Six male C57BL/6 mice (aged 6 weeks, 16–20 g; Slac, Shanghai, SCXK 2007-0005) were sacrificed for BMMSCs isolation. BMMSCs were isolated according to the protocol described by our previous study (Mei et al., 2015). Adherent primary BMMSCs were maintained in complete Dulbecco’s Modified Eagle’s medium with L-glutamine (DMEM-L) supplementing with 20% fetal bovine serum (FBS) (Gibco, EI Paso, Texas, USA) in an incubator at 37°C with 5% CO₂. The culture medium was changed every 48 h. Cells were used after passage 3. Once the cells were approximately 70–80% confluent, they were collected and seeded into a 6-well plate at 2 × 10⁵/well. After 3 days in sub-culture, cellular surface markers were identified with immunocytochemical staining using rabbit anti-mouse CD34, CD44, CD45, CD90, and CD105 primary antibodies (Abcam, Cambridge, United Kingdom) and secondary FITC-labeled goat anti-rabbit antibody (Chemicon; Temecula, CA, USA). BMMSCs were identified by positive CD44, CD90, and CD105 and negative CD34 and CD45 staining.

Astragaloside IV (AIV) (purity >98%) was obtained from Chengdu Desite Pharmaceutical Technology Co., Ltd., Chengdu, China. And it was dissolved in dimethyl sulfoxide (DMSO) (less than 0.1% in the final working concentration).

BMMSCs (1 × 10⁶/mL; passage 6) were seeded on 25 cm² cell culture plates (4 × 10⁴ cells/cm²) and cultured in Mcs differentiation medium containing Minimum Essential Medium Alpha Medium (MEM; Invitrogen, Carlsbad, CA, USA), 10% FBS and supplemented with 1.5 μg/mL hydrocortisone (Sigma-Aldrich, St. Louis, MO, USA), 10 μg/mL insulin (Sigma-Aldrich), 10 μg/mL transferrin (Invitrogen), 4 ng/mL basic fibroblast growth factor (FGF; Invitrogen) as a previous study described (Mei et al., 2015). The medium was changed every two days until the adherent cells reaching 90% confluent. The morphological characteristics of both primary and differentiated cells were observed under an inverted microscope (Leica, Wetzlar, Germany). In a preliminary experiment, we found that the induced BMMSCs demonstrated a lower survival rate in the culture media containing greater than 0.4 mg/L AIV (Suppl. Fig. S1). Thus, AIV was dissolved in Mcs differentiation medium at 0.4 mg/L.

The induced BMMCSs (Mcs) were stained with mixed solution, including 25 mL of 0.2% DOPA solution (Sigma-Aldrich), 6 mL of 11 mg/mL Na₂HPO₄ • 2H₂O, and 2 mL of 9 mg/mL KH₂PO₄ for 4 h at 37°C. Then, Mcs were washed three times with phosphate-buffered saline (PBS) and rinsed with water. First passage BMMSCs were used as negative controls, and B16 melanoma cells (ATCC, Manassas, VA, USA) were used as positive controls.

BMMSCs (passage 3) and differentiated cells (2 × 10⁵/well) cultured in a 6-well plate were fixed with 4% paraformaldehyde in PBS for 10 min at 4°C, washed with PBS, and blocked in buffer containing 5% bovine serum albumin (Gibco) for 30 min at room temperature. The BMMSCs were incubated with rabbit anti-mouse CD34, CD44, CD45, CD90, CD105 antibodies (all at 1:50, Abcam), and rat anti-mouse TRP-1, MART-1 (1:100, Abcam), rat anti-mouse Mitf (1:150, Abcam) and rabbit anti-mouse TRP-2 (1:200, Abcam) antibodies overnight at 4°C in 1% bovine serum albumin (BSA)/PBS, in a dark chamber. After two PBS washes, the cells were incubated with the appropriate secondary antibodies (1:100) for 2 h. Nuclei were visualized by staining with 4,6-diamidino-2-phenylindole (DAPI; Sigma). Immunocytochemical staining was evaluated according to a previously established method (Suleiman et al., 2019).

Total RNAs from B16 melanoma cells, induced cells, and BMMSCs (passage 6) were extracted with Trizol reagent (Invitrogen) according to the manufacturer’s instructions. The RNA (1 μg) was reversely transcribed into cDNA with M-MLV (Takara, Tokyo, Japan) and random primers in a final reaction volume of 20 μL. qPCR assay was performed using Sybergreen Master Mix (Takara). The primers for the qPCR are listed in Tab. 1. The thermal cycles of qPCR reaction were listed as followings: 95°C for 30 s, 45 cycles at 95°C for 5 s, 60°C for 30 s, 95°C for 1 min, and 79 cycles at 55–95°C for 4 s. The relative transcriptions of the targeting genes (TRP-1, TRP-2, MART-1, and Mitf) were quantified using the FTC3000HT Real-Time PCR System (Funglyn Biotech, Richmond Hill, Ontario, Canada). Each sample was analyzed at least for three repeats, and mouse ACTB (β-actin) was used as an internal reference gene for semi-quantitative analysis.

