Eupafolin and Ethyl Acetate Fraction of Kalanchoe gracilis Stem Extract Show Potent Antiviral Activities against Enterovirus 71 and Coxsackievirus A16

Enterovirus 71 (EV71) and coxsackievirus A16 (CoxA16) are main pathogens of hand-foot-and-mouth disease, occasionally causing aseptic meningitis and encephalitis in tropical and subtropical regions. Kalanchoe gracilis, Da-Huan-Hun, is a Chinese folk medicine for treating pain and inflammation, exhibiting antioxidant and anti-inflammatory activities. Our prior report (2012) cited K. gracilis leaf extract as moderately active against EV71 and CoxA16. This study further rates antienteroviral potential of K. gracilis stem (KGS) extract to identify potent antiviral fractions and components. The extract moderately inhibits viral cytopathicity and virus yield, as well as in vitro replication of EV71 (IC50 = 75.18 μg/mL) and CoxA16 (IC50 = 81.41 μg/mL). Ethyl acetate (EA) fraction of KGS extract showed greater antiviral activity than that of n-butanol or aqueous fraction: IC50 values of 4.21 μg/mL against EV71 and 9.08 μg/mL against CoxA16. HPLC analysis, UV-Vis absorption spectroscopy, and plaque reduction assay indicate that eupafolin is a vital component of EA fraction showing potent activity against EV71 (IC50 = 1.39 μM) and CoxA16 (IC50 = 5.24 μM). Eupafolin specifically lessened virus-induced upregulation of IL-6 and RANTES by inhibiting virus-induced ERK1/2, AP-1, and STAT3 signals. Anti-enteroviral potency of KGS EA fraction and eupafolin shows the clinical potential against EV71 and CoxA16 infection.

This study investigated antiviral effects of K. gracilis stem (KGS) extract, ethyl acetate (EA), and n-butanol (BuOH) fractions against EV71 and CoxA16, ferreting out potential antiviral compounds of K. gracilis stem extract. EA fraction, better than BuOH fraction, effectively inhibited virusinduced cytopathicity and viral replication in vitro. Eupafolin, rich in EA fraction, showed a potent antiviral activity, with IC 50 values of less than 10 M, inhibiting production of proinflammatory cytokines via suppressing ERK1/2 and AP-1 mediated signaling pathways.

Materials and Methods
2.1. Fractionation of K. gracilis Stem (KGS) Extract. K. gracilis was collected from farmlands and gardens in Chiayi County, as detailed in our prior report [14]; its stem juice filtered by Whatman No. 1 paper, and then lyophilized in an IWAKI FDR-50P freeze dryer. Powder of stem extract was stored in sterile bottles at −20 ∘ C, dissolved in distilled deionised water, then partitioned with ethyl acetate (V/V = 1/1). Water fraction was mixed with n-butanol (V/V = 1/1), with ethyl acetate (EA), n-butanol (BuOH), and aqueous (H 2 O) fractions evaporated under reduced pressure by BUCHI Rotavapor R-114.

Viruses and Cells
. EV71 and CoxA16 strains were amplified in RD cells grown in Dulbecco's Modified Eagle's Medium with 10% fetal bovine serum (FBS) at 37 ∘ C, 5% CO 2 , as detailed in our prior report [14]. HeLa-G2AwtR cells were maintained in Modified Eagle's Medium with 10% FBS and 20 g/mL zeocin, expressing FRET probe as well as fusion protein of red fluorescent protein (DsRed)-2Apro cleavage motif-green fluorescent protein (GFP) [16].

Cell Viability
Assay. In all, 3 × 10 4 RD cells were added to each well of 96-well plates, cultured at 37 ∘ C, 5% CO 2 overnight, then quintuplicate treated with KGS extract, indicated fraction (EA, BuOH, or H 2 O), eupafolin or caffeic acid for an additional 48-hour incubation. Cell survival rate was calculated as ratio of optical density (OD) 570 nm − OD 630 nm (OD 570−630 ) of treated cells to OD 570−630 of mock cell using MTT assay [14]. Data showed means ± SD from three independent experiments; 50% cytotoxic concentration (CC 50 ) yielding 50% toxic effect was determined via computer program (provided by John Spouge, National Center for Biotechnology Information, National Institutes of Health).

