Characterization of Rhamnolipids for the Inactivation of Enveloped Viruses

Ling Jin (  ling.jin@oregonstate.edu ) Department of Biomedical Sciences, Carlson College of Veterinary Medicine, Oregon State University, Corvallis, OR 97331 Wendy Black Department of Biomedical Sciences, Carlson College of Veterinary Medicine, Oregon State University, Corvallis, OR 97331 Teresa Sawyer Electron Microscopy Facility, 145 Linus Pauling Science Center, Oregon State University, Corvallis, OR, 97331

Similar to SARS-CoV2, bovine coronavirus (BCoV) is a member of the Coronaviridae family, which comprises enveloped positive-sense single-stranded RNA viruses. BCoV is a pneumoenteric virus that infects the upper and lower respiratory tract and intestine. Infection of BCoV can lead to calf diarrhea, winter dysentery, and respiratory infections in cattle of various ages (Saif, 2010). Human herpesvirus type 1 (HSV-1) is a member of the Herpesviridae family that include a large number of enveloped DNA viruses common in both animal and humans.
HSV-1 is the cause of cold sores or fever blisters in or around the mouth and encephalitis in newborns (Dhull et al., 2019;Patoulias et al., 2017). It is also the leading cause of corneal blindness in developing countries ( (Jones, 2003;Moerdyk-Schauwecker et al., 2009). Both are enveloped viruses, and their surface proteins on the envelopes are essential for virus entry in infection. Rhamnolipids have been shown to be capable of inactivating both bacteria and fungi (Elshikh et al., 2017;Gaur et al., 2020;Radlinski et al., 2017). In this study, the ability of rhamnolipids to inactivate BCoV and HSV-1 was investigated.

Material And Methods
Rhamnolipids are provided by AGAE Technologies, LLC (Corvallis, Oregon, USA).

Rhamnolipid cytotoxicity assay in vitro
Ninety-six-well plates were seeded with ~2X104 Vero cells per well and grown overnight at 37°C and 5% CO2. The cells were washed once with phosphate-buffered saline (PBS) and treated in 100 µl 222A or 222B with indicated concentrations diluted in PBS at 37°C for 1h. The rhamnolipid-treated cells were then washed once with PBS and then replenished with EMEM containing 5% FBS (GeminiBio, USA) and antibiotics (as described above), and further incubated for 24 h. After incubation, cell morphology was examined under light microscope and cell viability was examined by using colorimetric cell viability kit III (XTT) (PromoKine).
Brie y, 50 µl of the XXT reaction solution was added to each well and the plate was incubated at 37°C incubator for 3-4h. Absorbance at a wavelength of 450-500 nm was read on a microplate uoresce reader (BioTek) and recorded. Wells containing 222A or 222B at each concentration in media without cells served as a blank to ensure that rhamnolipids themselves were not registering uorescence (Moerdyk-Schauwecker et al., 2009).

HSV-1 and BCoV infection in vitro in the presence or absence of rhamnolipids.
Twelve-well plates were prepared the day before infection by seeding each well with approximately 1´105 Vero cells or Hrt18G cells. The cells were washed once with PBS and infected with 100µl of either ~2X106 PFU of HSV-1-GFP or ~1X106 PFU of BCoV in the presence or absence of rhamnolipids. Following 1-hr absorption, the cells were replenished with 1 ml culture medium containing 5% FBS and antibiotics (as described above) in the presence or absence of 222B at the indicated concentration. At 24 or 72 hr postinfection (hpi), the cytopathic effect (CPE) from virus infection was examined under a light microscope or uorescent microscope.

Immuno orescence staining
Hrt-18G cells infected with BCoV were washed with ice-cold PBS twice and xed in 1 ml ice-cold methanol at -20°C for 20 min. Following xation, the plates were washed again twice with cold PBS and blocked in 1% BSA in TBS at room temperature for 1 h, then stained with 1:40 diluted anti-BCoV monoclonal antibody conjugated with FITC at room temperature for 1 h on a rocker and washed twice with PBS before viewing under uoresce microscope.

