Aqueous Extracts of the Marine Brown Alga Lobophora variegata Inhibit HIV-1 Infection at the Level of Virus Entry into Cells

In recent years, marine algae have emerged as a rich and promising source of molecules with potent activities against various human pathogens. The widely distributed brown alga Lobophora variegata that is often associated with tropical coral reefs exerts strong antibacterial and antiprotozoal effects, but so far has not been associated with specific anti-viral activities. This study investigated potential HIV-1 inhibitory activity of L. variegata collected from different geographical regions, using a cell-based full replication HIV-1 reporter assay. Aqueous L. variegata extracts showed strong inhibitory effects on several HIV-1 strains, including drug-resistant and primary HIV-1 isolates, and protected even primary cells (PBMC) from HIV-1-infection. Anti-viral potency was related to ecological factors and showed clear differences depending on light exposition or epiphyte growth. Assays addressing early events of the HIV-1 replication cycle indicated that L. variegata extracts inhibited entry of HIV-1 into cells at a pre-fusion step possibly by impeding mobility of virus particles. Further characterization of the aqueous extract demonstrated that even high doses had only moderate effects on viability of cultured and primary cells (PBMCs). Imaging-based techniques revealed extract effects on the plasma membrane and actin filaments as well as induction of apoptosis at concentrations exceeding EC50 of anti-HIV-1 activity by more than 400 fold. In summary, we show for the first time that L. variegata extracts inhibit HIV-1 entry, thereby suggesting this alga as promising source for the development of novel HIV-1 inhibitors.


Introduction
Despite the implementation of the highly active anti-retroviral therapy (HAART) in 1996, infections with the human immunodeficiency virus 1 (HIV-1) still represent a global threat with more than 34 million infected individuals worldwide and 2.5 million new infections in 2011 (UNAIDS report on the global AIDS epidemic, 2012). Current treatment of HIV-1 infections still has several shortcomings with the emergence of resistant viruses, severe side effects, and high costs being the most pressing issues. Thus, there is an urgent need for novel anti-retroviral therapeutics. In recent years, marine algae have emerged as a rich source of bioactive molecules and have yielded several compounds with remarkable anti-HIV activity, e.g. lectins which inhibit HIV-1 entry and are potential HIV-1 microbicide candidates (reviewed in Huskens and Schols [1]). One of these, Griffithsin (GRFT), was isolated from the marine red alga Griffithsia sp., [2] and shows high inhibitory efficacy and safety [3].
Marine brown algal species have also been reported to produce molecules with promising anti-retroviral properties. Among these, sulfated polysaccharides, polyphenols and diterpenes are active constituents [4][5][6]. Examples include Reverse Transcriptase (RT)inhibiting diterpenes from Dictyota pfaffii and sulfated polysaccharides (fucoidans) from Adenocystis utricularis, which interfere with early events of HIV-1 replication [7,8]. Interestingly, as marine macroalgae are consumed by humans on a regular basis in large parts of Eastern Asia, the relatively low rates of HIV infected individuals in these areas may be correlated with the consumption of algae [9]. In fact, a recent proof-of-concept clinical study with two dietary algae indicated improvement of clinical outcomes of HIV-1 infection in terms of CD4 cell count and HIV-1 viral load [10].
Lobophora variegata is a common brown alga that is widely distributed in shallow water ecosystems of tropical and subtropical areas, including coral reefs of the Caribbean, the Indian Ocean, and the Red Sea [11,12]. In coral reefs, L. variegata can be an abundant part of the ecosystem and exhibits strong allelochemical . Cells were treated with serial dilutions of the aqueous L. variegata extract (ranging from 3.2 up to 100 mg/ml) in triplicates and exposed to the HIV-1 LAI strain. B: Effects of different concentrations of L. variegata extract (collected from Turneffe Reef, Belize) on viability of indicator cells were measured by the defense against potentially deleterious microorganisms [13]. It is able to induce bacterial assemblage shifts as well as sub-lethal effects on reef corals [13,14]. Kubanek et al. (2003) isolated a cyclic lactone, lobophorilide, showing strong activity against pathogenic and saprophytic marine fungi. Moreover, several studies demonstrated inhibiting effects of chemical constituents of L. variegata on several protozoans as well as anti-inflammatory and other health-promoting effects [5,[15][16][17][18][19][20]. L. variegata also contains high concentrations of phenolic compounds, mainly bromophenols [4]. Although L. variegata is a common and wellstudied organism, no specific antiviral effects have been described to date.
The high antibacterial and antiprotist potential of L. variegata, led us to investigate whether this organism also displays antiviral activity. Accordingly, we evaluated the capacity of extracts produced from L. variegata, specimens collected from coral reefs in different geographical regions (Caribbean and Red Sea) to inhibit HIV-1 infection, using various virus strains. In addition, we used live cell-based and microscopic imaging approaches to characterize mechanisms of activity and potential side effects of algal extracts from L. variegata.

