PER2 regulates odontoblastic differentiation of dental papilla cells in vitro via intracellular ATP content and reactive oxygen species levels

Immunohistochemistry

The present study was approved by the Institutional Animal Care and Use Committee, Sun Yat-sen University, China (No. SYSU-IACUC-2020-000511) and the Ethical Review Committee, Hospital of Stomatology, Sun Yat-Sen University, China (ERC-2013-15). This study adhered to the Ethical Principles of Animal Experimentation.

Normal-growth Sprague-Dawley (SD) rats, 3–5 days postnatal, provided by Sun Yat-sen University, Guangzhou, China, and routinely housed without experimental intervention, were euthanized by neck dislocation after CO2 over-inhalation. Their mandibles were isolated. Fix mandibles in 4% paraformaldehyde, followed by decalcification with 0.5 M EDTA. Tissue was dehydrated, paraffin-embedded, and serially sectioned at 4 µm. Next, sections were deparaffinized and rehydrated. The slices were immersed in Tris-EDTA heated at 70 °C (5 min), cooled (5 min), heated at 40 °C (10 min), followed by cooling and then incubation in 0.4% pepsin at 37 °C (30 min) for antigen retrieval. After incubating in 3% hydrogen peroxide solution (25 min protected from light), block the slices in 3% bovine serum albumin (BSA). Next, slices were incubated with anti-PER2 antibody (1:200) overnight at 4 °C, as well as in horseradish peroxidase-conjugated (HRP-conjugated) Goat Anti-Rabbit IgG (H+L) for 50 min. Sections were visualized via diaminobenzidine in darkness and counterstained with hematoxylin. Sections were visualized by an Aperio AT2 digital pathology scanner (Leica, Weztlar, Germany). Semi-quantitative analysis was performed using the ImageJ 1.51k (NIH, MD, USA) to assess mean optical density. Reagents not specified in this section were purchased from Servicebio, Wuhan, China.

Cell culture and differentiation

Postnatal 1–2-day SD rats from the above sources were euthanized using the method described above. Isolate their dental papilla tissues under stereomicroscope, as described previously (Zhang et al., 2018). Dental papilla tissues were microscopically cut and digested with 3 mg/mL of collagenase type I + 4 mg/mL of Dispase® II (Sigma-Aldrich, St. Louis, MO, USA) for 15 min to produce tissue fragments and cell suspensions of primary cells, which were inoculated in 10-cm culture dishes maintained in α-minimal essential medium (α-MEM) comprising 20% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin at 37 °C in 5% CO2 humidified air. Passage DPCs by TrypLE when they reached 90% confluence and purified by the difference digestion method. After passaging, DPCs were maintained in basal medium containing 10% FBS. Passages 2-4 were applied to the subsequent experiments. FBS, α-MEM, penicillin, streptomycin and TrypLE were products of Gibco, Grand Island, NY, USA.

For differentiation of DPCs, DPCs were grown in osteo/odontogenic induction medium (OS; α-MEM supplemented with 10% FBS, 10 mM β-glycerophosphate, 0.2 mM ascorbic acid and 0.1 mM dexamethasone; Sigma-Aldrich, St. Louis, MO, USA).

For all subsequent cell experiments, sample collection and detection were performed at the same time points by the same personnel.

Immunofluorescence

DPCs were seeded in 35 mm glass-bottom culture dishes. Fix DPCs with 4% paraformaldehyde for 15 min, and permeabilized using 0.5% Triton X-100 for 15 min. 5% BSA was applied to block DPCs. Cells were incubated with anti-cytokeratin antibody (1:100) or anti-vimentin antibody (1:100) or anti-PER2 antibody (1:100) overnight at 4 °C. Then, DyLight 488 or DyLight 594 conjugated secondary antibody was used to incubate cells for 30 min without light. If cell morphology was required, incubate DPCs with Actin-Tracker Green (Beyotime, Shanghai, China). Finally, DAPI was used to incubate cells for 5 min protected from light to stain cell nuclei. Cells were preserved in PBS and visualized by confocal laser scanning microscope (FV3000; Olympus, Tokyo, Japan).

Flow cytometry

Digest DPCs using TrypLE and prepare the single cell suspension with PBS comprising 2% FBS. DPC single cell suspensions were incubated with flow cytometry CD45, CD34, CD44, CD90, CD29 antibodies and their isotype controls for 30 min at 4 °C (Elabscience, Wuhan, China). LSRFortessa flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA) was for measurement of fluorescence intensity of cell samples.

