Smurf1 regulates ameloblast polarization by ubiquitination‐mediated degradation of RhoA

Abstract Cell polarity is essential for ameloblast differentiation and enamel formation. Smurf1 can mediate cell polarization through ubiquitination degradation of specific substrates. But it remains unclear whether Smurf1 could regulate ameloblast polarity and the underlying mechanism. Here, immuno‐fluorescence staining and RT‐qPCR were applied to detect the expression of Smurf1 and F‐actin. A mouse lower incisor defect model was constructed. Scanning electron microscope, rat lower incisor culture, western blot, wound healing assay and trans‐well migration assay were performed to detect the influence of Smurf1 knockdown on ameloblast. IF double staining, western blot and co‐immunoprecipitation were conducted to detect the interaction between Smurf1 and RhoA. The in vivo experiment was also performed. We found that Smurf1 was mainly expressed in the membrane and cell cortex of ameloblast, similar to F‐actin. Smurf1 expression increased along ameloblast polarization and differentiation. After knocking down Smurf1, the cytoskeleton and cell morphology changed and the cell polarity was damaged. Smurf1 regulated ameloblast polarity through ubiquitination degradation of activated RhoA in vitro. Local knockdown of Smurf1 in rat lower incisor ameloblast resulted in ameloblast polarity loss, enamel matrix secretion disorder and chalky enamel, but RhoA inhibitor Y‐27632 could reverse this effect. Collectively, Smurf1 could regulate the polarization of ameloblast through ubiquitination degradation of activated RhoA, which contributed to the knowledge of tooth development and provided new research ideas for cell polarity.


| INTRODUCTION
Ameloblasts (AB) are the key cells in tooth enamel formation. These cells are of epithelial origin and are involved in the whole process of enamel formation, including the synthesis and secretion of enamel matrix protein, the reabsorption and degradation of enamel matrix protein and the transport of calcium salts. 1 Abnormal interference in the differentiation and maturation of AB can lead to diseases related to enamel development such as enamel hypoplasia and enamel hypomineralization, 2 thus affecting patients' mastication, pronunciation and aesthetics. Cell polarity is crucial to the differentiation and maturation of AB and the formation of enamel.
Cell polarity refers to the asymmetric distribution of cytoskeleton, organelles and biomacromolecules due to the cooperation or Haoman Niu and Fei Bi contributed equally to this work. rejection of polarity protein complexes, 3 so that different regions of the cell can perform different functions, which plays an important role in embryonic development, 4 cell differentiation, 5 cell migration, 6 asymmetric cell division (ACD) 7 and tumour development. 8 According to the distribution characteristics of polarity protein complexes and cell morphology, cell polarity can be divided into four categories: Apical-basal polarity (ABP), 9,10 Planar cell polarity (PCP), 11,12 Front-rear polarity (FRP) [13][14][15] and ACD. [16][17][18]  Smad ubiquitination regulatory factor 1 (Smurf1), a ubiquitinprotein ligases belonging to the ubiquitin-proteasome system (UPS), has been found to regulate cell polarity for ubiquitination degradation of specific substrates through mediating the binding between Ubiquitin (Ub) and protein substrates via its C-terminal HECT domain. 19,20 In addition to cell polarity regulation, 21 Smurf1 has also been found to play an important role in a series of biological processes such as cell adhesion and migration, [22][23][24] nerve axon growth, 25 bone regeneration and bone homeostasis, 26,27 and autophagy. 28 Among the Smurf1 substrates found so far, RhoA, belonging to the Rho GTPases family, is involved in regulating cell polarity, 29 cell migration, 6 cell differentiation, 30 and tumour development. 31 RhoA is also essential during the odontogenesis process. 32 The expression of Rho GTPases is regulated by post-translational modifications, such as lipid modification, phosphorylation and ubiquitination. 33 Particularly, in the ubiquitination process, Smurf1 can catalyse the ubiquitination degradation of RhoA in activated state. 34,35 Smurf1 can mediate cell polarization through ubiquitination degradation of specific substrates, but whether Smurf1 is involved in AB polarity has not been reported. RhoA is one of the substrates of Smurf1 and also plays an important role in the polarity of AB. However, it remains unclear whether Smurf1 can regulate ameloblast polarity by degrading RhoA during tooth development. Therefore, we conducted in vitro and in vivo experiments to study the role of Smurf1 in regulating AB polarity during cell differentiation and the F I G U R E 1 Graphical abstract underlying mechanism specifically in a RhoA ubiquitination-mediated degradation way ( Figure 1).

