Abstract Rational: The mutating SARS-CoV-2 potentially impairs the efficacy of current vaccines or antibody-based treatments. Broad-spectrum and rapid anti-virus methods feasible for regular epidemic prevention against COVID-19 or alike are urgently called for. Methods: Using SARS-CoV-2 virus and bioengineered pseudoviruses carrying ACE2-binding spike protein domains, we examined the efficacy of cold atmospheric plasma (CAP) on virus entry prevention. Results: We found that CAP could effectively inhibit the entry of virus into cells. Direct CAP or CAP-activated medium (PAM) triggered rapid internalization and nuclear translocation of the virus receptor, ACE2, which began to return after 5 hours and was fully recovered by 12 hours. This was seen in vitro with both VERO-E6 cells and human mammary epithelial MCF10A cells, and in vivo. Hydroxyl radical (·OH) and species derived from its interactions with other species were found to be the most effective CAP components for triggering ACE2 nucleus translocation. The ERα/STAT3(Tyr705) and EGFR(Tyr1068/1086)/STAT3(Tyr705) axes were found to interact and collectively mediate the effects on ACE2 localization and expression. Conclusions: Our data support the use of PAM in helping control SARS-CoV-2 if developed into products for nose/mouth spray; an approach extendable to other viruses utilizing ACE2 for host entry. Keywords: SARS-CoV-2, cold atmospheric plasma (CAP), Plasma activated medium (PAM), Angiotensin-converting enzyme 2 Introduction In the process of virus infection, the binding of viruses to host receptors leading to their entry into host cells is the first, critical and rate-limiting step[57]^1. The two main viral entry pathways, i.e., receptor-mediated endocytosis[58]^2^,[59]^3 and TMPRSS2-mediated membrane fusion[60]^4^, [61]^5, have facilitated the entry of all pandemic viruses from the past 20 years, SARS-CoV (2002)[62]^2^,[63]^5, MERS-CoV (2013)[64]^3^,[65]^5, and SARS-CoV-2 (2019)[66]^4 into host cells. Angiotensin-converting enzyme 2 (ACE2) plays a vital role in both pathways, especially for SARS-CoV[67]^2^,[68]^5 and SARS-CoV-2[69]^4^,[70]^6. The viral spike (S) protein binds to ACE2 to attach viruses on the host cell surface[71]^4, activating endocytosis or membrane fusion. Importantly, the UK variant-B.1.1.7 (Alpha), South Africa variant-B.1.351 (Beta), Brazilian variant-P.1 (Gamma) and Philippines variant-P.3 (Theta) lineage show increased transmissibility due to enhanced ACE2 binding affinity as a result of mutation N501Y[72]^7^,[73]^8. Also, the mutation L452R carried by California coronavirus variant-B.1.427/B.1.429 (Epsilon) and Indian double mutant-B.1.617.1 (Kappa)[74]^9 exhibited enhanced interactions between ACE2 and the spike receptor binding domain (RBD)[75]^10. The high transmission rate of the India variant B1.617.2 (Delta) has been attributed to its high binding affinity to ACE2 as a result of three mutations in RBD, i.e., L452R, P681R and T478K[76]^11^,[77]^12. The spike protein of the SARS-CoV-2 lineage C.37 (Lambda) has a unique pattern of 7 mutations ( Inline graphic 246-252, G75V, T76I, L452Q, F490S, D614G, T859N) where L452Q shares a similar impact on virus-cell interactions through ACE2, leading to substantially improved transmissibility[78]^13. Vaccines have been proven effective in controlling the morbidity and mortality of viral diseases by activating the generation of neutralizing antibody (NAb)[79]^14. However, SARS-CoV-2 can still be detected on nasal turbinates even after systemic NAb treatment[80]^15, suggesting that vaccines do not completely block virus infection. In fact, mutated viruses could partly (E484K, Alpha, Beta, Gamma, Epsilon, Eta, Iota, Zeta and Theta) and (E484Q, Kappa) or completely (S477N, Iota) escape from NAb-mediated neutralization[81]^8^,[82]^16, and positive COVID-19 