation.

Fig. 5.

Identification of ID1 gene signatures. (A) ID1 is induced in a doxycycline-dependent manner and progenitor cells are differentiated and profiled (n=3). (B) Downregulated genes after ID1 overexpression and 24 h of differentiation. DEGs, differentially expressed genes. (C) Spinous gene expression is reduced in epidermal progenitors upon ID1 overexpression. (D) Identification of ID1 gene signatures by merging expression data from shId1 and ID1 overexpression profiling of epidermal progenitors reveals a cohort of statistically significant genes (marked in purple) (LogFC >1 and FDR >0.05). Q1 and Q3 represent genes for which expression correlates to modulation of ID1. Q2 and Q4 represent genes that are only induced (Q2) or repressed (Q4) when ID1 levels are altered. (E) ID1 signature genes are functionally linked to adhesion and extracellular matrix (ECM) modulation, differentiation or cell cycle regulation (derived from Q1/Q3). (F) HOMER transcription factor binding motif analysis show distinct enrichment binding motifs in promoters (+400 bp from transcription start) of genes found in Q1+Q3 compared with promoters in the expressed transcriptome. (G) bHLH motif enriched in Q1+Q3 gene promoters. (H) Immunoreactivity of CEBPA and K5 in E14.5 and E16.5 localizes CEBPA to suprabasal keratinocytes and scattered basal K5-positive progenitors. Arrowheads indicate CEBPA-positive basal cells. Dashed line indicates the boundary between the basal layer of the epidermis and dermis. (I,J) Overexpression of CEBPA significantly reduces EdU incorporation in vitro. (K,L) Proliferating K5-positive progenitors are largely CEBPA negative. (M,N) K14/CEBPA double-positive progenitors are significantly enriched in shId1-targeted epidermis when compared with shScr. Arrowheads highlight K14/CEBPA positive cells. Dashed line indicates the boundary between the basal layer of the epidermis and dermis. Data are represented as mean±s.e.m. (C,E,I,M) and mean±s.d. (L). *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test. n=3-6 for quantifications (C,E,L,M). Scale bars: 50 μm (H,K,N); 100 μm (J).

Fig. 5.

Identification of ID1 gene signatures. (A) ID1 is induced in a doxycycline-dependent manner and progenitor cells are differentiated and profiled (n=3). (B) Downregulated genes after ID1 overexpression and 24 h of differentiation. DEGs, differentially expressed genes. (C) Spinous gene expression is reduced in epidermal progenitors upon ID1 overexpression. (D) Identification of ID1 gene signatures by merging expression data from shId1 and ID1 overexpression profiling of epidermal progenitors reveals a cohort of statistically significant genes (marked in purple) (LogFC >1 and FDR >0.05). Q1 and Q3 represent genes for which expression correlates to modulation of ID1. Q2 and Q4 represent genes that are only induced (Q2) or repressed (Q4) when ID1 levels are altered. (E) ID1 signature genes are functionally linked to adhesion and extracellular matrix (ECM) modulation, differentiation or cell cycle regulation (derived from Q1/Q3). (F) HOMER transcription factor binding motif analysis show distinct enrichment binding motifs in promoters (+400 bp from transcription start) of genes found in Q1+Q3 compared with promoters in the expressed transcriptome. (G) bHLH motif enriched in Q1+Q3 gene promoters. (H) Immunoreactivity of CEBPA and K5 in E14.5 and E16.5 localizes CEBPA to suprabasal keratinocytes and scattered basal K5-positive progenitors. Arrowheads indicate CEBPA-positive basal cells. Dashed line indicates the boundary between the basal layer of the epidermis and dermis. (I,J) Overexpression of CEBPA significantly reduces EdU incorporation in vitro. (K,L) Proliferating K5-positive progenitors are largely CEBPA negative. (M,N) K14/CEBPA double-positive progenitors are significantly enriched in shId1-targeted epidermis when compared with shScr. Arrowheads highlight K14/CEBPA positive cells. Dashed line indicates the boundary between the basal layer of the epidermis and dermis. Data are represented as mean±s.e.m. (C,E,I,M) and mean±s.d. (L). *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test. n=3-6 for quantifications (C,E,L,M). Scale bars: 50 μm (H,K,N); 100 μm (J).

Combining transcriptional profiling results from both shId1 and ID1 overexpression in epidermal progenitor cells allowed us to define a set of 83 candidate Id1 target genes, the expression changes of which correlated significantly (FDR>0.05) with alteration of Id1 transcript levels (Fig. 5D, Q1 and Q3; Table S2). Within this relatively small group of candidates, we identified cohorts of genes representing aspects of the in vivo phenotypes described upon Id1 silencing (Fig. 5E), correlating alterations in Id1-associated gene signatures with ECM modulation, differentiation and cell cycle regulation.

ID1 interacts directly with bHLH transcription factors

Lacking a DNA binding domain, ID1 is unable to affect transcription through direct chromatin interaction at target genes (Massari and Murre, 2000). To mechanistically begin to understand how ID1 affects epidermal progenitor cells, we aimed to identify epidermal ID1 binding partners. ID1, as well as family members ID2 and ID3, were overexpressed in cultured epidermal progenitor cells, after which ID-binding protein complexes were isolated and analyzed using mass spectrometry (Fig. S5E,F). We identified three known class I bHLH transcription factors, TCF3 (E2A), TCF4 (ITF2) and TCF12 (HEB), to be bound to ID1, whereas ID2 and ID3 associated exclusively with TCF12 (Table S3). Other published ID1-interacting transcription factors (Roberts et al., 2001; Yates et al., 1999) were not identified, suggesting that ID1 predominantly affects gene expression in the epidermis through binding to the TCF bHLHs. TCF3/4/12 often heterodimerize with cell type specific class II bHLH factors (Murre et al., 1989a,b), thereby acquiring cell state and context specific target gene profiles. Expression and purification of TCF3 and TCF4 in primary epidermal progenitor cells, however, failed to identify additional bHLH interactors but could independently confirm the binding of TCF3 and TCF4 to ID1, respectively (Table S3). Collectively, these data suggest that ID1 binding to TCF3, TCF4 and TCF12 acts to modulate bHLH transcriptional programs during epidermal development.

Transcription factor binding motif analysis on promoter sequences in genes differentially expressed in shId1 and ID1 overexpression (Fig. 5D, Q1 and Q3) using HOMER motif discovery revealed enrichment of bHLH motifs in potential ID1 targets when compared with all genes expressed in epidermal progenitors (Fig. 5F,G). Interestingly, motif discovery in genes significantly altered, but uncorrelated to ID1 levels (Fig. 5D, Q2 and Q4), revealed distinct motif profiles highlighting known regulators of epidermal differentiation such as grainyhead-like (GRHL), basic leucine zipper domain (bZIP), T-box and Trp63 transcription factors (Fig. 5F,G; Fig. S5G). Taken together, these data suggest that bHLH transcriptional programs impact epidermal progenitor states through modulation of specific target genes.

CEBPA is ectopically expressed in epidermal progenitor cells in the absence of ID1

Having identified an in vitro ID1 gene signature harboring enrichment of bHLH binding motifs, we singled out CEBPA for further in-depth characterization. CEBPA is an interesting ID1 signature gene – the combined loss of CEBPA and CEBPB in the developing epidermis results in hyperproliferation and epidermal barrier defects (Lopez et al., 2009). We found prominent CEBPA expression in suprabasal layers at E14.5 and E16.5. In addition, a significant subset of K5-positive basal progenitors expressed lower levels of CEBPA (Fig. 5H). In vitro differentiation of epidermal progenitor cells mimicked the in vivo expression pattern with significant upregulation of Cebpa mRNA and protein concomitant with differentiation (Fig. S5H,I). In line with previous reports (Lopez et al., 2009), we found that doxycycline-mediated overexpression of CEBPA in cultured epidermal progenitors significantly reduced the number of EdU+ cycling progenitor cells (Fig. 5I,J; Fig. S5J). Quantification of EdU-incorporation in the K5-positive basal layer at E16.5 reveals that most proliferating cells are CEBPA negative (Fig. 5K,L), confirming a role for CEBPA in repressing epidermal progenitor cell proliferation. Supporting our in vitro sequencing data, shRNA silencing of Id1 resulted in an increase in the number of K14/CEBPA double-positive progenitor cells compared with shScr at E16.5 (Fig. 5M,N), indicating that CEBPA expression is repressed in the presence of ID1 in vivo. To mechanistically explore whether Cebpa is transcriptionally regulated through an ID1-TCF axis, we cloned a 2 kb fragment of the Cebpa promoter as well as a Cebpa enhancer (Cooper et al., 2015) containing reported functional bHLH binding E-box motifs (Cooper et al., 2015; Hu et al., 2016; Pfurr et al., 2017; Soleimani et al., 2012; Yoon et al., 2015). We found an increase in promoter and enhancer luciferase reporter activity in shId1 epidermal progenitor cells compared with shScr (Fig. S5K,L), correlating to the observed increase in mRNA and in vivo protein expression, suggesting that Cebpa regulatory elements are differentially engaged in the presence or absence of ID1.

