Fig. 1 depicts a general view of the design of the present study. Noni (Morinda citrifolia) was obtained from Ngaraard (Ngiwal, Palau) from a plant >15 years. The noni fruit was ripened at room temperature (20-28˚C) at a humidity of 60-85% for 11-12 weeks. The noni liquid was freeze dried using a lyophilizer (Ilshin Lab Co., Ltd.) (14).

Gingival stem cells were acquired from a healthy patient receiving periodontal treatment, as previously described (13). The present study was approved by the Institutional Review Board of The Catholic University of Korea (approval nos. KC19SESI0234 and KC20SISE0696; approval date, 07 May 2019) and written informed consent was provided prior to the study start. The stem cells were plated at 2x10³ cells/well into 96-well plates and incubated in osteogenic media comprising α-MEM (Gibco; Thermo Fisher Scientific, Inc.), dexamethasone, glycerophosphate disodium salt hydrate, ascorbic acid 2-phosphate and L-glutamine in the presence of noni, with final concentrations of 0, 10, 100 and 200 ng/ml (G1, G2, G3 and G4), respectively. The morphological characteristics of the cells were evaluated using an inverted light microscope at x40 magnification on day 7.

Qualitative analysis of cellular viability was performed on days 1 and 7 via the commercially available two-color assay (Live/Dead Kit assay, Molecular Probes; Thermo Fisher Scientific, Inc.), which determines both the plasma membrane integrity and esterase activity (15). Quantitative analysis of viable cells was performed via the Cell Counting Kit-8 assay (Dojindo Laboratories, Inc.), where absorbance was measured at 450 nm, using a spectrophotometer (BioTek Instruments, Inc.). Cells were incubated at room temperature for 1 h.

Alkaline phosphatase activity was determined on days 7 and 14, after detaching cells using trypsin and Alkaline Phosphatase Activity Colorimetric Assay kit (cat. no. K412-500; BioVision Inc.), according to the manufacturer's instructions at 405 nm (16). Cells were washed, fixed and stained with 2% Alizarin Red S solution (cat. no. 0223; ScienCell Research Laboratories, Inc.) for 30 min at room temperature, and quantification of the bound dyes was performed by applying 10% cetylpyridinium chloride (cat. no. C0732; Sigma-Aldrich; Merck KGaA) for 15 min at 560 nm (17).

Total RNA was extracted from the gingiva-derived stem cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). RNA quality was assessed using Agilent 2100 bioanalyzer (Agilent Technologies, Inc.) and RNA quantification was performed using a spectrophotometer (ND-2000; Thermo Fisher Scientific, Inc.). Libraries were prepared from total RNA using the NEBNext Ultra II Directional RNA-Seq kit (New England BioLabs, Inc.). The isolated mRNAs were reverse transcribed into cDNA, according to the manufacturer's instructions (New England BioLabs, Inc.). Quantification was performed using the library quantification kit (18) and library concentration was measured using Tape Station HS D1000 Screen Tape (Agilent Technologies, Inc.). The library loading volume and loading concentration were 150 µl and 300 pM, respectively. Sequencing was performed using Illumina Novaseq 6000 (Illumina, Inc.). The NovaSeq 6000 S4 Reagent kit v1.5 (300 cycles; Illumina, Inc.) was used as the sequencing reagent kit. Running format was set at PE100bP. Quality control of raw sequencing data was performed and low quality reads (<Q20) were removed (19). Estimation of gene expression levels and normalization of the values were performed (19,20). Pathway analysis was performed on differentially expressed genes using the Kyoto Encyclopedia of Genes and Genomes mapping tool (21). Data mining and graphic visualization were performed using ExDEGA (Ebiogen, Inc.) (22).

