Chemical characterization of Passiflora edulis extracts and their in vitro antioxidant, anti-inflammatory, anti-lipid activities, and ex-vivo vasodilation effect

2.1 Preparation of P. edulis f. flavicarpa Deg. extracts

Fruits of P. edulis were obtained from Chiang Mai, Thailand. A voucher specimen (BK No 082283) was authenticated by a taxonomist staff of plant varieties protection office and located at the Forest Herbarium, Royal Forest Department, Ministry of Agriculture and Cooperatives, Bangkok, Thailand. The seed of P. edulis was dried for 24 h in hot air oven (45 °C) before powdered. The powder of dried seed (1.2 kg) was extracted with 10 times (w/v) of 70% ethanol by maceration and shaken (150 rpm) at room temperature (RT) (30 ± 2 °C) in a orbital shaker (Ratek, Boronia, VIC, Australia) for 72 h and then filtered (Whatman®, No 1). The solvent was heated to dryness by evaporator (Buchi, Switzerland) under reduced pressure at 50 °C and preserved in a freezer. The yield of the seed P. edulis ethanolic extract (PSEE) was 4.55% (w/w) of the dry seed. P. edulis fruit juice and pulp (4.5 kg) was blent, filtered 3 times with cotton and then freeze dried. The percentage yield of the P. edulis fruit juice and pulp extract (PFWE) was 12.12% (w/w) of the fresh juice and pulp.

2.2 Thin-layer chromatography (TLC) fingerprint of PFWE and PSEE

The tested samples, PFWE and PSEE (10 mg/mL in methanol, 5 µL) were spotted to Aluminum TLC plates, silica gel coated with fluorescent indicator F254 (4 × 6 cm, Merck, Germany). There were 3 ratios (90:10, 95:5 and 98:2) of dichloromethane and methanol used in this study as developing solvents represented a polar, moderate polar and non-polar conditions, respectively. The spots appeared in TLC plate after developing in each mobile phase were observed under 4 conditions. There were 1) normal visual 2) observed under UV TLC cabinet (CAMAG, Switzerland) at the short wavelength (254 nm) 3) at the long wavelength (365 nm) and 4) after spaying with 10% sulfuric acid solution in ethanol then heat at 110 °C. The retention factor (Rf) values of each spot were calculated by dividing the ratio between the distance of the spot and the distance of solvent front.

2.3 Determination of piceatannol in PFWE and PSEE by high performance liquid chromatography (HPLC)

Reverse-phase HPLC with diode array detector was used to determine the amount of piceatannol in PFWE and PSEE. The extract solution was injected into a Hypersil ODS column (250 × 4.0 mm i.d.; 5 µm particle size). Deionized water with 0.1% formic acid was used as a mobile phase A, and acetonitrile with 0.1% formic acid was used as mobile phase B. Flow rate for all analysis was 0.75 mL/min. A wavelength of 320 nm was used to detect an optical absorbance. The amount of piceatannol was determined by comparing the peak area of the extract solution with the peak area from the standard curve.

2.4 Determination of beta-carotene and gamma-tocopherol in PFWE by HPLC

Beta-carotene and gamma-tocopherol in PFWE extract was examined by using HPLC with UV detector at 450 nm. A Hypersil ODS (250 × 4.0 mm i.d.; 5 µm particle size) and mobile phase system consisted of acetonitrile : dichloromethane : methanol (70:20:10) in isocratic condition were used for the analysis. Mobile phase system for gamma-tocopherol in an isocratic condition composed of methanol : acetonitrile (60:40) and the detection was set at 208 nm.

2.5 Determination of total phenolic, flavonoid, and carotenoid contents

Folin-Ciocalteu assay was utilized to evaluate the content of total phenolic with slight modifications at 760 nm (Spanos and Wrolstad, 1990). The amount of total phenolic was presented as mg gallic acid equals per gram extract (GAE). Aluminium chloride colorimetric method was utilized to detect the amount of total flavonoid content in PFWE and PSEE at 430 nm (Kamtekar et al., 2014) and presented as mg quercetin equivalents per gram extract (QE). The content of total carotenoids was detected at 450 nm and represented as mg β-carotene equivalents per gram of extract (mg β-carotene /g extract).

