Antiviral Research 92 (2011) 237–246

Contents lists available at SciVerse ScienceDirect

Antiviral Research

journal homepage: www.elsevier .com/locate /ant iv i ra l

In vitro inhibitory effect of carrageenan oligosaccharide on influenza A H1N1 virus

Wei Wang a,b,⇑, Pan Zhang a,b, Cui Hao a,b, Xian-En Zhang c, Zong-Qiang Cui c, Hua-Shi Guan a,b

a Key Laboratory of Marine Drugs, Ministry of Education, Ocean University of China, Qingdao 266003, PR Chinab Shandong Provincial Key Laboratory of Glycoscience & Glycotechnology, Ocean University of China, Qingdao 266003, PR Chinac State Key Laboratory of Virology, Wuhan Institute of Virology, Chinese Academy of Sciences, Wuhan 430071, PR China

a r t i c l e i n f o a b s t r a c t

Article history:Received 18 May 2011Revised 5 August 2011Accepted 10 August 2011Available online 16 August 2011

Keywords:Influenza A virusCarrageenan oligosaccharideAdsorptionInternalizationCell surface

0166-3542/$ - see front matter � 2011 Elsevier B.V. Adoi:10.1016/j.antiviral.2011.08.010

⇑ Corresponding author. Address: School of MedUniversity of China, Qingdao 266003, PR China. Tel.:532 8203 3054.

E-mail address: wangweidragon@yahoo.com.cn (W

Carrageenan polysaccharide has been reported to be able to inhibit the infection and replication of manydifferent kinds of viruses. Here, we demonstrated that a 2 kDa j-carrageenan oligosaccharide (CO-1)derived from the carrageenan polysaccharide, effectively inhibited influenza A (H1N1) virus replicationin MDCK cells (selectivity index >25.0). Moreover, the 2 kDa CO-1 inhibited influenza A virus (IAV) rep-lication better than that of 3 kDa and 5 kDa j-carrageenan oligosaccharides (CO-2 and CO-3). IAV multi-plication was suppressed by carrageenan oligosaccharide treatment in a dose-dependent manner.Carrageenan oligosaccharide CO-1 did not bind to the cell surface of MDCK cells but inactivated virus par-ticles after pretreatment. Different to the actions of carrageenan polysaccharide, CO-1 could enter intoMDCK cells and did not interfere with IAV adsorption. CO-1 also inhibited IAV mRNA and protein expres-sion after its internalization into cells. Moreover, carrageenan oligosaccharide CO-1 had an antiviral effecton IAV replication subsequent to viral internalization but prior to virus release in one replication cycle.Therefore, inhibition of IAV intracellular replication by carrageenan oligosaccharide might be an alterna-tive approach for anti-influenza A virus therapy.

� 2011 Elsevier B.V. All rights reserved.

1. Introduction

Influenza A virus (IAV) is a most formidable pathogen, whichhas been the cause of at least three pandemics in the last century.The most severe IAV pandemic caused more than 40 million deathsin the world during 1918–1919 (Lamb and Takeda, 2001; Yewdelland Garcia-Sastre, 2002). In late April 2009, a novel influenza A(H1N1) virus caused a pandemic in a short period of time, whichattracted great attention all over the world. Current anti-IAV drugsare directed against the M2 protein (adamantanes) and neuramin-idase (NA; zanamivir and oseltamivir) (Hay et al., 1985; Mendelet al., 1998). Despite these successes, the drug efficacy, resistance,toxicity, and cost remained to be unresolved issues (Hayden andPavia, 2006). Recently, several other approaches including inhibi-tors of viral RNA transcription, small interfering RNA, inhibitorsof virus-cell fusion or proteolytic processing of HA have beendeveloped, but so far none of inhibitors have been developed intoa drug (Lagoja and De Clercq, 2008). Hence, the development of no-vel antiviral agents that could be used alone or in combinationwith existing antiviral drugs is of high importance.

Carrageenan polysaccharide, extracted from red algae, showsdifferent inhibitory effects on different viruses (Buck et al., 2006;

ll rights reserved.

icine and Pharmacy, Ocean+86 532 8203 1980; fax: +86

. Wang).

Carlucci et al., 2004; Grassauer et al., 2008; Pujol et al., 2006;Talarico and Damonte, 2007). Previous studies have shown thatcarrageenan inhibits the replication of some enveloped virusesby interfering with virus adsorption and internalization into hostcells (Buck et al., 2006; Grassauer et al., 2008; Talarico and Da-monte, 2007). However, other reports have shown that carra-geenan does not inhibit virus adsorption but inhibits some stepsof virus life cycle in host cells (Gonzalez et al., 1987; Pujol et al.,2006). Recently, Leibbrandt et al. reported that i-carrageenan poly-saccharide could inhibit influenza A virus infection by directlybinding to the virus particles (Leibbrandt et al., 2010), while Tala-rico et al. found that carrageenan could inhibit dengue virus infec-tion in mosquito cells by targeting cellular proteins (Talarico et al.,2011). The inhibitory mechanism of carrageenans on virus replica-tion seems to be dependent on the type of polysaccharide (Da-monte et al., 2004) as well as the serotype of the virus and thehost cells (Girond et al., 1991). Despite having good inhibitory ef-fects on virus replication, the high molecular weight (MW) associ-ated poor tissue-penetrating ability of sulfated polysaccharideslimits their potential antiviral application in humans.

In recent years, marine oligosaccharides are attracting increas-ing interests in developing potential anti-viral drugs (Ji et al.,2011). Yamada et al. reported that O-acylated carrageenan oligo-saccharides with different MW had increased anti-HIV activitiesafter depolymerization and sulfation (Yamada et al., 1997, 2000).Ekblad et al. found that heparan sulfate (HS) mimetic PI-88, asulfomannan oligosaccharide of low MW, efficiently reduced the

238 W. Wang et al. / Antiviral Research 92 (2011) 237–246

cell-to-cell spread of herpes simplex virus (HSV) (Ekblad et al.,2010; Nyberg et al., 2004). The low-viscosity and high-solvencyof marine oligosaccharides at neutral pH also suggest their poten-tial use as anti-viral drugs. However, the molecular mechanismsunderlying the anti-viral effects of oligosaccharides have not beensystematically pursued.

In the present study, we investigated the capacity and molecu-lar mechanisms of j-carrageenan derived oligosaccharides ininhibiting IAV infection. Our results indicated that carrageenan oli-gosaccharide with low molecular weight (�2 kDa) could effectivelyinhibit IAV replication in vitro. IAV multiplication was suppressedby carrageenan oligosaccharide treatment in a dose-dependentmanner. Carrageenan oligosaccharide CO-1 did not bind to the cellsurface of MDCK cells but inactivated virus particles after pretreat-ment. CO-1 could enter into MDCK cells and did not interfere withIAV adsorption directly. Moreover, carrageenan oligosaccharidealso inhibited IAV mRNA and protein expression after its internal-ization into cells.

2. Materials and methods

2.1. Compounds and reagents

The j-carrageenan-derived oligosaccharides named CO-1, CO-2,CO-3 and FITC-labeled oligosaccharide (CO-1-FITC) were providedby Glycoscience and Glycoengineering Laboratory, school of Medi-cine and Pharmacy, Ocean university of China. Mouse anti-influ-enza A virus nucleoprotein (NP) and anti-b-actin monoclonalantibodies were purchased from Santa Cruz Biotechnology (USA).Rabbit anti-influenza A virus neuraminase (NA) polyclonal anti-body was purchased from Abcam (Hong Kong, China). FITC-labeledgoat anti-mouse and goat anti-rabbit secondary antibodies wereobtained from Boster (China). Alkaline phosphatase (AP)-labeledgoat anti-mouse secondary antibody was purchased from Cell Sig-naling Technology (USA). Ribavirin injection (50 mg/mL) was ob-tained from LuKang Cisen (China).