TCF3/4/12 localize to the developing epidermis and regulate progenitor cell proliferation

TCF3, TCF4 and TCF12 are established class I bHLH ID interactors (Lasorella et al., 2001), but the expression dynamics and function of TCFs in the developing epidermis have not been elucidated. Returning to the E13 epidermal single-cell profiling (Fan et al., 2018), we found all three Tcf genes to be ubiquitously expressed in a majority of epidermal progenitor cells in both cluster 1 and 2, contrasting the cluster 1 enrichment of Id1 (Fig. S6A,B). In situ hybridization confirmed the expression of Tcf3, Tcf4 and Tcf12 mRNA in K5-positive epidermal progenitors at E14.5 (Fig. 6A). Later in epidermal development (E16.5), Tcf3, Tcf4 and Tcf12 mRNA reactivity was apparent in both K5-positive progenitors as well as in suprabasal keratinocytes, only to become enriched in the K5 basal layer at E18.5 (Fig. 6A). We found decreasing expression of all Tcf genes during epidermal development (E14.5-E18.5) and higher epidermal mRNA reactivity for Tcf3 and Tcf12 compared with Tcf4. We also confirmed high expression of all three Tcf genes in the dermis at the time points analyzed (Rezza et al., 2016; Sennett et al., 2015). In vitro differentiation of cultured epidermal progenitors did not significantly alter the mRNA expression of Tcf3, Tcf4 or Tcf12 (Fig. S6C). Our data support a model in which the bHLH transcriptional output is controlled by spatially restricting ID protein expression.

Fig. 6.

TCF3/4/12 localize to the developing epidermis and regulate progenitor cell proliferation. (A) Visualization of Tcf3, Tcf4 and Tcf12 mRNA (green) in skin reveal expression in epidermal progenitors at E14.5 and E16.5, with subsequent enrichment in basal layer upon induction of differentiation (E18.5). Dashed line indicates the boundary between the basal layer of the epidermis and dermis. (B) Silencing of Tcf3, but not Tcf4 or Tcf12, promotes progenitor cell proliferation, n=12. (C) Overexpression of TCF3 and TCF4, but not TCF12, reduces proliferation in vitro, n=12. (D) Silencing of Tcf3, but not Tcf4 or Tcf12, reduces Cebpa mRNA levels, n=3. (E) ChIP-qPCR for FLAG-TCF3 compared with IgG control demonstrated enrichment of TCF3 at the Cebpa promoter (Prom) and enhancer (Enh). Cdkn1a is used as positive control, n=3. Data are represented as mean±s.d. *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test. Scale bar: 50 μm.

Fig. 6.

TCF3/4/12 localize to the developing epidermis and regulate progenitor cell proliferation. (A) Visualization of Tcf3, Tcf4 and Tcf12 mRNA (green) in skin reveal expression in epidermal progenitors at E14.5 and E16.5, with subsequent enrichment in basal layer upon induction of differentiation (E18.5). Dashed line indicates the boundary between the basal layer of the epidermis and dermis. (B) Silencing of Tcf3, but not Tcf4 or Tcf12, promotes progenitor cell proliferation, n=12. (C) Overexpression of TCF3 and TCF4, but not TCF12, reduces proliferation in vitro, n=12. (D) Silencing of Tcf3, but not Tcf4 or Tcf12, reduces Cebpa mRNA levels, n=3. (E) ChIP-qPCR for FLAG-TCF3 compared with IgG control demonstrated enrichment of TCF3 at the Cebpa promoter (Prom) and enhancer (Enh). Cdkn1a is used as positive control, n=3. Data are represented as mean±s.d. *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test. Scale bar: 50 μm.

To address the function of bHLH TCF transcription effectors in epidermal progenitor cells, we silenced Tcf3, Tcf4 and Tcf12 and assessed proliferation. Knockdown of Tcf3, but not Tcf4 or Tcf12, resulted in increased EdU incorporation in vitro (Fig. 6B; Fig. S6D,F). In contrast, overexpression of TCF3 and TCF4, but not TCF12, reduced proliferation in cultured epidermal progenitor cells (Fig. 6C; Fig. S6E,G), suggesting that ID1 regulates epidermal progenitor cell proliferation through TCF3 (and possibly TCF4) -dependent transcriptional programs.

Cebpa expression is bHLH-independent in epidermal progenitors

Our data suggest that activation of CEBPA is linked to silencing of Id1, in vitro and in vivo. To address whether ID1 acts to suppress Cebpa gene expression through sequestering of bHLH TCF transcriptional effectors, we assessed Cebpa mRNA expression after silencing or overexpression of TCF3, TCF4 and TCF12. Whereas silencing of Tcf3 reduced Cebpa levels, overexpression of TCFs did not change Cebpa mRNA expression (Fig. 6D; Fig. S6H). As downregulation of TCF3 also positively affects progenitor proliferation, we performed chromatin immunoprecipitation (ChIP) followed by qPCR to determine a direct TCF3-Cebpa interaction. Using TCF3 binding to the Cdkn1a promoter as a positive control (Calero-Nieto et al., 2014; Hu et al., 2016; Pfurr et al., 2017; Wohner et al., 2016) we confirmed TCF3 binding to Cebpa promoter and enhancer chromatin (Fig. 6E). Modulation of TCF3 expression did not, however, correlate to alterations in Cebpa promoter or enhancer luciferase reporter activity (Fig. S6I,J), arguing that even though TCF3 is able to bind Cebpa, enhancing TCF3 activity alone is not sufficient to drive Cebpa transcription.

As our data suggest that upregulation of the differentiation marker Cebpa in shId1-targeted epidermis could not be explained by TCF3 activity alone, we asked whether silencing of Tcf3 affects other genes associated with differentiation. Interestingly, epidermal progenitor cells targeted with shTcf3, but not consistently shTcf4 or shTcf12, downregulated markers associated with differentiation which were found upregulated in shId1-targeted progenitors (Fig. S6K). Taken together, these data suggest that silencing of Tcf3 affects markers of epidermal differentiation in progenitor cells.

pSMAD1/5 activation of the Id1 promoter is CEBPA-dependent

We noticed that modulating CEBPA expression affects Id1 mRNA and protein (Fig. 7A,B; Fig. S7A), where high CEBPA expression acts to suppress ID1. Id1 is a bona fide BMP target in the hair follicle stem cell lineage (Genander et al., 2014), and BMP signaling, as judged by phosphorylation of downstream effectors SMAD1/5, is active in epidermal progenitors and during in vitro differentiation (Fig. 7C; Fig. S7B-D). We assessed BMP sensitivity in epidermal progenitors and differentiated keratinocytes using a BMP-responsive pSMAD1/5 binding region located distally in the Id1 promoter (Genander et al., 2014; Korchynskyi and ten Dijke, 2002). Whereas the pSMAD1/5 binding region in the Id1 promoter responded to BMP in epidermal progenitor cells, keratinocytes differentiated for 24 h failed to efficiently activate Id1 luciferase reporter activity in the presence of BMP (Fig. 7D), suggesting that pSMAD1/5 engagement with the Id1 promoter is dynamically regulated during epidermal differentiation.

Fig. 7.

pSMAD1/5 activation of the Id1 promoter is CEBPA-dependent. (A) Relative Id1 mRNA levels upon 2 days of CEBPA overexpression (+Dox), n=3. (B) Protein level of ID1 is reduced upon forced CEBPA expression (+Dox). (C) pSMAD1/5 activity is pronounced in cultured epidermal progenitors and diminished upon differentiation. SMAD1 and SMAD5 are increased upon differentiation. (D) Luciferase reporter activity shows pronounced activity of a distal fragment of the Id1 promoter in response to BMP in progenitors when compared to differentiated keratinocytes, n=3. (E) Forced CEBPA expression in epidermal progenitor cells inhibits the BMP-mediated activity of the Id1 promoter, n=3. (F) Induction of Id1 mRNA is impaired after BMP4 treatment in the presence of CEBPA (+Dox, green bar). Statistical comparisons are made between control groups (*), within −Dox conditions (gray bar to gray bar, ***) and between BMP4-treated groups (gray BMP4 to green BMP4, *), n=3. (G) Protein levels of ID1 are suppressed after BMP4 treatment when CEBPA levels are high. (H) ChIP-qPCR localizes FLAG-CEBPA at the Id1 promoter. Cebpa promoter is used as positive control, n=3. (I) Statistical comparison made between shScr and shCebpa at 24 h of differentiation with BMP4, n=3. (J) Silencing of Cebpa enhances Id1 promoter activity in differentiated keratinocytes and leads to upregulation of ID1. Data are represented as mean±s.d. *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test.

