key: cord-0699259-1gwvp6yl
authors: Xu, Chuan; Wang, Annie; Hoskin, Eileen R.; Cugini, Carla; Markowitz, Kenneth; Chang, Theresa L.; Fine, Daniel H.
title: Differential effects of antiseptic mouth rinses on SARS-CoV-2 infectivity in vitro
date: 2020-12-01
journal: bioRxiv
DOI: 10.1101/2020.12.01.405662
sha: 9186fcf7c341722a00e776f33aa9b79432f1b211
doc_id: 699259
cord_uid: 1gwvp6yl

SARS-CoV-2 is detectable in saliva from asymptomatic individuals, suggesting a potential benefit from the use of mouth rinses to suppress viral load and reduce virus spread. Published studies on reduction of SARS-CoV-2-induced cytotoxic effects by antiseptics do not exclude antiseptic-associated cytotoxicity. Here, we determined the effect of commercially available mouth rinses and antiseptic povidone-iodine on the infectivity of SARS-CoV-2 virus and of a non-pathogenic, recombinant, SARS-CoV-2 infection vector (pseudotyped SARS-CoV-2 virus). We first determined the effect of mouth rinses on cell viability to ensure that antiviral activity was not a consequence of mouth rinse-induced cytotoxicity. Colgate Peroxyl (hydrogen peroxide) exhibited the most cytotoxicity, followed by povidone-iodine, chlorhexidine gluconate (CHG), and Listerine (essential oils and alcohol). Potent anti-viral activities of povidone iodine and Colgate peroxyl mouth rinses was the consequence of rinse-mediated cellular damage. The potency of CHG was greater when the product was not washed off after virus attachment, suggesting that the prolonged effect of mouth rinses on cells impacts anti-viral activity. To minimalize mouth rinse-associated cytotoxicity, mouth rinse was largely removed from treated-viruses by centrifugation prior to infection of cells. A 5% (v/v) dilution of Colgate Peroxyl or povidone-iodine completely blocked viral infectivity. A similar 5% (v/v) dilution of Listerine or CHG had a moderate suppressive effect on the virus, but a 50% (v/v) dilution of Listerine or CHG blocked viral infectivity completely. Prolonged incubation of virus with mouth rinses was not required for viral inactivation. Our results indicate that mouth rinses can significantly reduce virus infectivity, suggesting a potential benefit for reducing SARS-CoV-2 spread. Importance SARS-CoV-2 is detectable in saliva from asymptomatic individuals, suggesting the potential necessity for the use of mouth rinses to suppress viral load to reduce virus spread. Published studies on anti-SARS-CoV-2 activities of antiseptics determined by virus-induced cytotoxic effects cannot exclude antiseptic-associated cytotoxicity. We found that all mouth rinses tested inactivated SARS-CoV-2 viruses. Listerine and CHG were less cytotoxic than Colgate Peroxyl or povidone-iodine and were active against the virus. When mouth rinses were present in the cell culture during the infection, the potent anti-viral effect of mouth rinses were in part due to the mouth rinse-associated cytotoxicity. Our results suggest that assessing anti-viral candidates including mouth rinses with minimal potential disruption of cells may help identify active agents that can reduce SARS-CoV-2 spread.

Severe acute respiratory syndrome-related coronavirus (SARS-CoV-2), a non- To assess the effect of mouth rinses on the viruses, mouth rinse-treated SARS-CoV-2 192 viruses were concentrated by centrifugation at 14,000 rpm in a centrifuge (Eppendorf) at 193 4°C for 2 h as described previously (Holmes et al. 2015 

It is critical to assess antiviral agents under non-cytotoxic conditions as viruses depend 217 on viable host cells for productive infection. Therefore, we first determined the effect of 218 mouth rinses on cell viability. Note that percentage dilutions (v/v) of commercial mouth 219 rinse products are referenced in this study. For example, in Figure 1 , 50% (v/v) CHG 220 represents a solution composed of equal volumes of culture media and of the 221 commercial product, and does not indicate the final concentration of active ingredients. 222

HeLa-hACE2 cells were treated with 2-fold serial dilutions in medium of Listerine, CHG, 223 povidone-iodine, or Colgate Peroxyl for 20 sec, washed, cultured in fresh media, and 224 cell viability was determined. All 50% (v/v) dilutions of mouth rinses were highly toxic to 225

HeLa-hACE2 and oral epithelial cells (Fig 1) . Listerine was least cytotoxic, followed 226 closely by CHG. Both 0.5% (v/v) dilutions of povidone-iodine and Colgate Peroxyl were 227 highly toxic to cells. 228