Ethics statement
The study site of Rose Reef (Saudi Arabia) does not fall under any legislative protection or special designation as a marine/ environmental protected area. No special permit is required for the inshore coastal, reef, and intertidal areas around Thuwal. The Saudi Coast Guard Authority under the auspices of KAUST University issued sailing permits to the site, which included sample (algae) collection. All necessary permits for the collections at Turneffe Reef (Belize) were obtained for the described study by ALDEBARAN Marine Research&Broadcast (Hamburg, Germany) which complied with all relevant regulations. Prior to extraction, the algal material was ground into a fine powder using liquid nitrogen. Aqueous and methanolic extracts were prepared by addition of 1 ml distilled water or methanol to 100 mg of powdered algal material. Samples were briefly vortexed and extracted at 4uC overnight. Subsequently, samples were centrifuged at 13,000 g for 30 min to remove particulate material and then stored at 220uC until further use. Unless otherwise stated, the aqueous extract prepared from full-sunlight exposed thalli of L. variegata was used for all experiments.

Sample collection and extraction
In order to analyze the solubility of potentially active compounds of the aqueous L. variegata extract in organic solvents, liquid/liquid extraction was performed. Briefly, 3 ml of extract were mixed with 3 ml of hexane or chloroform, shaken for 2 min and centrifuged at 13,000 g for 20 min for efficient phase separation. The resulting solvent phases were dried in the rotation evaporator and re-suspended in ultrapure water (Chromasolv, Sigma Aldrich, Taufkirchen, Germany).

Virus stock preparation
Virus stocks were produced in HEK 293T cells by transfection with the following infectious molecular clones: pLAI.2, pNL(AD8) (HIV-1 AD8 Macrophage-Tropic R5), HIV-1 p7324-1 (multidrug resistant molecular clone with patient-derived mutations in Reverse Transcriptase encoding sequences) and pBR-NL4-3 V92th014.12-IRES-eGFP (HIV-1 NL4-3 Gag-iGFP-reporter virus, [21]) Briefly, HEK 293T cells were seeded in 6-well plates and transfected with 1 mg of the corresponding plasmid DNA per well using 3.8 ml of FuGene HD transfection reagent (Roche), following the manufacturer's instructions. After 72 h, the supernatant was harvested, centrifuged for 5 min at 2,000 g, aliquoted, and frozen at 280uC. Pseudo-typed viruses harboring the glycoprotein of the Vesicular Stomatitis Virus (VSVg) were prepared by co-transfection of HEK 293T cells with 0.5 mg of the HIV-1 env-defective molecular clone pNL4-3Denv and 0.5 mg of pMD2.G (Addgene, MA) for expression of VSV-G. The HIV-1 strain HIV-1 KIII was prepared from supernatants of the HIV-1 IIIB producer cell line KE37.1-IIIB. Briefly, KE37.1-IIIB cells were adjusted to 8610 5 per ml and cultured for 48 h. Viruscontaining culture supernatants were harvested by centrifugation at 2,000 g, passed through a 0.45 mm pore-sized filter (Sartorius, Goettingen, Germany), aliquoted and frozen at 280uC. The primary HIV-1 isolate HIV-1 PAT891 was isolated from cerebrospinal fluid (CSF) of an HIV-1 infected individual as described elsewhere [22]. Virus preparations were quantified by determining p24 levels (Gag-p24 enzyme-linked immunosorbent assay; Applied Biosystems, Carlsbad, CA), according to the manufacturer's instructions. In addition, viral preparations were tested on LC5-RIC cells in order to determine levels of infectious viruses. The volumes of virus stocks that were used for infection of LC5-RIC cells were determined according to the following critera: (i) more than 100-fold increase of the relative fluorescent signal by HIV-1 infection, (ii) relative signal induction levels below the plateau, and (iii) a reduction of the relative MTT signal by less than 10%. [22]. colorimetric MTT assay. LC5-RIC reporter cells were treated with serial dilutions (ranging from 3.1 up to 400 mg/ml) of the aqueous L. variegata extract for 48 hours. Data are given relative to untreated cells (100%). doi:10.1371/journal.pone.0103895.g001 HIV-1 Inhibition by Lobophora Aqueous Extract PLOS ONE | www.plosone.org Standard assay setup for testing of HIV-1 inhibitory activities A description of the full assay procedure can be found elsewhere [22]. Briefly, LC5-RIC cells were seeded into 96-well plates (mCLEAR-Plate Black; Greiner Bio-One, Kremsmuenster, Germany) at a density of 10 4 per well at 24 h prior to infection. L. variegata aqueous extracts were tested in serial dilutions in triplicates. Stock solutions of aequous L. variegata extracts, reference compounds and virus preparations were diluted with cell culture medium. For treatment and infection, 100 ml of compound solution and 20 ml of diluted virus inoculum were added to each test well, with each extract concentration tested in triplicate wells. Plates were incubated for 48 h after virus addition under standard conditions at 37uC in 5% CO and subsequently tested for fluorescent reporter signal intensity with a fluorescence microplate reader (Tecan infinite M200 (Tecan, Crailsheim, Germany)) at wavelengths 552 nm for excitation and 596 for emission. The influence of the extracts on the viability of LC5-RIC cells was determined by a standard MTT test [23]