Fig. 7.

pSMAD1/5 activation of the Id1 promoter is CEBPA-dependent. (A) Relative Id1 mRNA levels upon 2 days of CEBPA overexpression (+Dox), n=3. (B) Protein level of ID1 is reduced upon forced CEBPA expression (+Dox). (C) pSMAD1/5 activity is pronounced in cultured epidermal progenitors and diminished upon differentiation. SMAD1 and SMAD5 are increased upon differentiation. (D) Luciferase reporter activity shows pronounced activity of a distal fragment of the Id1 promoter in response to BMP in progenitors when compared to differentiated keratinocytes, n=3. (E) Forced CEBPA expression in epidermal progenitor cells inhibits the BMP-mediated activity of the Id1 promoter, n=3. (F) Induction of Id1 mRNA is impaired after BMP4 treatment in the presence of CEBPA (+Dox, green bar). Statistical comparisons are made between control groups (*), within −Dox conditions (gray bar to gray bar, ***) and between BMP4-treated groups (gray BMP4 to green BMP4, *), n=3. (G) Protein levels of ID1 are suppressed after BMP4 treatment when CEBPA levels are high. (H) ChIP-qPCR localizes FLAG-CEBPA at the Id1 promoter. Cebpa promoter is used as positive control, n=3. (I) Statistical comparison made between shScr and shCebpa at 24 h of differentiation with BMP4, n=3. (J) Silencing of Cebpa enhances Id1 promoter activity in differentiated keratinocytes and leads to upregulation of ID1. Data are represented as mean±s.d. *P<0.05, **P<0.01, ***P<0.001 using multiple unpaired t-test.

Chromatin binding of SMADs can be redirected upon expression of differentiation effectors (Trompouki et al., 2011). To this end, we overexpressed CEBPA in epidermal progenitor cells and found a dramatic reduction in BMP-induced response of the Id1 promoter fragment (Fig. 7E), which correlated to reduced induction of Id1 mRNA and protein (Fig. 7F,G). Comparing promoter activity and overall Id1 mRNA/protein induction, the difference in reduction suggests that additional Id1 promoter fragments drive Id1 expression in the presence of CEBPA. ChIP-qPCR for FLAG-tagged CEBPA in epidermal progenitor cells unveiled binding of CEBPA to the BMP-responsive fragment of the Id1 promoter [and Cebpa itself (Jakobsen et al., 2013; Meyer et al., 2016; Sun et al., 2018)], suggesting that CEBPA is able to modulate Id1 transcriptional activity (Fig. 7H).

To ask if CEBPA normally acts to repress pSMAD1/5-dependent Id1 expression, we silenced Cebpa in differentiated keratinocytes, a cell state in which CEBPA levels are endogenously high (Fig. 5). We found that downregulation of CEBPA potentiated BMP-mediated pSMAD1/5 activation of the Id1 promoter in differentiated keratinocytes, albeit at a low level (Fig. 7I,J; Fig. S7E-G). Collectively, these results suggest that CEBPA can desensitize the Id1 promoter elements to inductive BMP-mediated pSMAD1/5 transcription during epidermal commitment to differentiation.

The transcriptional networks underlying progenitor differentiation in developing tissues are beginning to clear as more regulatory circuits are being identified. Here, we characterize transcriptional regulators in the embryonic skin to identify a new role for ID1 in epidermal development and place ID1 in a context of previously known epidermal effectors. We exploit developmental transitions forming the stratified epidermis and identify transcriptional crosstalk impacting progenitor cell states during epidermal development. Using published transcriptomics describing E13 epidermis (Fan et al., 2018), we identify Id1 to be associated with progenitor cells committing to differentiation concurrent with stratification. Id genes have been previously implicated in human skin differentiation (Lopez-Pajares et al., 2015; Rotzer et al., 2006) and ID proteins are deregulated in human skin disease with impaired differentiation (Bjorntorp et al., 2003). Using in utero gene manipulation (Beronja et al., 2010), which, in contrast to transgenic recombination strategies (Andl et al., 2004; Vasioukhin et al., 1999), allows for targeting of the uncommitted single-layered epidermal progenitor population, we observe that ID1 promotes proliferative self-renewal (Fig. 4) and restricts commitment to differentiation during epidermal development in vivo (Fig. 3).

We find that progenitor cells devoid of ID1 are lost upon stratification (Fig. 2). Previous reports identified proliferation of basal progenitor cells as an inducer of epidermal stratification during development (Miroshnikova et al., 2018), where progenitor division results in local crowding, which induces differentiation and subsequent delamination. In contrast, stratification is suggested to precede progenitor proliferation in the adult epidermis (Mesa et al., 2018). Our work indicates that ID1-positive basal cells sustain progenitor states by coupling renewing proliferation with adhesion to the basement membrane (Figs 3 and 4). It is possible that Id1-silenced progenitors are outcompeted owing to additive defects in proliferation as well as adhesion. Although our in vitro transcriptional profiling cannot temporally delineate activation of differentiation markers from downregulation of anchoring proteins, our in vivo characterization indicates that delamination proceeds differentiation, identifying suprabasal K10-positive cells retaining cell cycle or basal K5 progenitor marker expression (Fig. 3). Interestingly, ID activity anchors neural stem cells to the ECM in the ventricular wall (Niola et al., 2012), suggesting that promotion of progenitor-to-niche adhesion is a common feature of ID-mediated transcriptional programs.

Although Id1 is responding to canonical SMAD-mediated BMP signaling in cultured epidermal progenitor cells (Fig. 7), ID1-dependent phenotypes are not necessarily mirrored when altering upstream BMP signaling. Paradoxically, inhibition of BMP signaling through transgenic overexpression of BMP signaling inhibitors (He et al., 2002; Plikus et al., 2004; Sharov et al., 2003) results in epidermal hyperproliferation and reduced differentiation. In contrast, suprabasal BMP6 affects basal progenitor cell proliferation in a dose-dependent manner (Blessing et al., 1996), where low and high BMP6 induces and represses proliferation, respectively. Epidermal BMP signaling thus generally stimulates differentiation, while the effect on progenitor proliferation is dose- or context-dependent. To reconcile these phenotypes, BMP must, in addition to inducing Id1 expression, exert ID1-independent effects on epidermal progenitor proliferation and differentiation, possibly through non-canonical BMP pathways (Botchkarev and Sharov, 2004).

ID1 promotes proliferation of immortalized keratinocytes in vitro (Rotzer et al., 2006) but antagonizes hair follicle stem cell (HFSC) activation in vivo (Genander et al., 2014). We establish ID1 as a positive regulator of epidermal proliferation in vivo (Fig. 4) and identify the interacting bHLH transcriptional factor TCF3 as a likely downstream effector. How ID1 promotes HFSC quiescence is unknown, but it is possible that HFSC-specific TCF heterodimerizing factors, lacking in the developing epidermis, direct the transcriptional response of TCF heterodimers in the absence of ID1. We were not able to identify additional heterodimerizing bHLH partners using cultured epidermal progenitors (Fig. 5), suggesting that TCF3 could regulate epidermal transcriptional targets as a TCF-TCF homodimer. Homodimerization of TCF3 has been demonstrated to control cell fate in pluripotent stem cells as well as in the B-cell and myogenic lineage (Neuhold and Wold, 1993; Rao et al., 2020; Shen and Kadesch, 1995). It would be interesting to functionally delineate the role of TCF dimerization in epidermal progenitors.

Focusing on CEBPA, an interesting regulator of cell cycle exit and lineage commitment (Lopez et al., 2009; Nerlov, 2007), we confirmed enrichment of CEBPA in shId1- compared with shScr-targeted progenitors in vivo (Fig. 5). Functional bHLH binding E-boxes are described in the Cebpa enhancer (Cooper et al., 2015) and promoter (Hu et al., 2016; Pfurr et al., 2017; Soleimani et al., 2012; Yoon et al., 2015) and, although we confirmed TCF3 binding to these chromatin regions (Fig. 6E), Cebpa luciferase reporter activity did not correlate to TCF3 levels. It is possible that overexpression of TCF3 induces a non-physiological response, explaining the lack of Cebpa activity, but we cannot exclude that ID1 acts upstream of Cebpa through TCF-independent mechanisms (Massari and Murre, 2000) and that CEBPA and TCF-mediated bHLH transcriptional programs act in parallel to cooperatively repress progenitor cell proliferation. However, it appears to be more likely, considering TCF3-binding to Cebpa chromatin, that additional so far unidentified co-factors are required to activate Cebpa transcription in a TCF-dependent manner, thereby acting to safeguard progenitor fate transitions towards differentiation.

