The fresh aerial parts of sea buckthorn (H. rhamnoides L.) were collected from the Karakoram-Indus valley of Pheh, Ladakh (3180 m above sea level), India in late July, 2018 (Fig. 1, upper panel). The plant was identified by a local herbalist-Amchi practitioner (Leh, Ladakh), and further validated by an expert taxonomist, Dr Tariq Husain (National Botanical Research Institute, Lucknow, India).

Analytical grade solvents (Sigma-Aldrich, Merck KGaA) were used for the extraction and fractionation of sea buckthorn, using standard methods. Briefly, the air-dried powder of the aerial parts (410 g) of sea buckthorn were extracted with 90% ethanol (3x250 ml) at room temperature with continuous shaking till exhaustion. The combined alcoholic-extracts were filtered (Whatman paper no. 1) and concentrated under reduced pressure at 40˚C, using rotary evaporator (Buchi Rotavapor; Model R-215, Thermo Fisher Scientific, Inc.). The obtained dried ethanol-extract (198.7 g) was dissolved in water and successively fractionated with chloroform (SB-Chl; 4.12 g), ethyl acetate (SB-Eac; 2.1 g) and finally with n-butanol (SB-But; 62.9 g), along with the aqueous part (SB-Aqu; 122.4 g) to yield the corresponding fractions. For each fraction, solvents were completely evaporated to dryness, and maintained at 4^˚C until analysis.

Based on the published literature, the sea buckthorn extract was subjected to HPLC analysis for the presence of the antiviral flavonoids: Quercetin, kaempferol and isorhamnetin (34,36). The Alliance chromatographic system (Waters Instruments, Inc.) was used for HPLC analysis equipped with built-in quaternary pump, dual wavelength absorbance detectors and autosampler at 25˚C. For separation, an Agilent HC-C18 column (5 µm, 250x4.6 mm) was maintained with the gradient flow rate (1 ml/min) of the mobile phase (A, acidified water:1% acetic acid, v/v; and B, methanol:acetonitrile, 80:20, v/v) and peaks were detected at 370 nm. The gradient was programmed as follows: 0-9 min, 0-35% B; 9-13 min, 35-60% B; 13-18 min, 60-80% B; 18-30 min, 80-90% B. For obtaining the calibration curve, standard stock solutions of quercetin, kaempferol, and isorhamnetin were prepared in methanol (0.5 mg/ml). Furthermore, gradient concentrations (0.01-100.0 µg/ml) of the three compounds were prepared by serial dilution for identifying and quantifying them in the sea buckthorn extract.

Two human hepatoma cell lines [HepG2 and its derivative, HepG2.2.15 (HBV-reporter cells)] were generously provided by Dr Shahid Jameel (International Center for Genetic Engineering and Biotechnology, New Delhi, India). The cells were cultured in T25 flasks (BD Biosciences, San Jose, CA, USA) using DMEM reconstituted with 10% bovine calf serum and 1X penicillin-streptomycin mix (all from Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) at 37^˚C with a 5% CO₂ supply. The anti-HBV drug lamivudine, as well as natural quercetin, kaempferol and isorhamnetin (all from Sigma-Aldrich, Merck KGaA) were procured.

The four sea buckthorn fractions (SB-But, SB-Chl, SB-Eac and SB-Aqu; 3 mg/ml each) were dissolved in 100 µl dimethyl sulfoxide (DMSO; Sigma-Aldrich, Merck KGaA) and reconstituted in DMEM to yield various working concentrations (50, 100 and 150 µg/ml). Likewise, isorhamnetin, quercetin and kaempferol (0.5 mg, each) were first dissolved in 50 µl DMSO and reconstituted in DMEM (1 mg/ml, final). HepG2 cells (0.5x10⁵/100 µl/well) in a 96-well cell culture plate (BD Biosciences) were grown overnight and treated with the sea buckthorn fractions (in triplicate) the following day after an incubation period of 72 at 37˚C h. DMSO (0.1%) served as a vehicle or untreated control. The cytotoxicity induced by the sea buckthorn fractions, if any, was evaluated using MTT assay (TACS MTT Cell Proliferation and Viability Assay kit, Tervigen; Thermo Fisher Scientific, Inc.) according to instructions provided with the kit. The optical density (OD) of each sample was recorded using a microplate reader (ELx800; BioTek Instruments, Inc.). Non-linear regression analysis was performed using Excel software (2010; Microsoft Corporation) to estimate the 50% maximal cytotoxic concentration (CC₅₀) in relation to the untreated control. The experiment was repeated twice under the same conditions for reproducibility.

