Chikungunya virus entry is strongly inhibited by phospholipase A2 isolated from the venom of Crotalus durissus terrificus

Chikungunya virus (CHIKV) is the etiologic agent of Chikungunya fever, a globally spreading mosquito-borne disease. There is no approved antiviral or vaccine against CHIKV, highlighting an urgent need for novel therapies. In this context, snake venom proteins have demonstrated antiviral activity against several viruses, including arboviruses which are relevant to public health. In particular, the phospholipase A2CB (PLA2CB), a protein isolated from the venom of Crotalus durissus terrificus was previously shown to possess anti-inflammatory, antiparasitic, antibacterial and antiviral activities. In this study, we investigated the multiple effects of PLA2CB on the CHIKV replicative cycle in BHK-21 cells using CHIKV-nanoluc, a marker virus carrying nanoluciferase reporter. The results demonstrated that PLA2CB possess a strong anti-CHIKV activity with a selectivity index of 128. We identified that PLA2CB treatment protected cells against CHIKV infection, strongly impairing virus entry by reducing adsorption and post-attachment stages. Moreover, PLA2CB presented a modest yet significant activity towards post-entry stages of CHIKV replicative cycle. Molecular docking calculations indicated that PLA2CB may interact with CHIKV glycoproteins, mainly with E1 through hydrophobic interactions. In addition, infrared spectroscopy measurements indicated interactions of PLA2CB and CHIKV glycoproteins, corroborating with data from in silico analyses. Collectively, this data demonstrated the multiple antiviral effects of PLA2CB on the CHIKV replicative cycle, and suggest that PLA2CB interacts with CHIKV glycoproteins and that this interaction blocks binding of CHIKV virions to the host cells.

www.nature.com/scientificreports/ of patients 14,15 . Unlike other arboviruses, CHIKV infection can result in chronic symptoms lasting for months or years, resulting in a disabling disease 16,17 . There are no approved antiviral drugs against CHIKV infection, as a consequence the treatment is often palliative and symptomatic, based on analgesics, non-steroidal antiinflammatory, rest, and hydration 18 . Given that many approved drugs employed in the treatment of infectious and chronic diseases originated or derived from natural sources 19,20 , it is reasonable to hypothesize that natural compounds may also be exploited to generate antiviral drugs. In this context, proteins isolated from snake venoms represent promising drug leads, since they are a complex mixture of lectins, oxidases, disintegrins, metalloproteins, and phospholipases A2 (PLA2s) 21,22 . From these, PLA2s, in its turn, are members of a secreted phospholipases, which can act in the cell membranes and play several roles in biological systems [23][24][25] .
The snake venom isolated from Crotalus durissus terrificus has numerous constituents such as crotoxin, crotamin, neurotoxin, among others 26,27 . Crotoxin is the major constituent of the C. d. terrificus venom. It is characterized as a protein complex composed by two noncovalent subunits, the basic subunit phospholipase A2 (PLA2 CB ), and the acid subunit crotapotin 28,29 . Subsequently, PLA2 CB is approximately 14 kDa protein which possess anti-inflammatory, antiparasitic, and antibacterial properties 30,31 . PLA2 CB has also presented activity towards viruses such as hepacivirus C (HCV) 32 , Rocio (ROCV), Mayaro (MAYV), Dengue (DENV) and Yellow Fever (YFV) 33,34 . Russo and coworkers expressed and purified two recombinant PLA2 CB (rPLA2 CB ) and partially assessed its anti-CHIKV activity. It was found that rPLA2 CB proteins possess lower antiviral activity and higher cytotoxicity profile than the native protein, probably due to nine additional amino acid residues present in their sequences 35 . Considering these previous results, herein we performed thorough in vitro evaluation of the effects of the native PLA2 CB on the CHIKV replication cycle.