Electron microscopy of virions
Six 75-cm2 asks seeded with 90% con uent Vero cells were infected with 0.01M.O.I. HSV-1-McKrae. The infected asks were harvested when 80-90% CPE developed at 3-4 days post-infection (dpi), and subjected to two-freeze and thaw cycles between -80°C and room temperature. The supernatant was cleared of cells and cell debris by centrifugation at 9,000 X g for 30 min at 4°C. The virions were then centrifuged through a 60% sucrose cushion in a Beckman Model XL-70 ultracentrifuge at 25,000 rpm for 1h in a SW28 rotor at 4°C as we described previously (Jin et al., 2008). The virion pellets were suspended in 100 µl dH2O rst, and then 10 µl of virion was mixed with 90 µl of 0.01% SDS, 0.01% Triton X-100, and 0.01% 222B, respectively. The treated virions were stained as we described previously (Jin et al., 2008). Brie y, treated virion suspensions were adsorbed to formvar-coated carbon-stabilized copper grids by oating grids on ~ 20 µl drops of the sample spotted on para lm. The grids were then blotted dry with Whatman lter paper and immediately oated on ~20 µl drops of 2% phosphotungstic acid (PTA) (pH 6.9) in water for 30 s. Excess PTA was removed by side-blotting and the grids were allowed to air dry (Jin et al., 2008). Images were obtained with a FEI Titan 80-200/ChemiSTEM Transmission Electron Microscope (TEM).
HSV-1 plaque reduction assay following rhamnolipid treatment Vero cells were seeded in 12-well tissue culture plates at approximately 2´105 cells/well on the day before infection. Three wells were infected with HSV-1 at each concentration in the presence or absence of rhamnolipids, 222B. Following a 1-h viral absorption, 3 ml of 2% methylcellulose overlay media was added to each well. Plates were incubated at 37°C with 5% CO2 in a humidi ed incubator. The plates were xed in 20% methanol, stained with 1% crystal violet, plaques were counted at 4-5 dpi (Moerdyk-Schauwecker et al., 2009).
Plastic surface coating with 222B and surface-contact assay The 48-well Corning Costar Flat Bottom Cell Culture Plate with a growth area of 0.95 cm2 was selected for 222B coating on the plastic surface. The bottom of the well was coated by adding 20 µl of 222B at 0.005% directly, which cover the entire surface of the well. The coated wells were air-dried overnight. Virus droplets in 10 µl containing 103 or 104 virions (PFU) HSV-1 were added to the bottom of the well for 1 min or 5 min. The contact between HSV-1 and 222B coated plastic surface was stopped by adding 1 ml of PBS. Viruses left on the plastic surface were eluted in 1 ml 1 X PBS for 1 hour at room temperature on a plate rocker. The control wells were coated with PBS only. The non-surface contact controls were wells that had 1 ml of PBS added rst, followed by adding 10 µl to the elution without contacting the coated surface. 100 µl of eluted viruses from each well were then titrated by standard plaque assay.
Fabric coating with 222B and surface-contact assay Approximately 1 cm2 square fabrics were cut from a commercial surgical mask, made of non-woven fabric (Tronex Healthcare, Mount Olive, US), and then soaked in 50-100µl of 0.005% 222B or PBS. The soaked fabrics were air-dried overnight after removing the excess liquid from the soaked fabrics. The dried fabrics were then placed in a well of 24-well plate (Corning). Virus droplets in 10 µl containing 103 or 104 PFU HSV-1 were added to the coated fabrics for 3-5 min. The contact between viruses and 222B coated fabrics was stopped by adding 1 ml of PBS directly to the well. Viruses left on the fabric surface were eluted at room temperature for 1h on a plate rocker. The non-surface contact controls were 222B coated fabrics were eluted rst with 1 ml of PBS, viral droplet was added to the elution media without contacting the fabrics. The mock controls were viral droplets added directly to 1 ml PBS in wells that were not coated. The eluted viruses were then titrated by standard plaque assay.

Statistics
All statistical analyses were performed using GraphPad Prism version 4 for Windows (GraphPad Software, San Diego, CA, USA). Plaque reduction results were analyzed by two-way ANOVA with Bonferroni post tests.

