An extract from Taxodium distichum targets hemagglutinin- and neuraminidase-related activities of influenza virus in vitro

Influenza virus remains an emerging virus and causes pandemics with high levels of fatality. After screening different plant extracts with potential anti-influenza activity, a water extract of Taxodium distichum stems (TDSWex) showed excellent activity against influenza viruses. The EC50 of TDSWex was 0.051 ± 0.024 mg/mL against influenza virus A/WSN/33. TDSWex had excellent antiviral efficacy against various strains of human influenza A and B viruses, particularly oseltamivir-resistant clinical isolates and a swine-origin influenza strain. We observed that the synthesis of viral RNA and protein were inhibited in the presence of TDSWex. The results of the time-of-addition assay suggested that TDSWex inhibited viral entry and budding. In the hemagglutination inhibition assay, TDSWex inhibited the hemagglutination of red blood cells, implying that the extract targeted hemagglutin-related functions such as viral entry. In the attachment and penetration assay, TDSWex showed antiviral activity with EC50s of 0.045 ± 0.026 and 0.012 ± 0.003 mg/mL, respectively. In addition, TDSWex blocked neuraminidase activity. We conclude that TDSWex has bimodal activities against both hemagglutinin and neuraminidase during viral replication.


Results
Antiviral activity and cytotoxicity of the TDSWex. First, we screened a panel of Chinese herbs and plants for their antiviral activity, using an assay to measure virus-induced cell death that shows whether a compound has potent antiviral activity. TDSWex revealed an EC 50 of 0.051 ± 0.024 mg/mL, a CC 50 of 0.287 ± 0.018 mg/mL, and a selectivity index (SI) of about 5.6 for inhibition of influenza A/WSN/33 virus in Madin-Darby canine kidney (MDCK) cells (Table 1). We demonstrated that the extract can protect cells from virus-induced cell death at a concentration of 0.1 mg/mL (Fig. 1). This concentration of TDSWex did not cause cytotoxicity in MDCK cells within 12, 24, or 72 h (row c of Figs 1 and S1a,b). When challenged by influenza A/WSN/33 virus in the absence of TDSWex, the MDCK cells, as observed under a microscope, appeared more rounded and detached from the dish, an effect defined as cytopathic effect (CPE) (row b, Fig. 1). The extract showed no cytotoxicity and inhibited the virus-induced CPE (rows c and d, Fig. 1). Therefore, we used 0.1 mg/mL TDSWex for the subsequent experiments in this study. This concentration was also below the CC50 for rhabdomyosarcoma (RD) cells (Table 1). TDSWex showed excellent anti-influenza virus activity. It could inhibit not only influenza A/H1N1 virus, but also influenza A/H3N2 virus, SOIV, and recent clinical strains of influenza, including oseltamivir-resistant strains ( Table 1). The raw dose/response curve of EC 50 , taking WSN as an example, is shown in Fig. S1c. Three influenza B viruses were tested, and TDSWex again showed anti-influenza virus activity (Table 1). However, it did not exhibit any activity against enteroviruses, adenovirus, or herpes simplex virus 1 (HSV-1), at concentrations up to 1.25 mg/mL (Table 1).

TDSWex does not suppress viral RNA or protein synthesis when added after viral adsorption.
Next, we assessed whether TDSWex blocked viral RNA and protein synthesis (Fig. 2). MDCK cells were infected with influenza A/WSN/33 virus (multiplicity of infection (MOI) = 0.01) for 1 h with or without TDSWex present during viral adsorption. After 1 h, the unbound virus was removed. The horizontal gray bars represent the time during which TDSWex was present. Alternatively, MDCK cells were treated with TDSWex after viral adsorption. The cells were harvested at the indicated time points for analysis using immunoblotting and reverse transcription-quantitative PCR (RT-qPCR) (Fig. 2). Viral RNA synthesis, as indicated by the RNA levels of M1, was suppressed when TDSWex treatment was present during -1 h to 0 h postinfection (p.i.), and the suppression became more noticeable at 12 h p.i. (Fig. 2a). However, viral RNA synthesis was not conspicuously different when MDCK cells were treated with TDSWex after viral adsorption (Fig. 2a) Fig. 2b).