Total RNA was extracted from the gingiva-derived stem cells using a commercially available kit (Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions (17). The quality of RNA was assessed using a bioanalyzer, and RNA quantity was evaluated using a spectrophotometer (ND-2000; Thermo Fisher Scientific, Inc.). RNA was reverse transcribed into cDNA using reverse transcriptase (SuperScript II; Invitrogen; Thermo Fisher Scientific, Inc.).

mRNA expression levels were detected via qPCR analysis on days 3 and 7. GenBank was used to design the sense and antisense primers for PCR. The following primer sequences were used: RUNX2 (accession no. NM_001015051.3; forward, 5'-CAGTTCCCAAGCATTTCATCC-3' and reverse, 5'-AGG TGGCTGGATAGTGCATT-3'), BSP (accession no. NM_004967.4; forward, 5'-CCTCTCCAAATGGTGGGTTT-3' and reverse, 5'-ATTCAACGGTGGTGGTTTTC-3'), OCN (accession no. NM_199173.6; forward, 5'-GGTGCAGAGTC CAGCAAAGG-3' and reverse, 5'-GCGCCTGGGTCTCTTCA CTA-3'), COL1A1 (accession no. NM_000088.4; forward, 5'-TACCCCACTCAGCCCAGTGT-3' and reverse, 5'-CCGAAC CAGACATGCCTCTT-3'); and β-actin (accession. no. NM 001101; forward, 5'-AATGCTTCTAGGCGGACTATGA-3' and reverse, 5'-TTTCTGCGCAAGTTAGGTTTT-3'). DNA synthesis and PCR amplification were performed using a qPCR detection system (StepOnePlus Real-Time PCR System, Applied Biosystems; Thermo Fisher Scientific, Inc.). qPCR was performed using 2X Power SYBR-Green Master Mix (Applied Biosystems; Thermo Fisher Scientific, Inc.). The following thermocycling conditions were used: 95˚C for 15 min, followed by 40 cycles of 95˚C for 15 sec and Tm at 59˚C for 30 sec. Relative expression levels were calculated using the 2^-ΔΔCq method (23) and normalized to the internal reference gene β-actin.

Statistical analysis was performed using SPSS 12 software (SPSS, Inc.). All experiments were performed in triplicate and data are presented as the mean ± standard deviation. The Shapiro-Wilk test was used to assess normality. One-way ANOVA followed by Tukey's post hoc test was used to compare differences between multiple groups. P<0.05 was considered to indicate a statistically significant difference.

The structures of cells grown in osteogenic media on day 7 are presented in Fig. 2. Treatment with noni at 10, 100 and 200 ng/ml did not yield noticeable morphological changes. The qualitative viability of stem cells was analyzed via a Live/Dead kit assay on days 1 and 7 (Fig. 3A and B). In all cases, most stem cells exhibited a fibroblast-like shape with intense green fluorescence, suggesting the presence of live cells on day 1 (Fig. 3A). Longer incubation of cells on day 7 did not exhibit noticeable decrease in green fluorescence (Fig. 3B).

Quantitative cellular viability on days 1 and 7 are presented in Fig. 3C. The absorbance values at 450 nm on day 1 were 0.636±0.074, 1.006±0.250, 1.178±0.275 and 1.002±0.257 for noni at 0, 10, 100 and 200 ng/ml, respectively (P>0.05). On day 7, the absorbance values were 0.562±0.093, 1.331±0.150, 1.811±0.286 and 1.652±0.155, for noni at 0, 10, 100 and 200 ng/ml, respectively (P<0.05). Treatment with noni produced significantly higher values in cellular viability compared with the 0 ng/ml group (P<0.05), with the highest value at 100 ng/l.

Significantly higher values for alkaline phosphatase activity were observed in the 10 and 100 ng/ml groups on day 7, with the highest value at 100 ng/ml (P<0.05; Fig. 4).

Fig. 5A shows the results for Alizarin Red S staining. Calcium deposits were clearly observed in each group on day 7. The absorbance values on day 3 were 0.123±0.006, 0.144±0.006, 0.153±0.007 and 0.147±0.006 for 0, 10, 100 and 200 ng/ml, respectively (Fig. 5B). Treatment with noni produced significantly higher values in Alizarin Red S staining, with the highest value from 100 ng/ml (P<0.05).

Differentially expressed mRNAs in the 10 ng/ml group vs. the control group are presented in Table SI (fold change >1.5 and log₃ normalized read counts >3 were selected). Comparison of the 10 ng/ml group vs. the control group revealed that 22 mRNAs were upregulated and 13 were downregulated. Clustering analysis of the differentially expressed mRNAs is presented in Fig. 6A. Bar charts depicting the top Gene Ontology terms are presented in Fig. 6B. The most highly associated pathway was ‘regulation of cell proliferation’.