| Animals and tissue sections
The C57BL/6 mice and Sprague-Dawley (SD) rats were purchased from the experimental Animal Laboratory of Sichuan University. All

| Ameloblasts' isolation
Carefully remove the surrounding bone tissues and acquire mandibular incisor of the adult mice under the stereo-microscope. Immerse the incisors into 2% collagenase type I solution at 37 C for 45 min and gently shake the container every 5 min to separate attached mesenchymal tissues from dental epithelium as much as possible. Carefully peel off the surrounding mesenchymal tissues and divide the mandibular incisor into three parts according to the anatomical points ( Figure S1A) under the stereo-microscope. The enlarged tail of the epithelium ends belonged to the cervical loop (CL) group. The epithelial tissues from the enamel deposition point to the crown side belonged to the AB group. Remaining tissues sited in the middle of CL and AB group belonged to the PAB group. 36

| Quantitative real-time PCR analysis (RT-qPCR)
The total RNA was extracted from tissues and cells using Trizol (Invitrogen, USA). The cDNA was synthesized by the Prime Script™ RT-PCR with gDNA Eraser Kit (TaKaRa, Japan). The SYBR ® Premix Ex Taq II (TaKaRa, Japan) was applied in the RT-qPCR tests. The reaction was performed with QuantStudio 3 Flex (Applied Biosystems, USA).
The primer sequences were listed in Table 1.

| Mouse lower incisor defect model
The left lower incisors of 8-week-old mice were grinded off with the dental emery drill. The right lower incisors were used for self-control.
Three days after the operation, the mice were executed and the teeth were observed under natural light. The dental epitheliums were isolated by CL, AB and PAB group as previously described.

| HAT-7 cells' transfection
The rat AB cell line namely HAT-7 cells were cultured with DMEM/ F-12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA) and 1% penicillin-streptomycin at 37 C with 5% CO 2 . The medium was changed every 2 days. The cells were seeded into 6-well plate. Once achieving 70% confluence, the cells were transfected with the following agents: the control siRNA, Smurf1 siRNA1, Smurf1

| In vitro rat lower incisor culture
The postnatal 5-7 days rat lower incisors were isolated and cultured in trans-well chambers of 6-well plate in DMEM/F-12, supplemented with 10% FBS, 1% penicillin-streptomycin and 200 mg/l ascorbic acid for 7 days. The medium was changed every 2 days. Collected tissues were fixed in 4% paraformaldehyde, and sliced for further H&E staining.

| Wound healing assay and trans-well migration assay
In the wound healing experiment, the migration area of cells was

| In vivo experiment
The PN 3d rats were divided into four groups (n = 5) randomly and injected with normal saline, lentivirus packed with control shRNA, sh-Smurf1 and mixture of sh-Smurf1 and Y-27632 at the mandible adjacent to the dental epithelium on PN 3d and PN 17d.
The mandibles were collected on PN 25d. After fixed with 4% paraformaldehyde overnight, the mandibles were demineralized with 10% EDTA (pH 8.0) and made into slices for H&E staining and IF staining.

| Statistical analysis
All experiments were performed independently at least three times.
The statistical analysis was carried out in SPSS 21.0 software using Student's t-test or one-way ANOVA. p < 0.05 was considered statistically significant. group had no statistical significance ( Figure 4C, C16, p > 0.05). These results suggested that Smurf1 knockdown had no effect on ameloblast proliferation and apoptosis.
3.4 | The mechanism of Smurf1 affecting ameloblast polarization by ubiquitination-mediated degradation of RhoA In Figure 5A, IF double staining assay exhibited that Smurf1 was mainly expressed in the membrane and nucleus, and was also positively expressed in the perinuclear cytoplasm. RhoA was strongly expressed in the nuclear membrane and nucleus region, and showed highly positive punctate expression on the cell membrane. The activated RhoA which Smurf1 was mainly responsible for degrading 34,37 were primarily located in the membrane while the inactive RhoA was located in the cytoplasm and nucleus. These results indicated that Smurf1 and activated RhoA co-located on the membrane, suggesting that there might be an interaction between them.
Compared with the blank group, the expression level of Smurf1 in the overexpression group was testified to be increased by 20.3 times via RT-qPCR and 1.98 times via western blot ( Figure 5B, p < 0.05).
RT-qPCR showed that RhoA expression significantly increased in RhoA T19N, RhoA WT and RhoA G14V groups compared with the blank and the vector group ( Figure 5C, p < 0.05). Western Blot results displayed the same trend of total RhoA expression ( Figure 5D). However, the expression level of activated RhoA was lowered in RhoA T19N group, slightly raised in RhoA WT group, and significantly elevated in RhoA G14V group ( Figure 5D). The above results suggested that the overexpression efficiency of Smurf1 and RhoA was high and the HAT-7 cells transfected with the plasmids could be used for further study. To testify that Smurf1 regulate cell polarity via degradation of RhoA, in vitro 3D culture of SD rat's mandibular incisor with transwell system had been employed. In the blank group, the negative control group and the rescue group, AB presented high columnar shape, palisade arrangement and the nuclei being far away from the basement membrane ( Figure 5H, H1, H2, H4). However, in the F I G U R E 4 Legend on next page. Smurf1 siRNA1 group, the AB were disordered and the polarity structure was damaged ( Figure 5H, H3). These results demonstrated that Smurf1 regulated ameloblast polarity through ubiquitination degradation of RhoA in vitro.

