Animal models

Epithelial Wnt10a knockout mice, K14-Cre;Wnt10afl/fl, were generated as previously described.18 The mice were housed under specific pathogen-free conditions and used for analysis regardless of sex. Newborn pups were recorded as postnatal stage PN0.5. Genotyping was conducted, on tail samples, using the One Step Mouse Genotyping Kit (Vazyme, Nanjing, China). Primer sequences are listed in Supplementary Table 1. All animal experiments were approved by the Ethics Committee of Peking University Health Science Center (LA2020191).

Histologic analysis

Mandibles of postnatal mice were dissected on day 0.5, 2.5, 3.5, and 4.5 (PN0.5, PN2.5, PN3.5, and PN4.5), and fixed in 4% paraformaldehyde (PFA) overnight at 4 °C. Subsequently, the samples were decalcified with 10% EDTA for 1 to 4 days, depending on mouse age in days. Decalcified mandibles were dehydrated and embedded in paraffin. The blocks were sectioned (8 μm) using a microtome and mounted on slides (CITOTEST, Nanjing, China) for H&E staining which was performed according to the standard protocol.41

Immunofluorescence assays

Paraffin-embedded sections were prepared as described above (Histologic analysis). The sections were then deparaffinized and rehydrated. Following antigen retrieval (ZSGB-BIO, Beijing, China), sections were blocked for 1 h at room temperature in blocking solution (ZSGB-BIO), and incubated with primary antibodies against E-Cadherin (1:100; ab76319, Abcam, Cambridge, UK), Acetyl-α-Tubulin (1:100; Cell Signaling Technology (CST), Danvers, USA), Cyclin D1 (1:100; CST), Notch2 (1:50; CST), or Jag1 (1:50; CST), overnight at 4 °C. After washing three times with 1x phosphate buffered saline (PSB), sections were then incubated with secondary antibodies (ZSGB-BIO) for 1 h at room temperature. DAPI (ZSGB-BIO) was used to stain cell nucleus. Images were taken (40x), using a confocal microscope (Leica SP8; Wetzlar, Germany).

EdU proliferation assays

For Keratin-14 (1:100; Abcam) and EdU (5 mg/kg; RiboBio, Guangzhou, China) dual immunofluorescence labeling, PN2.5 and PN3.5 mice (n = 5 per stage) were sacrificed 3 h after intraperitoneal EdU injection. Keratin-14 staining was performed as described above (Immunofluorescence assays). After incubated with secondary antibody, the EdU labeling was performed using the using Click-iT Apollo 567 Stain Kit (RiboBio), according to the manufacturer’s instructions and signals were detected using the Leica SP8.

TUNEL apoptosis analysis

HERS apoptotic cells, in the furcation region, were detected using the TUNEL BrightRed Apoptosis Detection Kit (Vazyme, A113), according to the manufacturer’s instructions.

Super-resolution microscopy imaging

To further observe the shape and long-axis orientation of the HERS epithelial cells, the mitotic spindle and the precise expression pattern of Notch2, Jag1, and Wnt10a, we photographed the HERS cells at super-resolution using an LSM880 with an Airyscan system (Zeiss, Jena, Germany).

For analysis of cell long axis orientation, we marked and traced the contour of HERS cells to define cell long axis according to the expression pattern of E-Cadherin. Then the angle between the cell long-axis and the horizontal plane was measured for each cell (0° ≤ α ≤ 90°). Radial histograms were generated to demonstrate the central tendency of the cell long-axis direction, using OriginPro (OriginLab Corporation, Northampton, USA). To analyze mitotic spindle orientation, we measured the angle between the long axis of the mitotic spindle and the horizontal plane.

