Extract of Scutellaria baicalensis inhibits dengue virus replication

Identification and quantitation of the active compound, baicalein, in AHPE-XA-09 was performed by liquid chromatography-mass spectrometry-mass spectrometry (LC/MS/MS). Briefly, the concentration of pure baicalein ranging from 0.5 μg/mL to 20 μg/mL (Indofine Chemical Company Inc., New Jersey, USA) was dissolved in methanol (BDH Laboratory Supplies, England) and injected (10 μl) into the Applied Biosystems 3200 LC/MS/MS System (California, USA). Compound separation was achieved on a Gemini C6-Phenyl 110A column (150 mm×2.00 mm, particle diameter of 5 μm) (Phenomenex, USA) with a mobile phase composed of 0.1% formic acid (BDH Laboratory Supplies, England): acetonitrile (BDH Laboratory Supplies, England) at a flow rate of 0.50 mL/min. The transition ions and retention times were used as a detection point for baicalein peak. The transition ions were monitored in the multiple reaction monitoring modes (MRM). The area under the curve based on the different intensity height in the chromatogram of the different concentrations of baicalein was used to plot a standard linear regression curve. The AHPE-XA-09 (100 μg/mL lyophilized powder) was dissolved in methanol, filtrated through 0.45 μm PVDF filter (Whatman, USA) and was subsequently injected into the LC/MS/MS System. Inverse prediction method for the best fit linear regression curve generated from the pure baicalein was used for quantitation of the concentration of baicalein in AHPE-XA-09.

C6/36 mosquito cell line was used for the propagation of all DENV isolates used in the investigation [16]. Vero (African green monkey kidney) cell line was used for the evaluation of antiviral activity as previously described [18]. The cell lines were maintained and propagated in EMEM (Gibco, NY, USA) containing 10% fetal bovine serum (FBS, Gibco, NY, USA). The C6/36 and Vero cells were incubated at 28°C and 37°C respectively, in the presence of 5% CO₂. At the time of virus inoculation and antiviral assays, the concentration of FBS was reduced to 2%. We used four different clinical DENV isolates representing the four serotypes of DENV (DENV-1, DENV-2, DENV-3 and DENV-4) in current investigation which identified by our group using full genome sequencing method. All the four clinical isolates propagated and maintained as previously described [17]. After titration of the virus isolates, the stocks were stored at −70°C until further use in the experiments.

The cytotoxic potential of AHPE-XA-09 extract against Vero cells was determined using the MTT assay as described previously [19]. Briefly, the confluent Vero cells in 96-well microplates were treated with different concentrations of AHPE-XA-09 in triplicates. The treated cells were incubated with the extract for four days, which was similar to the time period for the antiviral assay at 37°C followed by the addition of 15 μl of MTT solution to each well. The microplate was further incubated at 37°C for four hours and subsequently a solublization/stop solution was added. The optical density (OD) of all the wells including non-treated cells was read using a plate reader (TECAN, Mannendorf, Switzerland) at 570 nm. The cytotoxicity of AHPE-XA-09 was calculated using Graph Pad Prism 5 (Graph Pad Software Inc., San Diego, CA).

The effect of AHPE-XA-09 on intracellular replication of DENV was examined in Vero cells plated on 24-well cell culture microplate. After attaining 80% confluency, virus inoculum consisting of 200 FFU of each DENV serotype was added to each well. Viruses were allowed to adsorb to the cells for 1 hour at 37°C. Unadsorbed viruses were removed by rinsing the cells with sterile PBS twice. Different concentrations of AHPE-XA-09 were mixed with 1.5% carboxymethylcellulose (CMC) containing a cell-growth medium supplemented with 2% FBS and the plates were incubated at 37°C for four days. DENV foci were visualized as described previously [19]. The number of foci formed was expressed as foci-forming unit (FFU). The antiviral effects of AHPE-XA-09 was measured by calculating the percentage of foci reduction (% RF) against the controls maintained in parallel using the following formula [16]; RF (%) = (C-T) × 100/C, where, C is the mean of the number of foci from triplicates without AHPE-XA-09 added and T is the mean of the number of foci from triplicates of each treatment with AHPE-XA-09 extract.

The potential direct virucidal effects of AHPE-XA-09 on DENV were evaluated by treating the viral suspension with 200 FFU of DENV with increasing concentrations of AHPE-XA-09 for 2 h at 37°C. The confluent Vero cells in 24-well plate were infected with the AHPE-XA-09-treated viral suspensions. After 1 h of adsorption at 37°C, the cells were washed twice with PBS and overlaid with 1.5% CMC-containing cell culture medium. The microplate was incubated for 4 days in a humidified 37°C incubator in the presence of 5% CO₂. The viral foci were visualized as described above.

