key: cord-1011743-s0bni60d authors: Biasin, M.; Strizzi, S.; Bianco, A.; Macchi, A.; Utyro, O.; Pareschi, G.; Loffreda, A.; Cavalleri, A.; Lualdi, M.; Trabattoni, D.; Tacchetti, C.; Mazza, D.; Clerici, M. S. title: UV-A and UV-B Can Neutralize SARS-CoV-2 Infectivity date: 2021-05-31 journal: nan DOI: 10.1101/2021.05.28.21257989 sha: 6d0d1af969768c546fc056ea24c16934b13b3b7b doc_id: 1011743 cord_uid: s0bni60d We performed an in-depth analysis of the virucidal effect of discrete wavelengths: UV-C (278 nm), UV-B (308 nm), UV-A (366 nm) and violet (405 nm) on SARS-CoV-2. By using a highly infectious titer of SARS-CoV-2 we observed that the violet light-dose resulting in a 2-log viral inactivation is only 10-4 times less efficient than UV-C light. Moreover, by qPCR and fluorescence in situ hybridization (FISH) approach we verified that the viral titer typically found in the sputum of COVID-19 patients can be completely inactivated by the long UV-wavelengths corresponding to UV- A and UV-B solar irradiation. The comparison of the UV action spectrum on SARS-CoV-2 to previous results obtained on other pathogens suggests that RNA viruses might be particularly sensitive to long UV wavelengths. Our data extend previous results showing that SARS-CoV-2 is highly susceptible to UV light and offer an explanation to the reduced incidence of SARS-CoV-2 infection seen in the summer season. At the end of 2019 a new coronavirus, SARS-CoV-2, was firstly described in Wuhan, China, as being responsible for pneumonia within the scenario of a new disease: COVID-19. COVID-19 rapidly became a worldwide pandemic which is still raging and has been responsible for dramatic and unforeseeable Here, we used the TCID50 approach to quantify SARS-CoV-2 inhibition upon irradiation, using wavelengths spanning form UV-C to violet light and we build the corresponding UV-action spectrum. Next, we used PCR and FISH measurements to identify the UV-A and UV-B doses that completely inhibit the viral concentration comparable to the one present in the sputum of SARS-CoV-2-infected patients. Finally, we compared the obtained UV action spectrum on SARS-CoV-2, with those reported in the literature on other viruses and bacteria. Our data define the wavelengths and the doses of UV radiation that result in SARS-CoV-2 inactivation and show that SARS-CoV-2 infectivity can be completely blocked by UV-A and UV-B. These results offer an explanation to the seasonality of such infection, and provide fundamental information to build sterilization devices that avoid the toxicity of UV-C-based methods and that could be potentially effective in the inhibition of multiple RNA viruses. Table 1 ). For each UV-wavelength viral titers were determined by TCID50 endpoint dilution assay. Briefly: serial ten-fold dilutions, from 10 6 to 10 −4 TCDI50/mL (50ul), were plated onto 96-well plates, incubated at 37°C in 5% CO2 and checked daily to monitor the UV-induced . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint cytopathic effect. Seventy-two hours post infection (hpi) viral titer was determined as previously described 24 . The same UV-wavelengths were used to irradiate a SARS-CoV-2 viral stock at a concentration of 1.5 x 10 3 TCID50/mL. VeroE6 cell cultures were incubated with the virus inoculum in quadruplicate for one hour at 37°C and 5% CO2 and viral replication was assessed by qPCR, as previously described 14 , at 24, 48 and 72 hpi. SARS-CoV-2 that was not exposed to UV light was used as control. Fluorescence in-situ hybridization (FISH). It was performed using previously described 96 smiFISH probes 25 that target the whole (positive-strand) genome of the virus. According to the smiFISH protocol 26 , a FLAP sequence is appended at the 3' of each primary probe. Prehybridization with a complementary Cy3-labelled Cy3 probe was performed as using a PCR machine, as described 25 . A SARS-CoV-2 viral stock (1.5 x 10 3 TCID50/mL) was exposed to the UV-doses reported in Table 1 and an in vitro infection assay was performed on 1x10 5 Vero E6 cells grown on a 13mm glass coverslip (0.17mm thickness) as described in the previous section. Twenty-four hpi the supernatant was collected to quantify viral replication by qPCR, while cells were fixed by 4% PFA for 10 minutes, washed twice with PBS, and then stored until labeling