Results
Cytotoxicity assays of rhamnolipids on cultured cells Given cell membranes are made of phospholipid bilayers, it is possible that surfactants may have negative effects on cell membranes as well. Therefore, the toxicity of rhamnolipids was examined in uninfected Vero cells by the light microscope and XTT viability assay. Fig. 1 shows the morphology of Vero cells treated with 222A over a concentration range from 0.001% to 0.1%. Cells treated with 0.1% or 0.05% 222A were mostly detached, and very few live cells were visible under the light microscope. Similar results were observed with cells treated with 222B (data not shown). However, cell monolayers looked similar to the uninfected cells in wells treated 222A or 222B at or below 0.01%. To determine whether rhamnolipid affects cell viability, treated cells were tested by colorimetric cell viability kit III (XTT). The assay is based on the ability of mitochondria enzyme from metabolic active cells to reduce the tetrazolium salt XTT to orange-colored formazan compounds. The dye formed is water-soluble, and its intensity is proportional to the number of metabolic active cells. The greater the number of active cells in the well, the greater the activity of mitochondria enzymes, the higher the concentration of the dye formed, the higher absorbance on the ELISA reader. As shown in Fig. 2, cells treated with rhamnolipids 222A or 222B at or above 0.05% were less than 10% viable compared to the mock-treated cells. Cells treated with rhamnolipids 222A at 0.01% and 0.009% were about 75% viable compared to the mock-treated control, respectively. Likewise, Cells treated with rhamnolipids 222B at 0.001% and 0.009% were about 50% and 70% viable compared to the mock-treated control, respectively. In contrast, cells treated with rhamnolipids 222A or 222B at or below 0.005% had similar viability as the mock-treated cells. Therefore, rhamnolipids at 0.005% were considered as biologically safe in this study.
The effect of rhamnolipid treatment on HSV-1 and Coronavirus (BCoV) replication in vitro To determine whether rhamnolipids have anti-viral effects speci c to enveloped viruses, about 2X106 PFU HSV-1 was diluted in PBS containing 0.009% or 0.001% of 222B or 222A rst, then inoculated to a 12-well plate seeded with Vero cells, and cultured in the presence or absence of rhamnolipids. Fig.3 shows the cells infected with HSV-1 in the presence or absence of rhamnolipids. In HSV-1 infected wells without rhamnolipid, 90% cytopathic effect (CPE) was observed at 24 hour post-infection (hpi) (Fig. 3). In the presence of 0.001% 222A or 222B, 80-90% CPE was observed, which was similar to those seen in HSV-1 control infection at 24 hpi.
However, no or less CPE was observed at 24 hpi in the presence of 0.009% 222B or 222A (HSV-1+222B or 222A). Since HSV-1 infection in the presence of 0.009% 222B did not produce any visible CPE at 24hpi, 222B is selected for the remaining study.
To prove the rhamnolipid's anti-viral activity is not HSV-1 speci c, but speci c to enveloped viruses, 222B was tested against bovine coronaviruses (BCoV). Hrt-18G cells in a 12-well plate were infected with BCoV at ~1X106 PFU/well in the presence of 0.009% or 0.001% of 222B. At 72 hpi, BCoV CPEs, such as cell morphology change, presence of inclusion bodies, were visible in wells infected with BCoV in the presence of 0.001% 222B and BCoV only wells (Fig. 4A).
However, no CPE was visible in wells infected with BCoV in the presence of 222B at 0.009%. To further demonstrate the replication of BCoV, infected cells were stained with FITC labeled mAb speci c to BCoV at 72h post-infection. As shown in Fig.4B, positive FITC staining was visible in BCoV only wells and wells infected with BCoV containing 0.001% 222B. In contrast, little FITC staining was visible in wells infected with BCoV containing 0.009% 222B. Noticeably, the BCoV only wells had signi cantly more positive BCoV staining than those infected in the presence of 0.001% 222B, suggesting that 222B even at 0.001% still had damaging effects on this virus.