cases have been reported among vaccinated individuals[83]^17. The omicron variant, firstly being reported on Nov 25, 2021 and now imposing another round of global health threat, contains some deletions (e.g., 69-70 del, G142D/143-145 del) and over 30 mutations (e.g., T95I, K417N, T478K, N501Y, N655Y, N679K, P681H)[84]^18, which lead to enhanced transmissibility and extreme escape from immunological control[85]^19^-[86]^23. Therefore, the establishments of further rapid, broad-spectrum anti-viral therapies, including ACE2-based preventive approaches, are urgently called for. Cold atmospheric plasma (CAP), comprising partly ionized gas with a high electron temperature but low overall temperature, has been widely used in the biomedical sector[87]^24 for, e.g., skin disinfection[88]^25 and blood coagulation[89]^26, and has novel applications such as neural differentiation induction[90]^27 and emerging anti-cancer potential[91]^28^-[92]^36. Its roles in modifying the cell surface for safe yet desirable cellular responses have been well documented[93]^37. It had been proven that direct CAP treatment could fragment the viral capsid protein and damage the RNA of norovirus (NoV)[94]^38, bacteriophage MS2[95]^39, immunodeficiency virus (HIV-1)[96]^40 and SARS-CoV-2 [97]^41^, [98]^42. CAP pre-treated host cell, monocyte-derived macrophages (MDM), could escape viral infection by down-regulating CD4 and CCR5[99]^40. In this study, we investigated whether direct CAP or indirect CAP in the form of plasma-activated medium (PAM, produced from the plasma-liquid interactions) is preventive against SARS-CoV-2 infection by decreasing ACE2 surface prevalence using SARS-CoV-2 and bioengineered pseudovirus containing ACE2-binding spike protein domains. The proposed strategy, i.e., through effective control of ACE2 cell membrane expression in tissues with high exposure risk such as nose, throat and lung[100]^43, may represent an emerging and promising approach against the outbreak of viral diseases relying on this receptor for host entry[101]^44. By fabricating pseudoviruses with the ACE2-binding domain of spike protein, and conducting experiments using pseudoviral and SARS-CoV-2, we demonstrated the efficacy of PAM in protecting SARS-CoV-2 entry into Vero cells without cell toxicity. We found that CAP-triggered ACE2 relocation from cell surface to the nucleus via the ERα/STAT3(Tyr705) axis and reduced ACE2 expression through EGFR(Tyr1068/1086)/STAT3(Tyr705) signaling. Among the tested short- and long-lived reactive oxygen and nitrogen species (RONS) generated with CAP, hydroxyl radicals (·OH) and reactive species generated through its interactions with other species in PAM were found to be the most important CAP components triggering reduced ACE2 expression. Methods Cell culture MCF10A cells are normal epithelial breast cells and used here as the in vitro cell model to investigate the effect of CAP on ACE2 in healthy epithelial cells. MCF10A immortalized diploid quasi-normal human mammary epithelial cells obtained from ATCC (Manassas, USA) were cultured in DMEM with 20 ng/mL EGF, 100 ng/mL Cholera Toxin, 5% horse serum, 10 µg/mL insulin, 1% streptomycin and 0.5 mg/mL hydrocortisone (Sigma, USA). Vero E6 cells (African green monkey kidney cells) have abundant ACE2 expression and have been extensively used as cell-based virus infection models for SARS-CoV research since 2003[102]^45, and are thus used as the cell model for SARS-CoV-2 virus infection in this study. Vero E6 cells (ATCC C1008, ECACC, Wiltshire, England; Sigma Aldridge, St. Louis, MO, USA; CRL-1586, Manassas, USA) were cultured in RPMI-1640 with 10% fetal calf serum (FCS) (Sigma, USA). Cells were passaged with trypsin and stored under liquid nitrogen. Cells were routinely subjected to STR profile validation and confirmed negative for