Cytoplasmic localization of GRHL3 upon epidermal differentiation triggers cell shape change for epithelial morphogenesis

Kimura-Yoshida, Chiharu; Mochida, Kyoko; Nakaya, Masa-aki; Mizutani, Takeomi; Matsuo, Isao

doi:10.1038/s41467-018-06171-8

Download PDF

Article
Open access
Published: 03 October 2018

Cytoplasmic localization of GRHL3 upon epidermal differentiation triggers cell shape change for epithelial morphogenesis

Nature Communications volume 9, Article number: 4059 (2018) Cite this article

5215 Accesses
24 Citations
3 Altmetric
Metrics details

Subjects

A Publisher Correction to this article was published on 20 November 2018

This article has been updated

Abstract

Epithelial cell shape change is a pivotal driving force for morphogenesis of complex three-dimensional architecture. However, molecular mechanisms triggering shape changes of epithelial cells in the course of growth and differentiation have not been entirely elucidated. Grhl3 plays a crucial role as a downstream transcription factor of Wnt/β-catenin in epidermal differentiation. Here, we show Grhl3 induced large, mature epidermal cells, enriched with actomyosin networks, from embryoid bodies in vitro. Such epidermal cells were apparently formed by the simultaneous activation of canonical and non-canonical Wnt signaling pathways. A nuclear transcription factor, GRHL3 is localized in the cytoplasm and cell membrane during epidermal differentiation. Subsequently, such extranuclear GRHL3 is essential for the membrane-associated expression of VANGL2 and CELSR1. Cytoplasmic GRHL3, thereby, allows epidermal cells to acquire mechanical properties for changes in epithelial cell shape. Thus, we propose that cytoplasmic localization of GRHL3 upon epidermal differentiation directly triggers epithelial morphogenesis.

GRHL2 regulates keratinocyte EMT-MET dynamics and scar formation during cutaneous wound healing

Article Open access 14 October 2024

USP39 is essential for mammalian epithelial morphogenesis through upregulation of planar cell polarity components

Article Open access 19 April 2022

Murine interfollicular epidermal differentiation is gradualistic with GRHL3 controlling progression from stem to transition cell states

Article Open access 28 October 2020

Introduction

Epithelial morphogenesis directed by multiple cellular processes, such as cell shape changes, proliferation, and migration, involves the formation of complex three-dimensional architecture as seen, for example, in the formation of the mammalian neural tube¹. Coordinated cell shape changes, including contraction and elongation along the apical–basal axis, play a pivotal role in epithelial morphogenesis^2,3. Such cell deformations are controlled by cellular mechanical stress and tension at the cell surface primarily through intracellular contractile actomyosin networks^1,4,5,6,7. However, the mechanisms initiating the morphogenesis of epithelial cells in coordination with the timing of specification, i.e. cell growth or differentiation during development, are relatively unknown.

The epidermis constitutes the outermost epithelial layer that wraps the entire body and changes the body form. During epidermal development, the single ectodermal sheet is largely specified into neural and surface ectoderm (SE), which is an immature embryonic state of epidermis⁸. Such temporal SE cells subsequently commit into periderm and finally form mature epidermis, the outer component of the skin⁸. Epidermal specification is initiated by the signaling of several growth factors. Bone morphogenetic protein signaling has been shown to direct epidermal specification in frog and zebrafish embryos⁹. Wnt signaling has also been suggested to be involved in epidermal specification in chick and mouse embryos^10,11,12,13. Notably, we have found that during neural tube closure the canonical Wnt signaling pathway progressively specifies SE fate at the neural plate border, where neither surface nor neural cells are specified as uncommitted ectodermal progenitors¹⁴. These findings have led to the hypothesis that cell fate specification of SE during neurulation may be intimately linked to the epithelial morphogenesis of primary neurulation, which is governed by the non-canonical Wnt pathway involving planar cell polarity (PCP) genes¹⁵. However, little is known about what, when, and how molecular mechanisms control the coordination of epidermal fate decision and PCP-mediated epithelial morphogenesis.

The Grainy head family of transcription factors plays a highly conserved role in epithelial tissue development and remodeling in the animal kingdom^16,17. Since epithelial morphogenesis is one of the major driving forces of neurulation^18,19,20, the mammalian Grainy head family, encoding Grainyhead-like (Grhl) factors, is crucial for neural tube formation^21,22. During mouse primary neurulation, Grhl3 acts as a downstream effector of Wnt/β-catenin signaling to direct the specification of SE, a developmentally temporal structure of epidermis^8,14,23,24. However, molecular mechanisms underlying Grhl3-dependent epidermal differentiation in any subsequent epithelial morphogenesis remain poorly understood.

In this study, we demonstrate how the differentiation of epidermal cells intimately links to cell shape changes through Grhl3. Mechanistically, GRHL3, which has been considered as a transcription factor, appears to localize in the cytoplasm upon differentiation over time to activate non-canonical Wnt signaling. Such GRHL3-induced epidermal cells acquire mechanical properties necessary for cell shape change through actomyosin networks. This finding provides unique insights into the temporal mechanisms involved in the initiation of epithelial morphogenesis in coordination with cell specification during animal development.

Results

Grhl3 induces large and mature epidermal cells

During primary neurulation, Grhl3 can specify cellular fate into SE from ectodermal progenitor cells in neural folds¹⁴. To analyze the precise molecular mechanisms underlying epidermal differentiation by Grhl3, a two-step in vitro culture system was exploited (Fig. 1a)²⁵. Embryonic stem (ES) cell-derived embryoid bodies (EBs) consisted of central undifferentiated progenitor cells and TROMA-1 (keratin 8) positive peripheral epidermal cells, which are formed without any additional factors (Fig. 1b). In this system, the overexpression of Grhl3 cDNA induced epidermal cells within central EBs that were distinct from those in the periphery of EBs by control vector (pgk-neo) transfectants (Fig. 1b, c, g; n = 33/33 100%). To explore the role of Grhl3 in more detail, Grhl3-induced epidermal cells were characterized (Fig. 1h, i, Supplementary Fig. 1a–g); histological examination revealed these to be large and multinucleated (Fig. 1h, i). Molecular analyses using epidermal markers indicated that such large multinucleated epidermal cells were more mature than epidermal cells in the periphery of EBs (Supplementary Fig. 1a–h). In this regard, a marker for the neural crest, SLUG, was not expressed in large, mature (LM)-epidermal cells (Supplementary Fig. 1i, j). Furthermore, LM-epidermal cells were highly enriched with actomyosin as well as keratin bundles (Supplementary Fig. 1g, l, m). These findings indicate that Grhl3 can induce distinct LM-epidermal cells in the central region of EBs.

To identify whether Grhl1, another member of the Grhl family, could induce LM-epidermal cells in a similar manner, we overexpressed Grhl1 cDNA in EBs and found that Grhl1 did not induce LM-epidermal cells efficiently (Fig. 1d, g). However, Grhl1 cDNA appeared to induce defective types of LM-epidermal cells, designated as “solitary” and “scattered” epidermal cells, in EBs (Supplementary Fig. 1n, o). Afterward, we defined these two epidermal cell types cytomorphologically as follows: “solitary” epidermal cells consisted of a single epidermal cell in isolation but not multinucleated in EBs, while “scattered” epidermal cells formed as aggregates in EBs but not in multinucleated cells (Supplementary Fig. 1n, o). “None” means that TROMA-1–positive epidermal cells were found in EBs but not in the periphery (Supplementary Fig. 1k, p, Fig. 1b). Concurrently, a dominant-negative form of Grhl3 cDNA (DN-Grhl3), which can inhibit the activity of the Grhl family²⁶, reduced the number of LM-epidermal cells induced by Grhl3 cDNA (Fig. 1e–g). These findings indicate that Grhl3 has a distinct role in inducing “LM-epidermal cells” in EBs in vitro.