Cells transfection

To overexpress Per2, DPCs were seeded at 1.2 × 10⁵ cells/well in 12-well plates and maintained in basal medium without antibiotics. Within 24 h, cells reached about 80% confluence for transfection. The transfection protocol for each well was as follows: Opti-MEM™ (62.5 µL) diluted with 1.25 µg of plasmid and 2.5 µL of P3000™ reagent was used. Plasmids included pcDNA3.1-PER2 plasmid or negative control pcDNA3.1-NC plasmid. Diluted DNA was added 1:1 to each tube of diluted Lipofectamine™ 3000 and incubated for 10–15 min. The DNA-lipid complex was added to DPCs. The medium (without antibiotics) was replaced after 4–6 h. The culture medium was changed to conditioned medium (OS or basal medium) without antibiotics after 24 h.

To knock down Per2, DPCs were seeded at 7 × 10⁴/well in 12-well plates. Small interfering RNAs (siRNAs) targeting Rattus Norvegicus Per2 were transfected into cells using a protocol similar to the overexpression protocol, but without the addition of P3000™ reagent. Final siRNA concentration was 50 nM. Opti-MEM™, P3000™ and Lipofectamine™ 3000 were supplied by Invitrogen (Carlsbad, CA, USA). Plasmids and siRNAs were provided by MHBIO (Guangzhou, China).

RT-qPCR

Real-time quantitative polymerase chain reaction (RT-qPCR) was used for assessment of DPC gene expression. Total RNA was extracted via an RNA extraction kit from Yishan Biotechnology, Shanghai, China. Qualified RNA samples were reverse transcribed into cDNA. After proportionate addition of SYBR Green, primers (listed in Table 1), and RNA samples, RT-qPCR was conducted by Roche LightCycler® 96 System. The reverse transcription reagents and SYBR Green were purchased from Yeasen (Shanghai, China). The amplification program was set up according to the procedures recommended in the instructions. Glyceraldehyde-3-phosphate dehydrogenase (GADPH) acted as an internal reference for normalization. Calculate relative expression of target genes according to the 2^−ΔΔCt method.

Western blotting

Total protein was extracted from DPCs using radioimmunoprecipitation assay (RIPA) lysis buffer containing phenylmethylsulfonyl fluoride (PMSF). The extracted proteins were quantified via a bicinchoninic acid (BCA) protein assay kit. RIPA, PMSF and BCA kit were purchased from Cwbio, Beijing, China. Equal amounts of protein lysates were dispersed by SDS-PAGE and transferred to PVDF membranes. The membranes were blocked in skim milk and incubated overnight at 4 °C with the following antibodies: anti-PER2 antibody (1:2000), anti-DSPP antibody (1:500), anti-DMP1 antibody (1:1000), and anti- β-actin (1:1000). Next, the membranes were incubated in the HRP-conjugated secondary antibodies corresponding to the primary antibody species (1:2000) for 45 min. Immunoreactive bands were detected using enhanced chemiluminescence (ECL; Millipore, Billerica, MA, USA), captured on the Bio Rad ChemiDoc chemiluminescence imaging system, and quantified using ImageJ 1.51k (NIH, MD, USA).

Quantitative alizarin red staining

For fixation, cells were treated with 4% paraformaldehyde for 30 min. Next, use Alizarin Red S Solution (ALIR-10001, Cyagen Biosciences, Santa Clara, CA, USA) to stain cells for 5 min, followed by rinsing in double distilled water. An Epson professional-grade scanner was used to capture the stained overview and an Olympus IX83 inverted microscope was applied for capturing mineralized nodules. For semi-quantification of mineralized nodules, 0.1 M hexadecylpyridinium chloride monohydrate was supplemented and shaken for 30 min, and the supernatant was collected to determine the absorbance at 562 nm with a Biotek Epoch 2 Microplate Spectrophotometer.

ROS assay

After preparation into single cell suspensions, DPCs were treated with CellROX® Green Reagent (Invitrogen, Carlsbad, CA, USA) for 1 h at 37 °C, followed by rinsing in PBS. A BD LSRFortessa flow cytometer was applied for measurement of fluorescence intensity of cell samples.

Intracellular ATP content assay

Use an ATP assay kit (S0026; Beyotime, Shanghai, China) for measurement of intracellular ATP content. Lysed cell samples or standards were mixed with ATP assay reagents and added to an all-black 96-well plate for chemiluminescence detection by BioTek Synergy H1 multimode microplate reader.