We noticed that overexpression of CEBPA reduced Id1 mRNA and protein levels, suggesting upstream regulation of Id1 by CEBPA (Karaya et al., 2005; Mori et al., 2000; Saisanit and Sun, 1997). Exploiting a known BMP-sensing and pSMAD1/5-binding element in the Id1 promoter (Genander et al., 2014; Korchynskyi and ten Dijke, 2002), we observed reduced BMP responsiveness upon differentiation or when CEBPA was introduced in progenitor cells and increased sensitivity in differentiated keratinocytes upon Cebpa silencing (Fig. 7). CEBPA can interact with SMAD4 (Coyle-Rink et al., 2002; Zauberman et al., 2001) and redirect chromatin binding of SMAD1 during lineage specification (Trompouki et al., 2011). We did not probe a direct CEBPA-SMAD interaction, but instead used ChIP to place CEBPA on the Id1 promoter, suggesting that BMP and CEBPA act antagonistically to balance Id1 transcription and thereby progenitor fate. Furthermore, the presence or absence of CEBPA determines the ability of pSMAD1/5 to activate BMP-sensing chromatin, potentially diversifying and balancing the transcriptional output in response to BMP and thereby fine-tuning epidermal lineage progression. More specifically, our work indicates that CEBPA-positive basal progenitors are refractory to BMP-induced Id1 promoter activation and consequently fated towards epidermal differentiation. We identify ID1 as a new transcriptional effector coordinating epidermal development.

Mouse husbandry

Animals were housed in pathogen-free conditions according to the recommendations of the Federation of European Laboratory Animal Science Association. All animal experiments were approved by Stockholms djurförsöksetiska nämnd (ethical permit no. N243/14, N116/16 and 14051-2019). Id1 floxed mice have been previously described (Nam and Benezra, 2009), and Swiss mice were ordered (Janvier Labs) for in utero lentiviral injections with shRNAs.

Genotyping

Ear or tail biopsies were lysed overnight at 55°C in DirectPCR lysis reagent (BioSite) in the presence of 0.1 mg/ml Proteinase K (Thermo Fisher Scientific). Lysis was stopped by heat inactivation (45 min, 85°C). Taq polymerase (5 U/µl, final concentration: 0.05 U/µl) was used for amplification in PCR buffer supplemented with 0.08 mM MgCl2, 0.2 mM dNTPs (all Invitrogen) and 0.4 µM primers (see Table S4). PCR products were analyzed with 2% agarose gel.

Antibody staining

Tissue samples used in this study were either fixed before paraffin embedding or snap frozen and fixed before antibody staining with 4% formaldehyde (Sigma-Aldrich). Cells cultivated in chamber slides were subjected to the same fixative. For immunohistochemistry, antigen retrieval (R&D Systems) was performed, then blocking (Bloxall and Vector stain kit) and finally antibody incubation as indicated by the respective manufacturer's instructions. Samples used for immunofluorescence staining were cut at a thickness of 10 µm, blocked with goat serum (2.5%) and bovine serum albumin (1%) and permeabilized with 0.3% Triton X-100. When EdU was analyzed, the samples were treated with click chemistry before the antibody labeling (Click-iT, Thermo Fisher Scientific). Antibodies used for immunofluorescence staining were: ID1 (Biocheck, #BCH-1, 1:500), cleaved caspase-3 (Cell Signaling Technology, #9661, 1:400), K10 (Biolegend, #905404, 1:1000), K5 (Biolegend, #905901, 1:1000), K14 (OriGene, #BP5009, 1:1000), involucrin (Biolegend, #924401, 1:1000), TGM1 (Abcam, #ab103814, 1:1000), Flag (Sigma-Aldrich, F1804, 1:500), Cebpa (Cell Signaling Technology, #2295, 1:500). Antibodies used for immunoblotting were: Gapdh (Abcam, #ab8245, 1:5000), actin (Sigma-Aldrich, #A2228, 1:4000), tubulin (Abcam, #ab7291, 1:5000), ID1 (Biocheck, #BCH-1, 1:1000), Flag (Sigma-Aldrich, F1804, 1:5000), Cebpa (Cell Signaling Technology, #2295, 1:1000), phospho-Smad1/5 (Cell Signaling Technology, #9516, 1:1000). Smad1 (Cell Signaling Technology, #6944, 1:1000), Smad 5 (Cell Signaling Technology, #12534, 1:1000), Krt 10 (Covance, 90546, 1:1000).

Plasmid information

Lentiviral plasmids used for silencing of Id1, Cebpa or Tcf3/4/12 were generated by ligating the annealed short hairpin oligos into the pLKO.1 puro (Addgene plasmid #8453) or H2BGFP (Addgene plasmid #25999) vector backbone. The short hairpin sequences are listed in Table S1. The lentiviral plasmid expressing CRE and H2BRFP was modified from Addgene plasmid #25997. Expression plasmids coding for FLAG-tagged TCF3, TCF4 and ID1/2/3 were purchased from GeneCopoeia and used for co-immunoprecipitation experiments. Plasmids for inducible overexpression of ID1 and CEBPA were generated by Genescript from the pCW57.1 backbone (Addgene plasmid #38240).

RT-qPCR

Cells and tissue samples were collected in Trizol LS Reagent (Thermo Fisher Scientific). RNA was extracted using RNAeasy (Qiagen) or Direct-zol (Zymo), and treated with DNaseI (Qiagen). After the quality check, 100 ng of total RNA was used for cDNA synthesis using SuperScript IV Vilo (Thermo Fisher Scientific). RT-qPCR was run with selected primer pairs and SYBR Green on a ViiA 7 device or a 7500 fast system (both Applied Biosystems). HPRT was used as an internal control. Primer sequences are listed in Table S1.

RNAscope

RNA in situ hybridization to probe Tcf3, Tcf4 and Tcf12 was conducted using a Multiplex Fluorescent V2 kit from Advanced Cell Diagnostics, according to the manufacturer's instructions. In brief, FFPE slides were baked at 60°C for 1 h, followed by deparaffinization with two washes of xylene (5 min each) and two washes of 100% ethanol (2 min each). The slides were then dried for 5 min at 60°C, followed by incubation with hydrogen peroxide for 10 min at room temperature. The samples were then treated with protease at 40°C in the oven for 30 min, followed by incubation with Tcf3, Tcf4 and Tcf12 probes for 2 h at 40°C in the oven. After amplification with AMP (Amplification Reagent) 1, AMP 2 and AMP 3, each channel was developed by incubating HRP, fluorophore and HRP blocker sequentially based on the channels of the probe that was incubated. The samples were counterstained with DAPI (or antibodies if there were other channels available). Images were taken using a Zeiss Axioplan microscope and analyzed with Zen blue software.

Single-cell RNA-sequencing

The datasets (E13_WT: 929 cells; E15_WT: 633 cells) provided by Fan et al. (2018) were obtained from the GEO database (GSE102086), transformed into a matrix and used as an input for the R package Seurat (version 2.3.0). For the E13 dataset, cells with less than 500 genes, more than 8500 genes and more than 5% mitochondrial genes were excluded from further analyses. Seurat was used for clustering of the dataset. Count data was log-normalized with a scaling factor of 10,000 before identification of variable features. Variable features were identified using variance stabilizing transformation and the top 2000 features were selected. Data scaling regressed out unwanted variation due to library size or mitochondrial gene content. Features were selected using principal component analysis (PCA). Nearest neighbors were identified using the first 17 principal components and clusters identified using a resolution of 0.5. Using the FindAllMarkers function, keratinocytes were identified as being positive for Krt5 and Krt15 expression and negative for Pdgfra, Vim and Cd31 (Pecam1) (154 cells at E13). Using the function subset, re-analysis of the cells identified as keratinocytes was performed (variance stabilized transformed selecting top 2000 features, re-scaled data, re-ran PCA, performed nearest neighbor identification with six dimensions and clustering with a resolution parameter of 0.3) to visualize the expression of features of interest.

