African green monkey kidney cells (Vero, ATCC CCL-81) and human epithelial carcinoma cells (HEp-2, ATCC CCL-23 and A549, ATCC CCL-185) were maintained in Dulbecco’s modified Eagle’s medium (DMEM), modified Eagle’s medium (MEM) and RPMI-1640 medium, respectively, supplemented with 10% fetal bovine serum (Gibco), 100 units/mL penicillin, and 100 μg/mL streptomycin. All the cell lines were cultured at 37°C with 5% CO₂. HSV-1F (ATCC VR-733) and HSV-2G (ATCC VR-734) stocks were propagated in Vero cells, aliquoted and stored at -80°C. ACV-resistant HSV-1 was generated in our previous study [34].

1,000 g dried H. cordata was boiled in 2 L water for 3 hours and the supernatant was filtered through a 0.2 um filter and lyophilized into dry powder. The powder was dissolved in double-distilled water to the concentration of 100 mg/ml.

1X 10⁶ Vero cells were seeded onto 6-well culture plates for 24 hours and infected with 100 pfu/well virus at 37°C for 1 hour. After incubation, the virus-infected cells were washed twice with phosphate-buffered saline (PBS) and overlaid with 1% methylcellulose containing different concentrations of HCWEs as indicated. After 3 days, the cells were fixed with 3.7% formaldehyde and stained with crystal violet. The plaques were counted.

1X10⁴ cells in 96-well plate were incubated with indicated concentrations of HCWEs at 37°C for 3 days. The cells were washed with PBS, added with 0.1 mg/mL MTT reagent [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; Sigma] and incubated at 37°C for 5 hours. After incubation, MTT reagent was removed and DMSO was added to each well. The absorbance was then determined by ELISA reader (SpectraMAX 340, Molecular Devices) at the wavelength of 550 nm.

For cell pre-treatment assay, 1X10⁶ Vero cells were pretreated with various concentrations of HCWEs at 37°C for 3 hours. After incubation, HCWEs-pretreated cells were washed with PBS and then infected with 100 pfu HSV. The inhibitory activities were analyzed by plaque assay. For virus pre-treatment assay, 100 pfu HSV was pretreated with various concentrations of HCWEs at room temperature. After 3 hours of incubation, 1X10⁶ Vero cells were infected with HCWEs-pretreated virus and then analyzed by plaque assay.

For virus binding assay, 1X10⁶ Vero cells were first cooled to 4°C for 1 hour and exposed to 100 pfu HSV in presence of HCWEs at 4°C for 2 hours. After incubation, the unbound virus and HCWEs were removed with ice-cold PBS and Vero cells were analyzed by plaque assay and Western blot assay. For virus penetration assay, 1X10⁶ Vero cells were cooled to 4°C for 1 hour and then infected with 100 pfu HSV at 4°C to achieve viral binding. The unbound viruses were removed and Vero cells were incubated with HCWEs at 37°C for 10 minutes to allow viral penetration. Vero cells were then washed with neutralization buffer (137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 2 mM KH₂PO₄, 0.5 M glycine, pH 3.0) for 1 minute to inactivate non-penetrated viruses and analyzed by plaque assay [34, 35].

1X10⁶ Vero cells were infected with 1 m.o.i. of HSV at 37°C for 1 hour. After absorption, HCWEs were added to medium at indicated hours-post infection. After 24 hours, total Vero cells were collected and frozen and thawed for three times. Virus titer was determined by plaque assay.

RNAs were extracted from virus-infected or mock-infected cells by Trizol reagent (Invitrogen) and treated with DNase I (Promega). 1 μg total RNA was subjected to reverse transcription by SuperScript II First-Strand Synthesis System (Invitrogen) according to the manufacturer’s instructions. The expression level of viral genes was measured by SYBR Green-based real-time RT-PCR. The PCR primer pairs sequences for ICP0 were cited as described [36]. The primer pairs for UL52 and UL13 were UL52F (5’-GACCGACGGGTGCGTTATT-3’)/UL52R (5’-GAAGGAGTCGCCATTTAGCC-3’) and UL13F (5’-GCGACCTGCTGGTCATGTG-3’)/UL13R (5’-TGCGAGCCAATCCTTGAAG-3’).

