The absence of muscle segment homeobox 2 leads to the pyroptosis of ameloblasts by inducing squamous epithelial hyperplasia in the enamel organ

Abstract Muscle segment homeobox 2 (MSX2) has been confirmed to be involved in the regulation of early tooth development. However, the role of MSX2 has not been fully elucidated in enamel development. To research the functions of MSX2 in enamel formation, we used a Msx2 −/− (KO) mouse model with no full Msx2 gene. In the present study, the dental appearance and enamel microstructure were detected by scanning electron microscopy and micro‐computed tomography. The results showed that the absence of Msx2 resulted in enamel defects, leading to severe tooth wear in KO mice. To further investigate the mechanism behind the phenotype, we performed detailed histological analyses of the enamel organ in KO mice. We discovered that ameloblasts without Msx2 could secrete a small amount of enamel matrix protein in the early stage. However, the enamel epithelium occurred squamous epithelial hyperplasia and partial keratinization in the enamel organ during subsequent developmental stages. Ameloblasts depolarized and underwent pyroptosis. Overall, during the development of enamel, MSX2 affects the formation of enamel by regulating the function of epithelial cells in the enamel organ.

terminal cell differentiation and contributing to the initial patterning of dentition. 9,10 These studies suggest that MSXs play a pivotal role during organogenesis.
Previous studies have reported that MSX2 may act on dentinogenesis by controlling the formation of mineralized tissues. 11 The mutation of the human Msx2 gene causes craniosynostosis syndrome. 12,13 The absence of Msx2 leads to compound periodontal osteopetrosis, dentinogenesis imperfecta and amelogenesis imperfecta. 4,14 MSX2 not only regulates the expression of the enamel protein in ameloblast, but also participates in the regulation of the differentiation and the maintenance of function in the dental epithelium. 15,16 However, the developmental basis for amelogenesis imperfecta caused by Msx2-mutant remains unclear, including its role in the functional maintenance of the epithelial cells in the enamel organ, enamel defects and the fate of ameloblasts.

| Histology and immunohistochemistry (IHC)
The dissected mandibles from WT and KO mice and paraffin samples were prepared as previously described 4 for haematoxylin-eosin (HE) staining or IHC. For IHC analysis, the sections were treated overnight with primary antibody at 4˚C. Rabbit polyclonal antibody against MSX2, cleavage Caspase-1 rabbit polyclonal antibody, IL-1β polyclonal antibody and CK5/6 polyclonal antibody, all of the above antibodies were purchased from Abcam (Abcam, Cambridge, MA, USA).
IHC results were detected with the peroxidase substrate kit and the Vectastain ABC Elite kit (Vector Laboratories, Inc, Burlingame, CA, USA) according to the manufacturer's instructions, and haematoxylin stained the nuclei for 4 minutes at room temperature. The experiments were repeated for three times.

| Quantitative real-time PCR (qPCR)
Total RNA was extracted with TRIzol (Takara Biotechnology Co., Ltd., Shiga, Japan). The cDNA was reversely transcribed from total RNA using the MMLV-RT system (Promega Corporation, Madison, USA), and quantitative real-time PCR was performed using the SYBR® PrimeScript® kit (Takara Biotechnology) Relative mRNA expression levels were normalized to GAPDH. The primer sequences for realtime PCR are listed in Table 1. Relative gene expressions were calculated with the 2 −ΔΔCt method. Each experiment was performed in triplicate. All the above kits were performed in accordance with the manufacturer's protocol.

| Scanning electron microscopy (SEM)
Hemi-mandibles isolated from WT and KO mice (n = 3 per genotype) were fixed in 10% paraformaldehyde at 4℃ for 20 h. Alcohol gradient dehydration was operated and dried with the critical point drier (Quorum K850). In addition, the samples were embedded in epoxy resin for analysis 25s and air-dried at room temperature. The specimen was coated by palladium gold and observed under a scanning electron microscope (ZEISS EVO LS 15; Carl Zeiss AG, Jena, Germany) at 10 kV.