Cytopathic Effect (CPE) Reduction and Virus
Yield. RD cells cultured in 6-well plates were infected with EV71 or CoxA16 at multiplicity of infection (MOI) 0.1 in the presence or absence of various amounts of KGS extract, indicated fraction, eupafolin or caffeic acid for 24 or 48 hours. Cellular morphology was observed and photographed under microscope. To quantify virus yield, cultured supernatants from each treated/infected cells were harvested 12, 24, 36, or 48 hours postinfection, then counted by real-time RT-PCR with VP1-specific primers, as described in our prior report [14]. The Ct value for viral yield in cultured supernatant was monitored by ABI PRISM 7000 sequence detection system (Applied Biosystems), delta Ct value calculated by subtracting Ct value for viral yield in cultured supernatant of infected cells with indicated treatment from that of cultured media of infected cells without treatment.

Plaque Reduction
KGS extract (200 g/mL)   Evidence-Based Complementary and Alternative Medicine

EV71
CoxA16 * * * * * * * * * * * * * KGS extract ( g/mL) Figure 3: Plaque reduction by KGS extract. Monolayer of RD cells on six-well plates infected with EV71 or CVA16 at a titer of 100 pfu was treated simultaneously with indicated KGS extract concentrations. After 1 h incubation, monolayer was covered with 2 mL of agarose overlay medium and incubated for 2 days at 37 ∘ C in a CO 2 incubator. Finally, plaques were counted after crystal violet staining.

Cell Cycle Analysis Using Flow Cytometry.
A total of 2 × 10 5 RD cells were infected with EV71 or CoxA16 at MOI of 0.5 in the presence and absence of 10 g/mL EA or BuOH fraction, then collected 36 h postinfection, and washed with PBS buffer. After centrifuging at 8000 rpm for 3 min, cell pellet dissolved with 100 L PBS buffer was incubated 1 mL of 70% alcohol in −20 ∘ C overnight and washed twice with PBS buffer for propidium iodide (PI) staining (Biolegend).
After 30 min incubation with PI/RNAase solution, over 10,000 stained cells were analyzed by BD FACSAria (Becton Dickinson) with excitation at 488 nm and emission at 633 nm.

Quantification of RANTES and IL-6 Gene Expression
Using Real-Time RT-PCR. Total RNA was isolated from virus-infected RD cells treated with interferon-(IFN-) , EA fraction, or eupafolin by purification kit (PureLink. TM. Micro-to-Midi. TM. total. RNA purification system, Invitrogen) used for cDNA synthesis with oligo dT primer and SuperScript III reverse transcriptase kit (Invitrogen). To gauge mRNA expression in response to EA fraction or eupafolin treatment and/or virus infection, two-step RT-PCR with SYBR Green I was used. Oligonucleotide primer pairs included 5 -TCCCCATATTCCTCGGAC-3 and 5 -GATGTACTCCCGAACCCA-3 for human RANTES, 5 -GATGGATGCTTCCAATCTGGAT-3 and 5 -AGTTCT-CCATAGAGAACAACATA-3 for IL-6, and 5 -AGCCAC-ATCGCTCAGACAC-3 and 5 -GCCCCAATACGACCA-AATCC-3 for GAPDH. Real-time PCR reaction was carried out using ABI PRISM 7700 sequence detection system, as described in prior study [14]. Relative mRNA expression levels of indicated were normalized by housekeeping gene GAPDH.

In Vivo Anti-EV71
Assay. The 1-day-old suckling mice were intraperitoneally infected with 1.7 × 10 7 pfu EV71 then intraperitoneally injected with K. gracilis stem extract (5 mg/kg) once on days 1, 3, 5, and 7. Three mice from each group were sacrificed on days 2, 4, 6, and 8; their intestine samples were collected for detection of virus loads using realtime RT PCR, described as in Section 2.4.

Statistical Analysis.
Data from three independent experiments were represented as mean ± standard deviation (SD) and statistically analyzed, using SPSS program (version 10.1, SPSS Inc., IL) via one-way ANOVA analysis by Scheffe's test.

Antiviral Activity of KGS Extract against EV71 and
CoxA16. KGS extract has a CC 50 value of 1622 g/mL to RD cells 48 h posttreatment, showing low cytotoxicity ( Figure 1, Table 1). Subsequently, antienterovirus ability of KGS extract was rated with cytopathicity, virus yield, and plaque reduction assays. KGS extract (200 g/mL) reduced cytopathicity of RD cells induced by EV71 and CoxA16 (Figures 2(a)-2(b)). Virus titer assay of cultured supernatants using real-time RT PCR assay indicated KGS extract in vitro significantly inhibiting EV71 and CoxA16 replication in both time-and concentration-dependent manner ( Figures  2(c)-2(d)). For determining potency and selectivity, plaque reduction assay was further performed (Figure 3) Figure 7: Virucidal activity and attachment inhibition of eupafolin, EA, and BuOH fractions. In virucidal assay (a), eupafolin or each fraction was mixed with EV71 or CoxA16 (10 6 pfu), then incubated at 4 ∘ C for 1 h. Residual infectivity was performed by plaque assay with 1000-fold dilution of virus/compound mixture. In the attachment assay (b), EV71 or CoxA16 (50 pfu) was mixed with EA, BuOH fraction, or eupafolin, then immediately added onto RD cell monolayer for 1 h at 4 ∘ C. After washing twice with PBS, monolayer was overlaid with 2 mL of agarose medium for 2 days at 37 ∘ C in CO 2 incubator. Attachment inhibition was calculated as residual plaques after crystal violet staining.
http://dx.doi.org/10.1155/2013/591354). In mock group, EV71 in intestine samples was detectable 2, 4, and 6 days postintraperitoneal inoculation. By contrast, intraperitoneal treatment with KGS extract resulted in the decrease of EV71 loads compared to the mock group 2 days postinoculation, as not detectable 4, and 6 days postinoculation. Results demonstrate KGS extract consisting of active anti-EV71 and CoxA16 components in vitro and in vivo.