ChIP and ChIP-qPCR

ChIP from mouse epidermal progenitor cells overexpressing FLAG-tagged CEBPA or TCF3 was performed using the MAGnify chromatin immunoprecipitation system (Invitrogen) according to the manufacturer's instructions. In brief, cells were fixed with 1% formaldehyde at room temperature for 10 min. The cross-linking was stopped by incubation with glycine (125 mM) for 5 min at room temperature. The cells were washed with ice-cold PBS twice and lysed in lysis buffer supplemented with protease inhibitor (1×106 cells/50 µl lysis buffer), followed by sonicating into 200-500 bp fragments using a Bioruptor (Diagenode). The fragments were next incubated with magnetic beads, which were previously coupled with anti-Flag (Sigma-Aldrich, F1804) or IgG isotype (provided by the ChIP kit) antibodies at 4°C overnight. After washes with IP buffer1/2, the chromatin (including the input) was reverse cross-linked by incubation in reverse cross-linking buffer supplemented with proteinase K at 55°C for 15 min. The samples were then collected and purified with DNA purification magnetic beads, followed by eluting with 100 µl of elution buffer. The ChIP-qPCR was run with selected primer pairs and SYBR Green on a ViiA 7 device, and the calculations were based on the input and IgG isotype controls. Primer sequences are listed in Table S1.

In vitro establishment and expansion of primary epidermal progenitors

Primary epidermal progenitor cells were established from postnatal day (P) 0 pups and maintained according to published protocols (Nowak and Fuchs, 2009) using 3T3-J2 feeders and E-Low media [75% DMEM (Gibco, 21068028), 25% F12 (Gibco,21700026), supplemented with 15% chelated fetal bovine serum (FBS) (Hyclone, #SH30071.03), 0.45 µg/ml hydrocortisone (Merck, 386698-25MG), 0.1125 nM Cholera toxin (Sigma-Aldrich, D0564-1MG), 50 µg/mL transferrin (Sigma-Aldrich, T-2252), 50 µg/ml insulin (Sigma-Aldrich, I-5500), 0.02 nM 3T (3,3′,5-triiodo-L-thyronine, Sigma-Aldrich, T-2752), 1× penicillin-streptomycin (Gibco, 15140122), 8 mM L-glutamine (Gibco, 25030081) and 0.03% sodium bicarbonate (Gibco, 25080094)]. E-low media contained 63.53 µM calcium and in vitro differentiation media contained 1.5 mM calcium. Epidermal progenitor cells expressing shId1, shCebpa, shTcf3/4/12, control shScr or protein coding sequences for ID1 or CEBPA were generated by lentiviral infection followed by puromycin (1 µg/ml) selection. ID1 and CEBPA expression was induced by doxycycline (1 µg/ml) for 3 days using the pCW75.1 backbone (Addgene plasmid #38240). Primary epidermal progenitors were within 20 passages.

Production of high-titer lentivirus

Lentivirus was produced in 293TN cells (System Bioscience, LV900A-1) expanded in DMEM with 10% FBS, as previously described (Beronja et al., 2010), using pMD2/pPAX packaging plasmids (pMD2.G, Addgene plasmid #12259; psPAX2, Addgene, #12260) and calcium and the pLKO.1 (Addgene plasmid #10878) backbone. Epidermal progenitors were infected by spinoculation (1100 g for 30 min at 37°C) in the presence of 40 µg/ml hexadimethrine bromide (Beronja et al., 2010).

In utero lentiviral injections

High titer lentivirus was injected into the amniotic cavity of E9.5 embryos in accordance with previous publications (Beronja et al., 2010) and with the help of Infinigene (Karolinska Institutet).

Immunoprecipitation and mass spectrometry

The protein of interest (ID1, ID2, ID3, TCF3 or TCF4) was overexpressed by transfecting FLAG-tagged constructs into mouse epidermal progenitor cells (GeneCopoeia). Transfected cells were enriched using either puromycin or hygromycin and the FLAG-tag was precipitated using Sepharose beads pre-incubated with anti-FLAG M2 (Sigma-Aldrich, A2220) or IgG antibody. After incubation, beads were washed in Tris-buffered saline (TBS) and the precipitated protein was eluted using excess amounts of FLAG peptide (Sigma-Aldrich, final concentration 200 µg/ml in TBS). Efficiency of precipitations was confirmed using silver stain (Thermo Fisher Scientific, #24612) and, if approved, the eluate was forwarded for mass spectrometry analyses to the Proteomics Core facility, Karolinska Institutet. Interacting proteins were scored as hits based on the lack of signal in the IgG control, coverage and overall protein score.

Luciferase reporter assay

The pGL3 basic backbone (Promega) was linearized by enzyme digestion with NheI and XhoI. Then the Id1 promoter region (336 bp), previously described by Genander et al. (2014), was cloned into the pGL3 backbone with Infusion (Takara). Mouse Cebpa promoter (−2 kb) and enhancer regions (Cooper et al., 2015) were amplified using the following primer pairs: Cebpa_Prom_Infu_F TCTTACGCGTGCTAGCGCAGGGACATTTCTCACGAACC; Cebpa_Prom_Infu_R GATCGCAGATCTCGAGGAGTTAGAGTTCTCCCGGCA; Cebpa_Enh_Infu_F TATCGATAGGTACCGAGCTCCACCCCCTGATTTGCCATTCAT; Cebpa_Enh_Infu_R GCCTATCGAGCCCGGGTCCCAGGTCCCACCATACCTG.

The empty pGL3 backbone was used for background control. Epidermal progenitor cells were co-transfected with the luciferase construct and Renilla-luciferin 2-monooxygenase RLuc transfection baseline control (in a ratio of 1:10) using Lipofectamine LTX (Thermo Fisher Scientific). Lysates were measured for luciferase activity and RLuc using the Dual Glo luciferase assay system on a Glomax reader (both Promega). To assess BMP sensitivity, cells were serum starved overnight and treated with BMP4 (200 ng/ml) for 3 h before being lysed.

RNA-sequencing and bioinformatic analyses

Total RNA was analyzed on a bioanalyzer and used for library preparation. Samples were sequenced using a NovaSeq6000 platform (Illumina). A quality check (Fast QC/0.11.5) was run on raw sequencing reads. Raw sequencing reads were processed to obtain counts per genes for each sample. The EdgeR package was used to normalize for the RNA composition by finding a set of scaling factors for the library sizes that minimize the log-fold changes between the samples for most genes, using a trimmed mean of M values (TMM) between each pair of samples. Homer/4.10 was used to analyze promoters of genes and look for motifs that were enriched in the target gene promoters relative to all other promoters in the genome. Analyses, statistical computing and graphics were performed using R and in collaboration with National Bioinformatics Infrastructure Sweden (NBIS) at Karolinska Institutet.

Statistical analyses

GraphPad Prism and R were used for statistical analyses, including the use of multiple unpaired t-test and ANOVA when applicable. Error bars show the standard deviation or standard error of the mean as specified in the figure legends. Gene enrichment was analyzed using Gene Ontology (Ashburner et al., 2000). The n of in vivo quantifications is specified in the corresponding figure legend. In vitro experiments have been performed multiple times to verify the conclusions displayed in the representative figure.

We are grateful for technical assistance from Karolinska Institutet core facilities Infinigene and Proteomics Biomedicum as well as NBIS, SciLifeLab, Stockholm. We thank current and past members of the Genander lab for discussions, especially Srivathsa Magadi and Débora Lucia Joppi.

Author contributions

Conceptualization: W.Y., M.G., C.G.K.; Methodology: C.G.K., W.Y., D.G., K.V.P., K.H.-M.M.; Validation: W.Y.; Formal analysis: W.Y., C.G.K., D.G., M.G.; Investigation: C.G.K., W.Y., V.S.; Data curation: M.G.; Writing - original draft: W.Y., C.G.K., M.G.; Writing - review & editing: W.Y., M.G.; Visualization: W.Y., M.G., C.G.K.; Funding acquisition: M.G.

Funding

This study was supported by Vetenskapsrådet (2015-03215), Ragnar Söderbergs stiftelse (M26/15), Stiftelsen för Strategisk Forskning (ICA14-0055), Svenska Sällskapet för Medicinsk Forskning Stora Anslag 2015 and Cancerfonden (2016/270). M.G. is a Ragnar Söderberg Fellow and Cancerfonden Junior Investigator. Open access funding provided by Karolinska Institutet. Deposited in PMC for immediate release.

Data availability

RNA-sequencing data has been deposited at NCBI GEO (accession number GSE215055).