| Micro-computed tomography (µCT) analysis
The mandibles from WT and KO mice were isolated, and the soft tissue (around the bone and teeth) was removed. Specimens were fixed

| Western blotting (WB) analyses and ELISA
The mandibles from WT and KO mice were separated and ground in liquid nitrogen and then lysed with cell lysis buffer (Beyotime Biotechnology), and lysates were centrifuged at 10 000 g for 5 minutes at 4˚C. Total protein from the supernatant was quantified using an enhanced BCA protein assay kit (Beyotime Biotechnology). Western blotting analyses of protein expression were also performed as previously described. 4 The primary antibodies were rabbit anti-cleavage Caspase-1 monoclonal antibody (Abcam), and β-actin antibody (Santa Cruz Technology, Dallas, TX, USA) was used as an internal control. The Amersham enhanced chemiluminescence kit (GE Healthcare Life Sciences, Little Chalfont, UK) was used for detection. The Western blotting analyses were performed in triplicate.

F I G U R E 1 Construction and validation of the
ELISA, 5 ng total proteins were used for ELISA to detect cleaved IL-1β activity (Beyotime Biotechnology) according to the manufacturer's protocol.

| Statistical analysis
The data were analysed and compared using two-way ANOVA analysis and represented as the mean ± SD. Student's t test was used for the significant differences between two groups. Dunnett's test evaluated multiple comparisons when the F test result was significant. A value of P < 0.05 was considered to indicate a statistically significant difference. Asterisks were considered to indicate statistically significant differences (*P < 0.05; **P < 0.01; ***P < 0.001) from the control groups.

| Validation of MSX2 ablation in teeth of KO mice
Msx2 +/mice were generated by Cyagen Biosciences. Figure 1A illustrates the details of the gene targeting and genotyping strategy.
The region from exon 1 and exon 2 was deleted in the Msx2 gene.  Figure 1C). However, no positive staining for MSX2 was observed in ameloblasts and the enamel epithelium in KO littermates ( Figure 1D). Moreover, we further confirmed that the Msx2 expression at the mRNA level was significantly decreased in the mandible of KO mice ( Figure 1E). These results indicated that the Msx2 KO mouse model was successfully established.

| Ameloblast pyroptosis resulted in the inability of enamel to mineralize during the transition stage in KO mice
To further clarify the cause of enamel defects in KO mice, HE staining was used to observe the histological changes during enamel formation ( Figure 4). The experimental results in KO mice showed that a small amount of enamel was secreted in the early stage of enamel secretion, but no mineralization occurred in the mature stage, which led to severe enamel defects ( Figure 4D). Compared with WT mice, squamous epithelial hyperplasia of the enamel epithelium has appeared in the epithelial cell layer formed by OEE, SR and SI of KO mice at P7 ( Figure 4D). At the same time, ameloblasts also changed from polarization to depolarization, and enamel secretion stopped ( Figure 4D). CK5/6, a marker of squamous epithelial cells, was used to confirm that the enamel epithelium had squamous epithelial proliferation in the enamel organ of KO mice by immunohistochemistry ( Figure 5B). Squamous hyperplasia proliferation of the enamel epithelium was serious, and the keratinization phenomenon in squamous epithelial cells appeared near ameloblast in KO mice ( Figures 4E and 5B). At P15, keratinization of squamous epithelial cells further intensified ( Figure 4F).
In addition, the HE staining results indicated that ameloblasts and epithelium adjacent to ameloblasts in the enamel organ were dead at the mature stage, and a large number of inflammatory cells had accumulated ( Figure 4D-F). At P10, a large number of ameloblasts underwent pyroptosis in KO mice, which was further confirmed by TUNEL staining (Figure 6A). In the process of pyroptosis, the activation of cleavage Caspase-1 further promotes the maturation of IL-1β.
IL-1β is released to the outside of the cell, and then, IL-1β collects inflammatory cells and triggers inflammatory response. Both cleavage Caspase-1 and cleaved IL-1β were higher in the enamel organ of KO mice than that of WT ( Figure 6B), as determined by WB (Figure 6C) F I G U R E 5 Squamous epithelial proliferation of the enamel epithelium in the enamel organ of KO mice.CK5/6 expression analysis was performed on the first molars of WT mice (A) and KO mice (B) at P10 by immunohistochemistry. WT mice were used as control, and squamous hyperplasia proliferation of the enamel epithelium was significant, and the keratinization phenomenon in squamous epithelial cells appeared near ameloblast in KO mice (arrows, b)