To determine whether Grhl3 could also induce LM-epidermal cells in vivo, we generated transgenic mice that misexpressed Grhl3 cDNA (Fig. 2a, b). To create Grhl3-misexpressing embryos, we crossed transgenic mice carrying CAG-loxP-lacZ-loxP-Grhl3 (CAG-lacZ-Grhl3) cDNA, in which the lacZ gene is flanked by two loxP sites and the Grhl3 cDNA is under the control of the CAG promoter, and β-actin cre transgenic mice (Tg(CAG-Grhl3); Fig. 2a, b). Morphological abnormalities in Tg(CAG-Grhl3) embryos were observed from E4.5 to E5.5 (Fig. 2c–f). Tg(CAG-Grhl3) embryos also failed to develop and grow after implantation (Fig. 2d, f). To explore the possibility that Grhl3 may have induced actomyosin enrichment and multinucleation, we analyzed actin and nuclear markers in Tg(CAG-Grhl3) embryos (Fig. 3g–j). Indeed, F-actin expression was drastically enriched in the cell membrane of the Tg(CAG-Grhl3) embryos (Fig. 2h; n = 3). Additionally, some cells became multinucleated (Fig. 2h, j; white arrows (n = 5,3, respectively)). Moreover, TROMA-1 expression appeared to be enhanced in Tg(CAG-Grhl3) embryos (Fig. 2l; n = 4). These findings suggest that Grhl3 misexpression promotes multinucleation and F-actin enrichment, being comparable to LM-epidermal cells derived from EBs in vitro (Fig. 1)

Furthermore, we misexpressed Grhl3 cDNA locally in SE cells during neurulation and analyzed phenotypes (Fig. 2m–u). Medium containing CAG-Grhl3 cDNA and CAG-EGFP plasmids was injected into the wild-type exocoelomic cavity at E8.5 (Fig. 2m) and, thereafter, the plasmids were electroporated into the superficial layer, i.e. SE cells of embryos. Then, electroporated embryos were cultured for 16–24 h and subjected to marker analysis (Fig. 2n–u). The plasmids, including CAG-EGFP, were apparently expressed in TROMA-1–positive SE cells after 16 h culture (Fig. 2n, o). In addition, these Grhl3-electroporated cells became multinucleated as a cluster and were enriched in F-actin (Fig. 2s–u; embryos n = 3; areas n = 8). These findings indicate that Grhl3 appears to induce similar characters of LM-epidermal cells in vivo.

Cytoplasmic GRHL3 transition for LM-epidermal induction

Since the transcriptional activity of CP2 transcription factors, which include the Grhl family, may be regulated according to their subcellular localization²⁷, we hypothesized that the activity of GRHL3 in the formation of LM-epidermal cells may also similarly be controlled according to its subcellular localization. To explore this idea, we examined GRHL3 expression and found that GRHL3 protein was located in the cytoplasm of LM-epidermal cells (Fig. 3a–c, Supplementary Fig. 2). To trace GRHL3 localization during epidermal maturation, a HaloTag system was designed to image protein localization in living cells (Fig. 3a–c)²⁸. After a 48 h culture, HaloTag-GRHL3, as visualized by TMR-ligand, was found mainly localized within the nucleus (Fig. 3a, c, Supplementary Fig. 2). Strikingly, a further 24 h culture (72 h in total) shifted GRHL3 expression from the nucleus to the cytoplasm (Fig. 3b, c, Supplementary Fig. 2).

To understand the mechanisms involved in the localization of GRHL3, we analyzed the subdomains directing cytoplasmic localization using a GRHL3-EGFP fusion protein with N-terminus, middle region, and C-terminus, respectively (Fig. 3d, Supplementary Fig. 3a, b). We found GRHL3 was able to localize to the cytoplasm through the N-terminus, CP2 domain, and C-terminus, and to the nucleus through the CP2 domain and C-terminus (Supplementary Fig. 3a, b). Immunohistochemistry with specific polyclonal antibodies against mouse GRHL3 confirmed that GRHL3 localized to both the nucleus and cytoplasm in LM-epidermal cells (Supplementary Fig. 3c–j). The above findings together support the hypothesis that GRHL3 localization shifts from the nucleus to the cytoplasm during epidermal maturation (Fig. 3a–c).

To examine if the subcellular localization of GRHL3 is crucial to the induction of LM-epidermal cells, we analyzed deletion constructs of mouse Grhl3 lacking C2, transcriptional activation (TA), and CP2 domains as well as Ub-like folding domains (Fig. 3d–f) and transfected these into ES cells. When the C2 (7–68 aa) or TA domain (30–96 aa) was deleted, induced LM-epidermal cells were significantly reduced in number compared to those induced in response to intact GRHL3 (Fig. 3d–f; 29.4% [n = 5/17], 25.0% [n = 3/12], respectively). These findings suggest that the N-terminus is necessary for the induction of LM-epidermal cells. Additionally, the removal of the Ub-like folding domain also reduced the emergence of LM-epidermal cells (Fig. 3d–f). Furthermore, CP2 deletion severely reduced the induction of all three types of epidermal cells: LM-, scattered, and solitary epidermal cells (Fig. 3e, f). Collectively therefore, in addition to the CP2 DNA-binding domain, N- (C2 and TA) and C-termini (Ub-like folding) that involve cytoplasmic localization are necessary for the formation of LM-epidermal cells.

To estimate whether the subcellular localization of GRHL3 itself can directly control the induction of LM-epidermal cells, we modified GRHL3 localization by the inclusion of nuclear import and export signals (nuclear localization signal [NLS] and nuclear export signal [NES], respectively) in the GRHL3 protein and then analyzed the formation of LM-epidermal cells from EBs (Fig. 3e, g–i)²⁹. By adding an NLS sequence, the induction of LM-epidermal cells was markedly reduced (Fig. 3g, i). Conversely, GRHL3 carrying the NES sequence did not affect the formation of LM-epidermal cells (Fig. 3h, i). The above results together support the hypothesis that cytoplasmic GRHL3 (C-GRHL3) is crucial for the induction of LM-epidermal cells.

Canonical and non-canonical Wnt pathways in LM-epidermal induction

The previous finding that Grhl3 acts as a downstream effector of Wnt/β-catenin signaling¹⁴ led us to study whether this canonical Wnt pathway is involved in the induction of LM-epidermal cells (Figs. 1, 2, 3). To test this hypothesis, the constitutively active form of β-catenin (β-cateninS37A) was transfected into ES cells (Fig. 4a). However, β-cateninS37A failed to induce LM-epidermal cells (Fig. 4a; n = 0/31). Additionally, chemical modulators of canonical Wnt signaling, a Wnt agonist and an inhibitor were also employed. First, the Wnt agonist was not able to induce LM-epidermal cells (Fig. 4b; n = 0/9). Next, the effects of a combination of chemical reagents and cDNA constructs (Grhl3 cDNA) on the induction of LM-epidermal cells were analyzed. Strikingly, neither an activator nor an inhibitor of the canonical pathway (Wnt agonist and FH535, respectively) affected the emergence of LM-epidermal cells (Fig. 4e, f; n = 15, n = 17, respectively). Collectively, these findings suggest that canonical Wnt signaling is neither a downstream nor parallel pathway of Grhl3 for the induction of LM-epidermal cells. This led us to explore additional signaling pathways that may be necessary for the induction of LM-epidermal cells.

We subsequently tested the non-canonical Wnt pathway by means of chemical modulators such as an activator (Rho/Rac/Cdc42 activator I) and inhibitors (H-89, SP600125, and NSC23766; Fig. 4c, d, g–i). Notably, activation of the non-canonical Wnt pathway significantly induced much larger sized LM-epidermal cells compared to those induced by Grhl3 alone (p < 0.05; Fig. 4c, d, g, h; n = 46). Moreover, inhibition of the non-canonical pathway repressed the formation of LM-epidermal cells induced by Grhl3 (Fig. 4i). To explore the involvement of the non-canonical Wnt pathway, the dominant negative form of Dsh (CAG-Dsh-DEP), which specifically repressed the PCP pathway, was exploited (Fig. 4i). Consistently, the Grhl3-dependent formation of LM-epidermal cells was blocked by CAG-Dsh-DEP (Fig. 4i). Thus collectively, the non-canonical Wnt pathway appears to be involved in the induction of LM-epidermal cells as downstream or parallel pathways of Grhl3.