Andl
,
T.
,
Ahn
,
K.
,
Kairo
,
A.
,
Chu
,
E. Y.
,
Wine-Lee
,
L.
,
Reddy
,
S. T.
,
Croft
,
N. J.
,
Cebra-Thomas
,
J. A.
,
Metzger
,
D.
,
Chambon
,
P.
et al. 
(
2004
).
Epithelial Bmpr1a regulates differentiation and proliferation in postnatal hair follicles and is essential for tooth development
.
Development
131
,
2257
-
2268
.
Ashburner
,
M.
,
Ball
,
C. A.
,
Blake
,
J. A.
,
Botstein
,
D.
,
Butler
,
H.
,
Cherry
,
J. M.
,
Davis
,
A. P.
,
Dolinski
,
K.
,
Dwight
,
S. S.
,
Eppig
,
J. T.
et al. 
(
2000
).
Gene ontology: tool for the unification of biology. The Gene Ontology Consortium
.
Nat. Genet.
25
,
25
-
29
.
Bao
,
X.
,
Tang
,
J.
,
Lopez-Pajares
,
V.
,
Tao
,
S.
,
Qu
,
K.
,
Crabtree
,
G. R.
and
Khavari
,
P. A.
(
2013
).
ACTL6a enforces the epidermal progenitor state by suppressing SWI/SNF-dependent induction of KLF4
.
Cell Stem Cell
12
,
193
-
203
.
Beronja
,
S.
,
Livshits
,
G.
,
Williams
,
S.
and
Fuchs
,
E.
(
2010
).
Rapid functional dissection of genetic networks via tissue-specific transduction and RNAi in mouse embryos
.
Nat. Med.
16
,
821
-
827
.
Bjorntorp
,
E.
,
Parsa
,
R.
,
Thornemo
,
M.
,
Wennberg
,
A. M.
and
Lindahl
,
A.
(
2003
).
The helix-loop-helix transcription factor Id1 is highly expressed in psoriatic involved skin
.
Acta Derm. Venereol.
83
,
403
-
409
.
Blanpain
,
C.
and
Fuchs
,
E.
(
2006
).
Epidermal stem cells of the skin
.
Annu. Rev. Cell Dev. Biol.
22
,
339
-
373
.
Blanpain
,
C.
,
Lowry
,
W. E.
,
Pasolli
,
H. A.
and
Fuchs
,
E.
(
2006
).
Canonical notch signaling functions as a commitment switch in the epidermal lineage
.
Genes Dev.
20
,
3022
-
3035
.
Blessing
,
M.
,
Schirmacher
,
P.
and
Kaiser
,
S.
(
1996
).
Overexpression of bone morphogenetic protein-6 (BMP-6) in the epidermis of transgenic mice: inhibition or stimulation of proliferation depending on the pattern of transgene expression and formation of psoriatic lesions
.
J. Cell Biol.
135
,
227
-
239
.
Botchkarev
,
V. A.
and
Sharov
,
A. A.
(
2004
).
BMP signaling in the control of skin development and hair follicle growth
.
Differentiation
72
,
512
-
526
.
Boxer
,
L. D.
,
Barajas
,
B.
,
Tao
,
S.
,
Zhang
,
J.
and
Khavari
,
P. A.
(
2014
).
ZNF750 interacts with KLF4 and RCOR1, KDM1A, and CTBP1/2 chromatin regulators to repress epidermal progenitor genes and induce differentiation genes
.
Genes Dev.
28
,
2013
-
2026
.
Calero-Nieto
,
F. J.
,
Ng
,
F. S.
,
Wilson
,
N. K.
,
Hannah
,
R.
,
Moignard
,
V.
,
Leal-Cervantes
,
A. I.
,
Jimenez-Madrid
,
I.
,
Diamanti
,
E.
,
Wernisch
,
L.
and
Gottgens
,
B.
(
2014
).
Key regulators control distinct transcriptional programmes in blood progenitor and mast cells
.
EMBO J.
33
,
1212
-
1226
.
Cooper
,
S.
,
Guo
,
H.
and
Friedman
,
A. D.
(
2015
).
The +37 kb cebpa enhancer is critical for Cebpa myeloid gene expression and contains functional sites that bind SCL, GATA2, C/EBPalpha, PU.1, and additional Ets factors
.
PLoS One
10
,
e0126385
.
Coyle-Rink
,
J.
,
Sweet
,
T.
,
Abraham
,
S.
,
Sawaya
,
B.
,
Batuman
,
O.
,
Khalili
,
K.
and
Amini
,
S.
(
2002
).
Interaction between TGFbeta signaling proteins and C/EBP controls basal and Tat-mediated transcription of HIV-1 LTR in astrocytes
.
Virology
299
,
240
-
247
.
Fan
,
X.
,
Wang
,
D.
,
Burgmaier
,
J. E.
,
Teng
,
Y.
,
Romano
,
R. A.
,
Sinha
,
S.
and
Yi
,
R.
(
2018
).
Single cell and open chromatin analysis reveals molecular origin of epidermal cells of the skin
.
Dev. Cell
47
,
21
-
37.e25
.
Flora
,
P.
and
Ezhkova
,
E.
(
2020
).
Regulatory mechanisms governing epidermal stem cell function during development and homeostasis
.
Development
147
,
dev194100
.
Genander
,
M.
,
Cook
,
P. J.
,
Ramskold
,
D.
,
Keyes
,
B. E.
,
Mertz
,
A. F.
,
Sandberg
,
R.
and
Fuchs
,
E.
(
2014
).
BMP signaling and its pSMAD1/5 target genes differentially regulate hair follicle stem cell lineages
.
Cell Stem Cell
15
,
619
-
633
.
He
,
W.
,
Li
,
A. G.
,
Wang
,
D.
,
Han
,
S.
,
Zheng
,
B.
,
Goumans
,
M. J.
,
Ten Dijke
,
P.
and
Wang
,
X. J.
(
2002
).
Overexpression of Smad7 results in severe pathological alterations in multiple epithelial tissues
.
EMBO J.
21
,
2580
-
2590
.
Hopkin
,
A. S.
,
Gordon
,
W.
,
Klein
,
R. H.
,
Espitia
,
F.
,
Daily
,
K.
,
Zeller
,
M.
,
Baldi
,
P.
and
Andersen
,
B.
(
2012
).
GRHL3/GET1 and trithorax group members collaborate to activate the epidermal progenitor differentiation program
.
PLoS Genet.
8
,
e1002829
.
Hu
,
Y.
,
Zhang
,
Z.
,
Kashiwagi
,
M.
,
Yoshida
,
T.
,
Joshi
,
I.
,
Jena
,
N.
,
Somasundaram
,
R.
,
Emmanuel
,
A. O.
,
Sigvardsson
,
M.
,
Fitamant
,
J.
et al. 
(
2016
).
Superenhancer reprogramming drives a B-cell-epithelial transition and high-risk leukemia
.
Genes Dev.
30
,
1971
-
1990
.
Jakobsen
,
J. S.
,
Waage
,
J.
,
Rapin
,
N.
,
Bisgaard
,
H. C.
,
Larsen
,
F. S.
and
Porse
,
B. T.
(
2013
).
Temporal mapping of CEBPA and CEBPB binding during liver regeneration reveals dynamic occupancy and specific regulatory codes for homeostatic and cell cycle gene batteries
.
Genome Res.
23
,
592
-
603
.
Jankovic
,
V.
,
Ciarrocchi
,
A.
,
Boccuni
,
P.
,
DeBlasio
,
T.
,
Benezra
,
R.
and
Nimer
,
S. D.
(
2007
).
Id1 restrains myeloid commitment, maintaining the self-renewal capacity of hematopoietic stem cells
.
Proc. Natl. Acad. Sci. USA
104
,
1260
-
1265
.
Karaya
,
K.
,
Mori
,
S.
,
Kimoto
,
H.
,
Shima
,
Y.
,
Tsuji
,
Y.
,
Kurooka
,
H.
,
Akira
,
S.
and
Yokota
,
Y.
(
2005
).
Regulation of Id2 expression by CCAAT/enhancer binding protein beta
.
Nucleic Acids Res.
33
,
1924
-
1934
.
Klein
,
R. H.
,
Lin
,
Z.
,
Hopkin
,
A. S.
,
Gordon
,
W.
,
Tsoi
,
L. C.
,
Liang
,
Y.
,
Gudjonsson
,
J. E.
and
Andersen
,
B.
(
2017
).
GRHL3 binding and enhancers rearrange as epidermal keratinocytes transition between functional states
.
PLoS Genet.
13
,
e1006745
.
Korchynskyi
,
O.
and
ten Dijke
,
P.
(
2002
).
Identification and functional characterization of distinct critically important bone morphogenetic protein-specific response elements in the Id1 promoter
.
J. Biol. Chem.
277
,
4883
-
4891
.
Koster
,
M. I.
and
Roop
,
D. R.
(
2007
).