RNA and DNA quantification
For the quantification of viral RNA and DNA in infected LC5-CD4 cells, cells were seeded in 12-well plates. 24 hours later medium was replaced and cells were treated with either 65 mg/ml L. variegata aqueous extract or PBS and exposed to HIV-1 LAI . For the RNA-quantification, 200 nM Efavirenz was added to each sample to inhibit RNA-turnover by HIV-1 reverse transcriptase. Samples for RNA and DNA quantification were collected 4 or 24 hours p.i., respectively. After washing of cells with PBS before (three times) and after (twice) trypsination, RNA and DNA were isolated using the RNeasy kit or the DNA Mini (QIAGEN) according to the manufacturer's manuals. RNA and DNA quantification was performed as described in Helfer et al. (2014) [24]. Briefly, RNA levels were quantified by relative qPCR with specific primers for HIV-1 and calculation was performed with the 22 2DDCT method. Expression of RNA-Polymerase II was used as reference. For determination of HIV-1 DNA loads, an absolute quantification was done by qPCR using the TH4-7-5 cell line as external standard. b-globin was used as reference gene (Kabamba-Mukadi et al., 2005) [25].

PBMC Assay
Peripheral blood mononuclear cells (PBMCs) were prepared from whole blood of healthy donors (obtained from the blood bank of the German Red Cross, Munich, Germany) according to a standard protocol (Current protocols in immunology; 7.1.1-7.1.8 April 2009, Supplement 85). 72 h prior to infection, PBMC (pools of cells from 4 donors) were stimulated by the addition of Interleukin-2 (50 units per ml of cell culture medium; Sigma-Aldrich, Taufkirchen, Germany). For infection experiments, 5610 5 cells per well were seeded into 96-well plates. Treatment and infection of PBMC was carried out according to the standard assay setup as described above by using the HIV-1 LAI strain for infection. Four days after infection, 20 ml of supernatant from each well was transferred to LC5-RIC indicator cells in order to assay the production of infectious HIV-1 progeny from treated or untreated PBMC. The reporter signal intensity of the LC5-RIC indicator cells was measured at 48 h after addition of PBMC supernatants. Vitality of PBMCs was determined by an MTT test as described for LC5-RIC cells.

High Content Analysis
HeLa cells were transferred into 96-well plates, treated with different concentrations of L. variegata-extract for 24 h, then fixed and nuclei stained with Hoechst 33342 and b-tubulin stained with b-tubulin Alexa Fluor 488 (Life Technologies, Darmstadt, Germany). Images were acquired with a ZEISS Cell Observer (Carl Zeiss, Jena, Germany) system and a 10x Plan-NEOfluar objective with filters sets No. 49 (360/40; FT 400; BP 460/50 and 38 (475/40; FT 500; BP 530/50). Images were analyzed using the Physiology Analyst of the ASSAYbuilder module of AxioVision 4.6. Following the manufacturers protocol, nuclei were used to automatically detect cells. A ring mask around the nucleus was generated and used to detect b-tubulin fluorescence. The cell numbers and the nucleus shape of individual cells was determined for at least 4 randomly selected fields of view per well. The number of untreated control cells was set to 100%, and a treatment-caused alteration was set in relation to this. The average nuclear form factor of the control cells was used as a reference for a normal state. Increased form factor represents a tendency towards fragmented and irregular nuclei, which is usually an indicator of apoptosis, whereas a decreased form factor typically results from rounded up cells, indicating necrosis.