Data were represented as mean ± standard deviation (SD) and analyzed using SPSS software (version: 20.0, IBM Corporation, Armonk, NY, USA). qPCR analysis was normalized to β-actin. Gene expression was analyzed by Student’s t-test. Statistical significance was considered at P-values < 0.05 (two-sided).

The isolated BMMSCs were identified using the cellular immunostaining method (Fig. 1A). The results showed that the BMMSCs demonstrated CD44, CD90, CD105 positive staining and CD34, CD45 negative staining (Fig. 1B). Therefore, the isolated cells were characterized as BMMSCs.

During the initial induction period with 0.4 mg/L AIV-containing differentiation buffer, primary BMMSCs were adherent and round in shape and were mixed with some other suspending cells (Fig. 2A). The adherent cells presented as clonal expansion. At culture passage 3 (20-day induction), the cells were approximately 80–90% confluent and tended to gather in a monolayer on the bottom, which displayed a spindle-like shape with strong light refraction (Fig. 2B). At 70-day induction, the cells still displayed fibroblast morphology with a spindle-like shape, large cell body, and abundant cell nuclei (Fig. 2C). In this phase, the cells proliferated rapidly, with a passage cycle of approximately 4–5 days. At 90-day induction, the proliferation of the differentiated cells gradually slowed, and the passage cycle extended to 7 days. The differentiated cells formed networks, which displayed a polygonal shape with distinct outlines and prolonged dendrites, typical morphology of Mcs (Fig. 2D).

The DOPA assay was used to identify active tyrosinase (TYR). TYR is a rate-limiting oxidase that controls melanin production in Mcs. At 90-day induction with 0.4 mg/L AIV-containing differentiation buffer, the Mcs (Fig. 3A) and positive control B16 cells (Fig. 3B) demonstrated DOPA positive staining. However, the pre-differentiated BMMSCs demonstrated DOPA-negative staining (Fig. 3C). Meanwhile, there was no significant difference for DOPA-positive staining cells between the Mcs group and positive control B16 group (Fig. 3D, p > 0.05). The differentiated Mcs induced by 0.1 mg/L and 0.2 mg/L AIV differentiation buffer at 90-day induction demonstrated similar findings (Suppl. Fig. S1B).

The specific Mcs marker proteins, TRP-1, TRP-2, MART-1, and Mitf, were identified in the differentiated cells by immunofluorescence staining. Mitf was primarily detected in the nuclei, while the other markers were predominantly expressed in the cytoplasm (Fig. 4). These results further confirmed that cells undergoing 90-day induction with 0.4 mg/L AIV differentiated into Mcs. The Mcs differentiated in 0.1 mg/L and 0.2 mg/L AIV differentiation buffer for 90 days showed similar results (Suppl. Fig. S1C).

We used qPCR to measure the expression of the Mcs specific marker TRP-1 to determine the maturation state of the differentiated Mcs treated with different concentrations of AIV (Fig. 5A). At 150-day induction, TRP-1 expression was significantly increased in the 0.1, 0.2, and 0.4 mg/L AIV-treated cells compared to the control group (p < 0.01) and the differentiation medium group (p < 0.05), but still lower than that of the B16 positive control cells (p < 0.01) (Fig. 5A). The TRP-1 level in the 0.4 mg/L AIV group was higher compared to the 0.1 mg/L and 0.2 mg/L AIV groups (p < 0.05). TRP-1 expression levels between the 0.1 mg/L and 0.2 mg/L groups were not considerably different (p > 0.05). These data show that 150-day incubation with 0.1–0.4 mg/L AIV improves Mcs differentiation as evidenced by increased TRP-1 expression.

At 90- and 120-day induction, TRP-1 expression was significantly elevated in the 0.4 mg/L AIV-treated cells, which was not only higher compared to the negative control group (p < 0.01) and the differentiation medium group (p < 0.01 in 90-day group and p < 0.05 in 120-day group), but also higher compared to the 0.1 mg/L AIV-treated (p < 0.01 in 90-day group and p < 0.05 in 120-day group) and 0.2 mg/L (p < 0.01 in 90-day group and p < 0.05 in 120-day group) AIV-treated groups, respectively (Fig. 5A). TRP-1 expression was still lower compared to the B16 positive control group (p < 0.01). TRP-1 expression in the 0.1 mg/L AIV and 0.2 mg/L AIV treated groups was higher compared to the non-induction (negative control) group (p < 0.01), however, they were not significantly different compared to the differentiation medium group (p > 0.05), and lower compared to B16 positive control group (p < 0.01; Fig. 5A).