To assess the inhibitory effects of the sea buckthorn fractions on HBsAg expression levels, the HepG2.2.15 cells were seeded (0.5x10⁵ cells/100 µl/well) in a 96-well plate and grown overnight. The cells were treated with the non-cytotoxic SB-But, SB-Eac and SB-Aqu fractions (25, 50 and 100 µg/ml, each) as well as with isorhamnetin (10, 20 and 50 µg/ml), quercetin (10 µg/ml) and kaempferol (10 µg/ml). Lamivudine (2 µM) served as the standard, whereas DMSO (0.1%) functioned as a negative or untreated control. Following incubation at 37^˚C, culture supernatants were collected on days 1, 3 and 5, and stored at -20˚C for further analysis. HBsAg production was quantified using the Monolisa HBsAg ULTRA Elisa kit (Bio-Rad Laboratories, Inc.) according to the instructions provided with the kit. The OD (λ=570 nm) of the samples was recorded and analyzed in relation to the untreated control, and non-linear regression was performed using Excel software (2010; Microsoft Corporation) to estimate percent inhibitions. All samples were tested in triplicate and the experiment was repeated twice.

Based on the HBsAg inhibition data, the SB-But, SB-Eac and SB-Aqu fractions (50 µg/ml, each) as well as IRM, QRC and KMP (10 µg/ml, each) were further tested for their time-course inhibitory effects on HBeAg synthesis in HepG2.2.15 cells. The secretion of HBeAg was analyzed using the HBeAg/Anti-HBe Elisa kit (DIAsource ImmunoAssays, Belgium) according to the instructions provided with the kit. The OD (λ=570 nm) of the samples was recorded and analyzed in relation to the untreated control, and non-linear regression was performed to determine percent inhibitions. All samples were tested in triplicate and the experiment was repeated twice.

All data from all the analyzed samples (in triplicate) are expressed as the mean ± SEM. In a set of data, the determination of total variation was performed using one-way analysis of variance (ANOVA), following the Dunnett's test. Statistical analyses were performed using Excel software (2010; Microsoft Corporation). A value of P<0.01 was considered to indicate a statistically significant difference.

HPLC analysis of the sea buckthorn extract identified three antiviral flavonols isorhamnetin, quercetin and kaempferol (Fig. 2), and were quantified to be 138.75, 93.09 and 44.19 µg/g, respectively. This was in line with the previous demonstration of these flavonoids in the phenol-rich fraction of sea buckthorn leaves (34). Flavonols are natural flavonoids with variations in their heterocyclic (C₆-C₃-C₆) carbon ring (36). The detected flavonols are structurally similar in having hydroxyl groups at same positions on their A and B rings; however, isorhamnetin has an additional 3'-methyl group on the B ring as compared to quercetin and kaempferol (Fig. 1, lower panel).

The present study first examined the cytotoxic effects of the SB-But, SB-Chl, SB-Eac and SB-Aqu fractions on HepG2 cells. Of these, while SB-Chl exhibited significantly high cytotoxicity (CC₅₀, 32.58 µg/ml; P<0.01), SB-Eac, SB-But and SB-Aqu exhibited non-toxicity at concentrations of up to 150 µg/ml dose as compared to the untreated control (Fig. 3). Notably, HepG2 and HepG2.2.15 cells are biologically and physiologically the same, apart from the fact that HepG2.2.15 cells allow for HBV DNA replication and gene expression. In addition, while the HepG2 cells are widely distributed and easily available, the accessibility of HepG2.2.15 cells, even commercially, is very limited worldwide (37). In view of this, in the present study, MTT assay was performed on HepG2 cells, whereas the HepG2.2.15 cells were only used for anti-HBV assays.

The concentration-dependent analysis of the SB-Eac, SB-But and SB-Aqu fractions revealed marked inhibitory effects on HBsAg production at 50 and 100 µg/ml (P<0.01) in relation to the untreated control on 48 h (Fig. 4, left panel). As no further considerable enhanced effect was observed at 100 µg/ml when compared to 50 µg/ml, the 50 µg/ml concentration was selected for time-course analysis. In the time-course assay of the fractions (50 µg/ml), the optimal inhibition of HBsAg was observed by SB-Eac (~42.6%; P<0.01) followed by SB-But (~37.5%) and AB-Aqu (~36.2%) in relation to the untreated control on day 5 (Fig. 4, right panel). The concentration- and time-dependent analysis of isorhamnetin revealed its optimal, but comparative HBsAg inhibitory activity at the 10 and 20 µg/ml concentrations on day 5 (Fig. 5). At the selected 10 µg/ml concentration, isorhamnetin suppressed HBsAg levels by ~30.5% as compared to the high activities of quercetin (~67.5%; P<0.01) and kaempferol (~62.3%; P<0.01) in relation to the untreated control on day 5 (Fig. 5). The reference drug lamivudine (2 mM) suppressed HBsAg synthesis by ~87.4% (P<0.001) in relation to the untreated control on day 5.