Finally, the above findings led us to hypothesize that simultaneous activation of canonical and non-canonical Wnt pathways may lead to the formation of LM-epidermal cells. To test this concept, two cDNA vectors, harboring β-cateninS37A and Rac1, respectively, or the constitutively active form of Disheveled associated activator of morphogenesis 1 (Daam1)^19,30,31,32, which links the PCP pathway to PDZ-Rhogef, were transfected simultaneously into ES cells (Fig. 4j–m). Consequently, both transfected EBs were able to form LM-epidermal cells without Grhl3 cDNA (Fig. 4k, m; n = 3,8, respectively). Similarly, the simultaneous activation of both canonical and non-canonical pathways using chemicals was able to induce the formation of LM-epidermal cells (Fig. 4n; n = 7). These results indicate that the simultaneous activation of these two pathways was involved in the formation of LM-epidermal cells. Moreover, these findings imply that the canonical Wnt pathway facilitates the differentiation of epidermal cells while activation of the non-canonical Wnt pathway facilitates cell shape changes such as increased cell size, multi-nucleation, and the enrichment of actomyosin networks.

To explore how GRHL3 activated the non-canonical Wnt pathway, the expression of VANGL2, a PCP component, was analyzed using a HaloTag system (Fig. 5a). The expression of HaloTag-Vangl2 was localized to the cytoplasm of an LM-epidermal cell (Fig. 5a). Moreover, considerable amounts of cytoplasmic VANGL2 apparently co-localized with GRHL3 (Fig. 5b, asterisks). To confirm the co-localization of GRHL3 and VANGL2, we analyzed protein–protein interaction in NIH3T3 cells using fluorescent protein fragments of monomeric Kusabira–Green (mKG; n = 17; Supplementary Fig. 4). Such expression studies demonstrated that GRHL3 was located in close proximity to VANGL2 so that GRHL3 may indirectly or directly interact with non-canonical Wnt molecules in LM-epidermal cells.

In addition, expression of Scrib, another PCP protein³³, was specifically induced in the LM-epidermal cells (Fig. 5c). Coincidentally, the phosphorylation of non-muscle myosin light chain (pMLC), a marker for stress fiber formation³⁴, was exclusively found in LM-epidermal cells (Fig. 5d). However, although the expression of HaloTag–β-catenin was localized to both the nucleus and cytoplasm of LM-epidermal cells, cytoplasmic β-catenin did not co-localize with GRHL3 (Fig. 5e, f). The aforementioned studies support the notion that GRHL3 may contribute to non-canonical Wnt signaling partly through expression of PCP proteins in the cytoplasm.

C-GRHL3 is necessary for expression of PCP components

The aforementioned features of LM-epidermal cells allowed us to investigate if such cells reflect the character of in vivo GRHL3-positive SE cells during embryonic development. To evaluate this, several molecular markers were examined by immunohistochemistry (Fig. 6). Grhl3-positive cells, labeled with β-galactosidase antibody in Grhl3^cre/+ embryos, were evident in neural folds prior to neural tube closure and thereafter in dorsal SE cells¹⁴. In contrast, Grhl3-negative SE cells were distributed to the more lateral side of the epidermal layer (Fig. 6a–c)¹⁴. Further, F-actin and pMLC staining revealed that Grhl3-positive cells possessed assemblies of actomyosin bundles (Fig. 6a–d, o). Additionally, endogenous GRHL3 apparently localized to the nucleus and cytoplasm, as well as the cell membrane (Supplementary Fig. 5). Furthermore, multinuclear cells were sometimes found in Grhl3-positive (β-gal positive) SE cells (Fig. 6a–c; white arrows). Such studies suggest that Grhl3-positive dorsal epidermal cells during neurulation may be compatible with LM-epidermal cells derived from EBs in vitro (Figs. 1–4).

Given that GRHL3 appeared to co-localize with VANGL2 in LM-epidermal cells in vitro (Fig. 5a, b), it was reasonable to postulate that Grhl3 would genetically interact with Vangl2 during neurulation (Fig. 6e–h). When we analyzed VANGL2 localization, we found that VANGL2 protein was detected in dorsal SE cells, specifically in the cell membrane and partly in the cytoplasm, but not in the nucleus (Fig. 6e–g). Similarly, expression of other PCP components, CELSR1, RHOA, and SCRIB, were also evident in the cell membrane of dorsal SE cells (Fig. 6i, k, m). In Grhl3^cre/cre mutant embryos, however, membrane-associated VANGL2 expression was not evident (Fig. 6h) despite Vangl2 transcripts not being dysregulated in Grhl3^cre/cre embryos (Fig. 6q–t). Concordantly, the expression of other PCP components, CELSR1, RHOA and SCRIB, was reduced in Grhl3^cre/cre embryos, suggesting that expression of PCP components was down-regulated by the Grhl3 mutation (Fig. 6j, l, n). These findings suggest that GRHL3 is necessary for the correct membrane-associated localization of PCP components at the protein level. Consistent with a reduction of PCP-related molecules, pMLC expression was also down-regulated in Grhl3^cre/cre embryos (Fig. 6o, p), suggesting the crucial role of Grhl3 in the formation of actomyosin networks through PCP molecules. Concurrent with the above Grhl3-deficient phenotypes, expression of the PCP molecules, RHOA and CELSR1, was markedly up-regulated at E3.5 in Tg(CAG-Grhl3) transgenic embryos (Fig. 6u–x; n = 6, 2, respectively). These findings indicate that GRHL3 is necessary for the correct expression of PCP components.

To evaluate whether cytoplasmic GRHL3 was involved in epithelial morphogenesis, we generated genetically modified mice (Grhl3^NLS), in which GRHL3 localized to the nucleus rather than the cytoplasm, by inserting an NLS sequence into the translational start site of the Grhl3 locus (Fig. 7a, b, Supplementary Fig. 6). Grhl3 transcripts were normally generated and full-length GRHL3 was translated in Grhl3^NLS/NLS embryos (Fig. 7c–g, Supplementary Fig. 7,8). Additionally, NLS-fused GRHL3 was mainly localized in the nucleus of SE cells in Grhl3^NLS/NLS embryos as compared with GRHL3 in the wild type (Fig. 7h–k). Strikingly, Grhl3^NLS/NLS embryos displayed neural tube defects (NTDs) at E12.5 (Fig. 8c). Moreover, differences in the degree of abnormalities were apparent between Grhl3 null (Grhl3^Δex2/Δex2) and Grhl3^NLS/NLS embryos (Fig. 8a–c). While spina bifida developed in all Grhl3-deficient embryos (Fig. 8b), the spina bifida phenotype of Grhl3^NLS/NLS appeared to be milder than that of the Grhl3-deficient mutation; a much lower sacral spina bifida emerged in 57.1% of Grhl3^NLS/NLS embryos (n = 12/21); a curved tail phenotype alone was present in 19%, which manifested as incomplete penetrance (Fig. 8c, o). Therefore, collectively, cytoplasmic GRHL3 is essential for correct neural tube closure.

In order to verify whether defects in Grhl3^NLS/NLS embryos were a result of canonical Wnt–dependent epidermal differentiation or non-canonical Wnt–dependent epithelial morphogenesis, we analyzed Grhl3^NLS/NLS embryos more closely (Fig. 8d–n). In E9.25 wild-type embryos, ectodermal progenitor cells that did not express epidermal or neural markers were seen at the neural plate border (Fig. 8d; arrowheads)¹⁴. Conversely, in Grhl3-deficient embryos, progenitor cells were abolished (Fig. 8e)¹⁴. However, in Grhl3^NLS/NLS embryos, such progenitor cells appeared to be present and were similar to those in wild-type embryos (Fig. 8f; arrowheads). These results indicate that epidermal specification occurred normally in Grhl3^NLS/NLS embryos. Next, we found that expression of PCP molecules and pMLC were abolished in Grhl3^NLS/NLS embryos as found in Grhl3^cre/cre embryos (Fig. 5, Fig. 8g–n). Additionally, in agreement with the demonstration that Grhl3 is essential for wound repair (Supplementary Fig. 9)³⁵, Grhl3^NLS/NLS embryos did not show repair of the hind-limb amputation unlike wild-type embryos (Supplementary Fig. 9; n = 5). These findings suggest that defects in Grhl3^NLS/NLS embryos are brought about by failure of cytoplasmic GRHL3–dependent epithelial morphogenesis involving the non-canonical Wnt pathway, but not nuclear GRHL3 (N-GRHL3)–dependent transcription involving the canonical Wnt pathway.