Cells were harvested in RIPA lysis buffer (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% SDS, 1% sodium deoxycholate, 1 mM PMSF, and protease inhibitor cocktail) and the protein concentration was measured by the BCA protein assay (BioRad). Proteins were resolved by 12.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred onto PVDF membrane, blocked with 5% skimmed milk in Tris-buffered saline (TBS) (20 mM Tris-HCl [pH 7.5], 150 mM NaCl, and 0.5% Tween-20) and reacted with primary antibodies for β-actin (1:5000; Sigma), VP16 (1:200; Santa Cruz), ICP4 (1:200; Santa Cruz) and NF-κB (1:200; Santa Cruz). β-actin acts as an internal control.

HSV-1 ICP0 promoter region was amplified from HSV-1 viral DNA using the following primer set (ICP0pF: 5’-CTCGAGTTATGCTAATTGCTTTTTTGG-3’ and ICP0pR: 5’-AAGCTTTCGTATGCGGCTGGAGGG-3’). The primer set (mutF: 5’-GGAAGGCAAAGGAAAAAGGGGCAC-3’ and mutR: 5’-GTGCCCCTTTTTCCTTTGCCTTCC-3’) was used to perform PCR-based mutagenesis by which ten mutant nucleotides were introduced in the NF-κB binding site within viral ICP0 promoter. Both PCR fragments were cloned into pGL3.0 basic vector (Promega).

All transfections were carried out in triplicate in 96-well plates. 2X10³ cells were seeded 24 hours prior to transfection. The luciferase reporter constructs along with the control plasmid (pRL-TK Vector; Promega) were co-transfected into cells. After 24 hours, transfectants were infected with HSV at 1 m.o.i.. The luminescent signals were measured at 48 h.p.i. by Victor³ multilabel counter (PerkinElmer) according to the manufacturer’s instructions. The activity of Renilla luciferase was used as an internal control to normalize transfection efficiency.

The pure compounds, quercetin, isoquercitrin, quercitrin, hyperin, rutin and acyclovir (ACV), were purchased from Sigma-Aldrich, dissolved in dimethyl sulfoxide (DMSO) or absolute ethanol and diluted with sterile deionized distilled water before use.

Previous studies have indicated HCWEs exhibit a promising anti-virus activity against HSV-1 and HSV-2 in Vero and BCC-1/KMC cells [27, 28]. These studies shed light on a new direction for development of anti-HSV drugs, however, the MOA is not well understood. Before we performed this study we first measured the anti-HSV activity of HCWEs because the activity may be different from previous reports due to the different sources of H. cordata and different extraction procedures [27, 28]. The cytotoxicity of HCWEs was measured by MTT assays. Vero cells were treated with serial dilutions of HCWEs at 37°C for 72 hours. As shown in S1 Table, HCWEs showed low cytotoxicity on Vero cells (CC₅₀ was greater than 100 mg/ml). Next, the anti-HSV activity of HCWEs was determined by the plaque reduction assay. The results indicated that EC₅₀ of HCWEs against HSV-1, Acyclovir-resistant HSV-1 (HSV-AR) and HSV-2 were 0.692 mg/ml, 1.11 mg/ml and 0.3 mg/ml, respectively (S1 Table). Intriguingly, the EC₅₀ of HCWEs against HSV-AR was similar to that against wild-type HSV virus, and this fact implied that the antiviral mechanisms of HCWEs and ACV might be different. The selective index (SI) scores of HCWEs to HSV-1, HSV-AR, and HSV-2 were 144.51, 90.09, and 333.33 respectively. The inhibitory effects of ACV on HSV were also analyzed, and the EC₅₀ of ACV against HSV-1, HSV-AR, and HSV-2 were 0.74 μg/ml, 14.02 μg/ml, and 0.82 μg/ml respectively (S1 Table). We demonstrated self-prepared HCWEs used here could inhibit HSV infection as previous reports [27, 28].