PLA2 CB strongly impairs CHIKV infection in vitro.
We investigated the anti-CHIKV activity of the PLA2 CB (Fig. 1A) using BHK-21 cells and a recombinant CHIKV that expresses a nanoluciferase reporter (CHIKV-nanoluc) (Fig. 1B) 36,37 . First, the PLA2 CB antiviral activity was evaluated by performing a dose-response assay to determine the effective concentration of 50% (EC 50 ) and cytotoxicity of 50% (CC 50 ). BHK-21 cells were infected with CHIKV-nanoluc and simultaneously treated with PLA2 CB at concentrations ranging from 0.195 to 200 µg/mL in two-fold serial dilutions, and viral replication was assessed 16 h post-infection (h.p.i.) (Fig. 1C). In parallel, cell viability was assessed by an MTT assay. PLA2 CB was found to be able to inhibit virus replication to greater than 99%, while the cell viability at the highest concentration tested was 43%. It was determined that PLA2 CB has the EC50 of 1.34 µg/mL, CC50 of 172 µg/mL, and the Selectivity Index (SI) of 128 (Fig. 1D). Thus, PLA2 CB acts as strongly inhibitor of CHIKV infection with high SI value.   1D).
To assess the protective effects of PLA2 CB against CHIKV infection, cells were pretreated with PLA2 CB for 1 h at 37 °C, washed extensively with PBS to remove the compound and infected with CHIKV-nanoluc for 1 h. Then, the supernatant was removed, cells were added of fresh medium and luciferase levels were measured 16 h.p.i. (Fig. 2A). PLA2 CB significantly reduced CHIKV-nanoluc infection by 84% (p < 0.01), demonstrating a robust protective effect ( Fig. 2A). The protective effect did not increase when the compounds was present for all duration of the experiment (Fig. 2B), demonstrating that pre-treatment of cells with PLA2 CB inhibited CHIKV replication. This data suggests that PLA2 CB acts by protecting cells against infection and/or by affecting early stages of CHIKV infection.
To further evaluate the PLA2 CB effect on CHIKV entry to the host cells, virus and PLA2 CB were simultaneously added to BHK-21 cells for 1 h at 37 °C, cells were washed with PBS and replaced with fresh medium (Fig. 3A). PLA2 CB demonstrated to decrease 95.3% of CHIKV replication (p < 0.0001), indicating that this compound strongly inhibited the CHIKV-nanoluc entry (Fig. 3A). Combining this treatment with 1 h pre-incubation of the inoculum containing PLA2CB and CHIKV at 37 °C further increased inhibition that reached over 99% (Fig. 3B), indicating that PLA2 CB also possesses virucidal activity. To analyze the effect of PLA2 CB on CHIKV attachment, virus and compound were first incubated with the cells at 4 °C for 1 h. At this temperature, virus particles were able to attach to the cellular receptors, but not entry into the host cells. Cells were then washed with PBS, fresh medium added, and incubated at 37 °C (Fig. 3C), to allow the continuation of the entry process. Data obtained from this assay also showed strong inhibition of CHIKV attachment by reducing virus entry by 98.2% to the cells (p < 0.0001) (Fig. 3C). Post-attachment was evaluated by including an additional incubation of 30 min at 37 °C to the previous protocol (Fig. 3D), showing that the inhibition reminded strong reaching 95.2% (p < 0.0001) (Fig. 3D). Taken together this data indicates that PLA2 CB possesses a robust virucidal activity and the ability to block virus entry to host cells.
PLA2 CB moderately affect post-entry steps of CHIKV infection. Two assays were used to analyze effects of PLA2 CB on post-entry stages of CHIKV infection. Using CHIKV-nanoluc it was found that, if added after virus infection, compound cause relatively modest, 64% reduction of CHIKV replication (p < 0.0001)
PLA2 CB causes molecular changes in CHIKV glycoprotein. To further investigate the interactions between PLA2 CB and CHIKV particles, infrared spectroscopy spectral analysis and vibrational analysis among the virus and PLA2 CB was performed. Representative means of the infrared spectrum of CHIKV, PLA2 CB, and CHIKV plus PLA2 CB , which is the bio fingerprint region representing proteins, lipids, nucleic acids, and glycoproteins are shown in Fig. 8A. A representative infrared average spectrum of second derivative analysis from CHIKV virions, PLA2 CB, and CHIKV virions plus PLA2 CB is displayed in Fig. 8A. In the second derivative    (Fig. 8B). Furthermore, the Stacked Walls (Fig. 9A) and split heat map (Fig. 9B) reinforces the additional expression of vibrational mode at 1068 cm −1 under CHIKV virions plus PLA2 CB association.