Single Cell Analysis
By controlling locations of adhesive (fibronectin) and nonadhesive areas (PEG) on a coverslip, micro-patterns are generated and result in a highly reproducible and polarized cell organization. Using this technique, many thousands of living cells can be positioned on a single coverslip chip, and cellular feature analysis can subsequently be performed with optimal reproducibility and accuracy. HeLa cells were plated on CYTOO chip (Cytoo, Grenoble, France) coverslip with Y-shaped fibronectin micropatterns. Cells were either left untreated (control) or treated for

Calculation of values and curve fits
Curve fits and EC 50 calculations were performed by SigmaPlot, version 11.0 (Systat Software, Chicago, IL) or Prism, version 4 (GraphPad Software, La Jolla, CA). Statistical analysis of data was carried out with the GraphPad Prism program. Significances of differences between data sets were determined by calculating twotailed P values using the Mann-Whitney-U test.

Aqueous extracts of Lobophora variegata specimens from different geographical origins inhibit HIV-1 replication
Several aqueous extracts of Lobophora variegata collected from various geographical origins were tested in a cell-based fullreplication assay for anti-HIV-1 activity. All aqueous extracts inhibited HIV-1 replication in a clearly dose-dependent manner. ( Figure 1A). Liquid/liquid extractions of the L. variegata extracts with organic solvents revealed anti-HIV-1 activity solely in the aqueous and not in the organic phases, pointing to a highly hydrophilic nature of the active components (data not shown).

L. variegata aqueous extracts show only minor toxic effects at even high concentrations
Effects of L. variegata extracts on cells were first assessed by MTT assay, which measures cellular metabolic activity and is often used to assay cell viability. Viability of the indicator cells was not affected by treatment with L. variegata extracts at concentrations up to 400 mg ml 21 , which exceeded the concentration required for full HIV-1 inhibition (approx. 50 mg ml 21 ) (Figure 1B). Even a long-term exposure of Hela cells with a single dose of 400 mg ml 21 for up to seven days did not result in any detectable toxicity (Fig. S1 in File S1). Toxic effects were also studied in HeLa cells treated with extract concentrations up to 2 mg ml 21 by High Content Analysis of cells with stained nuclei and microtubuli. Whereas no significant effects on cell cycle state (ploidity) and b-tubulin were observed (data not shown), mild cell loss of approximately 20% and a slight induction of apoptosis was observed in cells treated with the highest extract concentration ( Figure 2A).
Next, effects of the L. variegata aqueous extract were analyzed on the single cell level. Microscopic images clearly depicted pronounced changes of several cellular features at higher extract concentrations, including a reduced membrane signal, re-arrangement of actin filaments, and a less well-defined Y-shaped morphology ( Figure 2B). Automated cell feature extraction by High Content Analysis revealed a significant increase of cell size (mean cell area), reduction of mean actin intensity, increase in the number of actin fiber bundles, and substantially increased number of membrane aggregates ( Figure 2C).  (Table 1). L. variegata extracts inhibited infection by the two X4-tropic (lymphotropic) strains, HIV-1 LAI and HIV-1 KIII as well as the R5-tropic (macrophage-tropic) strain, treated (50 mg ml 21 , nuclei in blue, actin in green, membrane in orange/red HIV-1 AD8 . The corresponding EC 50 values ranged from 2.6 to 8.1 mg extract per ml and did not correlate with virus tropism. Moreover, we found a potent inhibitory activity against the multidrug-resistant HIV-1 strain and the primary HIV-1 virus isolate. Furthermore, L. variegata extracts also inhibited infection by the VSV-G pseudotyped HIV-1 particles with an EC 50 value similar to those obtained for native HIV-1 strains. As HeLa cells are not the natural target cells of HIV-1, we further tested the potency of L. variegata aqueous extract to protect primary cells from HIV-1 infection. We found that L. variegata aqueous extract also inhibited HIV-1 replication in peripheral blood mononuclear cells ( Figure 3A) with a moderate decrease of cell vitality at intermediate extract concentrations (Figure 3B).

