Untargeted metabolomic analyses of fermented unpolished black rice with melanogenesis inhibition activity

Materials & chemicals

The microorganisms in the defined De-E11 microbial starter were obtained from the Department of Microbiology, Faculty of Science, Chulalongkorn University. All media for cell growth, such as Dulbecco’s modified Eagle’s medium (DMEM) high glucose, fetal bovine serum (FBS), and trypsin-EDTA (0.25%) were purchased from Gibco-BRL Inc., USA. The high-performance liquid chromatography (HPLC)-grade chemicals were purchased from Merck, Germany. Folin-Ciocalteu reagent, 2,2-diphenyl-l-picrylhydrazyl (DPPH), 2,2-azino-bis-(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), and mushroom tyrosinase enzyme were purchased from Sigma-Aldrich, USA. Kojic acid and l-3,4-dihydroxyphenylalanine (L-DOPA) were purchased from Tokyo Chemical Industry Co., Ltd., Japan. The UBR was purchased from L.H. Rice International Co., Ltd. (Nakhon Pathom, Thailand).

Sample preparations

The sample preparation was performed as previously described (Sangkaew & Yompakdee, 2020). Briefly, 20 g of UBR was added to 40 mL of distilled water and autoclaved at 121 °C for 15 min. After being cooled, the defined De-E11 microbial starter, comprised of R. oryzae E1101, S. cerevisiae E1103, S. fibuligera E1102, and P. pentosaceus E1104 at 1  ×  10⁴, 2  ×  10⁴, 1  ×  10³, and 3  ×  10⁸ colony forming units (CFU)/g, respectively, was mixed with cooked UBR and incubated at 30 °C for 12 days in a closed sterilized bottle. The fermented liquid was then collected (called FUBRS). For the Un-FR, 20 g of UBR was mixed with 40 mL of distilled water and boiled at 80 °C for 15 min, then centrifuged as above and the supernatant (Un-FR) was harvested. Both the Un-FR and FUBRS samples were then evaporated, resuspended in 20 mL of distilled water, and kept at −20 °C for further study.

Physicochemical analysis

The pH of the samples was measured using a pH meter (S-20K; Mettler-Toledo, Greifensee, Switzerland). The titratable acidity was measured by titrating the samples with 0.1 N NaOH using phenolphthalein as the color indicator (Woldemariam et al., 2014). The total reducing sugars was measured using the 3,5-dinitro salicylic acid (DNS) method and expressed as mg glucose equivalents per mL sample (mg GE/mL of sample) (Zeng et al., 2017).

Determination of total phytochemical contents

Biological activity analysis

Anti-mushroom tyrosinase activity (in vitro)

The samples of Un-FR and FUBRS were examined for tyrosinase inhibiting activity by evaluating the mushroom tyrosinase activity via the reported dopachrome method (Zengin et al., 2015) except with slight modification. The reaction mixture, which consisted of 80 µL of 50 mM potassium phosphate buffer (pH 6.8), 40 µL of the sample, and 40 µL of mushroom tyrosinase (50 U/mL), was mixed and then incubated at room temperature for 10 min. Next, 40 µL of 1.5 mM L-DOPA was added and incubated at room temperature for 10 min. The amount of dopachrome in the reaction mixture was then measured by spectrophotometer at 492 nm (A₄₉₂). Kojic acid (0.03 mg/mL) was used as the positive control. The percent inhibition of tyrosinase activity was calculated from Eq. (3): (3)${\displaystyle \mathrm{Mushroom tyrosinase inhibition}\left(\text{\%}\right)=\left[\left(\mathrm{A}-\mathrm{B}\right)/\mathrm{A}\right]\times 100}$

where A and B are the A₄₉₂ values of the without and with test sample, respectively.

Melanogenesis inhibition activity

Melanogenesis inhibition activity was determined by measurement of the melanin content in the B16F10 melanoma cell line as previously described (Sangkaew & Yompakdee, 2020). The B16F10 cell line (ATCCCCL-6475™) was cultured in DMEM supplemented with 10% (v/v) FBS, penicillin (100 U/mL), and streptomycin (100 mg/mL) in six-well plates (5 × 10⁴ cells/well) for 24 h at 37 °C under a humidified 95:5 (v/v) air: CO₂ atmosphere. The cells were then treated with Un-FR or FUBRS at the indicated final concentration and incubated for 72 h. After incubation, the cells were harvested and solubilized in 1 N NaOH at 60 °C for 60 min, and then the absorbance of the cell suspension was measured by spectrophotometer at 405 nm (A₄₀₅). The relative melanogenesis inhibition (%) was calculated from Eq. (4): (4)${\displaystyle \mathrm{Relative melanogenesis inhibition}\left(\text{\%}\right)=\left[1-\left(\mathrm{A}÷\mathrm{B}\right)/\left(\mathrm{C}÷\mathrm{D}\right)\right]\times 100}$

where A and C are the A₄₀₅ values of the treated cells and untreated cells, respectively, and B and D are the protein concentrations (mg/mL) of the treated cells and untreated cells, respectively.