In summary, these data suggest that from 90- to 120-day induction in differentiation medium supplemented with 0.4 mg/L AIV improves Mc differentiation, as evidenced by increased expression of TRP-1.

We analyzed expression of the Mcs markerTRP-2 in the differentiated Mcs treated with different concentrations of AIV using qPCR assay (Fig. 5B). At 120- to 150-day induction, TRP-2 expression was augmented in the 0.1, 0.2, and 0.4 mg/L AIV-treated groups compared to the control group (p < 0.01). In addition, the TRP-2 level in the 0.4 mg/L AIV-treated group was higher compared to the 0.1 mg/L and 0.2 mg/L AIV groups (p < 0.05 and p < 0.01, respectively) and the differentiation medium group (p < 0.01), but TRP-2 expression was still lower compared to the B16 positive control group (p < 0.01) (Fig. 5B). The TRP-2 levels in the 0.1 mg/L and 0.2 mg/L AIV groups were higher compared to the differentiation buffer group (p < 0.05 and p < 0.01, respectively). The levels between the 0.1 mg/L and 0.2 mg/L groups were not considerably different (p > 0.05). These data suggest that 120 to 150-day incubation with 0.2–0.4 mg/L AIV improves Mcs differentiation, as evidenced by increased TRP-2 expression.

At 90-day induction, TRP-2 expression was significantly increased in the 0.4 mg/L AIV-treated cells, which was higher compared to the negative control group (p < 0.01) and the differentiation medium group (p < 0.01), and also higher compared to the 0.1 mg/L (p < 0.01) and 0.2 mg/L (p < 0.01) AM-treated groups (Fig. 5B). However, TRP-2 expression was still lower compared to the B16 positive control group (p < 0.01). TRP-2 expression in the 0.2 mg/L AIV-treated group was higher compared to the negative control group (p < 0.01). There was no significant difference between the 0.1 mg/L and 0.2 mg/L groups (p > 0.05) (Fig. 5B).

In summary, these results indicate that 90-day induction with 0.2 and 0.4 mg/L AIV improves Mc differentiation, as evidenced by increased TRP-2 expression.

Gene expression of MART-1, another melanocytic marker protein, in the differentiated Mcs treated with different concentrations of AIV was measured using qPCR assay (Fig. 5C). At 120 to 150-day induction, MART-1 levels were amplified in the 0.1, 0.2 and 0.4 mg/L AIV-treated groups compared to the negative control group (p < 0.05, p < 0.01, and p < 0.01, respectively). Furthermore, MART-1 levels in the 0.4 mg/L AIV-treated group were elevated compared to the differentiation medium group (p < 0.05). However, there were no differences for MART-1 levels among the 0.1, 0.2 and 0.4 mg/L AIV-treated groups (p > 0.05). MART-1 levels in all AIV-treated groups were lower compared to the B16 positive control group (p < 0.01) (Fig. 5C). At 90-day induction, MART-1 expression was significantly elevated in the 0.4 mg/L AIV-treated cells, which was higher compared to the negative control group (p < 0.01) and the differentiation medium group (p < 0.05) but was not different compared to the 0.1 mg/L (p > 0.05) and 0.2 mg/L (p > 0.05) AIV-treated groups (Fig. 5C). However, MART-1 was lower compared to the B16 positive control group (p < 0.01). MART-1 expression in both the 0.1 mg/L and 0.2 mg/L AIV-treated groups was higher compared to the non-induction (negative control) group (P < 0.01) and differentiation medium group (p < 0.05). There was no significant difference between 0.1 mg/L and 0.2 mg/L groups (p > 0.05) (Fig. 5C).

In summary, these results indicate that 90-day induction with 0.4 mg/L AIV improves Mc differentiation, as evidenced by increased MART-1 expression.

Finally, Mitf gene expression, which is limited in Mcs, osteoclasts, and mast cells (Hershey and Fisher, 2004), was measured in the differentiated Mcs treated with different concentrations of AIV using qPCR assay (Fig. 5D). At 90 to 150-day induction, Mitf levels were augmented in all AIV-treated groups compared to the negative control group (p < 0.01, except the 0.1 mg/L AIV-treated group for 90-day induction, where the p-value was less than 0.05). However, none of the AIV supplemental groups was significantly different compared to the differentiation medium group (p > 0.05) (Fig. 5D). These results indicate that Mitf expression is not increased by AIV supplementation.