in 70% ethanol at -20°C (minimum ethanol incubation: overnight). The day of the labeling, the coverslips were brought to room temperature, washed twice with 10% SSC-20× in RNase-free water (Buffer I) for 5 minutes, followed by a wash in 10% SSC-20× and 20% formamide in nuclease-free water (Buffer II). Cells were incubated overnight with the hybridized FLAP-duplex in a humidified chamber at 37°C. The probes were diluted 1:100 in the hybridization buffer (10% (w/v) of dextran sulfate, 10% of SSC-20× buffer and 20% formamide in Rnase-free water). Following hybridization, cells were washed twice in Buffer II in the dark for 30 min at 37°C, then washed in PBS for 10 min and stained with 1g/ml Hoechst 33342 in PBS. The coverslips were then mounted on glass slides using Vectashield (Vector Laboratories, Peterborough, UK). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. LED illumination conditions. To irradiate the virus we placed the above mentioned virus titers in 6well plates, which were then irradiated using LEDs with the spectral features reported in Figure 1A . While these sources are not monochromatic as in the case of low-pressure mercury lamps, they show a bandwidth of about 10 nm, similar to the sources used to generate the action spectrum on other viruses 29 . The sources were calibrated using an Ocean Optics HR2000+ spectrometer (Ocean Optics Inc., Dunedin, USA). The spectrometer was calibrated against a reference deuterium-halogen source (Ocean Optics is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint negligible and needs to be accounted to calculate the real dose delivered to the virus 14 . It is important to note that the virus itself does not contribute significantly to the source attenuation instead, as its concentration is too low to have a significant effect (see results). To compute the fraction of UV light that is absorbed by the medium we used the absorption spectrum of DMEM reported in Figure 1B and calculated the transmittance trend as function of the liquid thickness ( Figure 1C ). The resulting fraction of UV light delivered on average in the well (average) and at the bottom of the well (minimum) as function of the wavelength is reported in Table 1 . . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint receive the same UV dose. At 278 nm (UV-C), the discrepancy between the minimum and average value became more evident. The consequence is that the viruses at the bottom of the well receive 20% less UV photons than the average. According to the spectral and irradiance calibrations and the DMEM transmittances, the exposure times were calculated to provide the doses reported in Table 1 . Statistical Analyses. In the bar plots in Figure 3 , the mean ± standard error of the mean (SEM) are indicated. Data were analyzed using two-way ANOVA test by GRAPHPAD PRISM version 5 (Graphpad software, La Jolla, Ca, USA), and p-values of 0.05 or less were considered to be significant. We first applied the TCID50 approach to measure how effective UV light of different wavelengths is in inactivating the virus. To achieve this result, we infected cells with an initial viral concentration (6 x 10 6 TCID50/mL), which is significantly higher than that observed in infected patients. Next, we performed experiments on lower initial viral loads, comparable to the ones present in the sputum of COVID-19 samples 30 . Here, we used qPCR to evaluate the capability of the irradiated virus to replicate . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint and FISH to quantify the ability of the different UV doses to impair SARS-CoV-2 infectivity at the single-cell level. Wavelength-dependent inhibition of SARS-CoV-2 infectivity. Results reported in Figure 2 describe the trend of the viral titer as function of the dose for the four wavelengths considered. For all the analyzed wavelengths, we observed a dose-response relationship between the administered UV-dose and viral titer, indicating that virus is susceptible to different extents to a wide range of UV wavelengths. Notably, the decreasing trend was observed to be exponential, as predicted for such kind of mechanism 31 . Importantly, while a few mJ/cm 2 of UV-C light are sufficient to cause a several order of magnitude drop in the effective viral concentration, 10x to 1000x higher doses were needed to observe a similar viral load reduction following UV-B and UV-A irradiation. These data confirm earlier findings indicating that UV-C is more effective in virus inactivation compared to both UV-B and UV-A, and reinforce the possibility of using UV-C irradiation as an . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint efficient method for a fast inactivation of SARS-CoV-2 14-16 . Further, the data indicate that UV-B and UV-A also have virucidal potential, although at higher doses than UV-C. Action spectrum. To better characterize these results, we built the action spectrum at a 2-Log inactivation of SARS-CoV-2. We extrapolated the data fitting the effective concentration curves of Figure 2 with a simple single exponential trend, normalized the action spectrum by setting at 1 the inhibition efficiency at 278 nm and calculated the relative effectiveness at the other wavelengths ( Table 2 and Figure 3 ). The data and the plot recapitulate the results obtained by TCID50, namely that the measured relative effectiveness of UV irradiation on SARS-CoV-2 reaches a plateau at longer wavelengths, with UV-C being 10 4 times more effective than violet light. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. To examine the efficacy of UV-inactivation in a real-world scenario, we used the UV-doses reported in Table 1 on a SARS-CoV-2 viral concentration equivalent to the one found in the sputum of SARS-CoV-2 infected patients (1.5x10 3 TCID50/ml) 30 . QPCR was employed to quantify viral replication over time (Figure 4 ). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05. 28.21257989 doi: medRxiv preprint to note that the viral concentration in such case is expected to be much lower than those used in our experimental tests. Cell-by-cell distribution of the FISH signal displayed a bi-modal distribution, with 20-40% of cells being clearly infected at 24 hours post infection when non-irradiated virus was seeded on cells ( Figure 5C ). In contrast with these data, all the analyzed UV-wavelength reduced the fraction of positive cells, as probed by smFISH, resulting in a complete inhibition of the virus when SARS-CoV-2 was irradiated with wavelength specific critical doses (4mJ/cm 2 for UV-C, 100mJ/cm 2 for UV-B, 4000 mJ/cm 2 for UV-A and 24000 mJ/cm 2 for violet light) ( Figure 5D ). Notably at lower UV doses some cells with high levels of vRNA could be detected, suggesting that, rather than inhibiting the capability of the viral genome to replicate, UV-irradiation might affect either virus entry in the cells or the assembly of functional viral particles following the replication of the viral genome. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint Table 1 . Quantification is provided as mean +/standard deviation of the fraction of vRNA positive cells on at least two 6x6 mosaics. Scale bar: 10 m. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint Action-spectrum comparison of the UV-susceptibility of different microorganisms. We finally compared the results obtained in the TCID50 assay with those reported in the literature for other viruses and bacteria by building the UV action spectra for each of the different pathogens ( Figure 6 ). Data were normalized to the UV-C inactivation dose (for SARS-CoV-2 it has been assumed the same level of inactivation at 253 and 278 nm). It should be noted that whereas for both bacteria and DNA viruses the inactivation observed in the UV-A is 10 -5 -10 -6 times lower than that observed in the UV-C region, for RNA viruses the inactivation power is only marginally lower in the UV-A than in the UV-C region (just about 100-1000 times lower at 366 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint nm). Taking into account that the Solar illumination in the UV-A region is much larger than in the UV-B (95% of UV-A and 5% of UV-B), these results might justify the acknowledged seasonal behavior of the outbreaks of airborne viruses, including corona [39] [40] [41] and influenza viruses 42, 43 . In this respect, we would like to stress how it is crucial to have the specific data of the target microorganism to develop reliable solar inactivation and/or seasonal models and how Lytle and Sagripanti action spectrum cannot be considered as a reference trend. The ability of UV light to inactivate SARS-CoV-2 