TDSWex also inhibited the accumulation of viral proteins, as determined by western blot, when inoculated with 3 pfu of virus/cell in a single cycle of viral replication (lower panel of Fig. 2c). This result was confirmed by immunofluorescence microscopy analysis (Fig. 3). In the mock-infected control, no immunofluorescent foci of viral M1 or NP protein were observed, suggesting that the antibodies used are specific (row a, Fig. 3a,b). In the absence of TDSWex, M1 and NP proteins became apparent at 12 h p.i. (row b, Fig. 3a,b). The results were similar, in terms of viral protein synthesis and subcellular distribution, in cells treated with TDSWex during 0-12 h p.i.
(row e, Fig. 3a,b). In conclusion, synthesis of M1 and NP proteins was suppressed by TDSWex treatment during -1 h to 0 h p.i. but not during 0 h to 12 h p.i. (rows c and d, Fig. 3a,b). The ratio of M1-or NP-positive cells to the  The ratio of M1-or NP-positive cells to the total cell number was then determined. The data were presented as mean ± SD of the results of two independent experiments. ***p < 0.001 compared to virus only control.
total cell number was significantly reduced by TDSWex treatment during -1 h to 0 h p.i. but not during 0 h to 12 h p.i. (right panels of Fig. 3). Based on these results, we suggest that TDSWex blocks influenza virus replication by inhibition of the viral entry stage (-1 h to 0 h) and does not affect viral transcription, translation, or the subcellular localization of M1 or NP.

Inhibitory mechanism of TDSWex investigated by a time-of-addition assay.
To explore the mechanism of the antiviral effect of TDSWex, we used a time-of-addition assay to estimate the virus yields after TDSWex treatment was applied at different stages of viral replication (Fig. 4). The timing of the TDSWex treatment and viral infection is shown schematically in Fig. 4a. First, MDCK cells were preadsorbed with influenza virus during -1 h to 0 h p.i. on ice, and we harvested the culture supernatant at 12 h p.i. for a plaque assay. To assess whether TDSWex affected host cell surface, the extract was added before infection during the period from -3 h to -1 h (Fig. 4b). The viral yields in this sample showed no difference compared with the virus-only control (virus-only control vs. lane 1, Fig. 4b). When TDSWex was present at all stages of the viral replication cycle during the period from -3 h to 12 h p.i., the yields of progeny virus were markedly decreased (virus-only control vs. lane 3, Fig. 4b). When TDSWex was present during the period from -1 h to 0 h p.i. (during adsorption) or from 0 h to 12 h p.i. (after infection), the extract showed excellent protective ability (virus-only control vs. lane 2 and virus-only control vs. lane 4, Fig. 4b). We performed a time-of-addition assay with an MOI of 3. The results showed that TDSWex inhibited different stages of replication, similar to the findings obtained when an MOI of 0.01 was used (Fig. 4c). Based on these data, we suggested that TDSWex might act bimodally on both viral entry and release of progeny.
The bifunctionality of TDSWex prevents both viral entry and release of progeny. Before uptake of a virus particle by cellular endocytosis and its passage across the cell membrane, viral envelope protein HA must bind to receptors on host cells 17,29 . Therefore, we investigated whether TDSWex targeted the transmembrane viral envelope protein HA using a hemagglutination inhibition (HI) assay to detect the signature effect of HA binding to red blood cells (RBCs). In this assay, twofold serial dilutions of TDSWex were made from 0.125 mg/mL (lane 12, Fig. 5a) to 0.00024 mg/mL (lane 3, Fig. 5a). In the RBC-only control, the RBCs did not agglutinate (lane 1,   Fig. 5a). The higher concentrations of TDSWex caused hemolysis (lanes 7-12, lower row, Fig. 5a), but there was no hemolysis at concentrations of TDSWex between lanes 3-6 (lower row, Fig. 5a). Inhibition of hemagglutination was observed at TDSWex concentrations of 0.00195 mg/mL and 0.00098 mg/mL (lanes 5 and 6, upper row, Fig. 5a). This result Cells were precooled before infected with 3 × TCID 50 WSN on ice. After removing unbound virus, the infected cells were incubated with medium containing TDSWex at 37 °C for 1 h followed by acid wash to inactivate the virus on cell surface. The protection against cell death mediated by TDSWex was measured by MTT assay. The results are presented as mean viability relative to control ± SD from three independent experiments.