Untargeted metabolites profiling by gas chromatography-mass spectrometry (GC-MS)

GC-MS analysis

Chromatographic separation was performed on a GC-MS system (Agilent Technologies) using a HP-5 ms (30 m × 250 µm × 0.25 µm; Agilent Technologies) capillary column. The oven and injector temperatures were set at 80 °C. Helium was used as the column carrier gas at a constant flow rate of 0.8 mL/min. For fatty acid analysis, the temperature and constant flow rate were set at 240 °C and 2.25 mL/min, respectively. Identification of the untargeted compounds was performed by comparing the MS fragmentation patterns with those of reference compounds, including the mass spectra in the NIST database.

Quantitative determination of metabolites in the FUBRS sample

The FUBRS sample were extracted by solid-phase extraction using Sep-Pak C18 column cartridges (Sep-Pak Waters, Milford, MA, USA) according to Flores et al. (2012) with some modifications. Briefly, the cartridge was first conditioned with one mL of methanol and then with 1.0 mL of milliQ water. Next, the sample was loaded and washed with 50 mL of milliQ water. The analyte was eluted with five mL of 0.1% (w/v) HCl in methanol, evaporated to dryness in a rotary evaporator at 40 °C, re-dissolved in 50% (v/v) methanol, and then kept at −20 °C for further analysis.

The metabolites in the sample extract were determined by high-performance liquid chromatography (HPLC). Initially, 1.0 mL of flow through from FUBRS was performed using an Agilent HPLC 1260 Infinity II HPLC system (Agilent Technologies) with a UV/Vis detector (SPD-20A) monitoring at a wavelength of 210 nm, and a 5 µm, 4.6 × 150 mm Mightisil RP18 column (Waters). The sample injection volume was five µL and elution was performed in an isocratic mode with 0.02 M Na₂HPO₄, pH 2 at a flow rate of 0.1 mL/min.

Statistical analysis

All experiments were performed in triplicate, and the data in the tables are presented as the respective mean ± one standard error. The statistical analysis of data was performed using GraphPad Prism 9.0.2 (San Diego, CA, USA), and differences among groups were determined by one-way analysis of variance (ANOVA) with Dunnett’s Multiple Comparisons test. To elucidate the differential metabolites, the metabolites were imported into the MetaboAnalyst 6.0 (https://www.metaboanalyst.ca/) online software.

Physical characteristics and phytochemical content of FUBRS

Fermentation of UBR by the microorganisms in the De-E11 is known to effect the final product (Sangkaew et al., 2020; Sangkaew et al., 2023). The physicochemical properties and phytochemicals are important indicators that can be used to evaluate the degree of fermentation. Their characterization in the FUBRS and Un-FR samples revealed that the total acid contents increased markedly (93-fold) while the pH decreased in FURBS compared to in the Un-FR (Table 1). Moreover, the level of reducing-sugars was also markedly increased (8-fold) in the FUBRS compared to that in the Un-FR. In the fermentation process, starch is digested into the reducing sugar glucose, an important substrate for alcohol fermentation that greatly impacts the acidity, taste, and alcohol content (Kim et al., 2007).

The total phytochemical contents, as the TPC, TFC, and TAC, of the Un-FR and FUBRS are presented in Table 2. The results revealed that the Un-FR had a significantly higher phytochemical content than FUBRS. A similar result was reported in rice wine fermentation, and was suggested that the reduction in the TPC upon fermentation might be related to the utilization of phenolic compounds by microorganisms (Mu et al., 2019). Interestingly, the color of the FUBRS sample was lighter than that of the Un-FR sample (Fig. S1), which is consistent with the higher (over 3.5-fold) TAC in the Un-FR than in the FUBRS. Transformation of anthocyanins is mediated by its reaction with yeast metabolites, such as pyruvic acid and acetaldehyde (De Freitas & Mateus, 2011). Moreover, anthocyanins are also degraded into their aglycone forms by lactic acid bacteria during the fermentation process (Braga et al., 2018). Hence, these results suggested that the fermentation of UBR with the defined De-E11 microbial starter may convert the compounds in UBR into other compounds, which leads to the reduced phytochemical content in FUBRS.

