To achieve a good chromatographic resolution, chromatographic conditions were optimized. HPLC fingerprints of plasma samples collected at selected time points were obtained (Figure 1). As seen from Figure 1, a good separation of chemical constituents in plasma samples was achieved. It was found that more peaks with higher responses were detected in the chromatogram of plasma 4 at 2.5 h post-dose, whereas the chromatographic peaks of plasma 1 at 0 min pre-dose were not observed obviously at retention times from 5 min to 25 min. Thus, plasma 4 at 2.5 h post-dose was used to analyze the absorption compounds of hot-water extract from B. textilis McClure in rabbit.

Using LC-Q-TOF-MS, we acquired chromatographic profiles of control plasma, dosed plasma(2.5 h), and extract of B. textilis McClure (Figure 2). As shown in Figure 2, there were several peaks that appeared only in dosed plasma but do not appear in the chromatogram of control plasma. These compounds were absorbed into the blood and might be potential bioactive compounds from B. textilis extract. There are the absorbed components and metabolites in rabbit plasma. Except for the metabolites, the three selected peaks were clearly detected in both the leaves of B. textilis McClure and rabbit plasma. Thus, the peaks marked as I, II, and III in Figure 2(b) may be considered as the main bioactive compounds derived from B. textilis McClure.

To validate the three proposed potential bioactive compounds derived from B. textilis McClure, both hot-water extract of B. textilis McClure and dosed plasma sample at 2.5 h were analyzed by LC-Q-TOF-MS. The MS data and the tentative identification results are shown in Table 1 and Table 2.

LC-Q-TOF-MS technology can provide accurate mass, formula, UV spectra, retention time and main fragment information. As seen from Table 1, at least fifteen compounds were tentatively identified. Of the fifteen compounds, isoorientin, orientin, isovitexin, vitexin and p-coumaric acid were confirmed by comparing their accurate mass, UV spectra and retention times with those of standard compounds. Other compounds were identified by several means, utilizing their MS and MS/MS spectra, and comparing them with the literature data [13,14]. For example, a series of fragment ions of compound 5 are given in Table 1. In positive ion ESI mode, a [M+H]⁺ peak at m/z 611.1605 could confirm the molecular weight to be 610, so C₂₇H₃₀O₁₆was possible molecular composition deduced by the MassHunter software. A fragment at m/z 449.1074 [(M+H)-162]⁺, corresponding to loss of one hexose unit, also appeared. The absence of the aglycone ion is consistent with a C-hexosyl unit. Compound 5 showed fragments ions at m/z 431.0957, m/z 413.0855 and m/z 329.0655 characteristic for C-hexosyl luteolin. Therefore, compound 5 was inferred to have the structure of O-hexosyl-C-hexosyl luteolin.

The three potentially bioactive compounds in Table 2 were found to have the same MS data as No. 13, 14 and 19 in Table 1. The results further confirmed that the three compounds were derived from B. textilis McClure, so we next need to further identify the chemical structure of the potential bioactive compounds.

Compounds I and II in Table 2 were identified by comparing their UV spectrum, accurate mass, main fragment ions and retention time with those of (E)-p-coumaric acid and (Z)-p-coumaric acid standards (Figure 3, Figure 4 and Figure 5). The chemical structures of compounds Ι and ΙΙ are shown in Figure 6.

The UV spectrum of Compound III was a typical flavonoid UV spectrum according to its two major absorption bands in the UV region, band I (300–400 nm) and band II (220–280 nm) (Figure 7). Flavonoids are a large group of polyphenolic compounds possessing a basic flavan nucleus with two aromatic rings (the A and the B rings) interconnected by a three-carbon-atom heterocyclic ring (the C ring). The band I of compound III appear at 339.0 nm indicating that the compound III belongs to the flavone family unsubstituted at 3-position. The UV spectrum of the compound III showed a single peak at 270.1 nm, indicating that B ring contains only a 4′-OH group.

The ¹H-NMR spectrum of Compound III established the presence of five aromatic protons at δ 6.47 (1H, s), δ 7.81 (2H, d, J = 7.5, H-2′, 6′) and δ 6.81 (2H, d, J = 7.2, H-3′, 5′). Of these protons, the appearance of two doublets and their coupling constants values were δ 7.81 (J = 7.5) and δ 6.81 (J = 7.2), which were clearly assignable to ring B protons at H-2′, H-6′ and H-3′, H-5′, respectively. A single sharp peak at δ 6.72 ppm was assigned to the H-3 proton of the C ring. The ¹H-NMR spectrum of the compound exhibited signals at δ 4.52 (J = 9.9) applicable for sugar anomeric protons suggesting the presence of a β-glucopyranoside [15]. According to its ¹³C-NMR spectrum, a high-field signal at δ 17.9 (C-6′′′) was indicative for the presence of α-rhamnosyl. A downfield signal at δ 182.6 was clearly assigned to the carbonyl carbon C-4 of the pyrone ring. The three downfield signals appearing at δ 161.6, δ 163.1 and δ 156.7 were assigned to the C-5, C-7 and C-4′ carbon atoms bearing hydroxyl groups. Furthermore, a signal at δ 80.5 suggested that a rhamnosyl unit was attached to C-2′′.