replication was studied as function of the irradiation wavelengths. Results showed that irradiation of a viral stock with a high infectious titer irradiated with LEDs at 278, 308, 366 and 405 nm resulted in a significant decrease in the fraction of active virus proportional to the light dose for all the wavelengths. Considering a 2-Log decrease and fixing to 1 the efficiency at 278 nm, it was possible to build the action spectrum, which is characterized by an effectiveness ratio of 10 -4 at 405 nm, indicating that even violet light has a non-negligible efficiency in inactivating SARS-CoV-2. While such a relatively flat action spectrum might appear surprising, our comparison of multiple UV-action spectra on diverse microbes revealed that RNA viruses, including SARS-CoV-2, might be more susceptible than other pathogens to long wavelength UV-light. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ‡These authors contributed equally. These experiments were supported by the following grants: Bando Regione Lombardia DG Welfare cod. RL_DG-WEL20MBIAS_01; CARIPLO -EXTRABANDO E PROGETTI TERRITORIALI cod. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 31, 2021. ; https://doi.org/10.1101/2021.05.28.21257989 doi: medRxiv preprint World Health Organization Declares Global Emergency: A Review of the 2019 Novel Coronavirus (COVID-19) SARS-CoV-2): A Systematic Review for Potential Vaccines Understanding COVID-19 Vaccine Efficacy on behalf of the COVID-19 Commission of Accademia Nazionale dei Lincei, R. COVID-19 Vaccines: Where We Stand and Challenges Ahead Persistence of Coronaviruses on Inanimate Surfaces and Their Inactivation with Biocidal Agents Optimal Control Strategies for the Transmission Risk of COVID-19 The History of Ultraviolet Germicidal Irradiation for Air Disinfection Sunlight-Mediated Inactivation of Health-Relevant Microorganisms in Water: A Review of Mechanisms and Modeling Approaches Simulated Sunlight Rapidly Inactivates SARS-CoV-2 on Surfaces Predicted Inactivation of Viruses of Relevance to Biodefense by Solar Radiation A Note on the Inactivation of Influenza a Viruses by Solar Radiation, Relative Humidity and Temperature Potential Impact of Seasonal Forcing on a SARS-CoV-2 Pandemic Drivers of Infectious Disease Seasonality : Potential Implications for COVID-19 Seasonality and Uncertainty in Global COVID-19 Growth Rates Inactivation of Material from SARS-CoV-2-Infected Primary Airway Epithelial Cell Cultures SmiFISH and FISH-Quant -a Flexible Single RNA Detection Approach with Super-Resolution Capability Globally Optimal Stitching of Tiled 3D Microscopic Image Acquisitions An Open-Source Platform for Biological-Image Analysis Irradiation by Ultraviolet Light-Emitting Diodes Inactivates Influenza a Viruses by Inhibiting Replication and Transcription of Viral RNA in Host Cells Predicting Infectious Severe Acute Respiratory Syndrome Coronavirus 2 From Diagnostic Samples Ultraviolet Germicidal Irradiation Handbook: UVGI for Air and Surface Disinfection Ultraviolet A and B Wavelength-Dependent Inactivation of Viruses and Bacteria in the Water Inactivation of MS2 Phage and . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) Cryptosporidium Parvum Oocysts Using UV-A from High-Intensity Light-Emitting Diode for Water Disinfection Biological and Physical Dosimeters for Monitoring Solar UV-B Light Spectrum of Virucidal Activity from Ultraviolet to Infrared Radiation Solar UV Radiation and Microbial Life in the Atmosphere Bacterial Inactivation by Solar Ultraviolet Radiation Compared with Sensitivity to 254 Nm Radiation TEMIS UV Index and UV Dose Operational Data Products, Version 2 Coronavirus Occurrence and Transmission Over 8 Years in the HIVE Cohort of Households in Michigan Estimated Inactivation of Coronaviruses by Solar Radiation With Special Reference to COVID-19 Modulation of COVID-19 Epidemiology by UV-B and -A Photons from the Sun Inactivation of Influenza Virus by Solar Radiation We are grateful to De Sisti Lighting for having provided a custom designed multi-LED UV lamp used in the experiments. We are grateful to Dr. F Mueller and Dr. C. Zimmer (Institut Pasteaur, Paris) for kindly providing the smiFISH probes. We acknowledge F. Nicastro, J. R. Brucato, P. Tozzi, I. Ermolli, G. Sironi, E. Antonello from INAF for many useful discussions.