Biological activity of FUBRS

Fermented rice, especially colored rice, is rich in various bioactive components and has many health benefits (Ruksiriwanich et al., 2011). Hence, we investigated the biological activities, in terms of the antioxidant activity, mushroom tyrosinase activity, and melanogenesis inhibition activity, of the FUBRS and Un-FR samples.

Antioxidant characteristics of the FUBRS and Un-FR samples

The DPPH and ABTS assays are the most popular and convenient methods to determine the antioxidant activity. Specifically, the ABTS assay is based on the generation of blue/green ABTS⁺ ions that can be reduced by antioxidants, whereas the DPPH assay is based on the reduction of the purple DPPH to 1,1-diphenyl-2-picryl hydrazine (Floegel et al., 2011; Sánchez et al., 2007). Hence, the antioxidant capacity of the FUBRS and Un-FR samples was determined by both methods using ascorbic acid as the positive control. As depicted in Fig. 1, both the FUBRS and Un-FR samples possessed antioxidant activities with a clear concentration-dependent radical scavenging activity in both the DPPH and ABTS assays. However, Un-FR had a higher antioxidant activity than FUBRS in both assays. These results are in accord with a previous report that fermentation of black rice bran decreased the antioxidant activity, TPC, and TAC (Yoon et al., 2015). Moreover, our results revealed that the antioxidant activity of FUBRS and Un-FR (Fig. 1) was consistent with the TPC (Table 2). Thus, the reduction in the antioxidant activity after fermentation may be associated with the degradation of high antioxidant activity phenolic compounds into phenolic or other compounds of a lower or no antioxidant activity (Gan et al., 2017; Melini & Melini, 2021).

Mushroom tyrosinase inhibition activity

When skin is exposed to UV rays it causes an increase in tyrosinase expression, a key enzyme in melanin biosynthesis (Chan et al., 2014). Mushroom tyrosinase has been extensively utilized as a model system for the screening of tyrosinase inhibitors and in melanogenic studies (Zolghadri et al., 2019). Hence, the potential inhibitory effect of FUBRS and Un-FR on mushroom tyrosinase activity was evaluated using Kojic acid as the positive standard. The results (Fig. 2) clearly revealed that the tyrosinase inhibition activity was significantly increased following fermentation (FUBRS vs. Un-FR). In accord, several researchers have reported that fermentation could enhance the tyrosinase inhibition activity of raw materials (Abd Razak et al., 2017; Chan et al., 2014; Chen et al., 2013; Sangkaew & Yompakdee, 2023).

Melanogenesis inhibition activity

To determine the melanogenesis inhibition activity, 5% (v/v) Un-FR and FUBRS at three different concentrations (1%, 2.5% and 5% (v/v)) were evaluated with the B16F10 melanoma cell line. The results showed that FUBRS significantly inhibited melanin production in B16F10 melanoma cells in a dose-dependent manner, whereas Un-FR showed no such activity (Fig. 3). Interestingly, Un-FR and FUBRS both exhibited anti-mushroom tyrosinase activity (Fig. 2), but Un-FR showed no significant effect on the reduction of melanin content in B16F10 cells even though it was at a higher dose of 5% (v/v) than the FUBRS (1% (v/v)) (Fig. 3). Although mushroom tyrosinase is used for anti-tyrosinase activity screening, it does not appear to be sufficiently sensitive to measure the potential of the active substances, such as FUBRS, as a whitening agent. There are significant differences between mushroom tyrosinase and mammalian tyrosinase in terms of their structural, molecular, and kinetic properties, as well as their localization. Therefore, a more appropriate method might be to use the melanogenesis inhibition results from the cell-based assay (Promden et al., 2018; Song et al., 2009).

Untargeted metabolic analysis of Un-FR and FUBRS

Multivariate analysis of Un-FR and FUBRS

Metabolomics based on GC-MS has been successfully applied in the study of dynamic changes in the metabolite profiles in rice koji and rice wine (Lee da et al., 2016; Mu et al., 2019). Therefore, the metabolites in Un-FR and FUBRS were investigated using GC-MS. To initialize the systemic alteration of metabolites, multivariate analysis by orthogonal projections to latent structures discriminant analysis (OPLS-DA), a common classification method, was performed using the online MetaboAnalyst 6.0 software to identify the difference between the two groups.