2.2. Cell culture and virus infection

Madin-Darby canine kidney (MDCK) cells were grown inRPM1640 medium supplemented with 10% FBS, 100 U/mL of pen-icillin and 100 lg/mL of streptomycin. Human lung epithelial cells(A549 cells) were cultivated in Ham’s F-12 medium containing 10%FBS and 2 mM L-glutamine. Influenza virus (A/Puerto Rico/8/34[H1N1]; PR8) was propagated in 10-day-old embryonated eggsfor three days at 36.5 �C. For virus infection, virus propagationsolution was diluted in PBS containing 0.2% bovine serum albuminand was added to cells at the indicated multiplicity of infection(MOI). Virus was allowed to adsorb 60 min at 4 �C. After removingthe virus inoculum, cells were maintained in infecting media(RPMI1640, 4 lg/mL trypsin) at 37 �C in 5% CO2.

2.3. Cytotoxicity assays

The cytotoxicity of compounds was measured by the MTT(3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide;Sigma–Aldrich, USA) assay. Confluent MDCK cell cultures in 96-well plates were exposed to different concentrations of compoundsin triplicate, using incubation conditions equivalent to those usedin the antiviral assays. Next, 10 lL of PBS containing MTT (finalconcentration: 0.5 mg/mL) was added to each well. After 4 h incu-bation at 37 �C, the supernatant was removed and 200 lL of DMSOwas added to each well to solubilize the formazan crystals. Aftervigorous shaking, absorbance values were measured in a micro-plate reader (Bio-Rad, USA) at 570 nm. The CC50 was calculated

as the compound concentration necessary to reduce cell viabilityby 50%.

2.4. Infectivity antiviral assays

For the determination of antiviral activity, a virus yield reduc-tion assay was performed, as previously described (Basu et al.,2009; Leibbrandt et al., 2010). MDCK cell monolayers were firstlyincubated with IAV at an MOI of 1.0 for 1 h at 4 �C, then after re-moved the virus, the infecting media containing different drugswere added to cells and incubated at 37 �C for 48 h. Then the50% tissue culture infectious dose (TCID50) was determined inMDCK cells with 10-fold serially diluted viruses incubated at37 �C for 72 h. Virus titers in 50% tissue culture infectious doses(TCID50)/ml were determined according to Reed and Muench(1938).

Furthermore, the antiviral activity was also evaluated by theCPE inhibition assay as described previously (Hung et al., 2009).MDCK or A549 cells in 96-well plates were firstly infected withIAV (MOI = 1.0), and then treated with different compounds in trip-licate after removal of the virus inoculum. After 48 h incubation,the cells were fixed with 4% formaldehyde for 20 min at room tem-perature. After removal of the formaldehyde, the cells were stainedwith 0.1% crystal violet for 30 min. The plates were then washedand dried followed by solubilization of the dye with methanol,and the intensity of crystal violet staining for each well was mea-sured at 570 nm. The virus inhibition (%) was calculated by theequation: Virus inhibition (%) = [(Asample 570 � Avirus 570)/(Amock

570 � Avirus 570)] � 100; Where, Amock 570 was the absorbance with-out virus infection, Asample 570 was absorbance with virus infectionand drug treatment, Avirus 570 was absorbance with virus infectionbut without drugs.

2.5. Influence of time of treatment on antiviral activity

Time course analysis was performed as previously reported(Buck et al., 2006; Talarico and Damonte, 2007). The MDCK cellswere incubated with IAV (MOI = 3.0) for 1 h at 4 �C, washed to re-move unbound virus, and then added with infection media. Vari-ous doses of carrageenan oligosaccharide CO-1 or ribavirin wereadded to the cultures at the indicated time points (0, 1, 2, 4, 8 hpost-infection [p.i.]), respectively. For the 0 h time point, com-pounds were added to cultures simultaneously with the additionof infection media (after adsorption). Carrageenan oligosaccharidewas also added separately to the cultures at 1, 2, 4, 6, and 8 h p.i. At24 h p.i., the antiviral activity was evaluated by the CPE inhibitionassay, as described above.

2.6. Hemagglutination (HA) assay

The hemagglutination (HA) assay was performed as previouslyreported (Wolkerstorfer et al., 2009). Standardized chicken redblood cell (cRBC) solutions were prepared according to the WHOmanual 2002 (WHO, 2002). Virus propagation solutions (105 PFU/ml) were serially diluted 2-fold in round bottomed 96-well plateand 1% cRBCs were then added at an equal volume. After 60 minincubation at 4 �C, RBCs in negative wells sedimented and formedred buttons, whereas positive wells had an opaque appearancewith no sedimentation. HA titers are given as hemagglutinationunits/50 lL (HAU/50 lL).

2.7. Flow cytometry analysis and confocal imaging

Flow cytometry analysis was performed as previously described(Miao et al., 2004) with some modifications. MDCK cells (5 � 105)were harvested by trypsination and resuspended in 1 mL PBS.

W. Wang et al. / Antiviral Research 92 (2011) 237–246 239

FITC-labeled carrageenan oligosaccharide CO-1 (CO-1-FITC) wasadded to the cells at a final concentration of 250 lg/mL. After incu-bation for 1 h at 37 �C in the dark with gentle mixing every 10 min,the cells were harvested, washed with PBS for three times, andthen analyzed by FCM (BectonDickinson, USA), with a 488 nm laserexcitation and a 530 nm emission filter. Data were analyzed withWinMDI2.8 software. A minimum of 10,000 events were countedfor each sample. Results were expressed in mean fluorescenceintensity arbitrary units (au).

For FCM analysis of IAV binding, MDCK cells (5 � 105) were ex-posed to PR8 (MOI = 1.0) with or without CO-1 treatment at 4 �Cfor 60 min to allow virus binding. After that, the cells were washedthree times to remove unbound virus and then incubated with500 lL 2% BSA/PBS for 20 min. After washed with PBS twice, thecells were stained with an anti-IAV NA antibody followed by incu-bation with an FITC-labeled secondary antibody for 40 min. Thecells were fixed with 4% paraformaldehyde for 10 min, and thenwashed three times before FCM analysis. A minimum of ten thou-sand events were counted for each sample. The mean fluorescenceintensity of cells was used to determine the amount of IAV bindingto MDCK cells.

For confocal imaging, MDCK cells were firstly incubated with orwithout IAV (MOI = 1.0) for 1 h at 4 �C followed by three washeswith PBS to remove unbound virus, and then added with infectionmedia containing FITC-labeled CO-1 (CO-1-FITC) at the concentra-tion of 250 lg/mL. After incubation for 4 h at 37 �C, the media wereremoved, and the cells were washed with PBS for three times. Thegreen fluorescence of CO-1-FITC was measured at 520 ± 20 nm byLaser Scanning Confocal Microscope (Zeiss LSM 510, GER).