Discussion
Natural PLA2 CB has shown to have broad spectrum antiviral activity [32][33][34] . Here, we assessed the antiviral activity of the PLA2 CB against CHIKV, as well as sought comprehension on its mechanism of action. Our results demonstrated that PLA2 CB strongly inhibited CHIKV infection, corroborating with Russo and colleagues work, which demonstrated that rPLA2 CB impaired CHIKV infection 35 . Additionally, the results demonstrated that the pre-treatment of naïve cells with PLA2 CB protected host cells against CHIKV infection. In accordance with our study, Chen and coworkers have reported that phospholipase A2 isolated from the venom of the honeybee Apis mellifera was able to protect cells against Human immunodeficiency virus (HIV) and dengue virus (DENV) infections 41 . Fenard and colleagues also demonstrated that cells can be protected against HIV infection by different phospholipases A2 isolated from several mammalian species 42 . The PLA2s from snake venoms are classified  www.nature.com/scientificreports/ in the group II of a secreted family phospholipases and show homology to the mammalian inflammatory PLA 2 , which play different roles in the organism including in an immune response to infectious diseases [43][44][45][46] . Therefore, our data might also suggest that PLA2 CB plays a role in host cell metabolism and as a result protects cells against viral infection, by the possible mimicking effect of phospholipases found in host cells.  www.nature.com/scientificreports/ Our findings that PLA2 CB has strong virucidal effect and interferes with viral entry to the host cells are consistent with previous findings made for two flaviviruses, DENV and yellow-fever (YFV) 34 . The authors demonstrated that incubation of DENV or YFV virions with PLA2 CB results in inhibition of early steps of viral infection probably by disrupting virion envelope membrane and/or blocking virus adsorption 33,34 . Additionally, Russo and coworkers described that incubation of rPLA2 CB with CHIKV prior to the infection of cells significantly impaired CHIKV infectivity 35 . Therefore, our results are in agreement with previous data that suggested that the predominant activity of PLA2 CB is due to its virucidal effect, probably by acting on the virus particle. Several PLA2s isolated from snake venom have been described to possess antiviral activity against DENV, YFV, Herpes simplex types 1 and 2 and Influenza A (H3N2) by interacting with lipid membrane found in a pocket between glycoproteins and/or through attachment to the glycoproteins in the viral envelope surface 33,34,41,47 . Based on this data, we performed a blind molecular docking using PLA2 CB and the CHIKV glycoproteins complex (E1, E2, and E3) to assess the possible interaction among then. The results demonstrated that PLA2 CB most likely bonded to both E1 and E2 with E1 being the main target. These results are also consistent with the virucidal effect described here and corroborate previously published data 35 . The glycoproteins E1 and E2 are essential during the early stages of CHIKV infection. The CHIKV glycoprotein E2 is responsible for binding to cells receptors such as MXRA 48 ; the binding occurs in the "canyon" between protomers of CHIKV spike complex. In addition, E1 is a viral fusion protein that ensures the envelope fusion with host cells membranes 49,50 . Thus, molecules that can interact with E1 and/or E2 attachment sites can prevent virus entry 51 . We propose that PLA2 CB might act be binding to E1 and/or E2 glycoproteins and preventing virion entry by impairing attachment to the cells and/or membrane fusion. PLA2 CB /glycoproteins interactions, characterized using an infrared spectrum assay, did indeed support the hypothesis of interaction between CHIKV virions and PLA2 CB . On the other hand, it is important to emphasize that alphavirus virions exist in two state with different characteristics, a normal infectious state and pre-fusion state when E1 is activated prior to attachment [52][53][54] . In this context, the employment of cryogenic electron microscopy (cryo-EM) or atomic force microscopy presents as potential approaches to better characterize this interaction.
Moreover, PLA2 CB also demonstrated a modest yet significant ability of inhibit post-entry steps of CHIKV infection. Using CHIKV subgenomic replicon it was found that both RNA replication and subgenomic RNA synthesis were inhibited. Shimizu and coworkers have previously reported that PLA2 CB also affected the postentry steps of HCV infection 32 , corroborating with our findings. The mode how PLA2 CB affects RNA replication step remains currently unknown. However, it can be speculated that this may be associated with its effect on cell membranes that are required for formation of RNA replicase complexes. The exact mechanism of this as well as similarities and differences between anti-HCV and anti-CHIKV effects represent topics for additional studies.
In summary, our study evidenced that PLA2 CB isolated from Crotalus durissus terrificus inhibited multiple steps of CHIKV infection. The enzyme was able to protect the target cells against CHIKV infection, impaired virus entry to the host cells, mainly by virucidal activity, and also disturbed post-entry steps of the CHIKV infection. Therefore, this data might be useful for further development of new antiviral approaches against CHIKV and provide the potential for treatment of Chikungunya fever.

