Cell isolation and culture

This study was approved by the institutional review board of the Stomatological Hospital, Southern Medical University (201806). Healthy premolars and third molars were collected from patients aged 18–22 years for orthodontic reasons. Written informed consent was obtained from all participants. After extraction, the teeth were immediately placed in culture medium and kept at a low temperature for delivery to the laboratory for further processing. hDPSCs were isolated as previously reported [19]. Briefly, pulp tissue was extracted from the teeth, minced, and digested with collagenase type I (3 mg/mL; Invitrogen, Paisley, UK) and dispase (4 mg/mL; Invitrogen) for 1 h at 37 °C. After filtration with 70-mm strainers and centrifugation at 1000 rpm for 5 min, cells were resuspended and cultured in low-glucose Dulbecco’s modified Eagle’s medium (Gibco-BRL, Gaithersburg, MD, USA) supplemented with 10% foetal bovine serum (FBS; Gibco-BRL), 100 U/mL penicillin, and 100 mg/mL streptomycin (Invitrogen). The medium was refreshed every 2–3 days. Cells were passaged when they reached 70–80% confluency, and all experiments were performed at passages 3–5.

Tooth germs of mandibular first molars from prenatal and new-born C57BL/6 mice were dissected surgically under a stereomicroscope, followed by isolation of mouse tooth germ epithelial cells (mEpi) and mouse tooth germ mesenchymal cells (mMes) following the method described in our previous study [20]. Briefly, the tooth germs were washed with phosphate-buffered saline (PBS), and the enzymatic reaction was conducted with dispase for 10.5 min at 37 °C. Subsequently, the reaction was stopped, and the tooth germ epithelial and mesenchymal tissues were separated using a 25-gauge needle under a microscope. After filtration and centrifugation, some of the tooth germ epithelial cells (Epi) and Mes were collected for RNA extraction, and the rest were resuspended and separately cultured in medium. When tooth germ Epi reached 70–80% confluency, the medium was replaced with dermal cell basal medium (ATCC; Manassas, VA, USA) for purification, while tooth germ Mes was cultured in normal medium.

The ameloblast-derived cell line LS-8 was purchased from the State Key Laboratory of Military Stomatology, Department of Dental Technical Laboratory, School of Stomatology, Fourth Military Medical University (Xi'an, China). LS-8 cells were initially cultured in normal medium to induce differentiation, and when tooth germ Epi reached 70–80% confluency, the medium was replaced with induction medium, which contained 50 mg/mL ascorbic acid and 10 mM β-glycerophosphate, and the cells were further cultured for 2 weeks.

Co-culture system

Transwell permeable systems (Corning, New York, NY, USA) were used for the co-culture of Epi and Mes. LS-8 cells were seeded in the upper compartment, while mMes and hDPSCs were seeded separately in the lower compartment of the dish. After four days of co-culture, cells in the lower compartments were collected for subsequent experiments.

Short hairpin (sh)RNA lentivirus infection

The shRNA-targeting SCUBE3 lentivirus and shRNA negative control (NC) lentivirus were synthesized by Ribo Company (Guangzhou, China). For cell transfection, hDPSCs were seeded into six-well plates and transfected with lentiviral particles, including shSCUBE3#1–#3 and shNC. Untransfected cells were used as the blank control. After 72 h of transfection, the cells were collected to assess the efficiency of SCUBE3 silencing using quantitative reverse transcription polymerase chain reaction (RT-qPCR) and Western blotting. The shRNA sequences used in the present study are listed in Additional file 1: Table S1.

Migration assays

After starvation, 2 × 105 hDPSCs in medium with 1% FBS were plated in Transwell inserts (Corning), whereas the 10% FBS medium with rhSCUBE3 or SB431542 was added to the lower chambers. After 24 h of incubation, the migrated hDPSCs were fixed and stained using 0.5% crystal violet.

HDPSCs were seeded in a medium with rhSCUBE3 or SB431542. When they reached 90% confluency, hDPSCs were scratched using 200-μL pipette tips and photographed after 12 h.