Compound III had a pseudomolecular ion at m/z 579.1705 [M+H]⁺. A molecular formula of C₂₇H₃₀O₁₄ (calc. 578.1636) was generated using the MassHunter quantitative analysis software. A fragment at m/z 433.1129 [M+H-146]⁺, corresponded to loss of one deoxyhexose. Thus, compound III could have a deoxyhexosyl unit in the terminal position of a disaccharide unit. The presence of the ion at m/z 313.0703 [M+H-146-120]⁺ and the absence of the fragment [M+H-146-60]⁺ indicated a hexose as the C-glycosylation sugar. As analyzed above, the structure of compound III was determined as apigenin-8-C-β-D-(2"-O-α-L-rhamnosyl)-glucopyranoside, which was also confirmed by its MS fragmentation pathways (Figure 8).

Bamboo leaves are rich in flavonoids and the main functional components of extract from bamboo leaf are flavone C-glycosides [16,17]. According to the literature flavone C-glycosides, including isoorientin, orientin, vitexin and isovitexin, were poorly absorbed in the gastrointestinal tract of rat. Flavone C-glycosides largely reach the colon where they can be degraded into various aromatic acid metabolites by the microflora. The flavone C-glycosides cannot be absorbed in the rats’ blood after oral administration, whereas p-coumaric acid can be absorbed in blood [18]. The ability of p-coumaric acid to protect rat’s heart against doxorubicin-induced oxidative stress has been proved [19]. In this study, apigenin-8-C-β-D-(2"-O-α-L-rhamnosyl)-glucopyranoside belonging to the C-glycosides class was absorbed in rabbit blood. The absorption mechanism in vivo should be investigated in further research. This study provides helpful chemical information for further our understanding of the pharmacology and active mechanism research of B. textilis McClure.

Liquid chromatography/quadrupole time-of-flight mass spectrometry (LC-Q-TOF-MS) analysis was performed on an Agilent 1290 series LC system coupled with a Q-TOF (model 6540, Agilent Technologies, Santa Clara, CA, USA) and equipped with a diode array detector (DAD). The system was controlled under MassHunter B.04 software. Purification of bioactive compounds was performed with a Gilson preparative HPLC GX-281/322/156 system (Gilson, Middleton, WI, USA), using a C18 column (250 × 20 mm I.D., 5 µm, Hydrosphere, YMC Japan, Kyoto, Japan). The obtained compounds were dissolved in 0.5 mL of DMSO-d₆ and the solution was filled into a NMR tube. NMR analysis was carried out using a Bruker 300 MHz spectrometer operating at 300 (¹H) or 75 MHz (¹³C), respectively. Chemical shifts were reported in ppm on δ scale, and the coupling constants (J) were measured in Hertz (Hz). The compounds were identified by a combination of spectroscopic methods (¹H-, ¹³C-NMR), ESI-MS and comparison with the literature data.

The bamboo leaves of B. textilis McClure were collected from the bamboo garden of the Jiangxi Academy of Forestry, Nanchang, China. The species were authenticated by Professor Jiu-Sheng Peng from the Jiangxi Academy of Forestry. Bamboo leaves were dried in the shade, ground to powder, and stored at –20 °C.

The HPLC-grade acetonitrile and methanol were purchased from Fisher Scientific (Fair Lawn, NJ, USA). Acetic acid and (E)-p-coumaric acid were purchased from Sigma-Aldrich (St. Louis, MO, USA). (Z)-p-coumaric acid was purchased from J&K Scientific Ltd. (Beijing, China). Water was purified with an ultrapure water system (Purelab Plus, Pall, Port Washington, NY, USA). Analytical-grade chemical was obtained from Beijing Chemical Works (Beijing, China). The standards of isoorientin, orientin, isovitexin and vitexin were purchased from Shanghai Winherb Medical S & T Development Co., Ltd. (Shanghai, China).

The dried bamboo leaf powder (100 g) was extracted by percolation with water (600 mL) and the solution was heated to reflux and boiled gently for 20 min. The cooled solution was filtered and the residue of the bamboo leaves was carefully washed twice with water (600 mL). The combined extracts were concentrated under reduced pressure at 40 °C. The extracts were suspended in water (100 mL) and stored at 4 °C before use.