mycoplasma. Plasma-activated medium (PAM) generation The CAP generation device was a typical high-frequency (1.7 MHz, 2-6 kV), atmospheric pressure Argon plasma jet (model kINPen 09) developed at the Leibniz-Institute for Plasma Science and Technology (INP Greifswald, Germany). CAP was streamed out of the quartz tube via the Argon gas pressure with a flow rate of 5.0 Standard L/min (power: < 3.5 W in the hand-held unit). Aliquots (1 mL) of medium without serum in 2 mL centrifuge tubes were activated by CAP for 5 s and 30 s, namely '5 s-activated' and '30 s-activated' PAM. PAM used for SARS-CoV-2 virus treatment was CAP-activated for 10 min and diluted to 10% before being transferred to the PC3 lab. The distance between the kINPen jet and the liquid surface was 1 cm. Staining reagents and antibodies Primary antibodies: SARS-CoV-2 Spike S1 Antibody (HC2001, GenScript Biotech, USA), ACE2 for Western blot (WB; sc-20998, Santa Cruz, USA), ACE2 for immunofluorescence (IF) (sc-73668, Santa Cruz, USA), STAT3 (12640, Cell Signalling, USA), p-STAT3 (Try705) (9145, Cell Signalling, USA), EGFR (sc-120, Santa Cruz, USA), p-EGFR(Tyr1068) (44788G, Thermo-Fisher Scientific, USA), p-EGFR(Tyr1086) (369700, Thermo-Fisher Scientific, USA), ERα (ab14020, Abcam, USA), β-actin (sc-8432, Santa Cruz, USA). Secondary antibodies: goat anti-human IgG Alexa Fluor 555 (A-21433, ThermoFisher, USA), goat anti mouse IgG-FITC (sc-2010, Santa Cruz, USA), goat anti-chicken IgY H&L Alexa Fluro 647 (ab150175, Abcam, USA), goat anti-mouse IgG H&L Alexa Fluro 647 (ab150115, Abcam, USA), goat anti-mouse IgG-HPR (sc-2031, Santa Cruz, USA), goat anti-rabbit IgG H&L-HRP (ab6721, Abcam, USA). Phalloidin labelled with Alexa Fluor 647 for F-actin staining (A22287, ThermoFisher, USA). Hoescht 33342 for nuclear (DNA) staining (H3570, ThermoFisher, USA). Western blot Proteins were collected in RIPA buffer with phosphatase inhibitor (Thermo-Fisher Scientific), and levels normalized after Pierce BCA Protein Assay Kit (Thermo-Fisher Scientific) analysis. Each protein sample (10 mg) was diluted with 4×loading buffer, separated by a 10% SDS-PAGE gel, and wetly transferred to polyvinylidene fluoride (PVDF) membranes (Thermo-Fisher Scientific). Membranes were blocked (3% BSA in (Tris-HCl buffer and Tween; TBST)) for 1 h. The primary antibodies were diluted (3% BSA in TBST) at a ratio of 1:1000. All membranes were incubated with primary antibodies overnight at 4 °C with shaking. Membranes were washed for 10 min using TBST buffer three times with shaking, followed by incubation with HRP-conjugated secondary antibodies for 2 h at 4 °C with shaking. The secondary antibodies were diluted (3% BSA in TBST) at a ratio of 1:4,000. Membranes were washed 3×10 min with TBST and incubated by enhanced chemiluminescence (ECL) (Bio-Rad) for 5 min. The chemical fluorescence from blots was photographed and scanned by ChemiDoc XRS+ system (Bio-Rad). The images were analyzed and normalized by Image Lab program (Bio-Rad). SARS-CoV-2 stock production and titration SARS-CoV-2 infection studies at QIMR Berghofer were conducted in a dedicated PC3 (BSL3) suite, with safety approval from the QIMR Safety Committee (P3600). The SARS-CoV-2 virus was isolated from a patient and was a kind gift from Queensland Health Forensic & Scientific Services, Queensland Department of Health; the isolate (hCoV-19/Australia/QLD02/2020) has been sequenced, and is available at GISAID ([103]https://www.gisaid.org/). Virus stock was generated by infecting Vero E6 and after 3 days culture supernatant was clarified by centrifugation at 3000 Inline graphic g for 15 min at 4°C, and was aliquoted and stored at -80°C. Virus titer (10E6.79 TCID[50]/mL virus) was determined using standard CCID50 assay by infecting Vero E6 cells with 10-fold serial dilutions of virus stock and measuring cytopathic effect with titer calculation by the method of Spearman and Karber. Virus was determined to be mycoplasma free[104]^46 and FCS used for culture determined to be endotoxin free[105]^47. SARS-CoV-2 virus experiment Vero E6 cells were seeded in a 96-well-plate at 10,000/well (100 μL) for 24 h prior to PAM treatment. 