To investigate whether C-GRHL3 could genetically interact with Vangl2 or β-catenin, we generated Grhl3^NLS; Vangl2, and Grhl3^NLS; β-catenin compound mutants and analyzed phenotypes (Fig. 8o). Remarkably, spina bifida was detected in 29.1% of Grhl3^NLS/+; Vangl2^Lp/+ double heterozygous embryos (Fig. 8o). Thus, this suggests genetic interaction between Vangl2^Lp/+ and Grhl3^NLS. Conversely, spina bifida was not evident in mice carrying Grhl3^NLS/+; β-catenin^+/− genotypes (n = 0/24, Fig. 8o), although Grhl3^cre/+;β-catenin^+/− mice exhibited a high frequency of spina bifida¹⁴. Taken together, these findings suggest that N-GRHL3 is defective in the Vangl2-dependent non-canonical Wnt pathway, but not canonical Wnt pathway.

Since Grhl3 is known to activate the PCP pathway at the transcriptional level via Rhogef19 expression for promoting actin polymerization³⁵, we verified whether failure in epithelial morphogenesis observed in Grhl3^NLS/NLS embryos was due to a reduction of Rhogef19 expression at the level of transcription or not (Fig. 8p–u). At E9.25 and E11.5, Rhogef19 transcripts were found in the SE (Fig. 8p, s). In agreement with the previous report, Rhogef19 transcripts were reduced in the SE of Grhl3-deficient embryos (Fig. 8q, t). Conversely, transcripts were not reduced in the SE of Grhl3^NLS/NLS embryos (Fig. 8r, u). These additional findings support the concept that defects of epithelial morphogenesis in Grhl3^NLS/NLS embryos were not brought about by down-regulation of Rhogef19 transcripts through the transcriptional activity of N-GRHL3, but rather by the down-regulation of PCP components at the protein level through cytoplasmic GRHL3.

GRHL3 confers mechanical properties on epidermal cells

To explore if GRHL3-positive epidermal cells acquire mechanical cues necessary for epithelial morphogenesis, elastic properties of epidermal tissues were measured using micropipette aspiration experiments (Fig. 9a, b, Supplementary Fig. 10). Wild-type or Grhl3 heterozygous dorsal SE tissues, in which Grhl3 was expressed, were aspirated under 1.0 or 2.0 kPa with the same glass micropipette (100 μm diameter) for 1 min, and tissue deformation into the micropipette recorded (Fig. 9a, b, Supplementary Fig. 10). Consequently, plotted total lengths of aspirated SE aggregates indicated that the elastic property of Grhl3-deficient SE tissues was less stiff than that of the wild type (Fig. 9b).

Next, we measured a Young’s modulus on SE tissues during neurulation using an atomic force microscope (AFM; Fig. 9c–f). We determined that the Young’s modulus of SE sheets of wild-type embryos was an average of 8.328 kPa (embryos n = 4; areas n = 10), while that of Grhl3^cre/cre mutant embryos was an average of 2.847 kPa (embryos n = 2; areas n = 6; Fig. 9c–f). Additionally, Young’s modulus of SE cells in Grhl3^NLS/NLS embryos was an average of 5.796 kPa (embryos n = 3; areas n = 8). These data indicate that SE cells expressing Grhl3 or C-GRHL3 protein were stiffer than Grhl3-negative SE cells. Taken together, these results suggest that Grhl3 is essential for ensuring mechanical properties in epidermal cells against the tensile force required for epithelial morphogenesis.

Finally, to directly evaluate the cortical tension of LM-epidermal cells induced by Grhl3 cDNA in vitro, a Young’s modulus of TROMA-1–positive epidermal cells was measured with an AFM (Fig. 9g–l)³⁶. We determined that the Young’s modulus of epidermal cells formed in the periphery of the EBs was an average of 1.82 kPa (Fig. 9g–i; n = 5), while that of LM-epidermal cells was an average of 3.64 kPa (Fig. 9j–l; n = 5). The stiffness of LM-epidermal cells induced by Grhl3 cDNA was therefore significantly higher than that of epidermal cells formed in the periphery (p < 0.05; one-tailed t-test). Thus, concurrent with the above in vivo study, Grhl3 is able to confer proper physical properties on epidermal cells.

Discussion

Our current findings indicate that N-GRHL3 directs epidermal differentiation from ectodermal progenitors and that this specification process is controlled by canonical Wnt signaling (Fig. 10)¹⁴. Upon epidermal differentiation, GRHL3 is localized in the cytoplasm and cell membrane, and is necessary for non-canonical Wnt-dependent cell shape changes (Fig. 10). Consequently, up-regulation of PCP molecules generates biomechanical forces primarily mediated by actomyosin networks during epithelial morphogenesis, i.e. neurulation (Fig. 10). Thus, GRHL3 functions as a unique signaling trigger for both canonical Wnt-dependent epidermal differentiation in the nucleus and non-canonical Wnt-dependent epithelial morphogenesis in the cytoplasm and cell membrane.

Cytoplasmic GRHL3 confers mechanical properties on epidermal cells for cell shape changes in coordination with cellular differentiation. Mechanical forces contribute to apical constriction and convergence toward the midline of the neural plate through actomyosin networks³⁷. In this manner, mechanical forces drive neural tube closure processes such as bending and folding of the neural plate^15,38. Therefore, it is highly probable that activation of PCP molecules directed by cytoplasmic GRHL3 facilitates actomyosin networks in epidermal cells and the resultant mechanical properties can direct epithelial morphogenesis and remodeling for neurulation.

We have identified two separate functions for GRHL3: as a nuclear transcription factor and as a cytoplasmic transducer of cell shape changes that may be mediated by distinct protein domains, transcriptional activation, DNA-binding CP2 domains, and N- and/or C-terminal regions, respectively. The two protein functions of GRHL3 are apparently controlled during the course of embryonic development. For many years, the processes of specification and morphogenesis, crucial during development, were thought to be controlled independently by different sets of molecules. In this respect, GRHL3 may be a unique factor that plays two different roles in epidermal specification and cell shape changes at the same time through the different subcellular localization of a single protein molecule. In this manner, GRHL3 can transmit and convert those signals used in epithelial differentiation to be useful for cell shape changes during development. To date, numerous lines of evidence have revealed that specification processes are initiated by extracellular signaling, with such triggering mechanisms intensively analyzed. In contrast, molecular entities that trigger cell shape changes have not yet been clearly identified.

With respect to the polyploidy-like phenotype induced by GRHL3, it is well known that the epidermis of mammalian skin becomes polyploidy during homeostasis^39,40,41. In addition, wound-induced polyploidization of the epidermis has also only recently been discovered and analyzed in the mouse;⁴² epidermal cells after injury are multinucleated and become enlarged in size in order for epithelial structure to recover quickly. As Grhl3 is also crucial for skin development as well as wound repair processes³⁵, it may be rational that GRHL3 contribute to multinucleation and the enlargement of cell size during normal development as well as wound repair.

At this stage, it remains to be determined how LM-epidermal cells induced by Grhl3 in vitro are relevant to in vivo epidermal cells. Phenotypes of transgenic mice and electroporated cells mis-expressing Grhl3 in vivo may not be completely identical with those of in vitro LM-epidermal cells. However, clear similarities are present between in vitro and in vivo cells. For example, multinucleation and the up-regulation of F-actin appear to be common (Fig. 2). Additionally, the expression of PCP molecules was also up-regulated both in vivo and in vitro (Figs. 5, 6). To identify in vivo counterpart of LM-epidermal cells, further comprehensive studies between LM-epidermal cells in vitro and epidermal cells in vivo during mouse development would be necessary.