L. variegata aqueous extract inhibits HIV-1 infection at an early stage of virus replication
Our next aim was to characterize the mode-of-action by which the L. variegata extract inhibits HIV-1 replication. We first employed a time-of-addition assay [22] to identify which steps of the HIV-1 replication cycle are affected by the L. variegata extract ( Figure 4A). This assay revealed that L. variegata extract clearly inhibits an early step in the HIV-1 replication cycle which is different from and precedes the steps targeted by the entry inhibitor Griffithsin, the fusion inhibitor T-20 and the reverse transcription inhibitor Efavirenz. To evaluate the effect of L. variegata extract on HIV-1 attachment behavior, GFP-labelled virus particles were incubated with HIV-1 indicator cells in medium containing investigational inhibitors (i.e. L. variegata extract or Griffithsin) and the distribution of HIV-1 particles was analyzed by spinning-disc confocal microscopy ( Figure 4B). All samples also contained the fusion inhibitor T20 to prevent virus uptake. Virus particles associated predominantly with host cells in virus-containing samples lacking investigational inibitors (i.e. virus control in Figure 4B). Attachment of virus particles was increased even further by treatment with Griffithsin, which is in agreement with the reported enhancement of virus binding to the CD4receptor by Griffithsin [26]. In contrast, treatment with L. variegata extract led to more random distribution of virus particles with occurrence of virus particles in cell-free areas of the slide, indicating arbitrary attachment of the virus particles to noncellular surfaces. These results suggest that L. variegata extracts may change the attachment behavior of virus particles, possibly promoting their attachment to unspecific surfaces. Additional experiments were conducted to validate the inhibition of HIV-1 at an early stage of the viral replication cycle. Quantification of intracellular HIV-1 RNA levels in HIV-1 target cells (LC5-CD4) at 4 hours post infection showed a substantial reduction of viral RNA levels in cells treated with the L. variegata extract ( Figure 5A). Similarly, quantification of proviral copies revealed a clearly reduced viral load in extract-treated cells ( Figure 5B).

Environmental conditions affect anti-HIV-1 activity of the L. variegata aqueous extract
To test whether growth conditions of Lobophora affect its HIV-1 inhibitory potency we compared the activities of Lobophora samples collected from various settings. All samples showed anti-HIV-1 activity, regardless of their geographical origin (Caribbean and Red Sea). However, specimens from the same geographical origins that were exposed to different environmental conditions showed different potencies of anti-HIV-1 activity. Thus, extracts of Northern Red Sea specimens collected from a shady location showed higher anti-HIV-1 activity than extracts of specimens from a sunny location. (Figure 6A). Furthermore, extracts of Caribbean specimens that had been overgrown by epiphytes showed higher anti-HIV-1 activity than extracts of samples free of epiphytes. (Figure 6B).