To determine which stage HCWEs hit and identify which viral proteins or host factors mediated the anti-virus activity, we first explored whether HCWEs lead host cells to initiate anti-viral mechanisms or interfere virus particles directly to block further viral binding and penetration, Vero cells and HSV were pretreated respectively with HCWEs and the plaque reduction assay was performed to determine which stage HCWEs act on HSV infection. The pretreated Vero cells showed no obviously inhibitory effect on HSV-1 infection at a concentration of 10 mg/ml (S1 Fig.), which is 14 times higher than its EC₅₀ (0.692 mg/ml). However, in the cases of HSV-1, HSV-2, and even HSV-AR pretreated with 0.01 mg/ml HCWEs, the plaque formation was largely inhibited (Fig. 1A). These results implied that the anti-HSV effects of HCWEs might target virus particles directly and inhibit further stages of HSV infection. To examine this hypothesis, HCWEs-pretreated HSV-1 and HSV-2 were incubated with Vero cells at 4°C to allow virus to bind onto cells but not penetrate the cells. The unbound viruses were washed out with PBS, and the cell-bound viruses were analyzed by Western blot analysis. HCWEs obviously decreased the cell binding ability of HSV-1 and HSV-2 in a dose-dependent manner (Fig. 1B). The results implied that HCWEs might affected virus particles directly and result in the inhibition HSV-1 and HSV-2 infections.

HCWEs can interfere virus particles directly and inhibit HSV infection. To characterize the inhibitory mechanism of HCWEs and further identify which stage HCWEs act, the effects of HCWEs on viral binding and penetration were investigated separately. Vero cells were first cooled to 4°C and infected with viruses in presence of various concentrations of HCWEs for 2 hours. The unbound viruses and HCWEs were washed out with PBS, and the cell-bound viruses were then analyzed by plaque assays. As shown in Fig. 2A, HCWEs inhibited HSV-1, HSV-2, and HSV-AR binding ability in a dose-dependent manner. 0.1 mg/ml HCWEs inhibited 93.27%, 78.65%, and 92.44% of HSV-1, HSV-2, and HSV-AR plaque formation, respectively. ACV was served as a negative control because ACV is a guanosine analogue but does not block viral entry. The results indicated that the plaque formation was not obviously inhibited in the ACV-treated group. In addition to viral binding, we also examined whether HCWEs could influence viral penetration into host cells or not. Vero cells were cooled to 4°C and incubated with virus for 2 hours. The unbound viruses were washed out with PBS, and the cells were then incubated with HCWEs at 37°C for viral penetration. After 10 minutes of incubation, neutralization buffer was added to inactivate non-penetrated viruses and the virus-penetrated cells were analyzed by plaque assays. The percentages of inhibition on HSV-1, HSV-2, and HSV-AR penetration at a concentration of 0.1 mg/ml HCWEs were 93.62%, 97.76%, and 97.99%, respectively (Fig. 2B). To further precisely realize whether the reduction of plaque formation induced by HCWEs can be attributed to blockage of viral binding and penetration at least partly, we would determine the effect of HCWEs on viral entry directly. The translocation of viral tegument protein VP16 into the nucleus and the expression of viral immediate-early protein ICP4 have been used as the indicators to examine HSV entry [37]. HCWEs were added to the host cells at the viral binding or penetration stage as described in Fig. 2A and Fig. 2B, and the nuclear fraction and total proteins were extracted to examine the viral VP16 translocation and viral ICP4 expression by Western blot assays, respectively. As shown in Fig. 3A and Fig. 3B, viral VP16 translocation and viral ICP4 expression were inhibited by HCWEs in both the binding and the penetration stages.

