Chemical profiling of root bark extract from Oplopanax elatus and its in vitro biotransformation by human intestinal microbiota

Materials and reagents

The general anaerobic medium for bacteria culture was obtained from Shanghai Kayon Biological Technology Co. Ltd. (Shanghai, China). Formic acid and HPLC-grade acetonitrile were purchased from Merck (Darmstadt, Germany). Deionized water (18 MΩ·cm) was supplied with a Millipore Milli-Q water system (Milford, MA, USA). All other reagents were from standard commercial sources and of analytical purity.

Preparation of Oplopanax elatus extract

Root bark of O. elatus was obtained from Benxi city (Liaoning, China). The voucher samples were deposited at the Tang Center for Herbal Medicine Research at the University of Chicago (Chicago, IL, USA). The air-dried root bark of O. elatus was pulverized into powder and sieved through an 80-mesh screen. Eight g of the powder were extracted twice by heat-reflux with 70% ethanol for 2 h. The combined extract was evaporated under vacuum and lyophilized with a yield of 28%. The samples were stored at 4 °C until use.

Preparation of human intestinal microflora

The Institutional Review Board approved the present study protocol at the University of Chicago (IRB protocol number: 12536). Fresh fecal samples were collected from six healthy adult volunteers (male, aged 20–55, non-smokers without antibiotic consumption for more than 6 months, and written consent was obtained). All the fecal samples were mixed for analysis. A total of five g of samples were homogenized in 30 ml cold physiological saline, and centrifuged at 13,000 rpm for 10 min to obtain the resulting fecal supernatant.

Incubation of sample in intestinal bacteria

Two microliters of the fecal supernatant were added with eight ml anaerobic dilution medium containing five mg of O. elatus extract, which were then anaerobically incubated at 37 °C for 24 h in an anaerobic workstation (Electrotek, UK). The reaction mixtures were extracted three times with water-saturated n-butanol. All the n-butanol layers were mixed and dried under a nitrogen stream and then dissolved in one ml methanol. The solutions were centrifuged at 13,000 rpm for 10 min for analysis.

UPLC-Q-TOF/MS analysis

Data were collected as previously described (Wang et al., 2020). The Agilent 1290 Series UPLC system (Agilent Technologies, Santa Clara, CA, USA) was applied to perform the chromatographic analysis, and a binary pump, an online degasser, an auto plate-sampler, and a thermostatically controlled column compartment were also equipped for this system. The separation was carried out on UPLC ACQUITY HSS C₈ column (2.1 mm × 100 mm × 1.7 μm, Waters) with a constant flow rate of 0.4 mL/min, and the column temperature was kept at 40 °C. A gradient mobile phase system of 0.1% formic acid in water (phase A) and acetonitrile (phase B) was applied as follows: 5% B at 0–1 min, 5–20% B at 1–18 min, 20–30% B at 18–27 min, 30–35% B at 27–32 min, 35–60% B at 32–40 min, 60–95% B at 40–50 min, 95% B at 50–53 min, 95–5% B at 53–55 min. The injection volume of samples was set at two μL for MS mode and five μL for MS/MS mode.

The Agilent 6545 Q-TOF-MS system with a Dual electrospray ionization source was used to conduct the detection. Nitrogen (purity > 99.999%) served as a sheath gas and drying gas, and the flow velocities were set at 11 and 8 L/min. The temperatures of sheath gas and drying gas were set at 350 and 320 °C respectively. Positive and negative ion modes were both employed in this study. The other parameters were set as follows: nebulizer pressure, 35 psig; voltage, 3,500 V; fragmentor voltage, 175 V; mass range, m/z 100–1,700; data acquisition rate, 1.5 scans/s; MS/MS spectra collision energy, 50 eV (Wang et al., 2020).

Data analysis

Mass data were analyzed by the Agilent MassHunter Workstation software (Version B.06.01), based on the accurate measurements of m/z values with online databases (MassBank, etc.), to screen probable compounds. The empirical molecular formula was deduced by comparing the theoretical mass of molecular ions at the mass accuracy of less than five ppm.

Optimization of UPLC-Q-TOF/MS conditions

To obtain the chromatograms with better resolution and higher baseline stability of O. elatus extract and its primary metabolites, multiple mobile phases such as acetonitrile-water and methanol-water were detected. Acetonitrile-water was applied as the solvent, for its stronger separation ability, shorter retention time, and lower column pressure. Additionally, 0.1% formic acid added in the water as mobile phase adducts may help to achieve higher response and better peak sensitivity (Tao et al., 2016). Therefore, the optimal solvent system consisting of acetonitrile-water (0.1% formic acid), which remarkably enhanced the efficiency of ionization and satisfactory sensitivity, was ultimately selected as mobile phase with a gradient elution.

In addition, the factors related to MS performance, including ionization mode and collision energy, were further improved. The positive ion mode was ultimately employed to gain comprehensive data for structural characterization and metabolite assignment with much lower background noise. The collision energy was optimized to obtain the higher ionization efficiency and relative abundance of precursor and product ions.

Chemical profiling of O. elatus extract

In total, 18 ingredients of O. elatus were detected in this study, and their chemical structures are shown in Fig. 1. There are six types of compounds, including nine polyynes, three lignans, one phenylpropanoid, two sesquiterpenes, one triterpenoid, and two fatty acids. The total ion chromatogram (TIC) of O. elatus extract is shown in Fig. 2A in the positive ion mode by UPLC-Q-TOF-MS. Table 1 shows the detailed information, including retention time, signal intensity, molecular formula, calculated and experimental mass m/z, ppm error, and fragment ions of these 18 components (Schymanski et al., 2014; Wang et al., 2020).