Methods
Compound. The crude venom of Crotalus durissus terrificus was obtained from the "Animal Toxin Extraction Center" (CETA), duly registered and approved by the Ministry of the Environment under de process number 3002678. The venom was collected from 28 specimens from the Morungaba-SP collection under the Brazilian Institute for the Environment and Renewing Natural Resources (IBAMA) authorization: 1/35/1998/000846-1, and extraction was performed by Jairo Marques do Vale (CETA). All experiments were performed in accordance with relevant named guidelines and regulations available in the federal universities, IBAMA and the Ministry of Environment. The isolation and purification of phospholipase PLA2 CB (Fig. 1A) from the venom of Crotalus durissus terrificus snakes were carried out at the Toxinology Laboratory of the School of Pharmaceutical Sciences of Ribeirão Preto, University of São Paulo, as previously described 28,34 . The lyophilized protein was dissolved in PBS (phosphate buffer saline), filtered, and stored at -80ºC. Dilutions of the stock solution containing the protein were made immediately prior to the experiments. For all the performed assays, PBS was used as the untreated control. All authors complied with the ARRIVE guidelines. www.nature.com/scientificreports/ CHIKV-nanoluc. Cells were incubated with virus for 1 h 37 °C; after this, the inoculums were removed, cells were washed with PBS to remove the unbound virus, and fresh medium supplemented with 1% dilution of stock of penicillin and streptomycin, 2% FBS and 1% carboxymethyl cellulose (CMC) was added. Infected cells were incubated for 2 days in a humidified 5% CO 2 incubator at 37 °C, followed by fixation with 4% formaldehyde and staining with 0.5% violet crystal. The viral foci were counted to determine viral titer which was presented in plaque forming units per milliliter (PFU/mL).

Cell viability assay. Cell viability was measured by MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetra-
zolium bromide] (SIGMA-ALDRICH) assay as previously described 55 . BHK-21 cells were plated to 48 well plates at a density of 5 × 10 4 cells per well and incubated overnight at 37 °C. Medium containing two-fold serial dilutions of PLA2 CB (from 0.195 to 200 µg/mL) was added and cells were incubated for 16 h. After this, the medium was replaced with the MTT solution at 1 mg/mL, cells were incubated for 30 min, after which MTT solution was and replaced with 300 μL of DMSO (dimethyl sulfoxide) to solubilize the formazan crystals. The absorbance was measured at 490 nm on the Glomax microplate reader (PROMEGA). Cell viability was calculated according to the equation (T/C) × 100%, where T and C represent the mean optical density of the treated and untreated control groups, respectively. The cytotoxic concentration of 50% (CC 50 ) was calculated using GraphPad Prism 8.  (Fig. 2A). Alternatively, cells were treated for 1 h with the compound, washed with PBS and infected with CHIKV-nanoluc at the presence (Fig. 2B) of PLA2 CB for 16 h.