As HBeAg production is a serological hallmark of active viral DNA replication (18), the present study further tested the three fractions (50 µg/ml) and the flavonoids (10 µg/ml) for their effects on HBeAg synthesis. SB-Eac suppressed HBeAg production by ~43.2% in relation to the untreated control on day 5. In addition, SB-But suppressed this by ~35.5% and AB-Aqu by ~32.5% in relation to the untreated control on day 5 (Fig. 6). Isorhamnetin decreased HBeAg production by ~28.4%, whereas quercetin and kaempferol inhibited this by ~64.4% (P<0.01) and ~60.2% (P<0.01), respectively in relation to the untreated control (Fig. 6). Lamivudine (2 mM) suppressed HBsAg production by ~83.5% (P<0.001) in relation to the untreated control on day 5. In a previous study by the authors, the assessment of the synergistic effects of quercetin with other flavonoids, such as rutin and hesperidin revealed further suppressions of HBsAg and HBeAg by ~10% in HepG2.2.15 cells (28). Thus, the present study did not examine the synergistic effects of quercetin with kaempferol or isorhamnetin.

Of note, although isorhamnetin was quantified as the most abundant flavonol in the sea buckthorn extract, it exhibited low anti-HBV activity, as compared to that observed for quercetin and kaempferol (Figs. 5 and 6). As isorhamnetin has an additional 3'-methyl group on its B ring as compared to quercetin and kaempferol (Fig. 1, lower panel), the structure-activity association may be attributed to its weak antiviral activity. In contrast to the present data, in a previous comparative antiviral study of structurally-related flavonoids, isorhamnetin was reported to have the most potent activity against influenza virus, which is an RNA virus (38). Notably, HBV is a DNA virus that has different replication mechanisms than RNA viruses. Nonetheless, HBV has a unique replication mechanism similar to HSV and HIV, the retroviruses. Therefore, the majority of the anti-HSV and anti-HIV drugs are also potent anti-HBV drugs. In previous studies, while sea buckthorn leaf extract was reported for anti-DNV activity (32), the isolated compound, hiporamin, was found to exert potent antiviral effects against influenza virus, HSV and HIV (34). Of the three antiviral flavonoids identified in its leaves (34), quercetin and kaempferol, as well as their derivatives have been demonstrated to have marked HBsAg and HBeAg inhibitory potentials in HepG2.2.15 cells (27-30). Notably, although the broad antiviral potential of isorhamnetin remains unexplored, a previous study reported its in vitro activity against the influenza virus (37). Moreover, to further obtain insight into the association between the chemical structures and anti-HBV activities, molecular docking was previously performed for quercetin, kaempferol and lamivudine with HBV-polymerase in previous studies by the authors (28,29). As in the present study, isorhamnetin exhibited a comparatively low anti-HBV activity, molecular docking was not performed.

Notably, based on the safe usages of sea buckthorn in traditional medicine, several clinical studies have revealed the therapeutic effects of sea buckthorn against ulcerative stomatitis (39), chronic cervicitis and (40) and atopic dermatitis (41). In another study, sea buckthorn juice was reported to be beneficial in reducing the risk factors for coronary heart disease, possibly due to its high anti-oxidant property (42), Moreover, when used in cirrhotic patients, sea buckthorn was found to prevent the progression of liver fibrosis (8). In view of this, as well as in the absence of a primate model of chronic HBV infection, these data suggest that it is worthy of further clinical assessment.

In conclusion, the present study demonstrates that the in vitro anti-HBV potential of sea buckthorn is attributed to its well-known antiviral flavonols, quercetin and kaempferol. Notably, although with a comparatively minimal effect, isorhamnetin was found to exhibit anti-HBV activity, for the first time, at least to the best of our knowledge. Nonetheless, further molecular and pharmacological studies are warranted in order to validate and develop anti-HBV therapeutics.