Discussion
In this study, a combination of a cell-based HIV-1 full replication assay and imaging-based techniques was used to identify and characterize anti-HIV-1 activities in a crude aqueous extract of the brown alga Lobophora variegata.
All aqueous extracts of L. variegata showed potent and dosedependent inhibition of HIV-1 replication, although the specimens used in their production originated from different and far apart geographical locations, (i.e. Atlantic (Caribbean) and Indo-Pacific (Red Sea). This suggests that all L. variegata specimens are capable of inhibiting HIV-1 activity by a common principle.
Anti-HIV-1 activity of L. variegata was detected exclusively in aqueous but not in organic phases of the extract, indicating a highly polar nature of the active principle. Similarly, other algaederived molecules with high anti-retroviral activity, such as sulfated polysaccharides and the lectin Griffithsin, were also identified in aqueous extracts [2,6]. In contrast, anti-protozoal activities of L. variegata occurred exclusively in organic extracts. This indicates that anti-protozoal and anti-HIV-1 activities are mediated by different, unrelated components of L. variegata.
Evaluation of cellular toxicity of extracts by MTT assay, which is an indicator of cell metabolism [27], revealed no toxic effects of the extract on the HIV-1 indicator cells even at the highest concentrations. To evaluate effects of L. variegata on other cellular parameters we used High Content Analysis which allows simultaneous analysis of multiple cellular parameters. This technology, revealed no impairment of the cell cycle at any concentration (data not shown), at any concentration. However, changes in nuclear shapes associated with apoptosis, increased cell areas and changes in the actin cytoskeleton were apparent at the highest extract concentration of 2 mg ml 21 . Furthermore, treatment with L. variegata extract at this high concentration had a relatively pronounced effect on the plasma membrane, suggesting that the L. variegata aqueous extract also contains a membraneactive constituent. These cellular effects only occur at much higher  extract concentrations than those required for anti-HIV-1 activity and can most probably be attributed to other compounds. However, to further improve the safety of L. variegata extracts for humans both activities should be separated.
L. variegata extracts inhibited replication of all tested virus strains including HIV-1 strains with different cellular tropisms, a multi-drug resistant HIV-1 strain and a primary HIV-1 isolate obtained from an infected individual. In addition, anti-HIV-1 activity was confirmed with primary human HIV-1 target cells (PBMC). Together these data support the ability of the L. variegata aqueous extract to inhibit HIV-1 infection under physiologically relevant conditions.
To learn more about the mechanism-of-action by which the extract targets HIV-1 replication, we employed a strategy of timedependent addition of the test compound to infected cells. This method was recently demonstrated to be adaptable to crude biological extracts [22]. The resulting inhibition curve clearly points to an interference of the L. variegata aqueous extract with a very early step of virus replication. A substantial reduction of viral RNA levels and proviral copies in infected cells treated with the L. variegata extract further support this hypothesis. We next analyzed the effect of the extract on the attachment of GFPlabeled HIV-1 particles to target cells by microscopic imaging. The observed unspecific immobilization of viral particles on surfaces including non-cellular parts indicates a direct interaction of extract compounds with HIV-1 virions. Although viral particles can be found attached to cells, viral entry is obviously impaired. This may either be related to a virucidal effect or to an immobilization of the virus on the cell membrane making it impossible for the virus to further proceed in its replication cycle. The pronounced effect of the extract on the plasma membrane at high extract concentrations, including a reduced membrane signal and an increase of membrane aggregates, might be related to the presence of membrane-active compounds. These could interfere with the viral membrane leading to the observed immobilization and impaired viral entry into target cells. In addition, the strong inhibition of virions pseudotyped with a non-HIV-1 glycoprotein (VSV-g) makes a specific interaction of extract compounds with the HIV-1 glycoproteins unlikely and argues in favor of the viral membrane as the target structure. This inhibitory mechanism is different from the well-characterized HIV-1 entry inhibitors T-20 and Griffithsin or the plant-derived triterpene glycyrrhizin which may reduce the fluidity of the plasma membrane leading to an inhibition of virus/cell fusion [28,29]. Thus, the anti-viral effect of the L. variegata aqueous extract may represent a new inhibitory principle and should be characterized in more detail.
L. variegata inhabits a wide range of ecological settings and can have strong allelochemical effects on its environment [13,14]. It is likely that different growth conditions may have effects on the chemical composition of the thalli and thus on anti-viral activity.
Our results clearly showed pronounced differences in anti-HIV-1 activities of algae from different environmental settings. Apparently, environmental factors such as light conditions or overgrowth by epiphytes influence anti-retroviral activities of specimens. Environmental stress generally leads to the induction of diverse protective molecules in algae [30]. These include polyphenols, sulfated polysaccharides, or proteins and are used as sunscreens, scavengers of reactive oxygen species (ROS), deterrents, or antimicrobial agents by these organisms [30,31]. Thus, it is possible that the anti-HIV-1 activity observed in the aqueous extract may be part of the stress response of this alga.
In summary, our results reveal for the first time a strong and broad anti-retroviral activity of the common brown macroalga L. variegata and suggest a direct interaction of extract constituents with the viral particles. These findings further complete the rich anti-microbial spectrum of this algal species and suggest its extracts as a promising source for the development of novel HIV-1 inhibitors.

Supporting Information
File S1 Figure S1 and supplemental experimental procedures. Figure S1: Time-dependent effect of the L. variegata aqueous extract on the viability of Hela cells. A time series of the effect of the L. variegata aqueous extract on viability of Hela cells was determined by a colorimetric XTT assay. Hela cells were treated with a fixed concentration of 400 mg/ml of the extract for time periods of one day up to 7 days. Data are given relative to untreated cells measured at the same time point (100%). (DOCX)