2.6 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging assay

Free radical scavenging properties of each extracts was detected by DPPH assay as illustrated by Teixeira et al. (2013) at 517 nm. The activity of DPPH radical scavenging (%) was determined by the following formula: $$\mathrm{Radical}\mathrm{scavenging}\mathrm{activity}\left(\%\right)=\left({\mathrm{Abs}}_{\mathrm{control}}-{\mathrm{Abs}}_{\mathrm{test}}\right)/{\mathrm{Abs}}_{\mathrm{control}}\times 100$$

2.7 Determination of HMG-CoA reductase inhibition activity

The inhibitory activity of HMG-CoA reductase in PFWE and PSEE were evaluated by HMG-CoA reductase assay kit (Sigma-Aldrich Co., St Louis, MO, USA) under conditions recommended by the manufacturer at 340 nm. Pravastatin was utilized as a positive drug. The inhibition of HMG-CoA reductase was determined using the following formula: $$\mathrm{Inhibitory}\mathrm{activity}\left(\%\right)=\left({\mathrm{Abs}}_{\mathrm{control}}-{\mathrm{Abs}}_{\mathrm{test}}\right)/{\mathrm{Abs}}_{\mathrm{control}}\times 100$$

2.8 The pancreatic lipase inhibition assay

The inhibitory activity of pancreatic lipase was carried out following a previously described method with a slight modification (Od-Ek et al., 2020). Orlistat was used as a positive drug. All samples were carried out in triplicate, and determined at 405 nm with a microplate reader (Biotek, Vermont, USA). The percentage inhibition of pancreatic lipase was examined using the following formula: $$\mathrm{Inhibitory}\mathrm{activity}\left(\%\right)=\left({\mathrm{Abs}}_{\mathrm{control}}-{\mathrm{Abs}}_{\mathrm{test}}\right)/{\mathrm{Abs}}_{\mathrm{control}}\times 100$$

2.9 The pancreatic cholesterol esterase inhibition assay

The inhibitory activity of pancreatic cholesterol esterase was determined at 405 nm according to Makynen et al. (2013) with some modification. In this study, we used simvastatin (Sigma-Aldrich, MO, USA) as a positive control. All samples were performed in triplicate. The inhibition of pancreatic cholesterol esterase were measured using the following formula: $$\mathrm{Inhibitory}\mathrm{activity}\left(\%\right)=\left({\mathrm{Abs}}_{\mathrm{control}}-{\mathrm{Abs}}_{\mathrm{test}}\right)/{\mathrm{Abs}}_{\mathrm{control}}\times 100$$

2.10 In vitro model

2.11 Ex-vivo vascular relaxation model

A 8 weeks male Sprague-Dawley rats (180–200 g) were supported from Nomura Siam International (Bangkok, Thailand). All animals were kept under a controlled condition (22 ± 1.0 °C with a 12-hour light/dark cycle) in the center for animal research at Naresuan University. The study protocol was permitted by the Institutional Animal Care and Committee of Naresuan University, Thailand (approval number: 63-01-003). Intraperitoneal injection of sodium thiopental (100 mg/kg BW) was used to euthanize the rats. Then, the descending thoracic aorta were removed and saturated in ice-cold Krebs-bicarbonate solution. The thoracic aorta was then cut into ring segments in length of 2–3 mm and suspended between two stainless steel hooks. One hook was connected to an isometric force transducer (AD Instruments, Australia), which was linked to a Bridge Amplifier (Powerlab, AD instrument Ltd, Australia) for isometric tension recording. The resting tension of each aortic ring was controlled at 1 gm, and authorized to equilibrate for 60 min. The resting tension of each aortic ring was stabilized. The integrity of the endothelium was established by more than 80% relaxation to acetylcholine (10 μM) after pre-constriction with phenylephrine (PE, 0.1 μM). To measured the relaxant effect of PFWE and PSEE, aortic rings were declined with submaximal concentration of PE. After the contraction reached a plateau phase, PFWE, PSEE, or vehicle were added cumulatively to the endothelium-intact aortic rings. The relaxation effect was evaluated as a percentage of the contraction with PE.