We also determined the effect of 2h exposure of mouth rinses on cell viability for 230 comparison with the duration of viral attachment in the infection assay. We found that 231 6.25% (v/v) diluted Listerine and 1.5% (v/v) diluted CHG did not impact cell viability, 232

whereas 0.1% (v/v) diluted povidone-iodine or Colgate Peroxyl significantly affected cell 233 viability after 2 h exposure (Fig 2) . CHG, low cytotoxic dilutions of povidone-iodine, and highly dilute Colgate Peroxyl (with 245 cytotoxic effects) (Fig 2) , and were immediately added to Vero cells. Additional media 246

were added 2 h after infection, and cells were cultured overnight. The fluorescence 247 intensity from SARS-CoV-2 infection was determined, and cell morphology was imaged 248 at 24 h after infection. Diluted Listerine (3% v/v)CHG reduced SARS-COV-2 infection by 249 40%, and CHG (1.5% v/v) reduced infection by 70%, without apparent impacts on cell 250 morphology (Fig 3) . Diluted povidone-iodine (0.1% v/v) and Colgate Peroxyl (0.05% 251 v/v) appeared to have potent anti-viral activities; however, disruption of cell morphology 252 was apparent (Fig 3) , indicating that the putative anti-viral effect of these two agents 253 was likely a consequence of cytotoxicity. 254

We also used HIV pseudotyped luciferase virus particles expressing SARS-CoV2 256 surface protein (spike, S) to assess the effect of non-cytotoxic diluted Listerine and 257

CHG on viral infectivity. Unlike replication competent SARS-CoV-2 virus, which induces 258 cytopathic effects after prolonged culture, HIV pseudotyped luciferase viruses provide a 259 reliable, non-pathogenic, vector for assessing viral infectivity. We confirmed that 260 infection by pseudotyped SARS-CoV-2 was dependent on human hACE2, and that 261 infection was neutralized by anti-spike monoclonal antibody (Supplemental Figure 1) . 262

We determined the effect of diluted Listerine (1.5-6%v/v) and CHG (1.5% and 3%v/v), 263 which had no or little effect on cell viability (Fig 2) , on pseudotyped SARS-CoV-2 virus 264 infection without washing off the mouth rinses during the infection. We found that 6% 265

(v/v) Listerine had a moderate anti-SARS-CoV-2 activity, whereas 1.5% or 3% (v/v) 266 CHG suppressed viral infection by 88% and 97%, respectively (Fig 4A) . 267

We also determined whether pre-incubation of viruses with CHG affected the degree of 269

anti-viral activity. For this, pseudotyped SARS-CoV-2 viruses were pre-treated or not 270 with CHG 30 min at 37°C before being added to target cells. In contrast to the 271 experiment shown in Fig 4A, in which the mouth rinses were present during infection, 272

here, the mixture of virus and CHG was removed, fresh media were added, and cells 273

were cultured for 2 days before measuring luciferase activity (Fig 4B) . The effect of 274 viral effect of 1.5% (v/v) CHG was less potent when CHG was present only during viral 276 attachment (Fig 4B) compared to being continuously present during viral infection and 277 incubation for 2 days (Fig 4A) . The more pronounced anti-viral activity of 1.5% (v/v) 278 CHG in this experiment may be due to effects of prolonged contact with CHG on target 279 cells, indicating the importance of minimizing mouth rinse-associated cytotoxicity in the 280 infection assay. 281

To assess direct effects of the rinses on virus particles, pseudotyped SARS-CoV-2 283 viruses were incubated with mouth rinses for 30 min at 37°C, and were pelleted by 284 centrifugation (Holmes et al. 2015) prior to the infection assay. Note that centrifugation 285 did not impact infectivity of the virus (data not shown). After removal of the supernatant 286 containing the mouth rinse, viruses were resuspended in media and added to HeLa-287 hACE2 target cells (Fig 5A and 5B) . We also assessed the effect of centrifugation and 288 virus resuspension on infectivity cell viability to monitor potential cytotoxic effects of 289 residual mouth rinses on the cells (Fig 5C) . The result showed that all antiseptics tested 290 inactivated viruses. Anti-viral activity of 50% (v/v) Colgate Peroxyl was associated with 291 residual mouth rinse-induced cytotoxicity; 5% (v/v) Colgate Peroxyl and 5% (v/v) 292 povidone-iodine blocked viral infectivity (Fig 5B) ; 50% (v/v) Listerine and 50% CHG 293 inactivated viruses, but 5% of these rinses did not (Fig. 5B) . 294 Fig 5B) ; the apparent anti-viral 296 effect of 50% (v/v) Colgate Peroxyl was associated with cell toxicity (Fig 5C) . Treatment 297 with 5% (v/v) Listerine or CHG had a moderate anti-viral effect; whereas 5% (v/v) 298

Centrifugation and resuspension of the virus had no apparent impact on infectivity. All 300 mouth washes at non-cytotoxic levels exhibited antiviral activity. Colgate Peroxyl and 301 povidone-iodine had greater inhibitory effects on the viruses than CHG or Listerine. 302 303 Unlike high concentrations of Colgate Peroxyl and povidone-iodine, whose anti-viral 304 activities were associated with cytotoxicity, higher concentrations of Listerine and CHG 305 exhibited potent anti-viral effects without cytotoxicity. We asked whether preincubation 306 of the virus with Listerine or CHG was required to achieve their direct effect on the virus. 307