This study suggests that GRHL3 may trigger both cell differentiation and shape change, a hypothesis that may provide unique insights into why cell specification is coordinated with morphogenetic processes to construct a three-dimensional complex structure during animal development. The grainy head family plays a phylogenetically conserved role in epithelial morphogenesis in the animal kingdom^16,17,43. The capability of dual functions in the form of a nuclear transcription factor and a cytoplasmic PCP transducer of GRHL3 may also be conserved in other species. In summary, our study sheds light on some of the processes involved in the initiation of epithelial morphogenesis and its coordination with cell specification during development.

Methods

Epidermal differentiation from ES cells in vitro

Epidermal cells were differentiated from G4 embryonic stem (ES) cells⁴⁴ via embryoid body (EB) cells²⁵. Appearance frequency of LM-epidermal cells (%) is calculated from the number of EBs having LM-epidermal cells among total number of EBs we analyzed. The plasmids, CAG-Grhl1, CAG-Grhl3, CAG-DN-Grhl3, CAG-Grhl3ΔC2, CAG-Grhl3ΔTA, CAG-Grhl3ΔCP2, CAG-Grhl3ΔUb, CAG-Grhl3NLS, and CAG-Grhl3NES were constructed according to standard procedures (detailed procedures for plasmid constructions are available upon request). For the construction of cDNAs, the plasmids pgk-neo⁴⁵, CAG-β-cateninS37A⁴⁶, CAG-Dsh-DEP⁴⁷, CAG-mDaam1 (the carboxyl-terminal amino acids 421 to 1078 of the mouse Daam1 cDNA; GenBank AY426535), and CAG-Rac1 (human RAC1; pFN21AB8254 from the Kasuza DNA Research Institute, Kisarazu, Japan) were used. Each cDNA, fused with a CAG promoter⁴⁸, was ligated to a neo gene driven by a pgk1 promoter with a polyadenylation signal. These plasmids were transfected into ES cells with Lipofectamine LTX (Invitrogen). Then, ES cells were cultured with G418 (150 μg mL⁻¹) for 24 h. The following chemical reagents were added to EB cells after transfection of plasmids: Wnt agonist (canonical Wnt activator; Merck, Kenilworth, NJ, USA; cat. no. 681665; 10 μM), FH535 (inhibitor of β-catenin/Tcf pathway; Merck cat. no 219330; 15 μM), H-89 (RhoA kinase inhibitor; CST cat.no. 371963; 10 μM), SP600125 (JNK [Rac] inhibitor; Merck cat. no. 420119, 20 μM), Rho/Rac/Cdc42 activator I (PCP pathway activator; Cytoskeleton, Inc; cat. no CN04, 1 μg mL⁻¹) and NSC23766 trihydrochoride (RacI inhibitor; Funakoshi cat.no. A12546-10 100 μM).

HaloTag labeling

HaloTag labeling was performed according to the manufacturer’s directions (www.promega.co.jp). We obtained Flexi HaloTag clones, pFN21AE2519 (human GRHL3), pFN21AB6277 (human β-CATENIN), and pFN21ASDA1215 (human VANGL2) from the Kazusa DNA Research Institute and halotag TMR ligand (promega cat.no. G8252) as halotag ligand.

Cell lines

G4 ES cells were obtained from Dr. Andras Nagy (Lunenfeld-Tanenbaum Research Institute at Mount Sinai Hospital in Toronto) and NIH3T3 cells were obtained from the Japanese Collection of Research Bioresources (JCRB, Japan).

Immunohistochemistry and in situ hybridization

Information of primary antibodies used in this study is described in Supplmentary Table. Specifically, regarding the antibody against mouse GRHL3, since the aa478-493 antibody displayed less background staining and recognized cytoplasmic GRHL3, the aa478-493 antibody is suitable to detect the translocation of GRHL3 from the cytoplasm into nucleus, i.e. it is appropriate to examine how much cytoplasmic GRHL3 is reduced or not. Consequently, the aa478-493 antibody was used to analyze the distribution of GRHL3 in Grhl3^NLS/NLS mutant embryos. In situ hybridization involving digoxigenin-labeled probes was conducted in a manner identical to that of Wilkinson⁴⁹.

Generation of a nuclear localization signal knock-in mouse

All constructs used this study were generated by standard molecular cloning techniques. In brief, a fusion gene consisting of the 5′ noncoding region of the nuclear localization signal (NLS)²⁹, the translational start site of the Grhl3 gene, and a neo cassette flanked with loxP was constructed (detailed procedures for this construction are available upon request). In the targeting vector, the lengths of the homologous region were 6.3 and 2.5 kb at the 5′ and 3′ sides of the insert, respectively. Homologous recombinant G4 ES cells and chimeric mice were generated by microinjection. To delete the loxP-flanked neo cassette, male chimeras were mated with C57Bl/6 background females carrying a Flpe gene driven by the chicken β-actin promoter⁵⁰.

Generation of CAG-Grhl3 transgenic mice

Mouse Grhl3 cDNA fused to the CAG promoter was ligated to the LacZ cassette flanked by two loxP sequences⁵¹. The resultant construct, CAG-loxP-lacZ-loxP-Grhl3 cDNA (5 ng μl⁻¹) was microinjected into the pro-nucleus of the fertilized eggs from superovulated CD-1 female mice and the injected eggs were cultivated in KSOM overnight then transferred to the pseudopregnant CD-1⁴⁴. The resultant offsping are genotyped to identify Tg (CAG-lacZ-Grhl3) founders. Tg(CAG-Grhl3) embryos were obtained by crossing Tg (CAG-lacZ-Grhl3) mice with β-actin Cre mice¹⁴.

Mouse genotyping

Grhl3 NLS knock-in founders and their progenitors were routinely determined by PCR and confirmed, when necessary, by southern blots of genomic DNA samples prepared from mouse tails or yolk sacs. In PCR analyses, primers and lengths of products were as follows: knock-in NLS sequence in the Grhl3 allele Grhl3-cre (5′-AATTAAGAGACGAGTGGTCAGCAGCAGCGCCTG-3′) and Grhl3-cre wt rev (5′-ACCCTTACAAATTGCCGTGTGAATCCGGGC-3′), yielding 213 bp as the wild-type allele and 234 bp as the NLS knock-in allele.

Grhl3-Cre-IRES-nlsLacZ knock-in mice were obtained from the Mutant Mouse Regional Resource Center (MMRRC) and genotyped with the following primers: Grhl3-cre (5′-AATTAAGAGACGAGTGGTCAGCAGCGCCTG-3′), Grhl3-cre wt rev (5′-ACCCTTACAAATTGCCGTGTGAATCCGGGC-3′), and Grhl3-cre mut (5′-GCAGCCCGGACCGACGATGAAGCATGTTTA-3′), yielding 213 bp as the wild-type allele and 370 bp as the mutant allele⁵².

Grhl3^ΔEx2 mice were obtained by crossing Grhl3^{tm1a(EUCOMM)Wtsi} stock (International Mouse Phenotyping Consortium) with Cre-expressing mice⁵³. In PCR analyses, the primers and lengths of products were as follows: flox-Ex2-A2 (5′-TTTGGGTGATGATGGGATCAGACCCAGGTC-3′), flox-Ex2-B2 (5′-CCATTCAAGCAATGTGTGCTGTACACCAGG-3′) and flox-Ex-D2 (5′-CCCCGGATCTAAGCTCTAGATAAGTAATGA-3), yielding around 1050 bp as the wild –type allele and 300 bp as the Grhl3^ΔEx2alelle.

β-catenin mutant mice were obtained from the Jackson Laboratory (Bar Harbor, Maine, USA)⁵⁴. An β-catenin mutant allele was identified as follows: RM41-internal (5′-AAGTTGTTTGTACAGAGTGTGGAGTTACTA-3′), RM42-internal (5′-CTCTCTGCCCAAGTGTAAACTTATGAGGCC-3′), and RM43-internal (5′-GGTATGTACAACATTGTTGGAACTTAGACA-3′) yielding around 200 bp for the wild-type allele and around 500 bp for the muatant allele.