2.8. Real-time RT-PCR assay

Total RNA was extracted from IAV infected MDCK cells using anRNAiso™ Plus Kit (Takara, Japan), and analyzed by using the OneStep SYBR� PrimeScript� RT-PCR Kit (Takara, Japan). The real-timeRT-PCR was performed using the following primers: NS1 vRNAand NS1 mRNA, 50-CTTCGCCGAGATCAGAAATC-30 and 50-TGGACCATTCCCTTGACATT-30 (GenBank accession No. J02150); NP mRNA,50-CTCATCCTTTATGACAAAGAAG-30 and 50-AGATCATCATGTGAGTCAGAC-30 (GenBank accession No. EF190975); canine b-actinmRNA, 50-GGCATCCTGACCCTGAAGTA-30 and 50-GGGGTGTTGAAAGTCTCGAA-30 (GenBank accession No. XM536230).

The real-time RT-PCR was performed at 42 �C 5 min, 95 �C 10 s,40 cycles of 95 �C 5 s, 60 �C 34 s, followed by melting curve analy-sis, according to the instrument documentation (ABI PRISM 7500,Applied Biosystems, USA). All reactions were performed in tripli-cate and the results were normalized to b-actin. The relativeamounts of IAV RNA molecules were determined using the com-parative (2�DDCT) method, as previously described (Livak andSchmittgen, 2001).

2.9. Indirect immunofluorescence assay

IAV (MOI = 2.0) infected MDCK cells were added with 250 lg/mL of CO-1 or ribavirin after virus uptake at 37 �C for 1 h (at 1 hp.i.), and then incubated at 37 �C for 6 h. After that, cells werewashed with PBS and fixed with 4% paraformaldehyde for20 min. Then cells were permeabilized using 0.5% (v/v) Triton X-100 in PBS for 5 min before incubated with 2% BSA/PBS for 1 h at37 �C. Cells were washed and incubated with anti-influenza A virusNP antibody overnight at 4 �C. After washing, the cells were incu-bated with FITC-labeled secondary antibody for 50 min at 37 �C. Fi-nally, cells were washed and directly observed using an invertedfluorescence microscope (DMI6000B; Leica, Germany) equippedwith a cooled CCD camera.

2.10. Western Blot assay

Confluent MDCK cell monolayers were firstly infected with IAV(MOI = 2.0), and then treated with or without indicated com-pounds at 250 lg/ml after virus internalization (at 1 h p.i.). Afterincubation for 8 h, cell lysate was separated by SDS-PAGE andtransferred to nitrocellulose membrane. After being blocked inTris-buffered saline containing 0.1% Tween 20 (v/v) and 5% bovineserum albumin (w/v) at room temperature for 2 h, the membraneswere rinsed and incubated at 4 �C overnight with anti-influenza Avirus NP antibody or anti-b-actin antibody as control. The mem-branes were washed and incubated with AP-labeled secondaryantibody (1:2000 dilutions) at RT for 2 h. The protein bands werethen visualized by incubating with the developing solution [p-nitroblue tetrazolium chloride (NBT) and 5-bromo-4-chloro-3-indolylphosphate toluidine (BCIP)] at RT for 30 min. The relative densitiesof influenza virus NP protein were determined by using ImageJ(NIH) v.1.33u (USA).

2.11. Statistics

All data are representative of at least three independent exper-iments. Data are presented as mean ± SD. Statistical significancewas calculated by SPSS 10.0 software using one-way ANOVA, withP values <0.05 considered significant.

3. Results

3.1. Characterization of carrageenan oligosaccharides

Carrageenan oligosaccharides were prepared and depolymer-ized by acidic hydrolysis of j-carrageenan followed by fraction-ation using chromatographic separation. The molecular weightsof j-carrageenan oligosaccharides CO-1, CO-2 and CO-3 weredetermined by HPLC based on the calibration with dextran stan-dards as described before (Yamamoto et al., 1995). The averagemolecular weight of j-carrageenan oligosaccharide CO-1 wasabout 2 kDa. The degree of polymerization (DP) of CO-1 was about3–13 (Fig. 1). In contrast, the molecular mass of j-carrageenan oli-gosaccharides CO-2 and CO-3 was about 3 and 5 kDa, respectively.The structures of CO-1, CO-2 and CO-3 were determined by Fouriertransform infrared spectroscopy (FTIR) and nuclear magnetic reso-nance spectroscopy (NMR) analysis (data not shown). They havemainly 1,4-linked b-D-galactose and 1,3-linked a-D-galactose withabout 0.4 sulfate per sugar residue.

3.2. Inhibition of influenza A virus multiplication in MDCK cells bycarrageenan oligosaccharides

Carrageenan oligosaccharides CO-1, CO-2 and CO-3 were as-sayed for their ability to inhibit IAV multiplication in vitro usingTCID50 assay and CPE inhibition assay, as previously described(Basu et al., 2009; Hung et al., 2009). MDCK cells were firstly in-fected with PR8 virus (MOI = 1.0), and then treated with com-pounds at the indicated concentrations after removal of the virusinoculum. After 48 h, virus titers of the culture media were deter-mined by virus yield reduction assay, and the cell viability wasmeasured by CPE inhibition assay. As shown in Fig. 2A, the j-car-rageenan-derived oligosaccharide with the average molecularweight of 2 kDa (CO-1) could significantly reduce the virus titerand promote cell viability when used at the concentration of morethan 125 lg/mL (P < 0.05). The 90% inhibitory concentration (IC90

value) of CO-1 for virus yield reduction was about 215 lg/mL(Fig. 2A). Furthermore, the virus inhibition with different concen-trations of CO-1 was dose-dependent and the virus inhibition

Fig. 2. Carrageenan oligosaccharide CO-1 inhibited influenza A virus replication in adose-dependent manner. (A) MDCK cells were infected with influenza A virus (IAV)at an MOI of 1.0, and then treated with CO-1 at the indicated concentrations afterremoval of the virus inoculum. After incubation for 48 h, the supernatants werecollected and analyzed for TCID50 value by the method described before. Values aremeans ± SD (n = 3). Significance: ⁄P < 0.05 vs. non drug treated control group. (B)MDCK cells were firstly infected with IAV (MOI = 1.0), and then treated withindicated concentrations of CO-1 after removed virus inoculum. The antiviralactivity was determined by the CPE inhibition assay at 48 h p.i. Results areexpressed as percent of inhibition in drug-treated cultures compared withuntreated. Values are means ± SD (n = 5).

Fig. 1. Schematic diagram of the repeating saccharide units in carrageenanoligosaccharide CO-1. The disaccharide unit is mainly composed of 1,4-linked b-D-galactose and 1,3-linked a-D-galactose. The degree of polymerization (DP) of j-carrageenan oligosaccharide CO-1 is about 3-13.

240 W. Wang et al. / Antiviral Research 92 (2011) 237–246

percentage was >70% at 125 lg/mL (Fig. 2B). Moreover, the IC50 va-lue of j-carrageenan oligosaccharide CO-1 was only 32.1 lg/mL,and the selectivity index (CC50/IC50) for CO-1 was approximate26.7, very close to that of ribavirin (SI = 27.2) (Table 1). However,the carrageenan oligosaccharides with high molecular weights(CO-2, CO-3) had much lower inhibitory effect on IAV multiplica-tion (SI < 5.0) compared to CO-1 (Table 1). Thus, low molecularweight carrageenan oligosaccharide CO-1 could effectively sup-press the replication of influenza A virus in MDCK cells.