A a B b
and ELISA ( Figure 6D) separately, indicating that pyroptosis occurred in the enamel organ of KO mice.
The above results showed that the loss of Msx2 could induce the keratinization of a large number of epithelial cells near ameloblasts, which caused ameloblasts to separate from the epithelial layers. We speculate that the pyroptosis of ameloblasts may be due to the obstruction of nutrient access. It is suggested that the loss of Msx2 could induce the proliferation and keratinization of squamous epithelial cells and cause pyroptosis of ameloblasts and the aggregation of inflammatory cells. Therefore, MSX2 is a necessary factor to maintain the state and function of epithelial cells.

| D ISCUSS I ON
MSX2, as a transcriptional repressor, may establish a balance between the survival and apoptosis of neural crest-derived cells which are required for proper craniofacial morphogenesis. 13,17 MSX2 may also play a role in inhibiting cell proliferation and promoting apoptosis. 18,19 Under physiological and pathological conditions, MSX2 plays an im- There are many reasons for enamel defects in Msx2 -/mice. For example, the loss of Msx2 may cause the abnormal secretion of enamel matrix protein, which then affects the formation of enamel. This is consistent with the known repressive transcriptional activity of MSX2 on enamel matrix protein genes such as enamelin, amelogenin and ameloblastin in enamel. 4,14 We found that although a small amount of enamel matrix protein was secreted in the early stage, no mineralized enamel was eventually formed, which indicated that the key points of enamel defects caused by the loss of Msx2 are the mineralization deficiency of enamel at the mature stage. On the other hand, in the Msx2 -/mice, it was also found that the epithelial cell layer adjacent to ameloblasts had advanced squamous epithelialization in the later stage of enamel formation, which was consistent with previous research results. 16 F I G U R E 6 Absence of Msx2 induced the pyroptosis of ameloblasts. A, At P10, the pyroptosis of ameloblast in KO mice was shown with TUNEL staining in green (arrows), while no obvious staining was detected in ameloblasts of WT mice. The nuclei were stained with DAPI in blue. B, As a marker of pyroptosis, both cleavage Caspase-1 and cleaved IL-1β expression were significantly increased in the enamel organ of KO mice compared with that of WT mice by immunohistochemistry (arrows, B, c, d). Cleavage Caspase-1 expression was determined by WB (C), and cleaved IL-1β activity was performed for the quantification by ELISA (D). *P < 0.05. n = 3 We found that the epithelial cells nearing ameloblasts appeared to be hyperplasic squamous cells at postnatal day 15 in WT mice; however, these cells became hyperplasic squamous cells at postnatal day 7 in Msx2 -/mice. In addition, the squamous epithelial cells had severe keratinization at postnatal day 10 in Msx2 -/mice. Keratinization resulted in the separation of ameloblasts from the epithelial cell layer, which led to a high degree of cell death, presumably due to a lack of nutrition, and then a large amount of inflammatory cell aggregation, which defines the pyroptosis phenomenon. At this time, HE staining results also showed that the ameloblasts displayed homogeneous red staining of unstructured material, and nuclear staining disappeared.
Both cleavage Caspase-1 and IL-1β were higher in the enamel organ of Msx2 -/mice than that of wild type. These observations are characteristic of pyroptosis. [24][25][26][27] Our data delineate a putative mechanism: the absence of Msx2

ACK N OWLED G EM ENTS
The present study was supported by the National Nature Science

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest relevant to this article. Writing-original draft (lead).

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sets used and analysed during the current study are available from the corresponding author on reasonable request.