Comparison of different surfactants against an enveloped virus, HSV-1
The above study demonstrated that rhamnolipids have antiviral activity toward the enveloped viruses that we tested. It suggests that its mechanism of action is speci c to viral envelopes. To prove that viral envelopes are common targets of surfactants, different surfactants were tested against HSV-1 infection. An ionic surfactant, sodium dodecyl sulfate (SDS), and a non-ionic surfactant, such as Triton X-100, were examined in this study. We used HSV-1 with GFP expression to visualize the replication of HSV-1 directly. Vero cells seeded in 12 well plates were infected with ~1X105 PFU/well (i. e., 105 virions per well) in the presence of SDS, triton X-100, and 222B, respectively. The nal concentration for each reagent was set at 0.009%. At 24h hpi, neither HSV-1 CPE, nor GFP expression was observed in the presence of SDS or Triton X-100, or 222B (Fig. 5). Cells infected with HSV-1 looked normal in the presence of 222B. However, 50% and 90% of Vero cells were detached in wells treated with SDS and Triton X-100, respectively.
HSV-1 CPE and green GFP expression were only observed in HSV-1only infected wells. This proves all these reagents can inactivate the enveloped viruses.

Removal of HSV-1 envelopes by surfactants
To visualize the effect of surfactants on viral envelopes directly, HSV-1 virions were treated with different surfactants and examined under the transmission electron microscope (TEM). HSV-1 virions isolated from both tissue culture supernatant and cell lysates were suspended in H2O containing 0.009% of 222B, SDS, and Triton X-100, respectively, before TEM negative staining. Negative staining is a unique technique of "negative contrast" staining, the contrast is not applied to the object but its environment, using heavy metal salts, such as phosphotungstic acid (PTA). The electron beam can cross biological material easier than the surrounding space. The heavy metal staining forms a dark contrast around viral particles (virions). The enveloped viruses or complete virions are enclosed by phospholipid bilayers (Fig.  6, white arrow) surrounding the capsid (Fig. 6, black arrow). In the untreated viral samples, enveloped virions were visible in each view (Fig. 6A, white arrows).
However, no enveloped virion was visible in samples treated with 0.009% of 222B (Fig. 6B). Likewise, fewer enveloped virions were seen in samples treated with 0.009% of SDS (Fig. 6C) or Triton X-100 (Fig.  6D). This suggests that envelopes were dispersed or removed in the presence of surfactants. Viral proteins on the envelope are essential for the initial infection step. If the viruses lose their envelopes, they will not be able to bind to the host receptors, consequently, lose their infectivity. To prove that viral envelopes or viral enveloped proteins were lost once in contact with rhamnolipids, 100 µl of HSV-1 viruses at ~1X106 PFU/ml were mixed with either 0.9 ml of 0.01% of 222B or PBS alone brie y by vortexing. Afterward, 100 µl of the diluted HSV-1 were inoculated to a 12-well plate seeded with ~1X105 Vero cells.
Following virus inoculation, the cells were cultured in media either without 222B or with 222B. Fig. 7 shows the cells infected with HSV-1 that was diluted with 222B but cultured with media without 222B (HSV-1=222B-pre) or infected with HSV-1 without 222B but cultured in 222B media (HSV-1+222B-post).
No CPE was observed in cells infected with HSV-1 that was diluted with 0.009% 222B. On the other hand, 80-90 CPE was observed in cells infected with HSV-1 diluted only in PBS, but cultured with media containing 0.009% 222B. Likewise, GFP HSV-1 was observed in cells infected with HSV-1 diluted only in PBS, but cultured with media containing 0.009% 222B (Fig. 7, HSV-1+222B-post). This further demonstrated that rhamnolipids 222B acted directly on the surface of vial envelopes.

Plaque reduction assay with 22B treatment before infection
To determine the potency of 222B against enveloped viruses, different concentrations of HSV-1 were treated with different concentrations of 222B before inoculation. The contact between HSV-1 envelopes and rhamnolipids took place when HSV-1 was diluted to different concentrations of 222B in PBS. Therefore, the interaction was within 5 min. The treated viruses were then titrated immediately by standard plaque assay. As shown in Fig. 8, 222B at 0.009% completely inactivated ~2X106 PFU HSV-1 infections following brief treatment before viral inoculation. Similarly, 222B at 0.0045% reduced the input virus titer from ~2X106 PFU to ~2X102 PFU, ~2X105 PFU to 10-20 PFU, ~2X104 PFU to a few PFU, respectively. 222 B at 0.001% produced about 10-fold reduction. 222B at a concentration below 0.001% has little or no antiviral activity left (data not shown here). These results demonstrate that 222B at 0.0045% and 0.009% are capable of killing ~1X 104 and ~1X106 PFU of enveloped viruses, respectively.