3.3. Inhibition of IAV replication in human lung epithelial cells (A549)by carrageenan oligosaccharide CO-1

To explore whether the inhibitory effect of CO-1 on influenza Avirus was cell-specific or not, the antiviral assay was also performedin human lung epithelial cells (A549 cells). In brief, A549 cells wereinfected with IAV (A/PR/8/34) at an MOI of 1.0, and then treated withCO-1 at the indicated concentrations after removal of the virus inoc-ulum. After incubation for 48 h, the virus titers in A549 cell superna-tants were measured by hemagglutination (HA) assay. As shown inFig. 3A, viral replication in A549 cells was dose-dependently inhib-ited by CO-1 at the concentrations of 31.25–250 lg/mL and CO-1could significantly reduce the virus titer when used at the concen-tration of more than 125 lg/mL (P < 0.05). Furthermore, carra-geenan oligosaccharide CO-1 could reduce the cytopathic effect ofthe A/PR/8/34 virus by more than 50% at the concentration of62.5 lg/mL (Fig. 3B), and the protective effect was dose-dependentat concentrations ranging from 15.625 to 250 lg/mL. Moreover, theprotective effect of CO-1 in A549 cells was a little weaker than thatin MDCK cells when used at the concentration of less than 250 lg/mL (Fig. 3B). The IC50 value of CO-1 in A549 cells was about42.8 lg/mL, and the selectivity index (CC50/IC50) was approximate20.0, close to that of CO-1 in MDCK cells (SI = 26.7). These resultsindicated that CO-1 could also inhibit IAV replication in A549 cellsalthough not as effective as in MDCK cells.

3.4. Influence of treatment duration with CO-1 on influenza A virusmultiplication

A time course study was performed to analyze the influence ofdrug treatment during various stages of the virus replication cycle.The antiviral assay was carried out at an MOI of 3.0, so all the cellswere infected initially and thus only one round of replicationwould be influenced by drugs (Leibbrandt et al., 2010). In brief,MDCK cells were infected with PR8 virus at an MOI of 3.0, and thenafter removal of the virus inoculum, different concentrations ofCO-1 or ribavirin were added to the cells at indicated post-infec-tion time points (p.i.). The inhibition effect was evaluated by CPEinhibition assay at 24 h p.i. As shown in Fig. 4, when ribavirinwas added to MDCK cells at 0 h p.i., the inhibition effect was about90% at the concentration of 250 lg/mL (Fig. 4B). In contrast, whenribavirin was added at 1 or 2 h p.i., the inhibition effect decreasedto about 50%. Moreover, when ribavirin was added after 4 h p.i., theinhibition effect was no more than 30%. Consistent with previousreports of ribavirin inhibition of virus RNA synthesis (Browne,1978), we show that ribavirin mainly inhibits some early steps ofIAV life cycle in host cells.

However, there was little difference of the inhibitory effectwhen CO-1 added at different post-infection time points within4 hours (0, 1, 2, 4 h p.i.), and the virus inhibition percentage byCO-1 was all more than 80% at 250 lg/mL, which was different fromthat of ribavirin (Fig. 4A). Moreover, CO-1 treatment at time pointsafter 4 hours post-infection (6, 8 h p.i.) could not effectively inhibitIAV replication even at the concentration of 250 lg/mL (inhibitionpercentage <55%) (Fig. 4A). Furthermore, it was reported that a sin-gle replication cycle of the influenza A virus takes approximately 8–10 h (Alam, 2006), so another time course study was performedwithin 10 h to explore which stage CO-1 inhibits in one replicationcycle. IAV (MOI = 3.0) infected MDCK cells were treated with250 lg/mL of CO-1 for different time intervals (Fig. 4C). The resultsshowed that postinfection CO-1 treatment for 1 h (0–1 h p.i.) onlyproduced very slight antiviral effect (about 20%), which suggestedCO-1 might not prevent virus entry into the cells. In cells exposedto CO-1 for the first 4 h or 6 h after infection (0–4 h or 0–6 h p.i.),obvious inhibition (48 ± 1% or 56 ± 6%) of virus-induced CPE wasdetected (Fig. 4C). More substantial inhibition was observed when

Table 1Inhibition effects of different compounds on influenza A virus multiplication in vitroa.

Compounds IC50 (lg/mL)b CC50 (lg/mL)c SId

j-Carrageenan Oligosaccharide (CO-1, �2 kDa) 32.1 ± 3.5 857 ± 18.5 26.7j-Carrageenan Oligosaccharide (CO-2, �3 kDa) 239 ± 11.5 886 ± 21.0 3.7j-Carrageenan Oligosaccharide (CO-3, �5 kDa) 519 ± 16.5 1109 ± 16.5 2.1Ribavirin 27.4 ± 1.5 744 ± 8.5 27.2

a The inhibition effects on influenza virus A/PR8/34 (H1N1) (MOI = 1.0) multiplication in MDCK cells were evaluated byvirus yield reduction assay.

b Inhibition concentration 50% (IC50): concentration required to inhibit influenza virus A/PR8/34(H1N1) yield at 48 h post-infection by 50%. Values are means ± SD (n = 4).

c Cytotoxic concentration 50% (CC50): concentration required to reduce MDCK cell viability by 50%. Values are means ± SD(n = 3).

d SI: Selectivity index is defined as the ratio of CC50 to IC50 (SI = CC50/IC50).

Fig. 3. The inhibition effect of CO-1 on IAV was not cell type-specific. (A) A549 cellswere infected with IAV at an MOI of 1.0, and then treated with CO-1 at the indicatedconcentrations after removal of the virus inoculum. After incubation for 48 h, theantiviral activity was evaluated by hemagglutination (HA) assay. HA titers are givenas hemagglutination units/mL (HAU/mL). Values are means ± SD (n = 3). Signifi-cance: ⁄P < 0.05, ⁄⁄P < 0.01 vs. non drug treated control group. (B) A549 and MDCKcells were firstly infected with IAV (MOI = 1.0), and then treated with indicatedconcentrations of CO-1 after removed virus inoculum. The antiviral activity wasdetermined by CPE inhibition assay at 48 h p.i. Values are means ± SD (n = 5).

W. Wang et al. / Antiviral Research 92 (2011) 237–246 241

exposure was extended to the first 10 h (76 ± 1%). However, onlyslight inhibition was noted (about 27%) when CO-1 was added 4 hor 6 h after infection (4–10 h or 6-10 h p.i.). These data indicatedthat CO-1’s antiviral activity is largely related to its inhibition ofvirus life-cycle steps that occur 0–4 h after infection.

3.5. Carrageenan oligosaccharide CO-1 had inhibitory actions directlyon virus particles rather than on virus adsorption

To elucidate how carrageenan oligosaccharide protects cellsfrom IAV infection, we also investigated whether CO-1 elicits itsinhibitory actions directly on virus particles or on virus adsorption.MDCK cells were infected with IAV (MOI = 1.0) that had been pre-incubated with 250 lg/mL of CO-1 for 1 h at 37 �C, and then after1 h adsorption at 4 �C, cells were washed and overlaid with com-pound-free media. After 48 h virus titers of the culture media weredetermined by TCID50 assay. As shown in Fig. 5A, pre-incubation ofIAV with CO-1 could significantly reduce the virus titer (P < 0.05),suggesting that CO-1 could inactivate viral particles to some ex-tent. Moreover, CO-1 treatment after virus adsorption also signifi-cantly reduced the virus titer (P < 0.01) (Fig. 5A). However, thetreatment of CO-1 only during virus adsorption could not effec-tively inhibit IAV replication (Fig. 5A). Furthermore, we also evalu-ated whether CO-1 could neutralize IAV hemagglutinin (HA)resulting in inhibition of binding of HA to its receptors, Neu5A-ca2-6Gal-terminated sugar chains (Skehel and Wiley, 2000). Thus,virus and cRBC, respectively, were incubated with PBS or CO-1 (500and 250 lg/mL) for 1 h at room temperature and HA titers werethen determined. No difference of CO-1-treated virus or cRBC toPBS treatment was detected on HA titers (data not shown), sug-gesting that CO-1 does not inhibit IAV binding to its sialic acidreceptor.