Mouth rinses were added to the virus, mixed, and immediately centrifuged at 4°C. 308

Supernatants containing the mouth rinses were discarded. Viruses were then 309 resuspended in media and added to target cells. The viral inhibition profiles of Listerine 310 and CHG without preincubation (Supplemental Fig 2) were comparable to those with 311 30 min incubation (cf. Fig 5) . CHG had a moderate anti-viral effect (Fig 6A,B) . In contrast to infected cells with exposure to highly diluted povidone-iodine and Colgate Peroxyl during the infection 321 leading to cell death (Fig 3) , there was no apparent cell death in cells infected by 322 viruses after the removal of mouth rinses by centrifugation. Taken together, Listerine 323 and CHG may be better mouth rinse products for SARS-CoV-2 prevention. Highly Indeed, we found that anti-viral effects of highly diluted povidone-iodine and Colgate 346

Peroxyl were the consequence of cytotoxicity when the agents were present during a 24 347 h infection assay. Our results warrant concerns regarding reliability of findings in 348 previous studies in which infected cells were exposed to mouth rinses and antiseptics 349 for extended times. 350

We found that all mouth rinses tested (all products diluted 1:1 with culture medium, 50% 352 v/v) had cytotoxic effects on cells. We found the cytotoxicity of Colgate Peroxyl > 353 povidone-iodine > CHG > Listerine. Similar trends were observed in both HeLa-hACE2 354 and oral epithelial cells. Mouth rinse-induced cytotoxicity was more pronounced in cells 355 with 2h incubation than with 20 sec incubation. When CHG was present during a 2-day 356 infection period, 1.5 and 3% (v/v) CHG suppressed SARS-CoV-2 infection by nearly 357 99% (Fig 4) . However, 1.5% (v/v) CHG was less potent when the mouth rinse was only 358 present during viral attachment. Importantly, when assessing the effect of CHG on the 359 viruses after removal of mouth rinse during the infection, 5% (v/v) CHG had only a 360 moderate effect, reducing infection by 35-55%. Similarly, potent "anti-viral" effects of 361 0.1% (v/v) povidone-iodine and 0.05% (v/v) Colgate Peroxyl that were observed when 362 antiseptics were present during infection, were found by the cell image analysis to be 363 due to antiseptic-associated cytotoxicity (Fig 3) . In fact, we found that 0.5% (v/v) 364

if the povidone-iodine was removed from the virus before infection (Figs 5 and 6) . for 2 days before measuring luciferase activity. Significance of differences between 472 mouth rinse-treated viruses and mocked-treated controls was compared; *p < 0.05. 473

Data are means ± SD. 474 (Fig 6A) , and images of infected cells with 490 or without mouth rinse treatment were also acquired (Fig 6B) . Differences between 491 mouth rinse-and medium control (0%)-treated viruses were compared; *p < 0.05. Data 492 are means ± SD, and are representative of two independent experiments. 493 

Povidone-iodine demonstrates rapid in vitro 503 virucidal activity against sars-cov-2, the virus causing covid-19 disease

In vitro effect 506 of oral antiseptics on human immunodeficiency virus-1 and herpes simplex virus 507 type 1

Rapid in-509 vitro inactivation of severe acute respiratory syndrome coronavirus 2 (sars-cov-2) 510 using povidone-iodine oral antiseptic rinse

In vitro bactericidal and 512 virucidal efficacy of povidone-iodine gargle/mouthwash against respiratory and 513 oral tract pathogens

Reduction of 515 viable bacteria in dental aerosols by preprocedural rinsing with an antiseptic 516 mouthrinse

Assessing pre-procedural subgingival irrigation and rinsing with an antiseptic 519 mouthrinse to reduce bacteremia

Single-cell and single-cycle analysis of hiv-1 521 replication

Interventions to reduce aerosolized 523 microbes in dental practice: A systematic review with network meta-analysis of 524 randomized controlled trials

Clinical course and molecular viral shedding among asymptomatic and 527 symptomatic patients with sars-cov-2 infection in a community treatment center 528 in the republic of korea

Efficacy 530 of listerine antiseptic in reducing viral contamination of saliva

Virucidal efficacy of different 534 oral rinses against sars-cov-2

Potential role of oral rinses 537 targeting the viral lipid envelope in sars-cov-2 infection. Function

Effect of chlorhexidine on the in vitro and in vivo herpes 539 simplex virus infection. Oral Surg Oral Med Oral Pathol

Mechanisms of the action of povidone-iodine against human and avian 542 influenza a viruses: Its effects on hemagglutination and sialidase activities

We thank Eric C. B. Milner for editing the manuscript. This work was supported by NIH 410 grant NIH R01AI36948 to T.L.C. 411Author contributions: 412Xu, C: contributed to acquisition, analysis, and interpretation, drafted and critically 413 revised the manuscript. 414Wang, A: contributed to acquisition and analysis, and critically revised the manuscript. 415Hoskin, E: contributed to conception, interpretation, and critically revised the 416 manuscript. 417Cugini, C: contributed to conception, interpretation, and critically revised the manuscript.