implies that suppression of viral entry by TDSWex was caused by blockage of the interaction between HA and the sialic acid receptor. In parallel, the viral particle inactivation assay result showed that both WSN and A/Puerto Rico/8/34 (PR8) strains were significantly inhibited by TDSWex, suggesting that TDSWex might target to and inactivate viral particles (Fig. 5b). We then used biochemical assays to address whether TDSWex blocks viral attachment to and penetration of the host cells in the early stage of viral replication (Fig. 5c,d). In the attachment assay, when MDCK cells on ice were challenged by influenza virus, TDSWex was present in the medium during the adsorption step. The cells were protected from the binding of influenza virus if TDSWex was present, and the EC 50 was 0.045 ± 0.026 mg/mL (Fig. 5c). In the penetration assay, influenza virus was initially bound to the cellular surface on ice. Unbound virus was removed, followed by endocytosis of the virus at 37 °C in the presence or absence of TDSWex. TDSWex effectively inhibited viral penetration with an EC 50 of 0.012 ± 0.003 mg/mL (Fig. 5d). These results demonstrate that the mechanism of the antiviral effect of TDSWex is inhibition of the early steps of viral replication during endocytosis, e.g., the disruption of the virus-receptor interaction or of the fusion of the virus with the endosomal membrane.
To explore whether TDSWex inhibits virus replication through targeting NA activity, we performed NA assay to address this concern. The virus-only control showed NA activity, but the activity was reduced to baseline levels in the mock-treated control and after zanamivir treatment, demonstrating that the enzymatic activity of NA was specific (lanes 1, 2, and 4, Fig. 6a). The NA activity was significantly decreased by about 23% when TDSWex was added at 0.1 mg/mL (lane 6, Fig. 6a). However, lower concentrations of TDSWex (0.01 mg/mL and 0.001 mg/mL, lanes 8 and 10, Fig. 6a) did not inhibit NA activity. Based on these results, we suggest that TDSWex can inhibit NA activity. To confirm whether the antiviral effect of TDSWex correlated with its effect on NA function, we performed a plaque reduction assay 16,30 . The plaque numbers were significantly reduced by about 36% and 32%, respectively, when influenza viruses were treated with 0.1 mg/mL or 0.05 mg/mL of TDSWex during viral replication (#1 vs. #2 and #1 vs. #3, Fig. 6b,c). However, no inhibition of virus yield was detected at lower concentrations of TDSWex (#4 to #6, Fig. 6b), indicating that its inhibitory effect on plaque quantity was dose dependent. Overall, we demonstrated that TDSWex functions bimodally against influenza virus by inhibition of both viral entry and NA activity.

Discussion
Influenza viruses cause heavy morbidity and mortality in pandemics 9 . The emergence of viruses resistant to current therapeutic regimens has raised anxieties about new outbreaks in near future 22 . Therefore, it is crucial to develop new anti-influenza drugs. T. distichum is a coniferous tree whose biological activities have been rarely evaluated 31 . T. distichum cones showed antifungal and antitermitic activities 32,33 , but whether other parts of the plant have antiviral activity is unknown. We screened the extracts of cones, leaves and stems of T. distichum for anti-influenza virus activity, and the stem exhibited the best effectiveness against influenza A virus (data not shown). TDSWex was thus purified from the stems of T. distichum for our studies.