Polyynes have been found as the main constituents in the root of O. elatus (Yang et al., 2014). Among them, falcarindiol and oplopandiol were determined to have very high contents in the air-dried root bark. As shown in Table 1, polyynes exhibit the same elemental composition and similar MS/MS behaviors, with the characteristic fragment ions at m/z 79.05 in the positive ion mode.

For example, the typical protonated molecular ion [M+H]⁺ of FAD was observed at m/z 261.1848 in the mass spectrum. The fragment ion at m/z 105.0713 was formed by the losses of H₂O and C₁₀H₁₈, with m/z 79.0548 by further loss of C₂H₂. OPD was identified by the protonated molecular ion [M+H]⁺ at m/z 263.2010 compared with calculated m/z 263.2006. The fragment ion at m/z 107.0459 was produced by the losses of H₂O and C₁₀H₁₈, and m/z 79.0563 was formed by further loss of C₂H₄.

In addition, phenylpropanoid compound 4-(3-hydroxyprop-1-en-1-yl)-2,6-dimethoxyphenyl β-D-glucopyranoside was determined to be C₁₇H₂₄O₉ at m/z 373.1495 ([M+H]⁺, C₁₇H₂₄O₉⁺; calculated as 373.1493). The neutral loss of 1 × Glc moiety formed fragment ion at m/z 211.1526. Similarly, three lignans were determined by the neutral loss of 1 × Glc moiety in the [M+Na]⁺ mode.

Detection and identification of metabolites of O. elatus extract

The control sample was prepared in parallel, which used in the dilution medium and human fecal microflora, as shown in Fig. 2B. The biotransformed O. elatus sample by intestinal bacteria is shown in Fig. 2C. Samples were incubated, pretreated, and analyzed under the same conditions as mentioned in “Incubation of sample in intestinal bacteria”. The potential metabolites were detected from the TIC of the transformed O. elatus sample compared to the control group. All the metabolites were further confirmed by the extracted ion chromatograms (EICs) and their MS/MS corresponding fragments. A total of 62 metabolites were identified by UPLC-Q-TOF-MS in the positive mode. Table 2 shows the retention time, signal intensity, experimental and calculated mass m/z, difference between m/z and calculated m/z in ppm, and fragment ions in the MS/MS stage of these 62 metabolites (M1-M62). All these metabolites could not be observed or only in trace amounts in control samples (Schymanski et al., 2014).

Polyynes

A total of 46 metabolites of nine polyynes generated by the transformation of human intestinal microflora were detected and identified. For each polyyne, at least four types of metabolites were identified. Due to the high biological activities, FAD and OPD selected as the representative compounds of polyynes were stated in detail.

The EICs and MS/MS spectrums of metabolites of FAD are shown in Fig. 3. Five metabolites including M15, M17, M26, M46, M56 were detected. M15 was assigned to be the methylation product of FAD with the molecular formula C₁₈H₂₆O₂ at m/z 275.2008 ([M+H]⁺, C₁₈H₂₆O₂⁺; calculated as 275.2006). The fragment ion at m/z 105.0704 was generated by the neutral losses of H₂O and C₁₁H₂₀, and m/z 79.0548 was formed by further loss of C₂H₂. M17 was assigned to be the hydrogenation product of FAD with the characteristic fragment ions at m/z 105.0695 and 79.0535. In addition, metabolites M26, M46, and M56 were assigned as demethylation, dehydroxylation, and hydroxylation products of FAD, respectively.

Figure 4 presents the EICs and MS/MS spectrums of OPD metabolites (M27, M40, M43, and M49). M27 was assigned as the hydroxylation product of OPD, owing to the presence of [M+H]⁺ at m/z 279.1958. The characteristic fragment ion at m/z 107.0496 was formed by the neutral losses of H₂O and C₁₀H₁₈O, and m/z 79.0541 was formed by further loss of C₂H₄. Similarly, three other metabolites like M40, M43, and M49 were supposed to be the demethylation, dehydroxylation, and methylation products of OPD.

Proposed metabolic pathways of O. elatus extract

The proposed metabolic pathways of O. elatus extract by human intestinal microflora are presented in Fig. 5. Multiple major metabolite pathways can be observed in this study. The common pathways involved in the biotransformation of O. elatus extract include methylation, demethylation, hydroxylation, dehydroxylation, acetylation, hydrogenation, demethoxylation, and deglycosylation. Among them, polyynes were undoubtedly the most important compounds, as 46 out of 62 metabolites originated from polyynes. By comparing the signal intensity of metabolites, we could find that methylation, dehydroxylation and hydroxylation are major metabolic pathways of polyynes. Moreover, four metabolites of lignans and phenylpropanoid were produced by the loss of glucose. The other metabolites were generated from one triterpenoid and two fatty acids. This indicated that polyynes of O. elatus generated comprehensive biotransformation and were more readily metabolized than other compounds under the same conditions.

In summary, the main metabolic pathways of O. elatus refer to hydrolytic and reductive reactions by gut microorganisms. Because of the complexity of active ingredients or constituent concentrations, in vivo exposure, and individual differences, the metabolic profiles of O. elatus might be affected by several factors.