Contact inhibition against HSV-1 on 222B coated plastic surface
If a biologically safe antiviral product remains active on a coated surface, we could use it to create a physical barrier to prevent the spread of respiratory viruses, such as SARS-CoV2. To this end, 222B anti-HSV-1 activity was investigated on a 222B coated plastic surface. The well surface of a 48-well plate, 1cm2, was coated with 20 µl of 222B at 0.005%. Droplets in 10 µl containing ~2X104 or ~2X103 PFU of HSV-1 was directly applied to the surface. The contact between the coated surface and viruses was for 1 or 5 min, and then terminated by adding 1 ml of PBS. The post-contact viruses were eluted at room temperature for 1h on a plate rocker. Fig. 9 shows the number of HSV-1 PFU detected after the contact between HSV-1 and the 222B coated surface. When ~2X104 PFU HSV-1 per droplet was applied to the 222B coated plastic surface, around average 200 PFU per ml was detected following 1-minute contact, and about 10 PFU/per ml on average was detected following 5-minute contact (Fig. 9A). Likewise, when 2X103 PFU per droplet was applied to the 222B coated plastic surface, about 30 PFU/ml on average were detected in the elusion media following 1-min contact, no PFU was detected following 5-min contact (Fig.   9B). On the other hand, when ~2X104 or ~2X103 PFU HSV-1 in 10µl was added to the PBS coated surface, similar numbers of PFU were recovered as the input viruses in the mock-treated control. Similarly, in the non-contact control, where 2X104 or 2X103 PFU HSV-1 (in 10 µl) was added directly to the elution buffer without contacting the 222B coated surface directly, only10-fold reduction was observed compared to the mock-treated control and PBS control.

Contact inhibition against HSV-1 on 222B coated fabric surface
To determine whether 222B on the coated surgical mask can inactivate the enveloped viruses, about 1-cm2 mask fabrics were coated with 50-100µl of 0.005% 222B. The viral droplets in 10 µl were applied directly to the 222B coated mask fabric for 3-5 min. The interaction between viruses and 222B on the fabric surface was stopped by adding 1ml PBS, which diluted 222B to a concentration that has little or lacks antiviral activity. When 1.45X 104 PFU per droplet were applied to the 222B coated mask fabric, less than 10 PFU/ml on average were detected in the elusion media (Fig. 10 A).
Similarly, when 1.45X 103 PFU per droplet were applied to the 222B mask fabric, no viral plaques were detected in the elusion media (Fig. 10B). In contrast, when 1.45X 104 PFU per droplet were applied to the elusion buffer without contacting the 222B coated fabric directly, about 2.9X103 PFU were detected in the elusion media (Fig. 10A). Likewise, when 1.45X 103 PFU per droplet were applied to the elusion buffer without contacting the 222B coated fabric directly, about 2.2X102 PFU were detected in the elusion media (Fig. 10B). As seen in the 222B coated on the plastic surface, there was also a 10-fold reduction in the non-direct contact treatment compared to the mock-treated control and PBS control. To PBS coated fabric, about 2.2X103 PFU/ml and 2.2 X102 PFU/ml were detected in the elusion media with input viruses at 1.45X 104 PFU and 1.45X 103 PFU per droplet, respectively (Fig. 10). There was a little change in viral titer in the mock-treated control between input virus and post-contacting viral titer (Fig. 10). These results demonstrated that 0.005% of 222B coated on the mask is capable of inactivating about 103PFU enveloped viruses per cm2.