Functional Fractions of KGS Extract against EV71
and CoxA16. To evaluate potential antiviral fractions, KGS extract was further fractionated sequentially with EA, BuOH,  . Inhibitory activity was calculated as FRET ratio, that is, intensity of emission at 590/14 nm divided by that at 510/10 nm. * value < 0.05; * * value < 0.01; * * * value < 0.001 by Scheffe's test. and water; lyophilized powder of these three fractions was subsequently used to examine cytotoxicity and apoptosis to RD cells and antiviral effects on virus yield and plaque reduction (Figures 3-4). MTT assay elicited CC 50 values of KGS EA, BuOH, and H 2 O fractions above 400 g/mL (Table 1), indicating these as less cytotoxic to RD cells. Plaque reduction assay indicated EA and BuOH, but not H 2 O fraction, effectively inhibiting EV71 more than CoxA16 replication in vitro (Figures 4(a)-4(b), Table 1). Antienterovirus activity and selectivity of fractions in plaque reduction assay was ranked in the following order: EA (IC 50 = 4.21 ∼ 9.08 g/mL; SI = 45.14 ∼ 97.35) > BuOH (IC 50 = 11.88 ∼ 18.23 g/mL; SI = 23.34 ∼ 35.82) > H 2 O (IC 50 = >100 g/mL) (Figures 4(a)-4(b), Table 1). Moreover, cell cycle analysis using flow cytometry indicated EV71 and CoxA16 infection causing significant increases of apoptosis (sub-G1 phase) in RD cells as well as EA and BuOH treatment markedly reducing the apoptosis from about 70% to near 20% (Supplemental Figure 1). Results verify EA fraction exhibiting significantly higher antiviral activity against EV71 and CoxA16 than BuOH fraction, containing potential antienterovirus components.

Eupafolin as Potent Antiviral Component in KGS EA
Fraction against EV71 and CoxA16. To examine fingerprint of KGS fractions, ferulic acid, quercetin, kaempferol, caffeic acid, and eupafulin served as standard marker components; EA and BuOH fractions were analyzed using HPLC with C-18 reverse phase column (Figure 5(a)). The retention time of HLPC chromatograph at 345 nm was at 4.424 min for caffeic acid, 6.475 min for ferulic acid, 14.050 min for eupafolin (overlapped with quercetin), and 25.744 min for kameperfol, respectively. HPLC chromatogram indicated retention times of Peak 3 and Peak 7 of EA and BuOH fractions as consistent with caffeic acid and eupafolin (or quercetin), respectively. UV absorption spectra (200-400 nm) indicated Peak 3 as the similar profile to caffeic acid and Peak 7 as the same profile of eupafolin, not quercetin ( Figures 5(b)-5(c)). HPLC chromatogram and UV absorption spectra demonstrated EA and BuOH fractions consisting of caffeic acid and eupafolin. Based on the calibration curve of standard markers, amount of eupafolin was 3.10 ± 0.09 g/mg in EA fraction and 0.32 ± 0.01 g/mg in BuOH fraction, respectively. The amounts of caffeic acid in EA and BuOH fraction were 10.91 ± 0.34 g/mg and 0.80 ± 0.02 g/mg, respectively. Higher amounts of caffeic acid and eupafolin could account for better antienterovirus activity of EA versus BuOH fraction. Subsequently, antiviral activity of caffeic acid and eupafolin against EV71 and CoxA16 were rated by plaque reduction assay ( Figure 6).