10% FCS RPMI-1640 media was removed and replaced with 10% PAM (pre-warmed to 37°C for 5 min) for 10 min. PAM was removed and changed to the 10% FCS RPMI-1640 medium. The virus stock was diluted to 1/10^2, 1/10^3, 1/10^4, 1/10^5 or 1/10^6 in 10% FCS RPMI-1640 medium and added to cells (quadruplicate wells for each treatment group) for 2 additional hours, after which the medium was removed, cells were washed twice using phosphate buffered saline (PBS), and the cultures were incubated in 10% FCS RPMI-1640 medium, which also served as a no-virus control. Wells were fixed by 4% glutaraldehyde 24 h after virus addition, stained for ACE2, SARS-CoV-2 virus, actin, DNA as described under immunofluorescence and subjected to scanning electron microscopy (SEM). This SARS-CoV-2 virus experiment was independently conducted three times. Immunofluorescence For SARS-CoV-2 virus analysis, cells in four representative wells were fixed and permeabilized in 0.2% Triton X-100 for 5 min followed by blocking (3% BSA in PBS) for 1 h. Cells were then incubated with both ACE2 and SARS-CoV-2 Spike S1 antibodies diluted 1:1000 in PBS containing 3% BSA overnight, and washed 3×10 min with PBS. Goat anti-human IgG Alexa Fluor 555 (A-21433) and goat anti mouse IgG-FITC (sc-2010) were used as secondary antibodies at a dilution ratio of 1:4000 using PBS containing 3% BSA. After 2 h incubation, cells were washed 3×10 min with PBS and incubated with Phalloidin at the room temperature for 30 min, and again washed 3×10 min with PBS and incubated with Hoescht 33342 dye for 5 min. High-throughput analysis was performed using InCell 6500HS (60×, 120 fields per well) and quantified by IN Carta analysis software. For analysis of CAP and PAW effects on ACE2 levels and location, MCF10A or Vero E6 cells were plated at 10,000 cells per well for 12 h. After treatment, cells were fixed with 4% paraformaldehyde in PBS for 10 min at 4 °C. Cells were first permeabilized in 0.2% Triton X-100 for 5 min and then blocked (3% BSA in PBS) for 1 h, followed by incubation with primary antibodies overnight. After incubation, cells were washed 3×10 min with PBS before the addition of secondary antibodies followed by incubation at the room temperature for 1 h. Nuclear DNA was stained with Hoechst 33342 for 5 min at 1 μg/mL. Delta Vision Elite Live Imaging Microscope (Applied Precision, GE Healthcare Lifesciences, Parramatta NSW, AUS) and softWoRx analysis software (Applied Precision, USA) were used for cell imaging. Scanning electron microscopy All samples were fixed in 4% glutaraldehyde (Sigma-Aldrich, USA) in PBS buffer. Subsequently, an increasing gradient of ethanol solutions (0%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100%) was implemented for dehydration followed by 100% Hexamethyldisilazane (HMDS) (Merck, Australia) for 30 mins. All samples were dried in a biosafety hood for 24 h. Finally, all samples were loaded onto 12.6 mm SEM mounts (ProSciTech, AU) using 12 mm double-sided carbon tabs (ProSciTech, AU) and sputtered with a 10-nm-thick gold layer using Cool Sputter Coater Leica EM SCD005 (Leica, DE). SEM analysis was performed under 5 kV voltage and 8 A current using a JEOL JSM-7001F (JEOL Ltd. JP) at magnifications of ×7,500, ×10,000 and ×15,000. ROS scavenging assay Reactive species trapping tests were carried out to investigate the respective roles of different active species generated in PAM. Sodium pyruvate (10 mM), uric acid (100 μM), mannitol (200 mM), tiron (20 mM), hemoglobin (20 mM), monopotassium phosphate(1 mM) were purchased from Signa-Aldrich (Australia) and used to trap H[2]O[2], O[3], ·OH, ·O[2]^-, ·NO and e^-, respectively[106]^48^-[107]^50. MCF10A cells (10,000 cells per well) were seeded in the 96-well plate and cultured for 24 h, with 5,000 cells per well. 1 mL DMEM medium was activated by CAP for 30 s to prepare PAM following the protocol described in the 'PAM generation' section. PAM was divided into 7 tubes, 90 µL per tube. 10 µL of each ROS scavenger was added to 90 µL PAM followed by thorough mixing and 1 min incubation. 