| The in vivo study of Smurf1 regulating ameloblast polarization
Rats were divided into four groups then injected respectively with the transformed morphology and motility of a tumour cell. 40 However, so far, the role of Smurf1 has not been studied in enamel development.
To detect whether Smurf1 expresses in ameloblast, the slices of tooth germ and lower incisor of mouse were stained with IF. We found that Smurf1, similar to F-actin, mainly located in ABs' membrane and cell cortex during the whole process of enamel organ development. In order to quantitatively detect the expression changes of Smurf1 during AB differentiation in the lower incisor of mouse, the dental epitheliums in different regions were isolated into CL group, PAB group and AB group and the expression of Smurf1 was measured by RT-qPCR. Our data showed that with the polarity formation and differentiation maturation of ameloblast, the expression of Smurf1 was up-regulated, which suggested a correlation between Smurf1 and ameloblast polarization and development. In our study, the IF results showed that Smurf1 colocalized with F-actin, which is known to provide mechanical support for cells and control the cell morphology and cell migration through the rapid assembly and disassembly of the actin-network, 41 suggesting that Smurf1 might be able to interact with F-actin and play a role in cytoskeleton and cell morphology regulation.
There is a stem cell niche in labial CL of mouse's lower incisor, which renders the tooth continuing eruption for lifetime and repair after injury. 42 In our study, the formation of new enamel and the growth of the damaged mouse lower incisor were observed 3 days after injury, further confirming the existence and function of cervical stem cell niche. We found that 3 days after injury, the expression of AB lost polarity, the secreted enamel matrix was in disorder and the gross observation displayed chalky and opaque enamel, which identified that Smurf1 was essential for ameloblast polarity and enamel development.
Smurf1 also produces an influence on cell differentiation, cell adhesion and cell migration. For example, Smurf1 was required for the myogenic differentiation of C2C12 cells and played an important regulatory role in the BMP-2-mediated osteoblast conversion. 49 53 found that low levels of constitutively active RhoA expression associated with segmental foot-process effacement without changes observable by light microscopy, whereas higher levels of constitutively active RhoA expression associated with both extensive foot-process effacement and histologic features of focal segmental glomerulosclerosis. So, proper range of RhoA is essential to maintain AB polarity and form enamel, which is worth to be studied in future work.
Smurf1 has also been found to play an important role in regulating dental pulp stem cells (DPSCs) differentiation and odontoblasts' differentiation and homeostasis. Yang et al. 54 found that the inhibition of Smurf1 in DPSCs significantly increased RUNX2 at the protein level which was associated with Smurf1 level during odontoblastic differentiation. The knockdown of Smurf1 significantly up-regulated RUNX2 expression and down-regulated dentin sialophosphoprotein and dental matrix protein-1 expression in odontoblastic differentiation. Lee et al. 55 found that Smurf1 and Smurf2 induced NFI-C degradation and polyubiquitination in a TGF-β1-dependent manner and NFI-C was significantly degraded after TGF-β1 addition in odontoblasts. However, there is still a blank of in vivo verification of Smurf1's role on odontoblasts and dentin formation which is worth further exploring.
In our future study, we are dedicating ourselves to conduct further research on the upstream signalling pathway of Smurf1 to form a complete and thorough mechanism for better understanding of the ameloblast polarity regulation axis. We also intend to explore the role of epithelial mesenchymal interaction (EMI) in the polarity formation of AB and odontoblasts using rodent lower incisors as an animal model, and try to extend our findings to other human organs and tissues which resemble the development pattern.

| CONCLUSION
Collectively, we demonstrated that Smurf1 knocking down affected

FUNDING INFORMATION
This study was supported by the Nature Science Foundation of China (31971281).