TDSWex showed inhibitory activity against a broad spectrum of influenza A (including SOIV and resistant strains) and B viruses, but did not show antiviral activity against enteroviruses, adenovirus, or HSV-1, at concentrations up to 1.25 mg/mL (Table 1). A time-of-addition assay was utilized to explore the mechanism of the antiviral effect of TDSWex. The extract did not suppress viral replication when added before virus adsorption (lane 1, Fig. 4b), but we found that it had similar antiviral ability whether cells were treated during or after infection (lanes 2 and 4, Fig. 4b). We further demonstrated that the extract prevented viral binding and NA-related release of viral progeny.
The process of viral replication involves entry, transcription, translation, assembly, and budding 9 . In the time-of-addition assay, TDSWex had inhibitory activity throughout all these stages. First, the extract protected cells against the adsorption of influenza virus (lane 2, Fig. 4b). We used HI, attachment, and penetration assays to confirm that TDSWex inhibited viral entry. Therefore, the reduction in viral RNA and protein synthesis was the result of reduced viral entry (Fig. 2a, lane 5 of Fig. 2b, row d of Fig. 3a,b). In the HI assay, the RBCs did not display influenza virus-induced hemagglutination in the presence of TDSWex (lanes 5 and 6, Fig. 5a). Biochemical assays indicated that TDSWex inhibits both viral adsorption and viral penetration (Fig. 5b-d). We also demonstrated that TDSWex might inactivate both A/WSN and PR8 when applied together in the viral particle inactivation assay, suggesting that the mode of inhibition of TDSWex also applies to other strains of the influenza virus (Fig. 5b). These results imply that TDSWex might disrupt the interaction of HA and the cellular sialic acid receptor, or cause the viral particle to fail to fuse with the endosomal membrane. TDSWex treatment after infection (viral adsorption) also had antiviral activity (lane 4, Fig. 4b). However, because the synthesis of viral RNA and proteins was not significantly affected and there was no difference in their subcellular localization (Fig. 2a, lane 4 of Fig. 2b, row e of Fig. 3a,b), we examined the final stage of viral replication. The release of progeny virions from the host cell surface is dependent on the cleavage of sialic acid by NA 16 . Recent studies demonstrated that NA activity is also involved in viral entry [18][19][20]34 , because if sialic acid is removed, the influenza virus can penetrate host cells 34 . We first demonstrated that TDSWex blocked viral penetration (Fig. 5). In the NA assay, TDSWex treatment reduced the NA activity (lane 6, Fig. 6a). Next, we showed that viral plaque numbers were diminished when cells were treated with TDSWex (Fig. 6b). Based on these results, the extract may restrict both viral entry and release of progeny by inhibition of NA activity. Overall, we suggest that TDSWex possesses bimodal activity against influenza virus for inhibition of viral entry and release of progeny.
There have been reports concerning the chemical composition and biological activities of T. distichum. Taxodione and taxodone are two potent antitumor agents 35,36 . Shikimic acid is one component of the T. distichum extract and is classified as a carbocyclic sialic acid analogue 37,38 that is potentially able to inhibit NA function 38  and influenza virus (including H1N1 and H5N1) [39][40][41][42][43][44][45][46] . Quercetin exhibited protective efficacy during influenza virus induced-oxidative stress by restoring the concentrations of antioxidants including superoxide dismutase 47,48 . In a previous study, quercetin was found to have an inhibitory effect on NA activity 40 . Quercetin also shows significant inhibitory effects on viral translation 49 . However, our results showed that the inhibitory effect of TDSWex was mediated by its ability to block the entry and release of virus, and that it showed little effect on viral protein synthesis. This suggests that the level of quercetin in T. distichum may be too low to exert an antiviral effect.