Discussion
The above study demonstrated that rhamnolipids could inactivate enveloped viruses by targeting the envelopes. Viral envelopes are derived from cell membranes. Given that cell membranes are made of the phospholipid bilayer, we sought to examine whether rhamnolipids are toxic to the cells by XTT viability assay. The assay is based on the ability of mitochondria enzyme from metabolically active cells to reduce the tetrazolium salt XTT to orange-colored formazan compounds. The higher the number of active cells is, the higher the color intensity is. Unsurprisingly, rhamnolipids at or above 0.05% were toxic to the treated cells, and few cells were left following 1h exposure (Fig. 1). Cells treated with 0.1% or 0.05% rhamnolipids had less than 10% viable cells left following 1h exposure. When cells were exposed to 0.01% and 0.009%, around 50% and 70 % of cells were viable, respectively. Although, cells treated with 0.009% rhamnolipid looks similar to the untreated cells under the microscope (Fig. 1). However, when cells were treated with rhamnolipid at or below 0.005%, little or no viability loss was observed (Fig. 2). Therefore, rhamnolipids at or below 0.005% are completely biological safe within 1h exposure.
In this study, 222B, the predominant monorhamnolipids, were found to have better anti-HSV-1 activity than that of predomiant dirhamnolipid, 222A (Fig. 3). Cells infected with HSV-1 in the presence of 0.009% 222B had less CPE than those in the presence of 0.009% 222A. In line with this observation, a higher viral titer was produced in the presence of 222A than that in the presence of 222B (data not shown). The difference between 222A and 222B could be the difference between the extra rhamnose sugar group (rha) in the 222A, which may interfere with the interaction between the lipid tail with the phospholipid bilayers of envelopes. Therefore, the monorhamnolipids are better anti-enveloped virus agents.
Surfactants have antiviral envelope properties as they causes envelopes to disintegrate when their hydrophobic tails come to contact with phospholipid bilayers. When enveloped viruses lose their membrane, they lose their ability to infect. Both SDS (an ionic surfactant) and Triton X-100 (a non-ionic surfactant ) are surfactants commonly used for research. To prove envelopes are common targets of surfactants, the effect of SDS and Triton X-100 on viral envelopes was compared with rhamnolipids. As expected, virions treated with SDS, Triton X-100, and 222B, all had very few complete virions with envelopes left under TEM examination, while the untreated virions had complete virions with envelope in every view (Fig. 6). Although non-enveloped viruses were also visible in the untreated sample, this is because virions were isolated from both tissue culture supernatant and cell lysates. There are normally many incomplete virions inside of the infected cells, which lack envelopes. Only a small portion of virions are fully mature and enveloped. Here, to prove the principle of the action, the mixed virion preps were enough for us to tell the difference between treated and untreated virions. To support what we observed in TEM, we examined HSV-1 infection that was treated with 0.009% of 222B before infection. As expected, HSV-1 exposed to 222B before inoculation failed to infect the cells. However, untreated HSV-1 was able to infect cells and produced 80-90% CPE at 24hpi, although the infection was cultured with media contains 0.009% 222B (Fig. 7). Both studies con rmed the mechanism action of rhamnolipids against enveloped viruses is speci c to envelopes. The surfactant has little effect on virus replication inside of the infected cells.
A HSV-1 plaque reduction assay was used to evaluate the rhamnolipid anti-enveloped virus activity. This assay takes advantage of the observation that HSV-1 infections form large visible plaques at 4-5 days post-infection on cell monolayers (Eide et al., 2010). The number of plaques formed corresponds directly to the number of input infectious viral particles. Therefore, plaque-forming units (PFU) are used to quantify the infectious viral particles. In a 0.009% of 222B solution, ~2X106 PFU (or virions) per 100 µl can be inactivated within 2-5 min. Similarly, 0.0045% of 222B can kill ~ 2X104 PFU/100 µl. However, when 222B is at or below 0.001%, it can only reduce the infectivity to less than one log of the input viruses (Fig. 8). Since BCoV does not form visible plaques in the Hrt-18G monolayer, the plaque reduction assay was not performed in this study. Nevertheless, in a 0.009% of 222B solution, almost all 1X106 PFU of BCoV was inactivated as we saw in HSV-1, and no BCoV antigen was detected by direct antibody staining (Fig. 4). 222B at 0.001% was able to reduce the BCoV antigen production but did not completely block BCoV infection. Therefore, the rhamnolipid anti-enveloped virus activity is dose-dependent; the higher the concentration is, the higher the killing activity is.
Although 222B at 0.009% has some cytotoxicity, it is capable of killing ~1X106 PFU/100µl, or ~1X107 PFU/ml enveloped viruses. Therefore, 222B at 0.009% was used in the initial characterization of rhamnolipids against enveloped viruses. Since 222B at 0.005% has no cytotoxicity, and it would be a safe concentration to use medically against enveloped respiratory viruses, such as SARS-CoV2, SARS, In uenza viruses, common cold viruses.