Determination of the effective concentration 50% (EC 50 ).
In entry inhibition assay, cells were infected using media containing the compound-and virus for 1 h, washed with PBS and incubated with fresh medium for 16 h (Fig. 3A). The virucidal activity was assessed using the same setting except inoculum containing compound and virus was incubated for 1 h before it was added to the cells (Fig. 3B). The impact of compound on attachment step was analyzed using the same setting as in entry inhibition assay except cells were incubated with virus and compound at 4 °C (Fig. 3C). A variant of this assay where the incubation at 4 °C was followed by incubation for 30 min at 37 °C was used to analyze the effect of compound on post-attachment steps of infection (Fig. 3D).
In post-entry assay, cells were infected with CHIKV for 1 h, washed extensively with PBS, and the incubated in compound-containing medium for 16 h (Fig. 4A).

RNA replication assay using BHK-CHIKV-NCT cells. BHK-CHIKV-NCT cells that express CHIKV
nonstructural proteins, a selection marker (puromycin acetyltransferase, Pac) and Renilla luciferase and EGFP reporters 37 , were used to assess the activity of PLA2 CB on CHIKV RNA replication. Cells were seed at a density of 7 × 10 3 cells per well of a 96 well plate. After 24 h, cells were treated with the PLA2 CB at 12.5 µg/mL for 72 h (Fig. 4B). The impact of compound on CHIKV RNA replication was estimated by quantification of Renilla luciferase expression. In addition, EGFP fluorescence was monitored by placing plates directly using an EVOS (THERMO-FISCHER) fluorescence microscope and using 20 × lens and GFP filter.

Molecular docking analysis.
The interaction between PLA2 CB (PDB: 3R0L) and the envelope glycoproteins of the CHIKV (PDB: 3N42) was analyzed using blind docking performed in the PatchDock server 56 , using the parameters predefined by the program and refined by the FireDock algorithm 57 . The best docking positions were evaluated by the geometric complementarity score defined by PatchDock, with results refined and ranked by the global energy after refinement. The post-docking 3D image was generated in the DS Visualizer program, Dassault Systèmes BIOVIA, Discovery Studio Visualizer, version 17, San Diego: Dassault Systèmes, 2016, and a 2D diagram of the interactions interface between the molecules was generated with the aid of the LigPlot + program 58 .
Infrared spectroscopy spectral data analysis. An Fourrier Transform Infrared (FTIR) spectrophotometer Vertex 70 (BRUKER OPTICS, REINSTETTEN, Germany) connected to a micro-attenuated total reflectance (ATR) platform was used to record sample signature at 1800 cm −1 to 400 cm −1 regions as described by Oliveira and coworkers 55 . The ATR unit is composed of a diamond disc as an internal-reflection element. The sample dehydrated pellicle penetration depth ranges between 0.1 and 2 μm and depends on the wavelength, incidence angle of the beam, and the refractive index of ATR-crystal material. The infrared beam is reflected at the interface toward sample in the ATR-crystal. All samples (2µL) were dried using airflow on ATR-crystal for www.nature.com/scientificreports/ 3 min before sample spectra recorded in triplicate. The air spectrum was used as a background in all ATR-FTIR analysis. Sample spectra and background were taken with 4 cm −1 of resolution and 32 scans were performed for analysis. The spectra were normalized employing the vector method and adjusted to rubber band baseline correction. The original data were plotted in the Origin Pro 9.0 (ORIGINLAB, Northampton, MA, USA) software to create the second derivative analysis. The second derivative was obtained by applying the Savitzky-Golay algorithm with polynomial order 5 and 20 points of the window. The value heights indicated the intensity of the functional group evaluated.

Statistical analysis.
Individual experiments were performed in triplicate and all assays were performed a minimum of three times to confirm the reproducibility of the results. GraphPad Prism 8 software was used to assess statistical differences of means of readings using Student's unpaired t-test or Mann-Whitney tests. P values < 0.01 were considered to be statistically significant.