Proliferation assays

HDPSCs (1 × 103) were plated into 96-well plates with rhSCUBE3 or SB431542 and measured using Cell Counting Kit-8 (CCK-8; Sigma-Aldrich, St Louis, MO, USA) at 450 nm absorbance on the indicated days. HDPSCs were treated with rhSCUBE3 or SB431542. When cells reached 60% confluency, a 5-ethynyl-2′-deoxyuridine (EdU) kit was applied per the manufacturer’s instructions.

Alkaline phosphatase (ALP) staining

Normal medium was replaced with osteogenic inductive medium (OIM, Cyagen Biosciences, Guangzhou, China) in all groups when hDPSCs reached 70–80% confluency. After 2 weeks of osteogenic induction, the cells were fixed with 4% polyoxymethylene for 15 min, washed thrice with PBS for 5 min, and stained with BCIP/NBT ALP substrate (Beyotime Biotechnology, Guangzhou, China).

Alizarin red (ARS) staining

ARS staining was conducted to analyse the mineralized nodules of the cells in different treatment groups. After 2 weeks of culturing in OIM, hDPSCs were fixed, washed, and stained with 0.5% ARS solution (Sigma-Aldrich) to visualize calcium deposition.

Culture of dentin–pulp-like organoids

Dentin–pulp-like organoids were cultured as per a previous method [21]. According to the manufacturer’s instructions, hDPSCs at a density of 5 × 106 cells/mL were mixed with Matrigel (Nitta Gelatin, Osaka, Japan) at a ratio of 1:1, placed onto a sheet of Parafilm (Bemis, Oshkosh, WI, USA), and incubated for 40 min until polymerization of matrices. The constructs were transferred and cultured in ultra-low adhesion petri dishes (Corning) for 10 days. The medium was then changed to OIM for different treatment groups. After 21 days of culturing, all organoids were harvested for further experiments.

Immunofluorescent staining

For immunofluorescence staining, cultured cells and organoids were fixed for 20 min, permeabilized with 0.1% Triton X-100 for 15 min, washed thrice with PBS for 5 min, and blocked with 10% bovine serum albumin for 1 h. Cells were incubated with the primary antibody anti-SCUBE3 (ab189955; Abcam, Cambridge, UK), and the organoids were incubated with anti-DSPP (sc-73632; Santa Cruz Biotechnology, Santa Cruz, CA, USA) overnight. Sections of tooth root fragments were stained with DSPP (sc-73632; Santa Cruz) and α-tubulin (sc-8035; Santa Cruz). After washing thrice with PBS for 5 min, cells and organoids were, respectively, incubated with Dylight 594 (35560; Thermo Fisher Scientific, Waltham, UK) and CoraLite 488-conjugated (CL488-66122; ProteinTech Group, Chicago, IL, USA) secondary antibodies for 1 h in the dark. DAPI (Sigma-Aldrich) was used to stain the nuclei. Finally, the cells were observed under a Leica DMI4000 B fluorescence microscope (Leica Imaging Systems, Cambridge, UK). Organoids were transferred to observation dishes specified for confocal laser microscopy and examined with a confocal microscope (Carl Zeiss, Göttingen, Germany).

RT-qPCR

Total cellular RNA was extracted from cells and tooth germ tissues using TRIzol reagent (Invitrogen), as per the manufacturer's instructions. Reverse transcription was performed with 1 mg of RNA using the PrimeScript RT reagent kit (Takara, Dalian, China). Quantitative PCR was conducted using the SYBR Premix Ex Taq II kit (Takara) on a CFX96 Touch Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). The expression of the genes of interest was normalized to that of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in the same samples. The reactions were performed in triplicate, and three independent experiments were performed. The primer sequences used are listed in Additional file 1: Table S2.