In addition, the HPLC fingerprint of TDSWex was established, and compound 4 was identified as the chemical reference with a retention time (rt) of 55.4 min (Fig. S4). The analytic method validation of 4 was carried out, and we obtained a calibration curve with a good linearity ranging from 2.5 to 7.5 μ g/mL (Fig. S5). The regression equation and correlation coefficient were y = 92524x -6410 and r 2 = 0.9993, respectively. The validation data revealed the limit of detection (LOD) to be 1.25 μ g/mL, and this was calculated as 13.12 ± 0.89 times the signal to noise ratio (S/N) for the lowest concentration. In addition, the percentage coefficient of variation (%CV) of compound 4 at the LOD was 3.84 while its precision at five different concentrations (7.5, 6.25, 5.0, 3.75, and 2.5 μ g/mL) varied between 1.20 and 4.48%. Furthermore, the recovery of compound 4 at 3.0, 5.0, and 7.0 μ g/mL was 2.94 ± 0.03, 5.03 ± 0.05, and 6.99 ± 0.03 μ g/mL, respectively. These results suggest that the analytic method for compound 4 was reproducible and accurate. Accordingly, the content of compound 4 (2.75 ± 0.03 mg/g, %CV = 2.17) in TDW was determined using HPLC analysis.
Preparation of the TDSWex. The stems of T. distichum Rich. were collected from Taipei, Taiwan. The stems (3000 g) were extracted with ethanol (15 L × 2) at 70 °C for 4 h. After evaporation of the solvent in vacuo, the concentrated ethanol extract was partitioned between ethyl acetate and water to give ethyl acetate and water extracts. To prepare the stock of the water-soluble extract of T. distichum for assay, 200 mg of the dry water-soluble extract of T. distichum stem was first dissolved in 1 mL DMSO and incubated at 37 °C in a water bath for 16 h. After sedimentation at 3000 × g for 10 min at 10 °C, the supernatant of the extract was collected and subjected to one more round of centrifugation at 27,000 × g for 30 min. Finally, the aqueous TDSWex was obtained as the supernatant. TDSWex was characterized by HPLC fingerprint and the content of the marker compounds was quantified as described in Supplementary Information. The dried water-soluble extract of T. distichum stem used in the present study was recorded and will be stored for 10 years. The voucher specimen (TDSWex) used in the present study was deposited in the herbarium of Chang Gung University, Taoyuan, Taiwan. Cytotoxicity test by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Overnight cell culture was incubated with various concentrations of TDSWex in E0 (DMEM containing 100 U/mL penicillin, 0.1 mg/mL streptomycin, 2 mM L-glutamine, 0.1 mM NEAA mixture, and 2.5 μ g/mL trypsin) at 37 °C for 12 h or 72 h. The cells were then stained with rinsed MTT as previously described 20 . The 50% cytotoxic concentration (CC 50 ) calculated by Reed-Muench method was defined as the concentration of TDSWex that caused death of 50% of the cells. for 1 h in E0 with or without TDSWex (0.1 mg/mL), and were harvested at indicated time p.i. Total RNA extraction and RT-qPCR were performed as previously described 30 . To quantify changes in gene expression, the Δ Ct method was used to calculate the relative changes normalized against GAPDH. Ct is defined as the cycle where fluorescence is determined to be significantly greater than the background. The ratio of vRNA to the internal control was normalized to the 0-h p.i. control level, which was set arbitrarily to 1.0 Scientific RepoRts | 6:36015 | DOI: 10.1038/srep36015 Inhibition of viral protein synthesis by TDSWex by western immunoblotting. MDCK cells in six-well plates (5 × 10 5 cells/well) were infected with indicated MOI of influenza virus A/WSN/33 at 37 °C for 1 h and then washed twice with Hank's balanced salt solution (HBSS). The cells were treated with or without addition of TDSWex (0.1 mg/mL) in E0 and harvested at the indicated time points for western immunoblotting as previously described 30 . The antibodies against virus M1 antibody (catalogue #1321) were from ViroStat (Portland, ME). The antibody against GAPDH, as an internal control, was from Santa Cruz Biotechnology (SC25778, Santa Cruz, CA). The enhanced chemiluminescence western blotting detection system (Millipore, Billerica, MA) was used for color development.