Mechanisms regulating epithelial stratification
.
Annu. Rev. Cell Dev. Biol.
23
,
93
-
113
.
Kowanetz
,
M.
,
Valcourt
,
U.
,
Bergstrom
,
R.
,
Heldin
,
C. H.
and
Moustakas
,
A.
(
2004
).
Id2 and Id3 define the potency of cell proliferation and differentiation responses to transforming growth factor beta and bone morphogenetic protein
.
Mol. Cell. Biol.
24
,
4241
-
4254
.
Lasorella
,
A.
,
Uo
,
T.
and
Iavarone
,
A.
(
2001
).
Id proteins at the cross-road of development and cancer
.
Oncogene
20
,
8326
-
8333
.
Liu
,
H.
,
Jia
,
D.
,
Li
,
A.
,
Chau
,
J.
,
He
,
D.
,
Ruan
,
X.
,
Liu
,
F.
,
Li
,
J.
,
He
,
L.
and
Li
,
B.
(
2013
).
p53 regulates neural stem cell proliferation and differentiation via BMP-Smad1 signaling and Id1
.
Stem Cells Dev.
22
,
913
-
927
.
Loomis
,
K. D.
,
Zhu
,
S.
,
Yoon
,
K.
,
Johnson
,
P. F.
and
Smart
,
R. C.
(
2007
).
Genetic ablation of CCAAT/enhancer binding protein alpha in epidermis reveals its role in suppression of epithelial tumorigenesis
.
Cancer Res.
67
,
6768
-
6776
.
Lopez
,
R. G.
,
Garcia-Silva
,
S.
,
Moore
,
S. J.
,
Bereshchenko
,
O.
,
Martinez-Cruz
,
A. B.
,
Ermakova
,
O.
,
Kurz
,
E.
,
Paramio
,
J. M.
and
Nerlov
,
C.
(
2009
).
C/EBPalpha and beta couple interfollicular keratinocyte proliferation arrest to commitment and terminal differentiation
.
Nat. Cell Biol.
11
,
1181
-
1190
.
Lopez-Pajares
,
V.
,
Qu
,
K.
,
Zhang
,
J.
,
Webster
,
D. E.
,
Barajas
,
B. C.
,
Siprashvili
,
Z.
,
Zarnegar
,
B. J.
,
Boxer
,
L. D.
,
Rios
,
E. J.
,
Tao
,
S.
et al. 
(
2015
).
A LncRNA-MAF:MAFB transcription factor network regulates epidermal differentiation
.
Dev. Cell
32
,
693
-
706
.
Malaguti
,
M.
,
Nistor
,
P. A.
,
Blin
,
G.
,
Pegg
,
A.
,
Zhou
,
X.
and
Lowell
,
S.
(
2013
).
Bone morphogenic protein signalling suppresses differentiation of pluripotent cells by maintaining expression of E-Cadherin
.
Elife
2
,
e01197
.
Massari
,
M. E.
and
Murre
,
C.
(
2000
).
Helix-loop-helix proteins: regulators of transcription in eucaryotic organisms
.
Mol. Cell. Biol.
20
,
429
-
440
.
Mesa
,
K. R.
,
Kawaguchi
,
K.
,
Cockburn
,
K.
,
Gonzalez
,
D.
,
Boucher
,
J.
,
Xin
,
T.
,
Klein
,
A. M.
and
Greco
,
V.
(
2018
).
Homeostatic epidermal stem cell self-renewal is driven by local differentiation
.
Cell Stem Cell
23
,
677
-
686
.
Meyer
,
M. B.
,
Benkusky
,
N. A.
,
Sen
,
B.
,
Rubin
,
J.
and
Pike
,
J. W.
(
2016
).
Epigenetic plasticity drives adipogenic and osteogenic differentiation of marrow-derived mesenchymal stem cells
.
J. Biol. Chem.
291
,
17829
-
17847
.
Miroshnikova
,
Y. A.
,
Cohen
,
I.
,
Ezhkova
,
E.
and
Wickstrom
,
S. A.
(
2019
).
Epigenetic gene regulation, chromatin structure, and force-induced chromatin remodelling in epidermal development and homeostasis
.
Curr. Opin. Genet. Dev.
55
,
46
-
51
.
Miroshnikova
,
Y. A.
,
Le
,
H. Q.
,
Schneider
,
D.
,
Thalheim
,
T.
,
Rubsam
,
M.
,
Bremicker
,
N.
,
Polleux
,
J.
,
Kamprad
,
N.
,
Tarantola
,
M.
,
Wang
,
I.
et al. 
(
2018
).
Adhesion forces and cortical tension couple cell proliferation and differentiation to drive epidermal stratification
.
Nat. Cell Biol.
20
,
69
-
80
.
Miyai
,
M.
,
Hamada
,
M.
,
Moriguchi
,
T.
,
Hiruma
,
J.
,
Kamitani-Kawamoto
,
A.
,
Watanabe
,
H.
,
Hara-Chikuma
,
M.
,
Takahashi
,
K.
,
Takahashi
,
S.
and
Kataoka
,
K.
(
2016
).
Transcription factor MafB coordinates epidermal keratinocyte differentiation
.
J. Invest. Dermatol.
136
,
1848
-
1857
.
Mori
,
S.
,
Nishikawa
,
S. I.
and
Yokota
,
Y.
(
2000
).
Lactation defect in mice lacking the helix-loop-helix inhibitor Id2
.
EMBO J.
19
,
5772
-
5781
.
Mulder
,
K. W.
,
Wang
,
X.
,
Escriu
,
C.
,
Ito
,
Y.
,
Schwarz
,
R. F.
,
Gillis
,
J.
,
Sirokmany
,
G.
,
Donati
,
G.
,
Uribe-Lewis
,
S.
,
Pavlidis
,
P.
et al. 
(
2012
).
Diverse epigenetic strategies interact to control epidermal differentiation
.
Nat. Cell Biol.
14
,
753
-
763
.
Murre
,
C.
,
McCaw
,
P. S.
and
Baltimore
,
D.
(
1989a
).
A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD, and myc proteins
.
Cell
56
,
777
-
783
.
Murre
,
C.
,
McCaw
,
P. S.
,
Vaessin
,
H.
,
Caudy
,
M.
,
Jan
,
L. Y.
,
Jan
,
Y. N.
,
Cabrera
,
C. V.
,
Buskin
,
J. N.
,
Hauschka
,
S. D.
,
Lassar
,
A. B.
et al. 
(
1989b
).
Interactions between heterologous helix-loop-helix proteins generate complexes that bind specifically to a common DNA sequence
.
Cell
58
,
537
-
544
.
Nam
,
H. S.
and
Benezra
,
R.
(
2009
).
High levels of Id1 expression define B1 type adult neural stem cells
.
Cell Stem Cell
5
,
515
-
526
.
Nerlov
,
C.
(
2007
).
The C/EBP family of transcription factors: a paradigm for interaction between gene expression and proliferation control
.
Trends Cell Biol.
17
,
318
-
324
.
Neuhold
,
L. A.
and
Wold
,
B.
(
1993
).
HLH forced dimers: tethering MyoD to E47 generates a dominant positive myogenic factor insulated from negative regulation by Id
.
Cell
74
,
1033
-
1042
.
Niola
,
F.
,
Zhao
,
X.
,
Singh
,
D.
,
Castano
,
A.
,
Sullivan
,
R.
,
Lauria
,
M.
,
Nam
,
H. S.
,
Zhuang
,
Y.
,
Benezra
,
R.
,
Di Bernardo
,
D.
et al. 
(
2012
).
Id proteins synchronize stemness and anchorage to the niche of neural stem cells
.
Nat. Cell Biol.
14
,
477
-
487
.
Nowak
,
J. A.
and
Fuchs
,
E.
(
2009
).
Isolation and culture of epithelial stem cells
.
Methods Mol. Biol.
482
,
215
-
232
.
Oh
,
H. S.
and
Smart
,
R. C.
(
1998
).
Expression of CCAAT/enhancer binding proteins (C/EBP) is associated with squamous differentiation in epidermis and isolated primary keratinocytes and is altered in skin neoplasms
.
J. Invest. Dermatol.
110
,
939
-
945
.
Okuyama
,
R.
,
Nguyen
,
B. C.
,
Talora
,
C.
,
Ogawa
,
E.
,
Tommasi di Vignano
,
A.
,
Lioumi
,
M.
,
Chiorino
,
G.
,
Tagami
,
H.
,
Woo
,
M.
and
Dotto
,
G. P.
(
2004
).
High commitment of embryonic keratinocytes to terminal differentiation through a Notch1-caspase 3 regulatory mechanism
.
Dev. Cell
6
,
551
-
562
.
Pfurr
,
S.
,
Chu
,
Y. H.
,
Bohrer
,
C.
,
Greulich
,
F.
,
Beattie
,
R.
,
Mammadzada
,
K.
,
Hils
,
M.
,
Arnold
,
S. J.
,
Taylor
,
V.
,
Schachtrup
,
K.
et al. 
(
2017
).
The E2A splice variant E47 regulates the differentiation of projection neurons via p57(KIP2) during cortical development