RNA-Seq analysis and RT-qPCR validation of the dental epithelium

Three pairs of biological replicates (pooling dental epithelium from the five first mandibular molars as one sample) of the RNA samples were collected from PN2.5 K14-Cre;Wnt10afl/fl and Wnt10afl/fl mice. Samples were sent to Shanghai Majorbio Biopharm Technology Co., Ltd. (Shanghai, China) for mRNA extraction, cDNA library construction, and sequencing with an Illumina Novaseq 6000 platform (San Diego, USA). The raw sequence data was trimmed and quality assessed using SeqPrep (https://github.com/jstjohn/SeqPrep) and Sickle software (https://github.com/najoshi/sickle). Next, the acquired sequences were aligned to the reference genome using HISAT2 (http://ccb.jhu.edu/software/hisat2/index.shtml) software. The fragments per kilobase of exon per million mapped reads (FPKM) method was used to calculate transcript expression levels. RSEM (http://deweylab.biostat.wisc.edu/rsem) was used to quantify gene abundance. Genes differentially expressed between K14-Cre;Wnt10afl/fl and Wnt10afl/fl samples were detected based on the following criteria: P-value < 0.05, and the absolute difference was >20% of Wnt10afl/fl group. Heatmap and enrichment analyses of Gene Ontology and KEGG pathways were conducted using the Majorbio Cloud Platform (https://cloud.majorbio.com). The data are available at Gene Expression Omnibus GSE228411. To further confirm Notch signaling gene expression, quantitative RT-PCR was conducted, using the SYBR Green Master Mix (Vazyme) and a 7500 Real-Time PCR detection system (Applied Biosystems, Foster City, CA, USA). β-actin was used as the internal control to normalize expression. Primer sequences are listed in Supplementary Table 2.

Adenovirus infection and kidney capsule transplantation

A recombinant adenoviral vector (pAdEasy-EF1-MCS-CMV-EGFP), encoding the mouse intracellular domain of Notch2 (N2ICD), was used to overexpress Notch2 and activate Notch signaling. Recombinant adenovirus containing N2ICD (Ad-N2ICD-Over) and an empty adenovirus vector (Ad-Control) were constructed by Shanghai HanBio Technology. To verify the extent of adenovirus infection, the first mandibular molars (n = 3) of PN0.5 Wnt10afl/fl mice were dissected and infected with Ad-control and Ad-N2ICD-Over. The EGFP signals in the infected molars were detected using an inverted fluorescence microscope. For in vivo rescue experiments, kidney capsule transplantation was performed as previously described.42

Micro-CT analysis

After 21 days of growth in the kidney capsule, the collected mouse kidneys were fixed in 4% PFA overnight at room temperature. Then they were scanned using an Inveon MM system (Siemens, Munich, Germany) to observe the morphology of the root furcation. Images were obtained at 80 kV (voltage), 500 μA (current), 10 μm (pixel size), and 1500 ms (exposure time for each 360 rotational steps). Images were captured using the Inveon MM software.

Gene silencing and western blot analysis

LS8 cells were seeded in 6-well plates. The next day, after the cells were attached, the corresponding siRNA was transfected using the Ribo FECTTM CP Transfection KIT (RiboBio) according to the instructions. Mouse Wnt10a siRNA and negative control siRNA were generated by Tsingke Biotechnology (Beijing, China). The siRNA sequences are upon request.

The cell lysates were harvested and subjected to western blot analysis using primary antibodies including anti-Wnt10a (1:200; Abcam), anti-β-catenin (1:1 000, CST), anti-active-β-catenin (1:1 000, Abcam), anti-N2ICD (1:1 000, CST), anti-Jag1 (1:1 000, CST) or an anti-GAPDH (1:2 000, Abcam). The intensity of each band was visualized using ImageQuant LAS 500 (GE HealthCare) and semi-quantified using ImageJ software (National Institutes of Health). GAPDH was set as normalized control.

In situ mRNA hybridization of Wnt10a

The Wnt10a mRNA in situ hybridization was manually carried out by employing PinpoRNA multiplex Fluorescent RNA in-situ hybridization kit (GD Pinpoease Biotech Co.Ltd., Guangzhou, China). Briefly, the sections were deparaffinized and rehydrated, then the endogenous peroxidase was inhibited by Pre-A solution at room temperature. Then we used protease treatment to expose the target RNA molecular, and followed by probe hybridization for 2 h at 40 °C. Signals were amplified sequentially by reaction 1, 2 and 3 as described in manufacturer’s instructions. Finally, a tyramine fluorescent substrate was added and target RNA was labeled by green fluorescent. The signals were detected using an LSM880 with an Airyscan system.

Statistical analysis

Quantitative results are presented as mean ± SD values. Student’s t-test was employed to assess the statistical significance in this article. P < 0.05 (*), P < 0.01 (**), P < 0.001 (***) and P < 0.000 1 (****) was considered statistically significant. P > 0.05 was considered not significant.