Considering the worldwide involvement with the 2019 Coronavirus disease (COVD-19) pandemic where there is neither an effective vaccine nor treatment available at this time, developing an antiviral mask or barrier could help to prevent the spread of not only this virus, but others. If rhamnolipids on coated surface can kill enveloped viruses, it will allow the production of antiviral masks or shields. Hence, 222B contact killing was investigated on a ~1cm2 plastic or mask fabric surface in this study. To test the plastic surface, the 48-well plate with a growth area of 0.95 cm2 (~1 cm2) was coated with a 20 µl of 222B. To test the mask surface, the surgical mask was cut into ~1cm2 and then coated with 50-100 µl of 222B. We used 10 µl of viruses in the contact assay. Because enveloped viruses tend to cluster together due to the nature of their lipid surface, volumes of less than 10 µl will produce variation between replicates. Another reason is 10 µl is close to droplet size from running nose or coughing. Since 222B at 0.005% is biologically safe, we chose to coat the surface with this concentration. To our surprise, 0.005% of 222B on the plastic surface is capable of killing at least 1X103 PFU HSV-1/cm2 within 5 min contact (Fig. 9). Similarly, 0.005% of 222B on mask fabric is also capable of killing ~1X103 PFU HSV-1/cm2 within 3-5 min (Fig. 10). The contact killing is higher at 5 min than 1 min. In addition, we noticed that the coated fabric surface had higher killing capacity than the plastic surface. This could contribute to the fact that mask fabrics had higher absorption and were coated with more 222B than the plastic surface. Given the viruses were applied to the surface, then eluted in 1 ml PBS, it is not surprising to see that not all the viruses were recovered in the elution media (Fig. 10, PBS). To the plastic-coated surface, the non-contact control had a 10-fold reduction compared to the mock-treated control (Fig. 9). It is possible that there were variations during virus titration. Regardless, ~1cm2 surface coated with 222B can inactivatẽ 1X103PFU enveloped viruses.
Respiratory viruses are common in humans and are transmitted mostly via droplets or aerosols, or body secretions, such as tears, nasal or oral discharges. The amount of viruses shed in the nasal or oral secretion can range from 104 to 107 genome copies/ml depending on the severity of the infection and the types of viruses (Finberg et al., 2019;Liu et al., 2020a). For example, Middle East respiratory syndrome coronavirus (MERS-CoV) is a highly pathogenic coronavirus, can cause 34% mortality in infected humans (Ogimi et al., 2020). Around 5X106 genome copies/ml can be shed in sputum from patients with severe MERS (Corman et al., 2016). Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in Wuhan, China, in late 2019, is now causing a pandemic COVID-19. It was estimated that 104-107 genome copies/ml were present in nasal and pharyngeal swab specimens from COVID-19 patients (Liu et al., 2020a). In a similar study, 104-105 genome copies /ml SARS-CoV2 were detectable in sputum from severe infected COVID 19 patients (Huang et al., 2020;Liu et al., 2020b). In the case of In uenza A virus infection, 104-107 virions/ml can be detected from patients with uncomplicated acute in uenza A (Finberg et al., 2019). If these respiratory viruses came out in a 10µl droplet size, there would be 102-105 virions per droplet from an infected patient. If an infected person wears a 222B coated mask, all those viruses would be killed inside the mask. A 222B mask would reduce the amount of emission of infected saliva or droplets shed into the environment, which will stop the spread of COVID 19.
Mask wearing by itself can reduce the transmission of respiratory viruses, such as u and SARS-CoV-2.
Community-wide mask-wearing has been shown to contribute to the control of COVID-19 by reducing the amount of emission of infected saliva and respiratory droplets from individuals with subclinical or mild COVID-19 (Cheng et al., 2020). Countries like Japan and Slovenia, where people wear masks, were able to prevent the spread of COVID quicker than those non-mask wearing countries. In addition, recent reports on mask use found that the per-capita mortality tended to increase by 43% weekly in countries where people were not wearing masks, compared with a 2.8% increase in countries where people were wearing masks Lai et al., 2020). Due to the current pandemic, there is a shortage of both surgical masks and N95 masks available to the public. However, even home-made cloth masks can stop aerosol (<5 µm) or droplets (>5 µm) containing viruses (Clase et al., 2020;Ngonghala et al., 2020). The widespread use of 222B coated masks, could put an end of the COVID 19 pandemic much sooner.
Therefore, there is an urgent need that we apply rhamnolipids to masks or make 222B coated masks available to the public to stop the pandemic of COVID19.
In summary, this study demonstrated that rhamnolipids have antiviral activity against enveloped viruses. Many respiratory viruses are enveloped viruses, such as SARS-CoV2, in uenza viruses. Rhamnolipids not only can kill enveloped viruses in solution, but also can destroy enveloped viruses on coated surface.
Direct contact with rhamnolipids can destroy the envelopes within 5 min. Rhamnolipids, 222B at 0.005%, are completely biologically safe and can kill ~1X104 PFU enveloped viruses in solution and ~1X103 PFU enveloped viruses on the coated surface. The application of rhamnolipids 222B on masks will help us to end the pandemic of COVID19. Figure 1 Bright light images of Vero cells treated with rhamnolipids at different concentrations. The concentration of rhamnolipids is labeled above the image. The images were captured at 24h post-treatment. The bright light images were captured at 10X magni cation.