Differentially expressed mRNAs in the 100 ng/ml group vs. the control group are presented in Table SII (fold change >1.5 and log₂ normalized read counts >3 were selected). Comparison of the 100 ng/ml group vs. the control group revealed that 24 mRNAs were upregulated and 25 were downregulated. Clustering analysis of the differentially expressed mRNAs is presented in Fig. 7A. Bar charts depicting the top gene ontology terms are presented in Fig. 7B. ‘Collagen catabolic pathway’ was associated with mRNA expression.

Differentially expressed mRNAs in the 200 ng/ml group vs. the control group are presented in Table SIII (fold change >1.5 and log₂ normalized read counts >3 were selected). Comparison of the 200 ng/ml group vs. the control group revealed that 23 mRNAs were upregulated and 25 were downregulated. Clustering analysis of the differentially expressed mRNAs is presented in Fig. 8A. Bar charts depicting the top gene ontology terms are presented in Fig. 8B. The most highly associated pathways were ‘mRNA splication’ and ‘translation’. The signaling pathway associated with the target genes was transforming growth factor-β (TGF-β) pathway (Fig. 9).

Changes in the expression levels of bone morphogenetic protein 8b (BMP8B), parathyroid hormone like hormone (PTHLH), RUNX2 and catenin β-1 (CTNNB1) are presented in Fig. 10A. The results demonstrated that BMP8B expression was higher in the 10 and 100 ng/ml noni group compared with the control group (Fig. 10B). Furthermore, RUNX2 expression was higher in the 10 and 200 ng/ml noni groups compared with the control group. Similarly, CTNNB1 expression was higher in the 10 and 200 ng/ml noni groups compared with the control group. PTHLH expression was higher in the 100 and 200 ng/ml noni groups compared with the control group.

qPCR analysis demonstrated that RUNX2 mRNA expression levels on day 3 were 1.000±0.027, 1.123±0.060, 0.587±0.021 and 0.670±0.022 for noni at 0, 10, 100 and 200 ng/ml, respectively (P<0.05; Fig. 11A). RUNX2 mRNA expression levels on day 7 were 1.001±0.050, 0.956±0.085, 1.035±0.073 and 1.339±0.092 for noni at 0, 10, 100 and 200 ng/ml, respectively (P<0.05). Notably, treatment with 200 ng/ml of noni significantly increased RUNX2 expression at day 7. Furthermore, BSP mRNA expression levels on day 3 were 1.015±0.204, 1.434±0.370, 0.812±0.050 and 0.365±0.049, respectively (P<0.05) (Fig. 11B). BSP mRNA expression levels on day 7 were 1.001±0.050, 0.107±0.009, 2.391±0.172 and 0.852±0.033 for noni at 0, 10, 100 and 200 ng/ml, respectively (P<0.05). Notably, treatment with 100 ng/ml of noni significantly increased BSP expression at day 7. OCN mRNA expression levels on day 3 were 1.001±0.043, 1.773±0.146, 0.759±0.043 and 1.924±0.087 for noni at 0, 10, 100 and 200 ng/ml, respectively (P<0.05; Fig. 11C). OCN mRNA expression levels on day 7 were 1.001±0.051, 1.162±0.022, 1.042±0.069 and 1.607±0.059 for noni at 0, 10, 100 and 200 ng/ml, respectively (P<0.05). Notably, treatment with 10 (day 3) and 200 ng/ml (days 3 and 7) of noni significantly increased OCN expression. COL1A1 mRNA expression levels on day 3 were 1.000±0.011, 1.078±0.038, 0.911±0.050 and 0.951±0.011 for noni at 0, 10, 100 and 200 ng/ml, respectively (P<0.05) (Fig. 11D). COL1A1 mRNA expression levels on day 7 were 1.001±0.048, 0.735±0.050, 1.275±0.063 and 1.259±0.122 at 0, 10, 100 and 200 ng/ml, respectively (P<0.05). Notably, treatment with 100 and 200 ng/ml of noni significantly increased OCN expression at day 7.