2.12 Statistical analysis

Mean ± standard error of the mean was utilized to display the values of experiments. The differences of the mean values were examined by independent t-test and ANOVA, followed by LSD post hoc test (in vitro experiment) and Newman-Keuls post hoc test (ex-vivo experiment) to compare between groups. P values less than 0.05 was established for statistically significant. The data was analyzed by R program.

3.1 Chemical analysis of PFWE and PSEE by TLC and HPLC

The physical appearance and the percent yields of two tested extracts, PFWE and PSEE, were yellowish coarse powders (12.12%) and light brown slurry (4.55%), respectively. The TLC fingerprint in polar, moderate, and non-polar conditions were presented in Fig. 1. The TLC analysis showed normal visual of TLC plates in every conditions of both extract, and no spot were observed (Fig. 1A). However, the spots were observed under short and long wavelength, and also after spraying with 10% sulfuric acid solution and heating as demonstrated in Fig. 1B, 1C and 1D, respectively. The results indicated that PSEE appeared to have more spots than the PFWE. The Rf values of spots observed in TLC were calculated and shown in Table 1. These results suggested that PSEE which was extracted with 70% ethanol exhibited more chemical components than that of PFWE which was in aqueous solution. By the HPLC method, piceatannol was detected with retention time of 31.33 min. HPLC chromatogram from the standard solution (Fig. S1), and the extract solutions were demonstrated in Fig. 2A. Piceatannol content was detected only in PSEE (4.742 mg/g extract). Beta-carotene and gamma-tocopherol contents were found in PFWE (0.1917 mg/g extract and 20.03 μg/g extract, respectively) using HPLC as shown in Fig. 2B and 2C, respectively. HPLC chromatograms of the standard references of beta-carotene and gamma-tocopherol were snowed in Fig. S2 and Fig. S3.

Close

Fig. 1

TLC-fingerprints of the PFWE and PSEE in three mobile phases; dichloromethane : methanol = 90:10, 95:5 and 98:2 which (A) normal visual, (B) under UV light λ 254 nm, (C) under UV light λ 365 nm and (D) after spraying with 10% Sulfuric acid solution in ethanol then heated at 110 °C.

Export to PPT

Close

Fig. 2

HPLC chromatograms of piceatannol, beta-carotene and gamma-tocopherol in the extract. (A) piceatannol in PSEE, (B) beta-carotene in PFWE and (C) gamma-tocopherol in PFWE.

Export to PPT

3.2 The level of total phenolic, flavonoid and carotenoid constituents in PFWE and PSEE

Table 2 illustrated the contents of total phenolic, total flavonoid and total carotenoid level which were presented in terms of gallic acid equivalent, quercetin equivalent and β-carotene equivalents per gram of extract, respectively. Our result demonstrated that the levels of total phenolics, flavonoids, and carotenoids in PSEE were higher than PFWE (Table 2).

3.3 Percentage of radical scavenging activity of PFWE and PSEE

Fig. 3 demonstrated the radical scavenging effects of PFWE, PSEE and vitamin C. The antioxidant activities of the extracts were compared with a reference antioxidant vitamin C. The PFWE and PSEE had radical scavenging effects with values 56.32 ± 2.03 and 47.47 ± 0.26%, repectively. Vitamin C exhibited the most potent radical scavenging effect (90.94 ± 0.33%) compared with the extracts (Fig. 3).

Close

Fig. 3

The effect of PFWE and PSEE on percentage of radical scavenging activity. Data shown are mean ± SEM of three replicates. * P < 0.05 compared with PFWE and PSEE, # P < 0.05 compared with PSEE.