.
Development
144
,
3917
-
3931
.
Plikus
,
M.
,
Wang
,
W. P.
,
Liu
,
J.
,
Wang
,
X.
,
Jiang
,
T. X.
and
Chuong
,
C. M.
(
2004
).
Morpho-regulation of ectodermal organs: integument pathology and phenotypic variations in K14-Noggin engineered mice through modulation of bone morphogenic protein pathway
.
Am. J. Pathol.
164
,
1099
-
1114
.
Rao
,
C.
,
Malaguti
,
M.
,
Mason
,
J. O.
and
Lowell
,
S.
(
2020
).
The transcription factor E2A drives neural differentiation in pluripotent cells
.
Development
147
,
dev184093
.
Rezza
,
A.
,
Wang
,
Z.
,
Sennett
,
R.
,
Qiao
,
W.
,
Wang
,
D.
,
Heitman
,
N.
,
Mok
,
K. W.
,
Clavel
,
C.
,
Yi
,
R.
,
Zandstra
,
P.
et al. 
(
2016
).
Signaling networks among stem cell precursors, transit-amplifying progenitors, and their niche in developing hair follicles
.
Cell Rep
14
,
3001
-
3018
.
Roberts
,
E. C.
,
Deed
,
R. W.
,
Inoue
,
T.
,
Norton
,
J. D.
and
Sharrocks
,
A. D.
(
2001
).
Id helix-loop-helix proteins antagonize pax transcription factor activity by inhibiting DNA binding
.
Mol. Cell. Biol.
21
,
524
-
533
.
Romero-Lanman
,
E. E.
,
Pavlovic
,
S.
,
Amlani
,
B.
,
Chin
,
Y.
and
Benezra
,
R.
(
2012
).
Id1 maintains embryonic stem cell self-renewal by up-regulation of Nanog and repression of Brachyury expression
.
Stem Cells Dev.
21
,
384
-
393
.
Rotzer
,
D.
,
Krampert
,
M.
,
Sulyok
,
S.
,
Braun
,
S.
,
Stark
,
H. J.
,
Boukamp
,
P.
and
Werner
,
S.
(
2006
).
Id proteins: novel targets of activin action, which regulate epidermal homeostasis
.
Oncogene
25
,
2070
-
2081
.
Saisanit
,
S.
and
Sun
,
X. H.
(
1997
).
Regulation of the pro-B-cell-specific enhancer of the Id1 gene involves the C/EBP family of proteins
.
Mol. Cell. Biol.
17
,
844
-
850
.
Segre
,
J. A.
,
Bauer
,
C.
and
Fuchs
,
E.
(
1999
).
Klf4 is a transcription factor required for establishing the barrier function of the skin
.
Nat. Genet.
22
,
356
-
360
.
Sennett
,
R.
,
Wang
,
Z.
,
Rezza
,
A.
,
Grisanti
,
L.
,
Roitershtein
,
N.
,
Sicchio
,
C.
,
Mok
,
K. W.
,
Heitman
,
N. J.
,
Clavel
,
C.
,
Ma'ayan
,
A.
et al. 
(
2015
).
An integrated transcriptome atlas of embryonic hair follicle progenitors, their niche, and the developing skin
.
Dev. Cell
34
,
577
-
591
.
Sharov
,
A. A.
,
Weiner
,
L.
,
Sharova
,
T. Y.
,
Siebenhaar
,
F.
,
Atoyan
,
R.
,
Reginato
,
A. M.
,
McNamara
,
C. A.
,
Funa
,
K.
,
Gilchrest
,
B. A.
,
Brissette
,
J. L.
et al.  (
2003
).
Noggin overexpression inhibits eyelid opening by altering epidermal apoptosis and differentiation
.
EMBO J.
22
,
2992
-
3003
.
Shen
,
C. P.
and
Kadesch
,
T.
(
1995
).
B-cell-specific DNA binding by an E47 homodimer
.
Mol. Cell. Biol.
15
,
4518
-
4524
.
Soares
,
E.
and
Zhou
,
H.
(
2018
).
Master regulatory role of p63 in epidermal development and disease
.
Cell. Mol. Life Sci.
75
,
1179
-
1190
.
Soleimani
,
V. D.
,
Yin
,
H.
,
Jahani-Asl
,
A.
,
Ming
,
H.
,
Kockx
,
C. E.
,
van Ijcken
,
W. F.
,
Grosveld
,
F.
and
Rudnicki
,
M. A.
(
2012
).
Snail regulates MyoD binding-site occupancy to direct enhancer switching and differentiation-specific transcription in myogenesis
.
Mol. Cell
47
,
457
-
468
.
Sun
,
Y.
,
Zhou
,
B.
,
Mao
,
F.
,
Xu
,
J.
,
Miao
,
H.
,
Zou
,
Z.
,
Phuc Khoa
,
L. T.
,
Jang
,
Y.
,
Cai
,
S.
,
Witkin
,
M.
et al. 
(
2018
).
HOXA9 reprograms the enhancer landscape to promote leukemogenesis
.
Cancer Cell
34
,
643
-
658
.
Trompouki
,
E.
,
Bowman
,
T. V.
,
Lawton
,
L. N.
,
Fan
,
Z. P.
,
Wu
,
D. C.
,
DiBiase
,
A.
,
Martin
,
C. S.
,
Cech
,
J. N.
,
Sessa
,
A. K.
,
Leblanc
,
J. L.
et al. 
(
2011
).
Lineage regulators direct BMP and Wnt pathways to cell-specific programs during differentiation and regeneration
.
Cell
147
,
577
-
589
.
Vasioukhin
,
V.
,
Degenstein
,
L.
,
Wise
,
B.
and
Fuchs
,
E.
(
1999
).
The magical touch: genome targeting in epidermal stem cells induced by tamoxifen application to mouse skin
.
Proc. Natl. Acad. Sci. USA
96
,
8551
-
8556
.
Wohner
,
M.
,
Tagoh
,
H.
,
Bilic
,
I.
,
Jaritz
,
M.
,
Poliakova
,
D. K.
,
Fischer
,
M.
and
Busslinger
,
M.
(
2016
).
Molecular functions of the transcription factors E2A and E2-2 in controlling germinal center B cell and plasma cell development
.
J. Exp. Med.
213
,
1201
-
1221
.
Yates
,
P. R.
,
Atherton
,
G. T.
,
Deed
,
R. W.
,
Norton
,
J. D.
and
Sharrocks
,
A. D.
(
1999
).
Id helix-loop-helix proteins inhibit nucleoprotein complex formation by the TCF ETS-domain transcription factors
.
EMBO J.
18
,
968
-
976
.
Ying
,
Q. L.
,
Nichols
,
J.
,
Chambers
,
I.
and
Smith
,
A.
(
2003
).
BMP induction of Id proteins suppresses differentiation and sustains embryonic stem cell self-renewal in collaboration with STAT3
.
Cell
115
,
281
-
292
.
Yoon
,
S. J.
,
Foley
,
J. W.
and
Baker
,
J. C.
(
2015
).
HEB associates with PRC2 and SMAD2/3 to regulate developmental fates
.
Nat. Commun.
6
,
6546
.
Zauberman
,
A.
,
Lapter
,
S.
and
Zipori
,
D.
(
2001
).
Smad proteins suppress CCAAT/enhancer-binding protein (C/EBP) beta- and STAT3-mediated transcriptional activation of the haptoglobin promoter
.
J. Biol. Chem.
276
,
24719
-
24725
.
Zhang
,
N.
,
Yantiss
,
R. K.
,
Nam
,
H. S.
,
Chin
,
Y.
,
Zhou
,
X. K.
,
Scherl
,
E. J.
,
Bosworth
,
B. P.
,
Subbaramaiah
,
K.
,
Dannenberg
,
A. J.
and
Benezra
,
R.
(
2014
).
ID1 is a functional marker for intestinal stem and progenitor cells required for normal response to injury
.
Stem Cell Reports
3
,
716
-
724
.
Zhao
,
Q.
,
Liu
,
S.
,
Zhang
,
H.
,
Li
,
N.
,
Wang
,
X.
,
Cao
,
Y.
,
Ning
,
L.
,
Duan
,
E.
and
Xia
,
G.
(
2015
).
Spatiotemporal expression of p63 in mouse epidermal commitment
.
Int. J. Mol. Sci.
16
,
29542
-
29553
.
Zhu
,
S.
,
Oh
,
H. S.
,
Shim
,
M.
,
Sterneck
,
E.
,
Johnson
,
P. F.
and
Smart
,
R. C.
(
1999
).
C/EBPbeta modulates the early events of keratinocyte differentiation involving growth arrest and keratin 1 and keratin 10 expression
.
Mol. Cell. Biol.
19
,
7181
-
7190
.

Competing interests

The authors declare no competing or financial interests.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.

Supplementary information