Virucidal Activity by Eupafolin.
To evaluate possible direct-acting antiviral mechanism, eupafolin, EA, and BuOH fractions were tested for virucidal, attachment, and viral protease inhibition. In virucidal activity assay (Figure 7(a)), EA and BuOH fractions at 100 g/mL showed low virucidal activity for EV71 and CoxA16 (reduction less than 25%). Eupafolin at 1 or 10 g/mL reduced CoxA16 infectivity by 30%, but slightly affected EV71 infectivity (lower than 20%). In attachment inhibition assay (Figure 7(b)), only EA fraction at 10 or 100 g/mL inhibited above 30% of EV71 binding to RD cells. In cell-based viral 2A protease activity assay (Figure 8), EA and BuOH fractions at 10 g/mL inhibited CoxA16 2A protease activity by over 30%. Still, eupafolin had no significant inhibitory effect on EV71 and CoxA16 activity. Results show difference in direct-acting antiviral actions by KGS EA and BuOH fractions. Eupafolin had moderate virucidal effect against both enterovirus types.

Inhibition of Virus-Induced Proinflammatory Cytokines by Eupafolin.
To examine effect of eupafolin on virusinduced proinflammatory cytokine expression further, relative IL-6 and RANTES mRNA levels in virus-infected cells treated with/without eupafolin, IFN-, or EA fraction were derived by quantitative real-time PCR (Figure 9). Eupafolin at 1 g/mL decreased virus-induced IL-6 and RANTES expression by more than 10-fold, suppressing proinflammatory cytokines induced by EV71 and CoxA16 better than KGS EA fraction (10 g/mL) and IFN-(100 U/mL). Phosphorylation levels of cytokine induction-related proteins p38 MAPK,

Discussion
KGS extract, EA, and BuOH fractions, as well as its components eupafolin and caffeic acid, processed low cytotoxicity ( Figure 1, Table 1). KGS EA fraction had an antienterovirus activity with IC 50 values less 10 g/mL, more effectively inhibiting replication of EV71 and CoxA16 in vitro than KGS extract and BuOH fraction (Figures 2-4 [14]. Eupafolin was thus suggested as crucially active antienterovirus component in K. gracilis, at the same time showing similar anti-EV71 efficacy and selectivity with identified potential anti-EV71 compounds of natural products like allophycocyanin, aloe-emodin, gallic acid, chrysosplenetin, and penduletin [17][18][19][20]. KGS extract reduced by more than 90% both cytopathicity and virus yield 36 h postinfection (Figure 2), implying antiviral activity linked with direct and indirect antiviral actions: for example, virucidal activity, attachment blocking, targeting viral enzymes, and host factors, inducing host antiviral responses. Our prior study [14] demonstrated K. gracilis leaf extract inhibiting viral 2A protease activity, reducing virus-induced apoptosis, as well as suppressing IL-6 and RANTES upregulation by EV71 and CoxA16. This study averred both KGS fractions plus eupafolin exhibiting low virucidal activity and slightly blocking virus attachment (Figures 7(a)-7(b)). Both KGS fractions inhibited CoxA16 2A proteases of CoxA16, but eupafolin failed to inhibit viral 2A proteases (Figure 8). Eupafolin specifically inhibited upregulation of IL-6 and RANTES gene expressions induced by EV71 or CVA16 infection (Figure 9), which correlated with reduction of virus-induced ERK1/2, c-Jun, and STAT3 mediated signaling ( Figure 10). Both KGS fractions exhibited multiple inhibitory actions against EV71 and CoxA16, relating to decrease of viral infectivity, attachment, and protease enzymatic activity in vitro. Aside from virucidal activity and attachment inhibition, eupafolin significantly inhibited production of IL-6 and RANTES in enterovirus infection. EV71-infected patients' elevated levels of IL-1 , IL-6, and TNF-in CSF strongly correlate with clinical severity [21,22]. In addition, EV71 infection causes the upregulation of COX-2 and PGE(2) via activation of ERK1/2 and AP-1 signaling pathways [23]. Eupafolin significantly inhibited activation of ERK1/2, c-Jun, and STAT3 in both virus-infected cells, which correlates with suppressing upregulation of IL-6 and RANTES by eupafolin treatment. It thus processed potent antiviral and antiproinflammatory activities, displaying therapeutic potential against EV71 and CoxA16 infection. Combination of effective compounds of K. gracilis, including eupafolin, quercetin, and caffeic acid, could provide an alternative approach against enteroviral infection.
In sum, KGS extract contains potent antienteroviral components; fractionation augments antienteroviral effect. Eupafolin, a crucial antiviral component of KGS EA fraction, shows high selective index for EV71 and CoxA16 by greater than a 30-fold increase. Eupafolin is the potential enteroviral agent with anti-inflammatory activities via suppressing virusinduced activation of ERK1/2, AP-1, and STAT3-mediated signaling pathways.