AHPE-XA-09 effects on DENV RNA production was determined using qRT-PCR as previously described [22, 23]. Different primer sets for detection of the 4 different DENV serotypes were used based on previous designed primers [22, 23] with some minor modifications (Table 1). This experiment was undertaken to confirm the foci forming assay results.

The half maximal cytotoxicity concentration (CC₅₀) and half maximal inhibitory concentration (IC₅₀) were used as the main parameters in this investigation. Selectivity index value (SI) was determined as the ratio of CC₅₀/IC₅₀ of the plant extract. GraphPad PRISM for Windows, version 5 (GraphPad Software Inc., San Diego, CA, 2005) was used for all statistical analyses.

The toxicity of AHPE-XA-09 extract was measured against Vero cells using the MTT assay. We found that 72 ± 2.9% of the cells were viable after 4 days of treatment with AHPE-XA-09 at concentration 375 μg/mL. Also, we found that in presence of 187.5 μg/mL the viability of treated cells was 81.4 ± 1.3. The half maximal cytotoxic concentration (CC₅₀) value for AHPE-XA-09 was at 912.6 μg/mL (Figure 1). This suggests that AHPE-XA-09 at concentrations <375 μg/mL is in general non-cytotoxic and could be well-tolerated by Vero cells.

The foci-forming unit reduction assay (FFURA) together with qRT-PCR, were used to evaluate the in vitro anti-dengue activity of the aqueous extract of S.baicalensis. The extract was evaluated for its i) protective prophylactic effect ii) activity against virus adsorption to cells, iii) ability to inhibit replication after virus adsorption to the cells and iv) direct virucidal effect. The S. baicalensis extract, AHPE-XA-09, exhibited a dose-dependent inhibition effect against all the four DENV serotypes in Vero cells with IC₅₀ values as presented in Table 2. It was demonstrated that the plant extract showed antiviral activity against the different DENV serotypes used at all stages of in vitro replication, although the most significant effects were observed when the extract was evaluated during the time of virus adsorption (Table 2).

AHPE-XA-09 exhibited prophylactic effect against DENV replication in Vero cells, with the IC₅₀ values ranging from 269.9 to 369.8 μg/mL for all the four DENV serotypes tested (Table 2, Figure 2A). The SI values for all the four DENV serotypes ranged between 2.47 and 3.38 (DENV-1, SI= 3.4; DENV-2, SI= 2.5; DENV-3, SI= 2.8; DENV-4, SI= 2.6). When the cells when pre-treated with the extract at a concentration of 750 μg/mL, the RNA levels in all the DENV serotypes decreased by >70% as compared to untreated cells (Figure 2B). These observations could be due to the uptake of the plant extract by the cells during pre-treatment and their potential activity against the intracellular replication of DENVs and/or could be related to the masking of the viral receptors on the cell surface due to competitive binding of the bioactives present in the plant extract. Nonetheless, the anti-DENV activity of AHPE-XA-09 via prophylactic treatment in Vero cells was relatively less significant as compared to the other treatments (Table 2). Therefore, it could be further improved probably by optimizing the time of exposure and dose for effective pre-treatment.

The antiviral activity of AHPE-XA-09 against DENV was found to be the most effective as it acts by inhibiting the virus adsorption to cells. The IC₅₀ values ranged from 56.02 to 77.41 μg/mL for all the four DENV serotypes studied (Table 2, Figure 3A) with SI values ranged from 11.7 to 16.6 (DENV-1, SI= 13.2; DENV-2, SI= 16.3; DENV-3, SI= 11.8 and DENV-4 SI= 12.4). Interestingly, we also found that the DENV viral RNA reduced by ~68% to 82% at 187 μg/mL of AHPE-XA-09 added during the viral adsorption period (Figure 3B). This could be associated with the direct virucidal activity of the plant extract (Table 2), which possibly could have decreased the number of infectious DENVs particles during the adsorption time and/or binding to the cognate cellular receptors for DENVs. However, interfering with the virus attachment to the cells via binding to cell surface receptors and/or virus ligand could be another explanation for the anti-adsorption activity of the extract, which warrants further investigations.