Indirect immunofluorescence assay. MDCK cells (1 × 10 5 ) on the coverslips were infected or mock-infected with influenza virus A/WSN/33 at MOI = 0.01. After the cells were washed with HBSS, TDSWex (0.1 mg/mL) in E0 was added and the cells were harvested for fixation with 4% PFA for 1 h at room temperature. Permeabilization was done with 0.5% Triton X-100 in phosphate-buffered saline (PBS) for 5 min, followed by blocking with 0.5% bovine serum albumin in PBS for 1 h. The cells were incubation with anti-M1 (catalogue #1321, ViroStat) and anti-NP antibodies (catlogue #ab20343, Abcam, Cambridge, UK) for 1 h. The cell nuclei were stained with 4′ ,6-diamidino-2-phenylindole (DAPI). Fluorescence of the Alexa-Fluor-488-labeled secondary antibodies and DAPI was detected with a Zeiss Axiovert 200 M microscope with a 100× oil-immersion objective lens.
Time-of-addition assay. MDCK cells (5 × 10 5 cells/well) were seeded in six-well plates and then incubated overnight before challenged with influenza virus A/WSN/33 (MOI = 0.01) at -1 h p.i. and kept on ice for 1 h. After viral adsorption, the unbound virus was removed by washing with HBSS. TDSWex (0.1 mg/mL) in E0 was added at -3 to -1 h (preadsorption), -1 to 0 h (adsorption), 0 to 12 h, and -3 to 12 h. All the supernatants were collected at 12 h p.i. for titer determination by plaque assay 30 .

Viral hemagglutination (HAv) assay and hemagglutination inhibition (HI) assay. Serial twofold
diluted influenza A/WSN/33 were made in PBS and mixed with 2× volume guinea pig RBCs in round-bottom 96-well plates for 1 h. The lowest virus titer that caused RBC aggregation was defined as 1 × HAv, and 4 × HAv was used for the HI assay. In the HI assay, different concentrations of TDSWex were incubated with virus on ice for 1 h. This mixture was then incubated with RBCs for 1 h. Attachment assay. This assay was previously described with minor modification 30,57 . Briefly, MDCK cells were incubated on ice for 20 min before infected with 3 × TCID 50 of WSN in the presence of different concentrations of TDSWex on ice for 1 h. The medium containing the nonadsorbed virus was removed and the cells were then washed twice with HBSS. The cells were maintained in E0 and cell viability was determined with an MTT assay after incubation for 72 h.
Penetration assay. This penetration assay was modified from previous reports 30,58,59 . MDCK cells were chilled on ice for 20 min before infected with 3 × TCID 50 of influenza virus A/WSN/33 on ice for another 30 min. Different concentrations of TDSWex in E0 were added and the cells were incubated at 37 °C for 1 h. The infected cells were then treated with HBSS (pH 2) for 1 min to inactivate non-penetrated viruses before the addition of HBSS (pH 11) to neutralize the acidic buffer. After removing HBSS, E0 was added and the cells were incubated at 37 °C for 72 h. The effect of TDSWex was determined by the degree of cell viability through MTT assay. Viral particle inactivation assay. The WSN and PR8 viruses (5 × 10 5 pfu/mL) were incubated with DMSO or 0.1 mg/mL TDSWex for 1 h on ice. The resultant mixtures were diluted up to 30-fold (which was far below the EC 50 ) with DMEM for titer determination using a plaque assay. Plaque reduction assay. MDCK cells (5 × 10 5 cells/well) were seeded in six-well plates before infected with influenza virus A/WSN/33 (about 40 plaque-forming units [pfu]) and kept on ice for 1 h. After washing with HBSS, the cells were maintained in indicated concentrations of TDSWex in E0 containing 0.3% agarose. After incubation at 37 °C under 5% CO 2 for 72 h, the medium was removed and the cells were fixed with 4% PFA for 1 h at room temperature, followed by staining with 0.1% crystal violet for 20 min at room temperature. Statistical analysis. All data were presented as the mean or mean ± SEM (standard error of the mean) of three individual experiments. The statistical significance of the difference between two groups was analysed by one-way analysis of variance (ANOVA). A p value of < 0.05 was considered significant. The actual p value was marked in the figures, unless the p value was under 0.001, in which case it was expressed as ***.