A rabbit (3 kg) was obtained from the SFA Key Laboratory of Bamboo and Rattan Science and Technology (Beijing, China). The rabbit was fasted 16 h before the experiment and water was taken freely. Hot-water extracts from B. textilis McClure (100 g of bamboo leaves/100 mL of water) was administered orally to the rabbit at a single dose of 33.3 g/kg. Blood samples (2.5–3 mL) were collected in heparinized tubes pre-dose (0 min) and at selected time points (20 min, 1.3 h, 2.5 h, 4 h, 5.5 h, 8 h and 12 h) post-dose. The heparinized tubes were shaken gently as soon after collection as possible to prevent clotting. Then the blood samples were centrifuged at 3,000 rpm for 10 min. The supernatant layer of plasma (0.5 mL) was placed into a 5 mL polypropylene tube. Methanol (2 mL) was added to the tube to deposit proteins. Each tube was vortexed for 30 s and left for 12 h at –18 °C. Then plasma samples were centrifuged at 13,000 rpm for 10 min. The supernatant was transferred to a round-bottom flask, and evaporated to dryness at 40 °C in a rotary vacuum evaporator (Rotary Evaporator, EYELAN-1001D-W, Tokyo, Japan). The residue was reconstituted in 0.5 mL of methanol and then filtered through a 0.22 μm membrane. Experiments were carried out in accordance with the Guide for the Care and Use of Laboratory Animals [8,20].

LC separation was performed using a C18 column (4.6 × 250 mm, 5 µm, YMC pack R&D ODS-A, YMC Japan) at 30 °C. The solvent system consisted of a mixture of water with 0.5% (v/v) acetic acid and acetonitrile (85/15, v/v) at a flow rate of 1.0 mL/min. The LC effluent was split using a T-splitter to produce a flow of 0.25 mL/min [21].

The TOF/MS analysis worked using positive mode. Mass spectra were acquired in the range of 100 to 1,000 m/z for MS¹ and 20 to 1,000 m/z for MS². The MSⁿ data were collected in an auto MS/MS mode. The collision-induced dissociation (CID) voltage was set at 20 V. The conditions of ESI source were as follows: drying gas (N₂) flow rate, 8 L/min; drying gas temperature, 350 °C; nebulizer, 45 psig; capillary voltage, 4,000 V; skimmer voltage, 65 V; fragmentor voltage, 175 V; sheath gas temperature, 350 °C; nozzle voltage, 1,000 V. During the analysis, two reference masses were used: 121.0509 (C₅H₄N₄) and 922.0098 (C₁₈H₁₈O₆N₃P₃F₂₄). These masses are continuously infused to the system to allow constant mass correction. The chemical formula of the selected peaks were calculated by the accurate mass of the precursor and product ions. The formula predictor was set as follows: C (0-60), H (0-120), O (0-30). Other elements such as P, S, F and Cl were not considered, since they are rarely present in the components of bamboo leaves.

Powdered leaves (100 g) of B. textilis McClure were extracted again as above. After removing the solvent under reduced pressure and freeze-drying, the combined extracts were concentrated and yielded 18 g of a crude extract. The extract was suspended in water/acetonitrile (100 mL, 85/15, v/v) and filtered through a 0.45 μm membrane to remove the insoluble material. This crude solution was purified on a preparative HPLC system. The mobile phase consisted of H₂O (solvent A) and ACN (solvent B). The gradient elution program was as follows: 0–3 min, 0% B; 3–7 min, 0–15% B; 7–45 min, 15% B; 45–50 min, 15–50% B; 50–52 min, 50–15% B; 52–60 min, 15% B. The flow rate was set at 10 mL/min with UV detection at 342 nm to yield compound III (6.8 mg).

Apigenin-8-C-β-D-(2"-O-α-L-rhamnosyl)-glucopyranoside (III). Yellowish solid. UV λ_max(ACN-H₂O) 270, 339 nm; HRESIMS m/z 579.1705 [M+H]⁺ (calc. for C₂₇H₃₀O_14, 579.1708); ¹H-NMR (DMSO-d₆) δ: 6.72 (1H, s, H-3), 6.47 (1H, s, H-6), 7.81 (2H, d, J= 7.5, H-2′, 6′), 6.81 (2H, d, J= 7.2, H-3′, 5′), 4.52 (1H, d, J= 9.9, H-1′′), 4.97 (1H, s, H-1′′′), 3.40–4.50 (13H, m, Sugar); ¹³C-NMR (DMSO-d₆) δ: 163.8 (C-2), 103.1 (C-3), 182.6 (C-4), 161.6 (C-5), 94.7 (C-6), 163.1 (C-7), 109.5 (C-8), 156.7 (C-9), 109.0 (C-10), 121.5 (C-1′), 128.8 (C-2′), 116.2 (C-3′), 156.7 (C-4′), 116.2 (C-5′), 128.8 (C-6′), 75.0 (C-1′′), 80.5 (C-2′′), 76.2 (C-3′′), 71.7 (C-4′′), 81.9 (C-5′′), 62.2 (C-6′′), 103.1 (C-1′′′), 71.7 (C-2′′′), 71.1 (C-3′′′), 71.4 (C-4′′′), 72.0 (C-5′′′), 17.9 (C-6′′′).