FCM analysis was performed to investigate whether CO-1 couldbind to the cell surface of MDCK cells. After incubated with orwithout 250 lg/mL of FITC-labeled CO-1 (CO-1-FITC) for 1 h at37 �C, MDCK cells were washed three times, and then analyzedby FCM. As shown in Fig. 5B, the CO-1-FITC treated MDCK cellsexhibited nearly the same fluorescent intensity (12.74 ± 0.65 au)as that of untreated control cells (11.95 ± 0.27 au), indicating thatcarrageenan oligosaccharide CO-1 could not bind to the cell surfaceof MDCK cells.

Furthermore, FCM analysis was also performed to explore theeffect of CO-1 on IAV adsorption as described before (Wu et al.,2010). In brief, MDCK cells were exposed to PR8 (MOI = 1.0) withor without CO-1 treatment at 4 �C for 60 min to allow virus bind-ing. After that, the cells were washed and stained with anti-IAVNA antibody before FCM analysis. The fluorescence intensity ofNA protein was used to evaluate the mount of IAV particles ad-sorbed to MDCK cells. As shown in Figs. 5C and D, IAV infectedMDCK cells with CO-1 treatment (PR8 + CO-1) showed almost thesame fluorescent intensity (49.81 ± 6.65 au) as that of untreated

IAV infected cells (PR8) (47.75 ± 7.47 au), which suggested thatcarrageenan oligosaccharide CO-1 treatment did not affect virusbinding to MDCK cells.

3.6. CO-1 inhibited IAV mRNA expression after its internalization intoMDCK cells

The results above showed that CO-1 treatment after adsorptioncould also significantly inhibit IAV replication (Fig. 5A), so we then

Fig. 4. The influence of the timing of addition of CO-1 on virus infection. (A and B)MDCK cells were incubated with IAV (MOI = 3.0) for 1 h at 4 �C, washed to removeunbound virus, and then added with infection media containing different concen-trations of CO-1 (A) or ribavirin (B) (31.25, 62.5, 125, 250 lg/mL) at the indicatedtime points after infection. At 24 h p.i., media were discarded and the CPE inhibitionassay was performed. Results are expressed as percent of inhibition in drug-treatedcultures compared with untreated. Values are means ± SD (n = 6). (C) IAV(MOI = 3.0) infected MDCK cells were treated with 250 lg/mL of CO-1 for thespecified time period, and then the media were removed and cells were overlaidwith compound-free media. The antiviral activity was determined by CPE inhibitionassay at 24 h p.i. Values are means ± SD (n = 3).

242 W. Wang et al. / Antiviral Research 92 (2011) 237–246

explored whether CO-1 could enter into MDCK cells by using theFITC-labeled CO-1 (CO-1-FITC) probe. In brief, MDCK cells werefirstly incubated with IAV (MOI = 1.0) for 1 h at 4 �C, washed to re-move virus and then added with infection media containing FITC-labeled CO-1 (CO-1-FITC) at the concentration of 250 lg/mL. Afterincubation for 4 h at 37 �C, the cells were washed three times be-fore confocal imaging. As shown in Fig. 6A, high levels of greenfluorescence were detected in the CO-1-FITC treated cells, andthe fluorescence was almost found in the cytoplasm and around

the cell nucleus. Moreover, we also investigated the internalizationof CO-1-FITC in non virus infected cells (Mock), and the resultshowed that CO-1-FITC could also distributed to the cytoplasm ofuninfected MDCK cells, which was very similar to that in PR8 in-fected cells (Fig. 6A). The results showed that carrageenan oligo-saccharide could enter into MDCK cells, which meant thatcarrageenan oligosaccharide might be able to influence the intra-cellular replication events of IAV life cycle.

The inhibition effect of CO-1 on IAV transcription was then eval-uated by Real-time RT-PCR assay. IAV (MOI = 2.0) infected cellswere added with 250 lg/mL of CO-1 or ribavirin after virus uptakeat 37 �C for 1 h (at 1 h p.i.) and then incubated for 4 h. After that,total RNA was extracted for real-time RT-PCR. As shown inFig. 6B, IAV NS1 mRNA levels decreased to about 20% of that of un-treated cells after ribavirin treatment, consistent with the fact thatribavirin inhibits IAV RNA synthesis (Browne, 1978). CO-1 treat-ment also reduced the IAV NS1 mRNA to no more than 30% of thatof untreated control cells (Fig. 6B). Moreover, CO-1 was also able toreduce the IAV NP mRNA to about 45% of that of untreated cells(Fig. 6B). Furthermore, quantitative RT-PCR of viral RNAs (vRNAs)was also performed to evaluate the effect of CO-1 on IAV adsorp-tion and internalization, and the results showed that CO-1 treat-ment did not reduce the amount of IAV vRNAs adsorbed toMDCK cells or internalized into cells (Figs. 6C and D), which sug-gested that CO-1 did not interfere with IAV adsorption and inter-nalization into MDCK cells. In conclusion, these results indicatedthat carrageenan oligosaccharide could inhibit IAV mRNA expres-sion after its internalization into MDCK cells.

3.7. Carrageenan oligosaccharide CO-1 could also inhibit IAV proteinexpression

The effect of CO-1 on IAV protein expression was also moni-tored using an indirect immunofluorescence assay. IAV(MOI = 2.0) infected MDCK cells were added with 250 lg/mL ofCO-1 or ribavirin after virus internalization (at 1 h p.i.) and thenincubated at 37 �C for 6 h. After that, viral NP protein expressionwas detected by immunofluorescence assay. As shown in Fig. 7,in virus-infected cells without drug treatment, the fluorescenceof viral NP proteins could be obviously found in both the cell nu-cleus and cytoplasm (Fig. 7A), while nearly non fluorescence couldbe found in the non-infected cells (Fig. 7B). After treatment withCO-1 or ribavirin for 6 h, the number of virus antigen-expressingcells was drastically reduced, and only very few fluorescence couldbe found in cells (Fig. 7C and D).

Western blot assay was also performed to determine the effectof CO-1 on viral protein production. MDCK cells were firstly in-fected with IAV (MOI = 2.0), and then treated with or without indi-cated compounds at 250 lg/ml after virus internalization (at 1 hp.i.). After incubation for 8 h, viral NP protein production was de-tected by western blot assay. As shown in Fig. 7E, after treatmentwith CO-1 or ribavirin for 8 h, the level of viral NP protein was sig-nificantly reduced compared to that of the non drug treated controlgroup (P < 0.01). The treatment with CO-1 or ribavirin inhibited theproduction of IAV NP protein by 78% and 56%, respectively (Fig. 7F).All the results indicated that carrageenan oligosaccharide couldalso inhibit IAV protein expression although the intracellular tar-gets need to be investigated further.