Statistical analysis

Statistical analyses were undertaken with GraphPad Prism 8.0.2 (GraphPad, San Diego, CA, USA). All experimental data are provided as mean ± standard deviation (SD), derived from at least three independent experiments. A two-tailed Student’s t-test was applied for two-group comparisons. Multiple group comparisons were assessed by one-way analysis of variance. Fisher’s least significant difference test was used for post-hoc analysis. P <0.05 was considered statistically significant.

PER2 elevated during odontoblastic differentiation of rat DPCs in vivo

Immunohistochemistry was carried out on the incisors of Sprague-Dawley rats on postnatal day 5 (Fig. 1A) to assess whether PER2 has the potential to modulate DPC differentiation and dentin development in vivo. As shown in Fig. 1B, a vertical section of the incisor, DPCs close to the enamel organ gradually elongated, polarized and differentiated into odontoblasts. The differentiation of DPCs can be divided into four stages, and the typical morphologies of the cells at each stage are shown in Figs. 1C–1F. In the undifferentiated DPCs (Fig. 1C), little PER2 expression can be observed. As DPCs gradually differentiated into preodontoblasts (Fig. 1D) and immature odontoblasts (Fig. 1E), PER2 expression levels gradually increased. By the time the DPCs had fully differentiated into mature odontoblasts (Fig. 1F), the expression of PER2 was at its highest level. Semi-quantitative analysis revealed that PER2 expression was progressively upregulated as DPCs differentiated in vivo. Therefore, it is highly probable that PER2 modulates DPC differentiation.

Expression and intracellular localization of PER2 in DPCs in vitro

After identifying the potential role of PER2 in DPC differentiation in vivo, we performed in vitro experiments to further assess the regulatory functions of PER2 in DPC differentiation. DPCs were extracted and cultured in vitro according to previously described protocols (Zhang et al., 2018). After 3 days, primary DPCs migrated from pieces of dental papilla tissue (Fig. 2A) and were successfully amplified. DPCs were homogeneous, large polygonal fibroblast-like cells with abundant cytoplasm and centered nuclei (Fig. 2B). Immunofluorescence illustrated that the DPCs were strongly expressed vimentin (MSCs marker, Fig. 2C) but hardly expressed cytokeratin (epithelial cells marker, Fig. 2D). Flow cytometry showed DPCs positively expressing mesenchymal markers CD44 (92.1%), CD90 (94.8%), and CD29 (92.9%), but negatively expressing hematopoietic markers CD45 (3.15%) and CD34 (3.62%) (Figs. 2E–2F).

To determine the levels and intracellular localization of PER2 in DPCs in vitro, cellular immunofluorescence assays were performed (Fig. 3). Nuclei were stained by DAPI (Fig. 3A), while the outlines of the cells were stained by Actin-Tracker (Fig. 3B). PER2 (red fluorescence) was strongly expressed in the nucleus and perinuclear cytoplasm (Figs. 3C–3D).

Overexpression of Per2 enhances odontoblastic differentiation of DPCs in vitro

In vitro, Per2 was overexpressed in DPCs and relevant differentiation markers were examined to identify the regulatory role of PER2 in odontogenic differentiation of DPCs (Fig. 4). PcDNA3.1-PER2 (oe-PER2) was transfected to overexpress Per2, as well as pcDNA3.1-NC (oe-NC) for negative control. Overexpression efficiency of Per2 was measured via RT-qPCR and western blotting at 48 and 72-hours post-transfection, respectively (Fig. 4A). MRNA and protein levels of PER2 remarkably elevated in oe-PER2 in comparison to oe-NC, confirming the success of Per2 overexpression in DPCs (Fig. 4A).