We have demonstrated HCWEs could affect virus particles directly and inhibit virus infection while which viral envelope protein mediated the inhibitory activity was not identified. HSV glycoprotein D (gD) plays a critical role in viral binding onto host cells, cell-to-cell spreading, and viral entry into host cells [38–40]. To determine whether gD is a target of HCWEs, the recombinant gD proteins were expressed by baculovirus expression system and incubated with Vero cells at 37°C or 4°C for 1 hour to allow recombinant gD to bind onto Vero cells. The unbound recombinant proteins were washed out with PBS, and the bound recombinant proteins were detected by Western blot assays. As shown in Fig. 4A and Fig. 4B, HCWEs substantially inhibited the binding of recombinant gD onto Vero cells at concentrations of 0.05 mg/ml at both 37°C and 4°C. To rule the possibility that the decrease of recombinant gD binding is due to protein degradation caused by HCWEs, the recombinant gD proteins were incubated with different concentrations of HCWEs at 37°C for 3 hours, and analyzed by Western blot assay. Even in the highest concentration, HCWEs did not lead into gD degradation (S2 Fig.).

We demonstrated that HCWEs target HSV gD, interfere with viral binding and penetration, and finally block viral entry. To further explore whether is there other anti-HSV mechanism of HCWEs acting after viral entry, the time-of-addition experiments were performed. 1 mg/ml HCWEs were added at indicated hours post-infection (h.p.i.), and the cells were harvested at 24 h.p.i. The cell lysates were frozen and thawed three times and then titrated by plaque assay. Intriguingly, HSV was totally inhibited even if host cells were treated with 1 mg/ml HCWEs at 12 h.p.i. but not in the case of 0.01 mg/ml HCWEs (Fig. 5 and S3 Fig.). We next investigated whether HCWEs suppress virus replication through regulating viral late genes at this stage. The results indicated that the expression levels of viral late genes such as gB, gC, gD, gE, gH, and UL13 were downregulated in the presence of HCWEs (S3 Fig.).

To address the stages in which HCWEs suppress HSV replication after viral entry, we first determined whether HCWEs influence the expression of viral proteins. Vero cells were first infected with HSV-1 followed by treatment of HCWEs for 24 hours, and the viral protein expression was detected by Western blot assay. As shown in Fig. 6, 1 mg/ml HCWEs obviously inhibited HSV-1 protein expression at 24 h.p.i.. HSV-1 IE genes are responsible for viral E and L gene expression and also essential for the viral lytic cycle. Thus, we next hypothesized that HCWEs could affect viral IE gene expression and further suppress viral proteins after viral entry. The host cells were infected with HSV-1 followed by treatment of HCWEs, and the RNA was extracted from HSV-infected cells at 2, 10, and 16 h.p.i. for analysis of the expression of the IE, E, and L genes, respectively, by real-time PCR (Fig. 7A and Fig. 7B). ICP0 expression, the viral IE gene, was inhibited by HCWEs in a dose-dependent manner (Fig. 7A). Besides ICP0, the expression of ICP4, ICP22, ICP27 and ICP47 was also determined and the results indicated HCWEs could regulate other viral IE genes expressions slightly (S4 Fig.). Furthermore, the expression levels of HSV-1 E gene UL52 and L gene UL13 were inhibited by HCWEs up to 300 and 1,000 folds, respectively (Fig. 7B). These data suggested that HCWEs might suppress the expression of viral genes through inhibiting viral IE gene ICP0 expression.