4. Discussion

Recently, marine oligosaccharides and their derivatives havebeen attracting increasing interest in developing potential anti-vir-al drugs (Artan et al., 2010; Ji et al., 2011). In this study, carra-geenan derived oligosaccharide CO-1, a low molecular weight

Fig. 5. Influence of different treatment conditions of CO-1 on IAV infection. (A) MDCK cells were infected with IAV (MOI=1.0) by three different treatment conditions. (i)Pretreatment: IAV was treated with 250 lg/mL of CO-1 at 37 �C for 1 h before infection. Then the virus/compound mixture was added to cells for 1 h, and then the media wereremoved and replaced with compound-free media. (ii) Adsorption: cells were infected in media containing 250 lg/mL of CO-1 and, after 1 h adsorption at 4 �C, were overlaidwith compound-free media. (iii) After-adsorption: after removed unabsorbed virus the infecting media containing 250 lg/mL of CO-1 were added to cells. At 48 h p.i., virusyields were all determined by TCID50 assay. Values are means ± SD (n = 3). Significance: ⁄P < 0.05, ⁄⁄P < 0.01 vs. virus control group. (B) MDCK (5 � 105) cells were incubatedwith or without FITC-labeled CO-1 (CO-1-FITC) at the concentration of 250 lg/mL. After incubation for 1 h at 37 �C, the cells were harvested, washed and then analyzed byFCM. The result shown is a representative of three separate experiments. (C and D) MDCK cells were exposed to PR8 (MOI = 1.0) with or without CO-1 treatment at 4 �C for60 min to allow virus binding. After that, the cells were washed and stained with anti-IAV NA antibody before FCM analysis. A minimum of ten thousand events were countedfor each sample. The mean fluorescence intensity of each sample is also shown (D). The data shown are a representative of three separate experiments. Mock: non-infectedcells; PR8: IAV infected cells without drugs; PR8+CO-1: IAV infected cells with CO-1 treatment.

W. Wang et al. / Antiviral Research 92 (2011) 237–246 243

fragment of j-carrageenan (�2 kDa), was demonstrated to be ableto effectively suppress the replication of influenza A virus in MDCKcells (IC50 < 35 lg/mL). In contrast, the carrageenan oligosaccha-rides with higher molecular weights (CO-2, CO-3) had lower inhib-itory effect on IAV multiplication (IC50 > 200 lg/mL) compared toCO-1 (Table 1). The different inhibitory activities among carra-geenan oligosaccharides of varying chain lengths might be attrib-uted to the water solubility properties and their abilities to enterinto cells, as suggested by previous studies (Ji et al., 2011; Ninget al., 2003).

The antiviral activities of carrageenan against different kinds ofvirus have been studied during the past 20 years. However, therewere only a few studies about the anti-viral effects of carrageenanoligosaccharide and its derivatives (Yamada et al., 1997; Yamadaet al., 2000). Our results indicate that carrageenan derived oligo-saccharide CO-1 protects MDCK cells from IAV induced cell deatheven at an MOI of 3.0 in a dose dependent manner and effectivelyinhibits the multiplication of IAV in MDCK cells. Moreover, reduc-tion of the amount of input virus (MOI = 0.01) obviously increasedthe inhibitory effect of CO-1 on PR8 virus in MDCK cells(IC50 = 9.7 lg/ml) (data not shown), which suggests that the antivi-ral effect of CO-1 is dependent on the relative amount of input

virus, and the action of CO-1 might be related to the interactionwith IAV virions. Besides that, CO-1 could also promote survivalof IAV-infected A549 cells and inhibit IAV replication in a dosedependent manner. Compared to A549 cells, CO-1 showed a stron-ger antiviral effect on MDCK cells (Fig. 3B). Furthermore, the inhib-itory effect of CO-1 on another IAV strain A/WSN/33 (H1N1) wascomparable to that on PR8 strain in MDCK cells (data not shown),which indicates that CO-1 could effectively inhibit H1N1 strains ofinfluenza A virus.

Carrageenan polysaccharides were reported to be able to inhibitvirus replication by preventing virus adsorption (Buck et al., 2006;Grassauer et al., 2008; Leibbrandt et al., 2010; Talarico and Da-monte, 2007). However, different to the actions of polysaccharides,oligosaccharide CO-1 treatment only during virus adsorption couldnot inhibit virus replication and did not reduce the amount of IAVadsorbed to cells (Fig. 5). Pretreatment of IAV with CO-1 beforeadsorption could inhibit virus replication, which suggests thatCO-1 might have direct actions on IAV particles before adsorption.Sulfated polymannuroguluronate (SPMG), a marine sulfated poly-saccharide, was reported to be able to bind to the cell receptor oflymphocytes to interfere the interaction between HIV and cellreceptor (Miao et al., 2004). CO-1, the marine sulfated oligosaccha-

Fig. 6. CO-1 inhibited IAV mRNA expression after its internalization into MDCK cells. (A) MDCK cells were firstly incubated with or without IAV (MOI = 1.0) for 1 h at 4 �C,then after removed the virus, 250 lg/mL of FITC-labeled CO-1 (CO-1-FITC) was added to the cells. After incubation for 4 h at 37 �C, the cells were washed three times beforeconfocal imaging. Scale bar represents 20 lm. Mock: non-infected cells; PR8: IAV infected cells. (B) IAV (MOI = 2.0) was firstly adsorbed to MDCK cells at 4 �C for 1 h, and theninternalized at 37 �C for 1 h after removal of virus inoculums. After that, 250 lg/mL of ribavirin or CO-1 were added to cells and then incubated for 4 h. Then total RNA wasextracted for real-time RT-PCR assay of IAV mRNAs and cellular b-actin mRNA. The relative amounts of virus mRNAs were determined using the comparative (2�DDCT)method. RNA levels for non-drug treated cells (Control) were assigned values of 1. Values are means ± SD (n = 4). Significance: ⁄P < 0.05, ⁄⁄P < 0.01 vs. virus control group. (C)IAV (MOI = 2.0) was adsorbed to MDCK cells for 1 h at 4 �C in the absence or presence of 250 lg/mL CO-1. Then, total RNA was extracted and the amount of adsorbed IAVvRNA molecules was determined by real-time RT-PCR. The relative amounts of virus NS1 vRNAs were determined using the comparative (2�DDCT) method. The vRNA levelsfor untreated cells (Control) were assigned values of 100. Values are means ± SD (n = 3). (D) IAV (MOI = 2.0) was firstly adsorbed to MDCK cells at 4 �C for 1 h. After incubationat 37 �C for 1 h in media containing 250 lg/mL of CO-1, cells were treated with proteinase K to remove adsorbed but not internalized virus. Then the amount of internalizedviral RNA molecules was determined by real-time RT-PCR. Values are means ± SD (n = 3).

244 W. Wang et al. / Antiviral Research 92 (2011) 237–246

ride, however, could not bind to the cell surface of MDCK cells byFCM analysis, which meant that CO-1 might not directly interferewith the association of virus with cell receptor as SPMG do. More-

over, the time course studies indicated that there was little differ-ence of the inhibitory effect when CO-1 added at different post-infection time points within 4 h, which was different from the

Fig. 7. Inhibition of IAV protein production by CO-1 treatment. (A–D) IAV(MOI = 2.0) infected MDCK cells were added with 250 lg/mL of CO-1 or ribavirinafter virus internalization (at 1 h p.i.) and then incubated at 37 �C for 6 h. After that,immunofluorescence staining was performed using anti-influenza A virus NPantibodies. Scale bar represents 20 lm. (E) MDCK cells were firstly infected withIAV (MOI = 2.0), and then treated with or without indicated compounds at 250 lg/ml after virus internalization. After incubation for 8 h, Cell lysates were separatedby SDS–PAGE and blotted for virus NP protein expression by Western blotting. Blotswere also probed for b-actin protein as loading controls. (F) Quantification ofimmunoblot for the ratio of IAV NP protein to cellular b-actin. The ratio for non-drug treated cells (PBS) were assigned values of 1 and the data presented asmean ± SD (n = 3). Significance: ⁄⁄P < 0.01 vs. virus control group (PBS).