Next, we set out to investigate the intracellular anti-dengue credentials of the plant extract DENV-infected Vero cells. Following the addition of AHPE-Xa-09 subsequent to DENV infection of Vero cells, we noticed an inhibition of DENV replication with IC₅₀ values ranging from 86.59 to 95.19 μg/mL for the different DENV serotypes examined (Table 2, Figure 4A). The extract in this experiment showed SI values ranging from 9.5 to 10.5 (DENV-1, SI= 10.5; DENV-2, SI= 9.7; DENV-3, SI= 10.2 and DENV-4 SI= 9.6). The findings were convincingly supported by the results from qRT-PCR analysis, which showed a decrease of viral RNA yield by >50% at 93 μg/mL of the extract used, and a reduction of viral RNA by at least 90% at 750 μg/mL of the extract used as compared to the non-treated cells (Figure 4B). Similar findings have been reported by others where the extract from the roots of S. baicalensis have shown antiviral activity against respiratory syncytial virus (RSV) with a SI value of 10.7 [12]. It has also been established that a flavone from the roots of S .baicalensis possessed significant antiviral activity against influenza viruses when used after virus adsorption to the susceptible cell line [24].

It was demonstrated that AHPE-Xa-09 exhibited a potent extracellular anti-DENVs effect against all the four DENV serotypes. Inhibition of DENV replication in Vero cells was significant when AHPE-XA-09 was pre-incubated with the viruses, with IC₅₀ values ranging from 74.33 to 95.83 μg/mL for the different DENV serotypes (Table 2, Figure 5A). The SI values of the extract in this part of the investigation ranged between 9.5 and 12.3 (DENV-1, SI= 9.9; DENV-2, SI= 9.5; DENV-3, SI= 9.7; DENV-4, SI= 12.3). The extract at 187 μg/mL decreased the RNA production of DENVs at the range of 77–83% (Figure 5B) as compared to the non-treated virus inoculum.

LC/MS/MS of the pure baicalein (0.5–20 μg/mL) was initially performed to establish a standard curve using area under the curve based on the different intensity height in the chromatogram of the different baicalein concentrations. A linear plot with R² value of >0.996 was obtained. A transition ion with the MRM of mass-to-change-ratio (m/z) 270.9/122.9 atomic mass units (amu) and retention time of 3.14 minutes was monitored for pure baicalein. This enabled us to identify and confirm the presence of baicalein in AHPE-XA-09. Using the standard curve plotted from pure baicalein (Figure 6A), the amount of baicalein in 100 μg/mL of AHPE-XA-09, which showed intensity of 1.80 × 10⁴ counts per second (cps), was at 1.03 μg/mL (Figures 6A and 6B). The amount of baicalein present in AHPE-XA-09 was consistent with those reported from extraction of S. baicalensis roots using many other methods [25, 26].

Recent lines of evidence suggest that certain compounds from the S. baicalensis extract, such as baicalein or chrysin, showed antiviral activity against hepatitis A virus (HAV), which is an RNA virus. However, the mechanism(s) underlying the mode of action of these compounds remain elusive [27]. Indeed, antiviral activity of 5,7,4'-trihydroxy-8-methoxyflavone a flavonoid from the roots of S. baicalensis was shown against influenza viruses [24, 28].

Recently, we have shown the novel antiviral activity of pure commercial baicalein against DENV-2 (NGC strain) [18]. Therefore, our current investigations point to the potential use of cold water extract of the roots of S. baicalensis, AHPE-XA-09 as one of the main source for baicalein, although some other bioactives present in the extract could also possess anti-DENV activities. The anti-DENV activities of S. baicalensis extract were noticeable at all stages of in vitro infection; when the cells were treated after virus adsorption, during the virus adsorption or even prior to virus infection. The tested extract exerted significant direct virucidal effects on extracellular free DENVs particles, which is a notable ability for anti-dengue candidate to neutralize free viruses at viremic stages. The mechanism of how the extract precisely inhibits DENVs replication remains ambiguous. However, our findings suggest that one possible mechanism whereby the S. baicalensis extract acts against DENV replication could be attributed to its ability to bind and/or to inactivate certain important structural and/or non-structural protein(s) of DENVs. Such inhibitory mechanism has previously been reported with some flavonoids such as pinostrobin against DENV NS2B/NS3 protease [29]. Moreover, it has been demonstrated that baicalein, as a main flavonoid found in the tested extract, is able to bind to HIV-1 integrase [30]. On the other hand, the activity of flavonoids such as wogonin and baicalein has been demonstrated against cellular DNA and RNA besides their ability to inactivate the cellular RNA polymerases [31‐33]. Therefore, it may be another potential mechanism for the tested plant extract and its constituents to inhibit the DENV replication via interference with DENV RNA polymerase and/or viral RNA, which is open for future investigations.