An Lp mutant allele (Vangl2^Lp-2J obtained from the Jackson Laboratory) was identified as follows: Vangl2Alu-for1 (5′-CAACAGTATCTTCTCCCTTCCTCAGGCCT-3′) and Hpy166IIrev1: (5′-TCCTCAGAGAGTTTGAAGAAGGGCACCTTC-3′), yielding around 230 bp for the wild-type allele and the Lp mutant allele⁵⁵. Next, the 230 bp DNA fragments were purified with an illustra^TM GFX^TM PCR DNA and Gel Band Purification kit (GE Healthcare; cat. no 28903470). Subsequently, the purified fragments were digested with Hyp166II restriction enzyme and classified as wild type (125 bp and 100 bp), Lp heterozygous (125 bp, 115 bp, and 100 bp) or Lp homozygous (115 bp and 100 bp) after electrophoresis in UltraPure Agarose-1000 (ThermoFisher; scientific cat. no 16550-100).

Transgenic mice, CAG-lacZ-Grhl3 and CAG-Grhl3, were routinely determined by PCR. PCR primers and lengths of the products in the PCR analyses were as follows; in the transgenic mice, CAG-lacZ-Grhl3 was identified with primers CAG-Pro (5′-TAGAGCCTCTGCAACCTGTTCATGCCTT-3′) and CAG-lacZ (5′-AGTGTCCCAGCCTGTTTATCTACGGCTTAA-3′), yielding 270 bp. The CAG-Grhl3 allele excised by Cre protein was determined with primers CAG-Pro2 (5′-TACAGCTCCTGGGCAACGTGCTGGTTGTTG-3′) and CAG-Grhl3 (5′-CTTTCCTTGGTCATTCCGGCCACCAGTGC-3′), yielding 300 bp.

Whole embryo electropolation ex utero

Ex utero whole–embryo (E8.75) electroporation was conducted as follows⁵⁶. In brief, dissected tissues of ICR strains of mice were soaked in culture medium (50% rat serum and 50% Tyrode’s saline buffer). To prepare embryos for electroporation, decidua was trimmed away and two types of DNA solution, CAG-EGFP (1 μg mL^-1), with or without CAG-Grhl3 (1 μg mL^-1), were injected into the amniotic cavity using a glass capillary. Electroporation was conducted at a voltage of 70 (5 cycles of 50 msec pulse length and 950 msec interval) in Tyrode saline buffer using a CUY21EDITII electroporater and LF650P3 electrode (BEX, Tokyo, Japan). After electroporation, embryos were cultured in the above culture medium at 37 °C in a 5% CO₂ incubator.

Histology

For histological analysis, embryos and EB cells were fixed in Bouin’s fixative, dehydrated, and embedded in paraplast. Serial sections were generated and stained with hematoxylin and eosin.

Western blot analysis

Nuclear proteins from SE tissues of E16.5 were extracted using a Subcellular Protein Fraction Kit (ThermoFisher, cat. No.78840). Nucleoprotein extracts were quantified using a Qubit 3 Fluorometer (ThermoFisher), after which 40 μg of each extract was loaded onto a 10% bis-acrylamide gel and subjected to SDS-polyacrylamide gel electrophoresis (PAGE). Proteins were transferred to a polyvinylidene fluoride membrane using a Trans-Blot turbo transfer system (BioRad). The membrane was blocked using Block Ace Powder (DS Pharma Biomedical) for 2 h at RT. After primary incubation using blocking solution with anti-GRHL3 antibody (1:1000 antigen aa195-211) overnight at 4 °C, a second incubation was performed using blocking solution with anti-rabbit HRP antibody (Promega). Finally, the signal was revealed using a Western Lightning ECL Pro Kit (Perkin Elmer, Waltham, MA, USA; cat. No. NEL120001EA).

Reverse transcription (RT)-PCR analysis

Total RNA was isolated from wild-type, Grhl3^ΔEx2/ΔEx2 and Grhl3^NLS/NLS mutant embryos at E11.5 with TRIzol reagent (Thermo Fisher cat.no. 15596026), respectively. First-strand cDNA synthesis was performed with oligo-dT primers and Superscript III First-Strand synthesis System for RT-PCR (Invitrogen cat.no. 18080-51). The cDNA was utilized as a PCR substrate by standard protocols. The PCR primers, the forward primer (5′-ACCAGAGACGGATCGCTGGAACCTCGGAGA-3′) and the reverse primer (5′-CTGGAGAACCCTTTGACGGCTGCCACCAAA-3′) were employed for amplification.

Atomic force microscopy

Young’s modulus of epidermal cells in vitro was measured using atomic force microscopy (AFM). AFM and the following statistical analysis have been described previously^57,58. Epidermal cells differentiated from ES cells via EBs (embryoid bodies) were transfected with a Grhl3 plasmid and cultured on a glass substrate. AFM was used to measure Young’s modulus of epidermal cells. After the AFM measurement, cells were fixed and stained with TROMA-I–specific antibody. TROMA-I–positive cells were considered epidermal cells and used for statistical analysis. Cells with an area smaller than 3000 μm² were considered peripheral epidermal cells, and cells with an area larger than 10,000 μm² were considered large and mature (LM)-epidermal cells.

Young’s modulus of mouse embryonic SE cells was measured by AFM using a NanoWizard IV imaging system (JPK Instruments) combined with an inverted optical microscope (Olympus; IX73). AFM was operated in the QI mode in liquid by adjusting the contact force during imaging with a SQUBE type cantilever (type CP-cont-BSG-B). Post-processing was conducted using JPK Image Processing; in any case, the post-processing was minimized so to reduce processing artifacts.

Micropipette aspiration experiments

In the micropipette aspiration assay, two opposing glass micropipettes (about 100 μm diameter) were connected with a pressure control system through a manipulator (Narishige; cat. no. MMO-202ND and IM-11-2) and held against the dorso-ventral region of each embryo. Aspirations, (1 kPa and an additional 1 kPa to total 2 kPa) measured by digital pressure gauge (Krone, KDM30), were applied to the dorsal region of embryos to stably hold SE tissues. After 1 min aspiration, the height of absorbed SE tissues was recorded with an Olympus FluoView^TM FV1000 and × 40 zoom, and calculated by Olympus software.

Embryonic wound repair assays

Analysis of wound healing repair was performed on E12.5 wild type, Grhl3^cre/cre and Grhl3^NLS/NLS mutant embryos⁵⁹. After amputation of hind-limb, wound repairs were judged by the repaired epidermal region covered on the original wound diameter marked by TROMA-I expression with immunohistochemistry and scanning electron microscope (SEM) (Miniscope TM3030Plus Hitachi).

Animal experiments

All mouse studies followed the fundamental ethical guidelines for proper conduct of animal experiments and related activities in academic research institutions under the jurisdiction of the Ministry of Education, Culture, Sports, Science, and Technology of Japan and were approved by institutional committees at the Research Institute, Osaka Women’s and Children’s Hospital for animal and recombinant DNA experiments.

Statistical analysis

We analyzed cell-based assays and animal data by Student’s t-test or χ² test and defined statistically significant differences as p < 0.05.

Data availability

All relevant data are available from the corresponding authors upon reasonable request.

Change history

20 November 2018
The original version of this Article contained an error in the labelling of Fig. 4. In panel i, the sixth column was incorrectly labelled as NSC23766 negative, and should have been NSC23766 positive. This has now been corrected in both the PDF and HTML versions of the Article.