W. Wang et al. / Antiviral Research 92 (2011) 237–246 245

behavior of ribavirin. However, CO-1 treatment at time points after4 h post-infection (6, 8 h p.i.) could not effectively inhibit IAV mul-tiplication (Fig. 4). So CO-1’s antiviral activity is largely related toits inhibition of virus life-cycle steps that occur 0–4 h after infec-tion. Furthermore, the results of cell-free neuraminidase inhibitionassay indicated that CO-1 had no inhibition effect on IAV neur-

aminidase which is required for IAV release (data not shown). Ta-ken together, CO-1 may have an antiviral effect on IAV replicationsubsequent to viral adsorption but prior to virus release in one rep-lication cycle.

It has been reported that influenza virus binding and internali-zation might proceed by a series of distinct steps. Sialic acid,although acting as an efficient attachment factor on cells, is notsufficient as an influenza virus receptor, and influenza appears torequire additional postattachment factors such as N-linked glyco-protein for successful endocytosis into host cells (Chu andWhittaker, 2004). Macropinocytosis, the main route for the non-selective uptake of extracellular fluid by cells, could be used asan alternative IAV entry route (de Vries et al., 2011). In this study,carrageenan oligosaccharide CO-1 could enter into MDCK cellswith or without IAV infection (Fig. 6), and could not bind to the cellsurface specifically, which suggests that CO-1 might enter intoMDCK cells by macropinocytosis rather than the clathrin-mediatedendocytosis (CME). Moreover, the oligosaccharide CO-1 boundvirus particles might still bind to MDCK cells and enter into cellsby macropinocytosis, so CO-1 did not block virus adsorption andinternalization efficiently (Figs. 5 and 6). However, carrageenanpolysaccharides (100 KDa) were reported to be able to inhibitIAV adsorption rather than internalization (Leibbrandt et al.,2010). The carrageenan polysaccharide has long molecular chainand network structure (Funami et al., 2007), which binds IAV par-ticles to prevent IAV adsorption to cell receptors, but hardly enterinto MDCK cells. The different action modes between carrageenanpolysaccharides and oligosaccharides might be attributed to themolecular structures and their abilities to enter into cells.

Some reports indicated that sulfated oligosaccharides and theirderivatives exhibited the virus-inactivating activity or inhibitedthe entry of herpes simplex virus (Copeland et al., 2008; Ekbladet al., 2010). However, it is not clear if sulfated oligosaccharideshave intracellular inhibition effect on virus multiplication. Hereinwe reported that carrageenan oligosaccharide CO-1 could enterinto IAV infected MDCK cells within 4 h, and distribute to the cyto-plasm after internalization, which meant that it might influencethe intracellular events of IAV life cycle. CO-1 treatment after virusinternalization could also inhibit IAV mRNA expression and proteinproduction (Figs. 6B and 7), which suggested that CO-1 might inhi-bit IAV replication after its internalization into cells. Moreover, IAVreplicates in the nucleus but CO-1 did not enter into the nucleus ofMDCK cells, so the inhibition effect of CO-1 on viral mRNA levelsmight be due to the actions on some early steps of IAV life cycle,which was also verified in time course studies (Fig. 4C). Talaricoand Damonte reported that carrageenan could block the uncoatingof dengue virus and its escape from the endosome probably due tothe association of carrageenan with virion glycoprotein (Talaricoand Damonte, 2007). So there is the possibility that CO-1 could in-hibit virus uncoating or the shuttling of vRNP to the nucleus aftervirus internalization for the association of CO-1 with IAV particles.However, the inhibition effect on IAV protein expression after virusinternalization might be associated with the decreased productionof IAV mRNAs or interference of virus protein translation. Com-bined with the results that CO-1 inhibited a step in IAV replicationsubsequent to viral internalization but prior to virus release, wespeculate that the j-carrageenan oligosaccharide CO-1 may havean antiviral effect on intracellular replication although the intracel-lular targets need to be explored further.

In conclusion, carrageenan oligosaccharide could effectively in-hibit influenza A H1N1 virus replication in vitro. Carrageenan oli-gosaccharide CO-1 did not bind to the cell surface of MDCK cellsand did not interfere with IAV adsorption directly. Different tothe actions of polysaccharide, CO-1 could enter into MDCK cellsand inhibited IAV mRNA and protein expression after its internal-ization. Carrageenan oligosaccharide CO-1 might inhibit a replica-

246 W. Wang et al. / Antiviral Research 92 (2011) 237–246

tion step in IAV life cycle subsequent to viral internalization butprior to virus release. Furthermore, the antiviral activities of CO-1 against other IAV strains (H5N1, H3N2 or newer H1N1 strains)in vitro and in vivo need to be further investigated in order to facil-itate further development of CO-1 as a virostatic agent to preventIAV infection. In a word, interference of IAV intracellular replica-tion by carrageenan oligosaccharide might be an alternative ap-proach for anti-IAV therapy.

Conflict of interest statement

None declared.

Acknowledgments

We thank Prof. Lijuan Zhang (Ocean University of China, PRC)for her helpful advice and critical readings of the manuscript. Thiswork was supported in part by the Program for Changjiang Schol-ars and Innovative Research Team in University (IRT0944), theFundamental Research Funds for the Central Universities(201113013), Shandong Provincial Natural Science Foundation(ZR2011HQ012), and Special Fund for Marine Scientific Researchin the Public Interest (201005024). Z.-Q. C. is supported byNational Basic Research Program of China (2006CB933102).

References

Alam, M.I., 2006. Investigation of the role of PKC-alpha for influenza A virus-induced signaling and of the inhibitory effect of Verapamil on virus replication.J. Biol. Chem. 281, 16707–16715.

Artan, M., Karadeniz, F., Karagozlu, M.Z., Kim, M.M., Kim, S.K., 2010. Anti-HIV-1activity of low molecular weight sulfated chitooligosaccharides. Carbohydr. Res.345, 656–662.

Basu, D., Walkiewicz, M.P., Frieman, M., Baric, R.S., Auble, D.T., Engel, D.A., 2009.Novel influenza virus NS1 antagonists block replication and restore innateimmune function. J. Virol. 83, 1881–1891.

Browne, M.J., 1978. Mechanism and specificity of action of ribavirin. Antimicrob.Agents Chemother. 15, 747–753.

Buck, C.B., Thompson, C.D., Roberts, J.N., Muller, M., Lowy, D.R., Schiller, J.T., 2006.Carrageenan is a potent inhibitor of papillomavirus infection. PLoS Pathog. 2, e69.

Carlucci, M.J., Scolaro, L.A., Noseda, M.D., Cerezo, A.S., Damonte, E.B., 2004.Protective effect of a natural carrageenan on genital herpes simplex virusinfection in mice. Antiviral Res. 64, 137–141.

Chu, V.C., Whittaker, G.R., 2004. Influenza virus entry and infection require host cellN-linked glycoprotein. Proc. Natl. Acad. Sci. USA 101 (52), 18153–18158.

Copeland, R., Balasubramaniam, A., Tiwari, V., Zhang, F.M., Bridges, A., Linhardt, R.J.,Shukla, D., Liu, J., 2008. Using a 3-O-sulfated heparin octasaccharide to inhibitthe entry of herpes simplex virus type 1. Biochemistry 47, 5774–5783.