50 µL of each mixture was then added to MCF10A cells for 30 s followed by immunofluorescence imaging of ACE2. All scavengers were proven non-toxic at the working concentrations used in the assays[108]^51. Sodium pyruvate is not a specific scavenger for H[2]O[2] but can also react with peroxynitrite (ONOO^-)[109]^52. Thus, inhibition of any reaction by pyruvate can be explained by either the action of H[2]O[2] or ONOO^-. ROS quantification assay The titanium sulfate method was used to measure H[2]O[2], where H[2]O[2]and titanium oxysulfate (TiOSO[4]) could react and generate a yellow-colored complex (pertitanic acid) that can be quantified by spectroscopy at 410 nm[110]^53. ·OH concentrations were determined using benzene‐1,4‐dioic acid (TA) and measured by the fluorescence assay, where ·OH radicals react with TA to form 2-hydroxyterephthalic acid (HTA) with an irradiation wavelength of 310 nm and a fluorescence emission wavelength of 425 nm[111]^54. A multiparameter photometer (Hanna Instruments, HI83399) was used to determine the O[3] concentration using a certified ozone reagent kit from the supplier. The Griess Reagent method was used to quantify NO^2-; and a nitrate specific ion electrode was used to quantify NO^3-, with an Ionic Strength Adjuster (NH[4])[2]SO[4] (2 mol/L) being added to maintain a constant ionic strength before detection[112]^55. A spin trap comprised of a complex of N-methyl-D-glucamine dithiocarbamate (MGD) and iron (II) ion was used to react with ·NO to form (MGD)[2]Fe^2+-NO adducts, which were detected by ESR[113]^51. Plasmids for biological assembly of SARS-CoV-2 protein coated polyester particles The following DNA fragments encoding the SARS-CoV-2 proteins N (nucleocapsid) and RBD (receptor binding domain derived from spike protein) had been previously synthesized by Biomatik (Canada) and codon optimized for E. coli strains: N protein SDNGPQNQRNAPRITFGGPSDSTGSNQNGERSGARSKQRRPQGLPNNTASWFTALTQHGKEDLKFPRGQGV PINTNSSPDDQIGYYRRATRRIRGGDGKMKDLSPRWYFYYLGTGPEAGLPYGANKDGIIWVATEGALNTPK DHIGTRNPANNAAIVLQLPQGTTLPKGFYAEGSRGGSQASSRSSSRSRNSSRNSTPGSSRGTSPARMAGNG GDAALALLLLDRLNQLESKMSGKGQQQQGQTVTKKSAAEASKKPRQKRTATKAYNVTQAFGRRGPEQTQGN FGDQELIRQGTDYKHWPQIAQFAPSASAFFGMSRIGMEVTPSGTWLTYTGAIKLDDKDPNFKDQVILLNKH IDAYKTFPPTEPKKDKKKKADETQALPQRQKKQQTVTLLPAADLDDFSKQLQQSMSSADSTQA. RBD protein RVQPTESIVRFPNITNLCPFGEVFNATRFASVYAWNRKRISNCVADYSVLYNSASFSTFKCYGVSPTKLND LCFTNVYADSFVIRGDEVRQIAPGQTGKIADYNYKLPDDFTGCVIAWNSNNLDSKVGGNYNYLYRLFRKSN LKPFERDISTEIYQAGSTPCNGVEGFNCYFPLQSYGFQPTNGVGYQPYRVVVLSFELLHAPATVCGPKK. All clonings had been carried out using the E. coli Top10 strain (Table [114]1) following the protocol described in Reference [115]^56. The target DNA sequences of all pET plasmids were confirmed by the Griffith University DNA Sequencing Facility (Griffith University, Australia). Three genes, phaA, phaB, and phaC encoding enzymes PhaA, PhaB, and PhaC were required for polyester particle formation. The plasmid pMCS69 contains the genes phaA and phaB, and the pET plasmid contains the gene phaC. Both pMCS69 and pET-14b phaC- SARS-CoV-2 genes were transformed into the endotoxin-free production strain, ClearColi BL21(DE3) (Table [116]1). Table 1. Bacterial strains and plasmids used in this study. Strains/plasmids Characteristics/DNA sequences References