Western blot analysis

Total cell proteins were lysed using radioimmunoprecipitation assay (Beyotime) according to the manufacturer’s protocol. Proteins in the conditioned medium were extracted using a liquid sample total protein extraction kit (Solarbio, Beijing, China). Western blotting was performed as previously described [10]. The primary antibodies used were anti-SCUBE3 (ab189955; Abcam), anti-AMBN (orb155652; Biorbyt, Cambridge, UK), TGFβR1 (ab31013; Abcam), anti-DSPP (sc-73632; Santa Cruz), anti-DMP1 (ab103203; Abcam), anti-OPN (ab8448; Abcam), anti-OSX (ab13418; Abcam), anti-BMP2 (ab14933; Abcam), anti-BMPR1A (ab264043; Abcam), anti-p-SMAD1/5 (9516; Cell Signalling Technology, Boston, MA, USA), and anti-SMAD1 (D59D7; Cell Signalling Technology). Anti-GAPDH (Rayantibody, Beijing, China) was used as an internal control. Antibody binding was detected using an enhanced chemiluminescence kit (Millipore, Billerica, MA, USA). The intensity of the bands was quantified using the ImageJ software (NIH, USA).

Preparation of root fragments of human teeth

Decellularized tooth root fragments were prepared according to a previous report [21]. In brief, the tooth root fragments were sectioned into 5-mm-thick slices using a high-speed dental handpiece. The canals were enlarged to 3 mm in diameter using ProTaper files (Dentsply-Maillefer, Ballaigues, Switzerland) to remove pulp tissue and partial dentin, soaked in 5% ethylene diamine tetraacetic acid (EDTA) for 5 min, and ultrasonicated for 10 min, followed by sealing one of the open endings with mineral trioxide aggregate (Dentsply Sirona, Yorktown, VA, USA). Tooth root fragments were stored in PBS containing 2% penicillin/streptomycin (Invitrogen) at 4 °C and disinfected using UV sterilization before use. According to the manufacturer's instructions, hDPSCs at a density of 5 × 105 cells/mL were mixed with Matrigel (Nitta Gelatin) at a ratio of 1:1, injected into the canal cavity, and incubated at 37 °C for 30 min for the subsequent experiments.

Subcutaneous implantation into nude mice

Surgical procedures were performed as formerly reported [10]. In short, in 6-week-old immune-compromised nude mice (n = 12 in each group) which were purchased from Southern Medical University Laboratory Animal Science and Technology Development Co., Ltd (Guangzhou, CN), general anaesthesia was induced and maintained by intraperitoneal injection of 1.2 μL of 1% pentobarbital sodium (Merck, Darmstadt, Germany). A single hydrogel-hDPSC-filled fragment was transplanted subcutaneously into the dorsal side of each mouse. All nude mice were euthanized by cervical dislocation under anaesthesia four weeks post-surgery, and all fragments were harvested. This manuscript adheres to the ARRIVE guidelines for the reporting of animal experiment which improves the reporting of research involving animals.

Histological assessment

Tooth root fragments were fixed at 4 °C for 48 h and decalcified for 12 weeks using a 10% EDTA–2Na solution (pH 7.4) on an orbital shaker at room temperature. Embryonic day 12.5 (E12.5), E14.5, E16.5, E18.5, postnatal day 3 (P3), P7, and P14 mice were killed to collect mandible samples as reported previously [22]. Tooth root fragments and mandible samples were embedded in paraffin, sectioned at a thickness of 5 μm, and subjected to haematoxylin and eosin and Masson’s trichrome staining (Solarbio). Histomorphology was observed using the Scanscope CS and Image Scope software (Aperio, Sausalito, CA, USA). The ImageJ software was used for quantitative analysis.

Immunohistochemical staining

Immunohistochemical analysis was conducted according to a standard protocol. Briefly, samples were incubated with anti-SCUBE3 (ab189955; Abcam) and anti-mitochondria (ab92824; Abcam) as primary antibodies, followed by washing and incubation with horseradish peroxidase-conjugated secondary antibodies. Histometric observations were performed as described for the histological assessment.

Statistical analysis

Data were statistically analysed using SPSS software (version 20.0; IBM, Armonk, NY, USA) or GraphPad Prism 8 (GraphPad Software, Inc., La Jolla, CA, USA). Data are presented as mean ± standard deviation (SD) from at least three independent experiments. Differences among groups were compared using one-way analysis of variance (ANOVA) with t-tests. Statistical significance was set at P < 0.05.