Export to PPT

3.4 The inhibitory effects of PFWE and PSEE on HMG-CoA reductase, lipase, and cholesterol esterase activity

HMG-CoA reductase inhibitory activity of PFWE and PSEE was presented in Fig. 4A, compared to the pravastatin (a standard drug). PFWE displayed minimal HMG-CoA reductase inhibition, whereas PSEE showed no effect (Fig. 4A). Both PFWE and PSEE exhibited anti-lipase activity (IC₅₀ > 10 and 1.58 mg/mL, respectively) in a dose-dependent manner (Fig. 4B). PSEE were significantly higher inhibitory activity than PFWE. As shown in Fig. 4C, both PFWE and PSEE possess a significant anti-cholesterol esterase activity in a dose-dependent manner. The IC₅₀ values of PFWE and PSEE were 1.36 and 0.34 mg/mL, respectively.

Close

Fig. 4

The inhibitory effects of PFWE and PSEE on HMG-CoA reductase, pancreatic lipase and cholesterol esterase activity. (A) HMG-CoA reductase, (B) pancreatic lipase and (C) cholesterol esterase. Data shown are mean ± SEM of three replicates. * P < 0.05 compared with PFWE.

Export to PPT

3.5 Effect of PFWE and PSEE on RAW264.7 macrophage viability

The viability of RAW264.7 macrophages after receive to PFWE and PSEE for 24, 48, and 72 h were measured by the MTT assay. PFWE at 1–1000 µg/mL did not alter RAW264.7 macrophages viability after 24, 48 and 72 h of treatment (Fig. 5A and 5B). However, the exposure with PSEE (100–1000 µg/mL) for 48 and 72 h significantly suppressed RAW264.7 macrophages viability. These results indicated that PFWE and PSEE at 1–100 µg/mL are safe to the RAW264.7 macrophages at 24 h of treatment. Therefore, we selected the concentration of PFWE and PSEE at 1–100 µg/mL for further experiment.

Close

Fig. 5

Cytotoxicity of PFWE (A) and PSEE (B) on RAW264.7 macrophages. RAW264.7 macrophages were treated with 1–1000 μg/mL of PFWE and PSEE for 24–72 h. The viability RAW264.7 cell was detected by MTT assay. Data shown are mean ± SEM of four independent experiments. *P < 0.05 compared to the control.

Export to PPT

3.6 Effects of PFWE and PSEE on LPS-induced NO level in RAW264.7 macrophages

To explore the anti-inflammatory of PFWE and PSEE, LPS was utilized to activate the NO secretion into the medium of RAW264.7 macrophages. The exposure of LPS significantly enhanced NO secretion into the medium of RAW264.7 cells after 24 h. Pretreatment the cells with various doses of PFWE and PSEE prior to LPS administration demonstrated a markedly attenuated the augmentation of NO released into RAW264.7 medium. This indicated that PFWE and PSEE possess a potent anti-inflammatory activity against LPS-induced NO production (Fig. 6).

Close

Fig. 6

Effects of PFWE and PSEE on LPS-induced the secretion of nitric oxide in RAW264.7 macrophages. The cells were pre-treated with 25–100 μg/mL of PFWE and PSEE for 2 h and then treated with or without of LPS for 24 h. Data shown are mean ± SEM of four independent experiments. *P < 0.05 compared to the control. **P < 0.05 compared to the LPS alone group.

Export to PPT

3.7 Effects of PFWE and PSEE on vascular function of rat thoracic aorta

As shown in Table 3, PFWE and PSEE (10, 100, and 200 μg/mL) caused vasorelaxation in endothelium-intact aortic rings precontracted with PE in a dose-dependent manner. In addition, the maximum relaxation stimulated by the highest concentration of PFWE and PSEE (200 μg/mL) was 31 ± 5 and 86 ± 6%, respectively, indicating the vasorelaxant effect produced by PSEE was greater than those of PFWE.