Figure 2
Rhamnolipid cytotoxicity assay in vitro. Vero cells were incubated with the indicated concentration of rhamnolipids in PBS for 1 h. The treatment was then removed, and the cells were washed twice with PBS and further incubated for 24 h in DEMEM with 5% serum and antibiotics as described in materials and methods. Cell viability was evaluated with XTT cell viability kit III (PromoKine), which measures metabolic activity based on the extracellular reduction of XTT by NADH produced in the mitochondria, and expressed as a percentage of the mock-treated control (n = 3). A signi cant statistical difference from the mock-treated control is marked with **p < 0.001 (two-way ANOVA with Bonferroni post-test). The solid bar and open bar represents the percentage of cell viability following treatment of 222B and 222A, respectively.    Bright light (A) and uorescence (B) images of GFP-HSV-1 infected cells in the presence or absence of 222B. "HSV-1+222B-pre" represents cells infected with HSV-1 that were diluted with PBS containing 0.009% 222B, and cultured in a media without 222B. "HSV-1+222B-post" represents cells infected with HSV-1 that were diluted with PBS only and cultured in a media containing 0.009% 222B. Both bright light and uorescent images were captured at 10X magni cation.

Figure 9
Contact inhibition against HSV-1 on 222B coated plastic surface. A: 222B contacts with input viruses at 2X 104 PFU per well. B: 222B contacts with input viruses at ~2X103 PFU per well. The X-axis stands for contacts between coated surface and HSV-1; the Y-axis represents the number of PFU post-contact with the coated surface. The checked bar stands for direct contact between 222B and HSV-1 for 1 min (222B-1mn).; the solid bar stands for direct contact between 222B and HSV-1 for 5 min (222B-5min); the open bar stands for the no direct contact between HSV-1 and 222B coated surface (222B-NC); the striped bar stands for direct contact between PBS coated surface and HSV-1 for 5 min (PBS); and the gray bar stands for HSV-1 diluted in PBS directly (Mock-treated). Plaque formation was quanti ed at 3-4 dpi and expressed as average PFU post 222B treatment (n=3).

Figure 10
Contact inhibition against HSV-1 on a 222B coated fabric surface. A: 222B contacts with input viruses at 1.45 X 104 PFU per ~1cm2 fabric. B: 222B contacts with input viruses at ~1.45 X103 PFU per ~1cm2 fabric. The X-axis stands for contacts between coated surfaces and HSV-1; the Y-axis represents the number of PFU post-contact with the coated fabrics. The solid bar stands for direct contact between 222B and HSV-1 (222B); the open bar stands for the no direct contact between HSV-1 and 222B (222B-NC); striped bar stands for direct contact between PBS coated surface and HSV-1 (PBS), and the gray bar stands for HSV-1 diluted in PBS directly (Mock-treated). Plaque formation was quanti ed at 3-4 dpi and expressed as average PFU post 222B treatment (n=3).