References

Munjal, A. & Lecuit, T. Actomyosin networks and tissue morphogenesis. Development 141, 1789–1793 (2014).
Article CAS Google Scholar
Rodriguez-Boulan, E. & Macara, I. G. Organization and execution of the epithelial polarity programme. Nat. Rev. Mol. Cell Biol. 15, 225–242 (2014).
Article CAS Google Scholar
Baum, B. & Georgiou, M. Dynamics of adherens junctions in epithelial establishment, maintenance, and remodeling. J. Cell. Biol. 192, 907–917 (2011).
Article CAS Google Scholar
Clark, A. G., Wartlick, O., Salbreux, G. & Paluch, E. K. Stresses at the cell surface during animal cell morphogenesis. Curr. Biol. 24, R484–R494 (2014).
Article CAS Google Scholar
Begnaud, S., Chen, T., Delacour, D., Mege, R. M. & Ladoux, B. Mechanics of epithelial tissues during gap closure. Curr. Opin. Cell Biol. 42, 52–62 (2016).
Article CAS Google Scholar
Gorfinkiel, N. & Blanchard, G. B. Dynamics of actomyosin contractile activity during epithelial morphogenesis. Curr. Opin. Cell Biol. 23, 531–539 (2011).
Article CAS Google Scholar
Zulueta-Coarasa, T. & Fernandez-Gonzalez, R. Tension (re)builds: biophysical mechanisms of embryonic wound repair. Mech. Dev. 144, 43–52 (2017).
Article CAS Google Scholar
Koster, M. I. & Roop, D. R. Mechanisms regulating epithelial stratification. Annu. Rev. Cell Dev. Biol. 23, 93–113 (2007).
Article CAS Google Scholar
Stern, C. D. Neural induction: old problem, new findings, yet more questions. Development 132, 2007–2021 (2005).
Article CAS Google Scholar
Wilson, S. I. & Edlund, T. Neural induction: toward a unifying mechanism. Nat. Neurosci. 4 Suppl, 1161–1168 (2001).
Article CAS Google Scholar
Wilson, S. I. et al. The status of Wnt signalling regulates neural and epidermal fates in the chick embryo. Nature 411, 325–330 (2001).
Article ADS CAS Google Scholar
Pearton, D. J., Yang, Y. & Dhouailly, D. Transdifferentiation of corneal epithelium into epidermis occurs by means of a multistep process triggered by dermal developmental signals. Proc. Natl Acad. Sci. USA 102, 3714–3719 (2005).
Article ADS CAS Google Scholar
Mukhopadhyay, M. et al. Dkk2 plays an essential role in the corneal fate of the ocular surface epithelium. Development 133, 2149–2154 (2006).
Article CAS Google Scholar
Kimura-Yoshida, C., Mochida, K., Ellwanger, K., Niehrs, C. & Matsuo, I. Fate specification of neural plate border by canonical Wnt signaling and Grhl3 is crucial for neural tube closure. EBioMedicine 2, 513–527 (2015).
Article Google Scholar
Copp, A. J., Greene, N. D. & Murdoch, J. N. The genetic basis of mammalian neurulation. Nat. Rev. Genet. 4, 784–793 (2003).
Article Google Scholar
Traylor-Knowles, N. et al. The evolutionary diversification of LSF and Grainyhead transcription factors preceded the radiation of basal animal lineages. BMC Evol. Biol. 10, 101 (2010).
Article Google Scholar
Venkatesan, K., McManus, H. R., Mello, C. C., Smith, T. F. & Hansen, U. Functional conservation between members of an ancient duplicated transcription factor family, LSF/Grainyhead. Nucleic Acids Res. 31, 4304–4316 (2003).
Article CAS Google Scholar
Sullivan-Brown, J. & Goldstein, B. Neural tube closure: the curious case of shrinking junctions. Curr. Biol. 22, R574–R576 (2012).
Article CAS Google Scholar
Nishimura, T., Honda, H. & Takeichi, M. Planar cell polarity links axes of spatial dynamics in neural-tube closure. Cell 149, 1084–1097 (2012).
Article CAS Google Scholar
Dworkin, S., Jane, S. M. & Darido, C. The planar cell polarity pathway in vertebrate epidermal development, homeostasis and repair. Organogenesis 7, 202–208 (2011).
Article Google Scholar
Gustavsson, P. et al. Increased expression of Grainyhead-like-3 rescues spina bifida in a folate-resistant mouse model. Hum. Mol. Genet. 16, 2640–2646 (2007).
Article CAS Google Scholar
Gustavsson, P., Copp, A. J. & Greene, N. D. Grainyhead genes and mammalian neural tube closure. Birth Defects Res. A Clin. Mol. Teratol. 82, 728–735 (2008).
Article CAS Google Scholar
Croce, J. C. & McClay, D. R. The canonical Wnt pathway in embryonic axis polarity. Semin. Cell. Dev. Biol. 17, 168–174 (2006).
Article CAS Google Scholar
Logan, C. Y. & Nusse, R. The Wnt signaling pathway in development and disease. Annu. Rev. Cell Dev. Biol. 20, 781–810 (2004).
Article CAS Google Scholar
Troy, T. C. & Turksen, K. Commitment of embryonic stem cells to an epidermal cell fate and differentiation in vitro. Dev. Dyn. 232, 293–300 (2005).
Article CAS Google Scholar
Kudryavtseva, E. I. et al. Identification and characterization of Grainyhead-like epithelial transactivator (GET-1), a novel mammalian Grainyhead-like factor. Dev. Dyn. 226, 604–617 (2003).
Article CAS Google Scholar
Chae, J. H., Kang, H. C. & Kim, C. G. The relative cellular levels of CP2a and CP2b potentiates erythroid cell-specific expression of the alpha-globin gene by regulating the nuclear localization of CP2c. Biochem. Biophys. Res. Commun. 380, 813–817 (2009).
Article CAS Google Scholar
England, C. G., Luo, H. & Cai, W. HaloTag technology: a versatile platform for biomedical applications. Bioconjug. Chem. 26, 975–986 (2015).
Article CAS Google Scholar
Abkallo, H. M., Kawano, H., Watanabe, K. & Kobayashi, N. A new cell-based reporter system for sensitive screening of nuclear export inhibitors. Drug Discov. Ther. 5, 286–292 (2011).
Article CAS Google Scholar
Liu, W. et al. Mechanism of activation of the Formin protein Daam1. Proc. Natl Acad. Sci. USA 105, 210–215 (2008).
Article ADS CAS Google Scholar
Seo, H. S., Habas, R., Chang, C. & Wang, J. Bimodal regulation of Dishevelled function by Vangl2 during morphogenesis. Hum. Mol. Genet. 26, 2053–2061 (2017).
Article CAS Google Scholar
Habas, R., Kato, Y. & He, X. Wnt/Frizzled activation of Rho regulates vertebrate gastrulation and requires a novel Formin homology protein Daam1. Cell 107, 843–854 (2001).
Article CAS Google Scholar
Zigman, M., Trinh le, A., Fraser, S. E. & Moens, C. B. Zebrafish neural tube morphogenesis requires Scribble-dependent oriented cell divisions. Curr. Biol. 21, 79–86 (2011).
Article CAS Google Scholar
Kasza, K. E., Farrell, D. L. & Zallen, J. A. Spatiotemporal control of epithelial remodeling by regulated myosin phosphorylation. Proc. Natl Acad. Sci. USA 111, 11732–11737 (2014).
Article ADS CAS Google Scholar
Caddy, J. et al. Epidermal wound repair is regulated by the planar cell polarity signaling pathway. Dev. Cell 19, 138–147 (2010).
Article CAS Google Scholar
Kumar, S., Chaudhury, K., Sen, P. & Guha, S. K. Atomic force microscopy: a powerful tool for high-resolution imaging of spermatozoa. J. Nanobiotechnol. 3, 9 (2005).
Article Google Scholar
Takeichi, M. Dynamic contacts: rearranging adherens junctions to drive epithelial remodelling. Nat. Rev. Mol. Cell Biol. 15, 397–410 (2014).
Article CAS Google Scholar
Cearns, M. D., Escuin, S., Alexandre, P., Greene, N. D. & Copp, A. J. Microtubules, polarity and vertebrate neural tube morphogenesis. J. Anat. 229, 63–74 (2016).
Article Google Scholar
Zanet, J. et al. A mitosis block links active cell cycle with human epidermal differentiation and results in endoreplication. PLoS ONE 5, e15701 (2010).
Article ADS CAS Google Scholar
Orr-Weaver, T. L. When bigger is better: the role of polyploidy in organogenesis. Trends Genet. 31, 307–315 (2015).
Article CAS Google Scholar
Freije, A. et al. Inactivation of p53 in human keratinocytes leads to squamous differentiation and shedding via replication stress and mitotic slippage. Cell Rep. 9, 1349–1360 (2014).
Article CAS Google Scholar
Losick, V. P., Jun, A. S. & Spradling, A. C. Wound-induced polyploidization: regulation by hippo and JNK signaling and conservation in mammals. PLoS ONE 11, e0151251 (2016).
Article Google Scholar
Wang, S. & Samakovlis, C. Grainy head and its target genes in epithelial morphogenesis and wound healing. Curr. Top. Dev. Biol. 98, 35–63 (2012).
Article CAS Google Scholar
Nagy, A., Rossant, J., Nagy, R., Abramow-Newerly, W. & Roder, J. C. Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc. Natl Acad. Sci. USA 90, 8424–8428 (1993).
Article ADS CAS Google Scholar
Soriano, P., Montgomery, C., Geske, R. & Bradley, A. Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64, 693–702 (1991).
Article CAS Google Scholar
Zorn, A. M. et al. Regulation of Wnt signaling by Sox proteins: XSox17 alpha/beta and XSox3 physically interact with beta-catenin. Mol. Cell 4, 487–498 (1999).
Article CAS Google Scholar
Tada, M. & Smith, J. C. Xwnt11 is a target of Xenopus Brachyury: regulation of gastrulation movements via Dishevelled, but not through the canonical Wnt pathway. Development 127, 2227–2238 (2000).
CAS PubMed Google Scholar
Niwa, H., Yamamura, K. & Miyazaki, J. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108, 193–199 (1991).
Article CAS Google Scholar
Wilkinson, D. G. In Situ Hybridization: A Practical Apporach 2nd edn (Oxford University Press, Oxford, 1998).
Kanki, H., Suzuki, H. & Itohara, S. High-efficiency CAG-FLPe deleter mice in C57BL/6J background. Exp. Anim. 55, 137–141 (2006).
Article CAS Google Scholar
Kimura-Yoshida, C. et al. Canonical Wnt signaling and its antagonist regulate anterior-posterior axis polarization by guiding cell migration in mouse visceral endoderm. Dev. Cell 9, 639–650 (2005).
Article CAS Google Scholar
Camerer, E. et al. Local protease signaling contributes to neural tube closure in the mouse embryo. Dev. Cell 18, 25–38 (2010).
Article CAS Google Scholar
Lewandoski, M., Meyers, E. N. & Martin, G. R. Analysis of Fgf8 gene function in vertebrate development. Cold Spring Harb. Symp. Quant. Biol. 62, 159–168 (1997).
Article CAS Google Scholar
Brault, V. et al. Inactivation of the beta-catenin gene by Wnt1-Cre-mediated deletion results in dramatic brain malformation and failure of craniofacial development. Development 128, 1253–1264 (2001).
CAS PubMed Google Scholar
Guyot, M. C. et al. A novel hypomorphic Looptail allele at the planar cell polarity Vangl2 gene. Dev. Dyn. 240, 839–849 (2011).
Article CAS Google Scholar
Hashimoto, M. & Takemoto, T. Electroporation enables the efficient mRNA delivery into the mouse zygotes and facilitates CRISPR/Cas9-based genome editing. Sci. Rep. 5, 11315 (2015).
Article ADS CAS Google Scholar
Haga, H. et al. Elasticity mapping of living fibroblasts by AFM and immunofluorescence observation of the cytoskeleton. Ultramicroscopy 82, 253–258 (2000).
Article CAS Google Scholar
Mizutani, T., Kawabata, K., Koyama, Y., Takahashi, M. & Haga, H. Regulation of cellular contractile force in response to mechanical stretch by diphosphorylation of myosin regulatory light chain via RhoA signaling cascade. Cell Motil. Cytoskeleton 66, 389–397 (2009).
Article CAS Google Scholar
McCluskey, J. & Martin, P. Analysis of the tissue movements of embryonic wound healing—DiI studies in the limb bud stage mouse embryo. Dev. Biol. 170, 102–114 (1995).
Article CAS Google Scholar