Damonte, E.B., Matulewicz, M.C., Cerezo, A.S., 2004. Sulfated seaweedpolysaccharides as antiviral agents. Curr. Med. Chem. 11, 2399–2419.

de Vries, E., Tscherne, D.M., Wienholts, M.J., Cobos-Jimnéz, V., Scholte, F., Garcıá-Sastre, A., Rottier, P.J.M., de Haan, C.A.M., 2011. Dissection of the influenza Avirus endocytic routes reveals macropinocytosis as an alternative entrypathway. PLoS Pathog. 7 (3), e1001329.

Ekblad, M., Adamiak, B., Bergstrom, T., Johnstone, K.D., Karoli, T., Liu, L., Ferro, V.,Trybala, E., 2010. A highly lipophilic sulfated tetrasaccharide glycoside relatedto muparfostat (PI-88) exhibits virucidal activity against herpes simplex virus.Antiviral Res. 86, 196–203.

Funami, T., Hiroe, M., Noda, S., Asai, I., Ikeda, S., Nishinari, K., 2007. Influence ofmolecular structure imaged with atomic force microscopy on the rheologicalbehavior of carrageenan aqueous system in the presence or absence of cations.Food Hydrocolloid. 21, 617–629.

Girond, S., Crance, J.M., Van Cuyck-Gandre, H., Renaudet, J., Deloince, R., 1991.Antiviral activity of carrageenan on hepatitis A virus replication in cell culture.Res. Virol. 142, 261–270.

Gonzalez, M.E., Alarcon, B., Carrasco, L., 1987. Polysaccharides as antiviral agents:antiviral activity of carrageenan. Antimicrob. Agents Chemother. 31, 1388–1393.

Grassauer, A., Weinmuellner, R., Meier, C., Pretsch, A., Prieschl-Grassauer, E., Unger,H., 2008. Iota-Carrageenan is a potent inhibitor of rhinovirus infection. Virol. J.5, 107.

Hay, A.J., Wolstenholme, A.J., Skehel, J.J., Smith, M.H., 1985. The molecular basis ofthe specific anti-influenza action of amantadine. EMBO J. 4, 3021–3024.

Hayden, F.G., Pavia, A.T., 2006. Antiviral management of seasonal and pandemicinfluenza. J. Infect. Dis. 194 (Suppl. 2), S119–S126.

Hung, H.C., Tseng, C.P., Yang, J.M., Ju, Y.W., Tseng, S.N., Chen, Y.F., Chao, Y.S., Hsieh,H.P., Shih, S.R., Hsu, J.T.A., 2009. Aurintricarboxylic acid inhibits influenza virusneuraminidase. Antiviral Res. 81, 123–131.

Ji, J., Wang, L.C., Wu, H., Luan, H.M., 2011. Bio-function Summary of MarineOligosaccharides. Int. J. Biol. 3, 74–86.

Lagoja, I.M., De Clercq, E., 2008. Anti-influenza virus agents: synthesis and mode ofaction. Med. Res. Rev. 28, 1–38.

Lamb, R.A., Takeda, M., 2001. Death by influenza virus protein. Nat. Med. 7, 1286–1288.

Leibbrandt, A., Meier, C., König-Schuster, M., Weinmüllner, R., Kalthoff, D.,Pflugfelder, B., Graf, P., Frank-Gehrke, B., Beer, M., Fazekas, T., Unger, H.,Prieschl-Grassauer, E., Grassauer, A., 2010. Iota-carrageenan is a potentinhibitor of influenza A virus infection. PLoS One 5, e14320.

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data usingreal-time quantitative PCR and the 2�DDCT Method. Methods 25, 402–408.

Mendel, D.B., Tai, C.Y., Escarpe, P.A., Li, W.X., Sidwell, R.W., Huffman, J.H., Sweet, C.,Jakeman, K.J., Merson, J., Lacy, S.A., Lew, W., Williams, M.A., Zhang, L.J., Chen,M.S., Bischofberger, N., Kim, C.U., 1998. Oral administration of a prodrug of theinfluenza virus neuraminidase inhibitor GS 4071 protects mice and ferretsagainst influenza infection. Antimicrob. Agents Chemother. 42, 640–646.

Miao, B.C., Geng, M.Y., Li, J., Li, F.C., Chen, H.X., Guan, H.S., Ding, J., 2004. Sulfatedpolymannuroguluronate, a novel anti-acquired immune deficiency syndrome(AIDS) drug candidate, targeting CD4 in lymphocytes. Biochem. Pharmacol. 68,641–649.

Ning, J., Zhang, W., Yi, Y., Yang, G., Wu, Z., Yi, J., Kong, F., 2003. Synthesis of b-(1 ? 6)-Branched b-(1 ? 3) glucohexaose and its analogues containing an a-(1 ? 3) linked bond with antitumor activity. Bioorg. Med. Chem. 11, 2193–2203.

Nyberg, K., Ekblad, M., Bergström, T., Freeman, C., Parish, C.R., Ferro, V., Trybala, E.,2004. The low molecular weight heparan sulfate mimetic, PI-88, inhibits cell-to-cell spread of herpes simplex virus. Antiviral Res. 63, 15–24.

Pujol, C.A., Scolaro, L.A., Ciancia, M., Matulewicz, M.C., Cerezo, A.S., Damonte, E.B.,2006. Antiviral activity of a carrageenan from Gigartinatina skottsbergii againstintraperitoneal murine herpes simplex virus infection. Planta Med. 72, 121–125.

Reed, L., Muench, H., 1938. A simple method for estimating fifty percent endpoints.Am. J. Hyg. 27, 943–947.

Skehel, J.J., Wiley, D.C., 2000. Receptor binding and membrane fusion in virus entry:the influenza hemagglutinin. Annu. Rev. Biochem. 69, 531–569.

Talarico, L.B., Damonte, E.B., 2007. Interference in dengue virus adsorption anduncoating by carrageenans. Virology 363, 473–485.

Talarico, L.B., Noseda, M.D., Ducatti, D.R.B., Duarte, M.E., Damonte, E.B., 2011.Differential inhibition of dengue virus infection in mammalian and mosquitocells by iota-carrageenan. J. Gen. Virol. 92, 1332–1342.

WHO, 2002. WHO Manual on Animal Influenza Diagnosis and Surveillance.Wolkerstorfer, A., Kurz, H., Bachhofner, N., Szolar, O.H.J., 2009. Glycyrrhizin inhibits

influenza A virus uptake into the cell. Antiviral Res. 83, 171–178.Wu, S.H., Patel, K.B., Booth, L.J., Metcalf, J.P., Lin, H.K., Wu, W.X., 2010. Protective

essential oil attenuates influenza virus infection: an in vitro study in MDCKcells. BMC Complement. Altern. Med. 10, 69.

Yamada, T., Ogamo, A., Saito, T., Watanabe, J., Uchiyama, H., Nakagawa, Y., 1997.Preparation and anti-HIV activity of low-molecular-weight carrageenans andtheir sulfated derivatives. Carbohydr. Polym. 32 (1), 51–55.

Yamada, T., Ogamo, A., Saito, T., Uchiyama, H., Nakagawa, Y., 2000. Preparation of O-acylated low-molecular-weight carrageenans with potent anti-HIV activity andlow anticoagulant effect. Carbohydr. Polym. 41, 115–120.

Yamamoto, Y., Nunome, T., Yamauchi, R., Kato, K., Sone, Y., 1995. Structure of anexocellular polysaccharide of Lactobacillus helveticus TN-4, a spontaneousmutant strain of Lactobacillus helveticus TY1-2. Carbohydr. Res. 275, 319–332.

Yewdell, J., Garcia-Sastre, A., 2002. Influenza virus still surprises. Curr. Opin.Microbiol. 5, 414–418.