During HSV infection, there are two waves of NF-κB activation. The first-wave of NF-κB activation could enhance ICP0 transcription via binding onto viral ICP0 promoter while the second-wave of NF-κB activation could prevent host cells from apoptosis [17]. It has been reported that HCWEs block the second-wave of NF-κB activation in HSV-2 infection [18, 29]. We hypothesized whether HCWEs could suppress the first-wave of NF-κB activation and further eliminate NF-κB binding onto HSV ICP0 promoter. To test this hypothesis, host cells were infected with HSV-1 in the presence of different concentrations of HCWEs, and the cell lysates were harvested for determination of NF-κB activity. The nuclear fractions were extracted and applied to NF-κB translocation analysis by Western blot assay. As shown in Fig. 8A, HCWEs inhibited NF-κB translocation in a dose-dependent manner. To further explore whether HCWEs could inhibit HSV-1 ICP0 promoter activity via inhibition of NF-κB activity, the HSV-1 ICP0 promoter was constructed into the pGL luciferase reporter vector, named as pGL-ICP0 Wt. After transfection, luciferase activity of pGL-ICP0 Wt transfected cells increased up to 3,000-fold compared with the vector control, pGL, in HSV-1 infection. However, the induction of luciferase activity was suppressed by HCWEs (Fig. 8B). Although it is well known the expression of ICP0 is regulated by NF-κB, the exact NF-κB binding site on ICP0 promoter is not identified yet. Hence, we first used TRANSFAC software to predict the potential NF-κB binding sites within HSV-1 ICP0 promoter. Only one NF-κB binding site was predicted within the HSV-1 ICP0 promoter. To determine whether the HCWEs-mediated ICP0 suppression was NF-κB-dependent, the predicted NF-κB binding site within pGL-ICP0 Wt was mutated by PCR-based mutagenesis and the resulting plasmid was named as pGL-ICP0 Mut. Fig. 8B showed that the luciferase activity of ICP0-Mut was significantly decreased compared with ICP0-Wt in HSV infection, while there was no obvious suppression between HCWEs treatment and not. The results implied that HSV-1 ICP0 promoter activity was at least partly regulated by NF-κB. To our best knowledge, besides viral ICP0 promoter, there were no potential NF-κB binding sites had been identified within other viral IE genes promoters and that might partially reason why other viral IE genes such as CP4, ICP22, ICP27 and ICP47 did not regulate under HCWEs treatment.

We demonstrated that HCWEs inhibit HSV infection in Vero cells in multiple stages, including viral binding, viral penetration, and viral genes expression. We then demonstrated these anti-viral activities are ubiquitous phenomena in different human cell lines. Two additional human epithelial carcinoma cells HEp-2 and A549 cells were tested for cytotoxicity and anti-HSV activity by the plaque reduction assay. HCWEs showed no cytotoxicity on HEp-2 and A549 cells at 100 mg/ml (CC₅₀ > 100 mg/ml). Importantly, the plaque reduction assays showed that HCWEs inhibited HSV-1 and HSV-2 replication in both HEp-2 and A549 cells with an EC₅₀ similar to that of Vero cells (S1 Table).

The major components of HCWEs have been identified as chlorogenic acid, hyperin, rutin, quercetin, quercitrin, and isoquercitrin [41]. Quercetin, quercitrin, and isoquercitrin showed anti-HSV activity but which stage of HSV replication affected was not elucidated clearly [27]. To determine which stage these compounds act on, the pure components were analyzed by plaque reduction assays. Among these 6 compounds, rutin, isoquercitrin, and quercetin demonstrated anti-HSV-1 activities, with EC₅₀ of 187.58 μg/ml, 0.42 μg/ml, and 52.9 μg/ml, respectively (S2 Table). We further studied the anti-HSV mechanism of each active compound. The virus pretreatment assays and virus binding assays showed that quercetin, but not rutin or isoquercitrin, targeted virus particles directly and inhibited viral entry (Fig. 9A and Fig. 9B). Furthermore, we also identified which compound contributed to inhibition of NF-κB activity by Western blot assay. As shown in Fig. 9C, both isoquercitrin and quercetin inhibited NF-κB activation in HSV infection.

In summary, we discovered a new anti-HSV mechanism of HCWEs by which HCWEs target to virus particles and block viral binding and penetration in HSV infection. Additionally, HCWEs also inhibited ICP0 promoter expression through suppressing HSV-induced NF-κB activation, thereby suppress the expression of viral genes and block the synthesis of viral proteins (Fig. 10).