Download references

Acknowledgements

We are grateful to Drs. M. Tada, A.M. Zorn for the plasmids, Dr. A. Nagy for G4 ES cells, Dr. Y. Ueda for the technical support of AFM and Dr. M. Montcouquiol for the polyclonal VANGL2 antibody. The monoclonal antibody TROMA-I, developed by P. Brulet, and R. Kemler was obtained from the Developmental Studies Hybridoma Bank, created by the NICHD of the NIH and maintained at The University of Iowa, Department of Biology, Iowa City, IA 52242. The CAG-FLPe mouse strain (RBRC01834) was provided by RIKEN BRC through the National Bio-Resource Project of the MEXT, Japan. This work was supported in part by a grant-in-aid for Scientific Research on Priority Areas and on Innovative Areas (JP16H01456, JP18H04771, and JP17H05783) and Scientific Research (B)(JP26291053) and Challenging Exploratory Research (JP15K15408) from the Ministry of Education, Culture, Sports, Science, and Technology, Japan; and by the Kato Memorial Bioscience Foundation and the Takeda Science Foundation. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author information

Authors and Affiliations

Department of Molecular Embryology, Research Institute, Osaka Women’s and Children’s Hospital, Osaka Prefectural Hospital Organization 840, Murodo-cho, Izumi, Osaka, 594-1101, Japan
Chiharu Kimura-Yoshida, Kyoko Mochida & Isao Matsuo
Department of Molecular Biology, Yokohama City University Graduate School of Medicine, 3-9 Fukuura, Kanazawa-ku, Yokohama, 236-0004, Kanagawa, Japan
Masa-aki Nakaya
Department of Life Science and Technology, Faculty of Engineering, Hokkai-Gakuen University, Nishi 11-chome, Minami 26-jo, Chuo-ku, Sapporo, Hokkaido, 064-0926, Japan
Takeomi Mizutani

Authors

Chiharu Kimura-Yoshida
View author publications
Search author on:PubMed Google Scholar
Kyoko Mochida
View author publications
Search author on:PubMed Google Scholar
Masa-aki Nakaya
View author publications
Search author on:PubMed Google Scholar
Takeomi Mizutani
View author publications
Search author on:PubMed Google Scholar
Isao Matsuo
View author publications
Search author on:PubMed Google Scholar

Contributions

C.K-Y. and I.M. initiated the research and planned the experiments. C.K-Y. and K.M. performed most of the experiments. T.M. performed AFM measurement of epidermal cells. M.N. performed construction and characterization of the mouse Daam1 plasmids. C.K-Y. and I.M. contributed to the writing of the manuscript.

Corresponding authors

Correspondence to Chiharu Kimura-Yoshida or Isao Matsuo.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Supplementary Information

Peer Review File

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Kimura-Yoshida, C., Mochida, K., Nakaya, Ma. et al. Cytoplasmic localization of GRHL3 upon epidermal differentiation triggers cell shape change for epithelial morphogenesis. Nat Commun 9, 4059 (2018). https://doi.org/10.1038/s41467-018-06171-8

Download citation

Received: 15 November 2017
Accepted: 16 August 2018
Published: 03 October 2018
DOI: https://doi.org/10.1038/s41467-018-06171-8

This article is cited by

Long-term expansion of basal cells and the novel differentiation methods identify mechanisms for switching Claudin expression in normal epithelia
- Akihito Inoko
- Norihito Soga
- Kenji Kasai
Scientific Reports (2025)
Integrated profiling of metaplastic breast cancer identifies putative master regulators of intratumoral heterogeneity
- Yufan Feng
- Albert Xiong
- Amy E. McCart Reed
npj Breast Cancer (2025)
PALS1-dependent modulations of mRNA profiles in MDCK II cells grown in non-confluent monolayers and three-dimensional cysts
- Klaus Schughart
- Annika Möller-Kerutt
- Thomas Weide
BMC Genomic Data (2024)
USP39 is essential for mammalian epithelial morphogenesis through upregulation of planar cell polarity components
- Chiharu Kimura-Yoshida
- Kyoko Mochida
- Isao Matsuo
Communications Biology (2022)
A novel carbon nanotubes@porous carbon/sulfur composite as efficient electrode material for high-performance lithium-sulfur battery
- Lei Zhang
- Raja Arumugam Senthil
- Yanzhi Sun
Ionics (2019)