After Per2 was successfully overexpressed, the levels of indicators associated with odontogenic differentiation were determined to ascertain the role of PER2 in DPC differentiation in vitro. After inducing odontogenic differentiation for 3 days, RT-qPCR showed that odontogenic differentiation markers, including Dspp, Dmp1 and Alp, upregulated in OS group compared to the control group (Figs. 4B–4D). In OS-oe-PER2 group, Dspp, Dmp1 and Alp dramatically elevated compared to OS-oe-NC group, indicating that overexpression of Per2 resulted in significantly enhanced expression of differentiation markers of DPCs (Figs. 4B–4D). Similar results were obtained using western blotting. When protein levels were measured over seven days of odontogenic induction, DSPP and DMP1 went up significantly by 2-4 times in the OS group compared to Control group; the OS-oe-PER2 group revealed significantly higher levels of DSPP and DMP1 than the OS-oe-NC group, showing an approximately 2-fold increase (Figs. 4E–4G). Mineralized nodules were noted in OS group, but not in the Control group by Alizarin red staining (Fig. 4H). More mineralized nodules were observed in the OS-oe-PER2 group than in OS-oe-NC group, and semi-quantitative analysis also revealed that mineralized nodule formation was significantly enhanced in the OS-oe-PER2 group in comparison to OS-oe-NC group (Fig. 4H). The aforementioned results illustrate that overexpression of Per2 promotes odontogenic differentiation of DPCs.

Knockdown of Per2 attenuates odontoblastic differentiation of DPCs in vitro

We knocked down Per2 in DPCs and examined differentiation-related markers to further identify the contribution of PER2 to DPC differentiation (Fig. 5). Three different Per2-specific siRNAs (si-PER2-1,2,3) were transfected to knockdown Per2, as well as si-NC for negative control. Knockdown efficiency of Per2 was measured via RT-qPCR and western blotting at 48 and 72-hours post-transfection, respectively (Fig. 5A). Knockdown efficiency of si-PER2-2 was the highest among the three si-RNAs, for both mRNA and protein levels (>70% efficiency at the mRNA level, Fig. 5A). Therefore, si-PER2-2 was selected for subsequent assays and denoted as si-PER2 (Fig. 5).

After Per2 was successfully knocked down, levels of differentiation-related markers were monitored to determine the role of PER2 on odontoblastic differentiation in DPCs in vitro. Three days after induction of odontogenic differentiation, RT-qPCR showed that the odontogenic differentiation markers (Dspp, Dmp1 and Alp) dramatically upregulated in OS-si-NC group compared to Control-si-NC group (Figs. 5B–5D). Nevertheless, the aforementioned markers were significantly downregulated in the OS-si-PER2 group compared to the OS-si-NC group, with statistical significance (Figs. 5B–5D). Consistently, western blotting illustrated that DSPP and DMP1 considerably increased in the OS-si-NC group in comparison to the Control-si-NC group; nevertheless, a statistically significant decline in DSPP and DMP1 was observed in the OS-si-PER2 group compared to the OS-si-NC group (Figs. 5E–5G). Alizarin red staining indicated that mineralized nodules were present in the OS group after a 7-day-induced differentiation, whereas they were absent in the Control group (Fig. 5H). Reduced mineralized nodules were found in the OS-si-PER2 group compared with the OS-si-NC group, and semi-quantitative assays revealed a significant decrease in mineralized nodules in the OS-si-PER2 group compared to the OS-si-NC group (Fig. 5H). The aforementioned results indicated that Per2 knockdown attenuated odontogenic differentiation of DPCs.

Knockdown of Per2 affects intracellular ATP content and ROS levels during DPCs differentiation in vitro

ROS levels and intracellular ATP content were examined in DPCs with Per2 knockdown and odontogenic induction for 3/7 days to determine if the pro-differentiation effect of PER2 is associated with altered mitochondrial function (Fig. 6). The ROS levels of each group were determined using flow cytometry (Fig. 6A). Median fluorescence values for the Control-si-NC group are marked with a dashed line. Apparently, the median and mean fluorescence values of the OS-si-NC group is reduced compared to the Control-si-NC group, whereas the median and mean fluorescence of the OS-si-PER2 group is higher than that of the OS-si-NC group. As shown in Fig. 6B, ROS levels in the OS-si-NC group were reduced compared to the Control si-NC group; however, ROS levels in the OS-si-PER2 group were elevated compared with the OS-si-NC group with statistical significance. These results suggested that ROS levels were downregulated in DPCs during differentiation. However, Per2 knockdown may lead to an increase in ROS in DPCs during differentiation, which may affect differentiation. Intracellular ATP content of each group was also assessed (Fig. 6C). ATP content of the OS-si-NC group was visibly raised compared to the Control-si-NC group, whereas it was considerably declined in OS-si-PER2 group compared to OS-si-NC group (Fig. 6C). This suggests that intracellular ATP levels are upregulated in differentiated DPCs compared to undifferentiated DPCs, but knockdown of Per2 may downregulate the ATP content of DPCs during differentiation, thereby suppressing differentiation.