The OPLS-DA score plot of the GC-MS data from the Un-FR and FUBRS samples showed a clear separation (Fig. 4), implying that the metabolic profile was significantly altered after fermentation. Moreover, the R2X, R2Y, and Q2 values, which are the critical parameters used to evaluate the OPLS-DA model, were 0.772, 0.935 and 0.886, respectively, in the metabolites group (Fig. S2A) and 0.556, 0.868 and 0.794, respectively, in the fatty acid group (Fig. S2B). Note that R2 and Q2 represent the explanatory ratio and predictability, respectively, where the closer the value is to 1, the higher the model accuracy. Generally, when R2 and Q2 > 0.5, the model fits well.

Identification of significantly different metabolites in Un-FR and FUBRS

To investigate the changes in metabolite abundance between Un-FR and FUBRS, a heatmap was constructed using the MetaboAnalyst 6.0 software. In total, 42 metabolites were identified, comprised of 17 sugars and sugar alcohols, nine amino acids, 13 organic acids, and three phenolic acids (Fig. 5). The result indicated that most of these metabolites were increased after fermentation. In the fatty acids group, five fatty acids (stearic, oleic, linoleic, palmitic, and myristic acids) were detected (Fig. 6), of which oleic and myristic acids were increased while linoleic acid was decreased in FUBRS compared to Un-FR.

Furthermore, change in the abundance by more than a two-fold change of 47 metabolite (17 sugars and sugar alcohols, nine amino acids, 13 organic acids, three phenolic acids and five fatty acids) between both samples was analyzed, with the results shown in the volcano plot. The volcano plot illustrated that among all these metabolites, 25 were significantly increased and nine were decreased in FUBRS compared to Un-FR, whereas 13 metabolites were not significantly different (Fig. 7 and Table S1).

During the fermentation process, rice starch is hydrolyzed and then saccharified to sugars and then sugar alcohols by enzymes from the microorganisms. Sugars are generally consumed as carbon sources, providing energy for proliferation and growth of the microorganisms via carbohydrate metabolic pathways that led to the production of other metabolites, such as amino acids, organic acids, and fatty acids (Zhu et al., 2004). The level of detected sugars and sugar alcohols, namely: ribopyranose, arabinose, glucose, myo-inositol, trehalose, lactose, mannobiose, and melibiose in FUBRS was higher than those in Un-FR. Five amino acids (alanine, leucine, proline, serine, and phenylalanine) were significantly increased after the fermentation process.

Generally, the acidity of a sample results from organic acids, which can dramatically affect the organoleptic quality of the fermented product and can inhibit the growth of microbes. They mainly come from microbial metabolism (Mato, Suárez-Luque & Huidobro, 2005). Among the detected 13 organic acids, 10 (lactic, acetic, succinic, α-hydroxy glutaric, glyceric, phosphoric, 3-phenyllactic, tartronic and pyruvic acids as well as levoglucosan) were increased in the fermentation process (FURBS vs. Un-FR).

Phenolic compounds in the fermented product are mainly derived from plant materials that are associated with the quality of product (Martins et al., 2011). The level of 3-hydroxybenzoic acid was higher in the FUBRS than in the Un-FR samples. Moreover, the fatty acid (oleic acid) content was also increased after fermentation (Fig. 7). Table S2 illustrates the metabolites found in FUBRS which have previously been reported to have an inhibitory activity of melanogenesis or tyrosinase, where the most active metabolites were increased after fermentation.

Our results are consistent with previous studies, which reported that the microorganisms in De-E11 starter expressed the genes or proteins involved in the synthesis of many melanogenesis inhibitors, such as amino acids, organic acids, sugars, phenolic acids, and fatty acids (Sangkaew et al., 2020; Sangkaew et al., 2023).

Quantitative determination of some selected metabolites in the FUBRS sample

To determine the quantitative level of metabolites in the FUBRS sample, 1-phenolic acid (p-hydroxybenzoic acid) and three organic acids (lactic, acetic, and succinic acids), from the untargeted metabolite results (Figs. 5 and 7) that were significantly increased after fermentation (Fig. 8) were used as representative compounds for quantitative determination in the FUBRS sample (Table 3). These metabolites were selected because they have been reported to have a tyrosinase or melanogenesis inhibition activity (Chan et al., 2014; Yamamoto et al., 2006). Among the identified organic acids, lactic acid showed the highest level at 947 ± 10 ng/mL, followed by acetic acid (420 ± 13 ng/mL) and succinic acid (41.7 ± 7.6 ng/mL), respectively, while p-hydroxybenzoic acid was at 193 ± 1 ng/mL.