The VeroE6 (African green monkey, Cercopithicus aethiops, kidney) and HEK293T (Human embryonic kidney) cell lines were cultivated in Dulbecco’s modified Eagle (DMEM) (Thermo Scientific #12100-046) medium, supplemented with 10% fetal bovine serum (FBS # 12657029) and 1% penicillin/streptomycin (Gibco #15140-122). Cells were maintained at 37°C in a humidified atmosphere containing 5% carbon dioxide.

An aliquot of SARS-CoV-2 SP02.2020 (GenBank accession number MT126808) isolate was kindly donated by Dr. Edison Luiz Durigon (Institute of Biomedical Sciences, University of São Paulo). Vero E6 cells were used for virus propagation in the Biosafety Level 3 Laboratory (BSL-3) of the Laboratory of Emerging Viruses (Institute of Biology, State University of Campinas). Viral infections were performed in Vero cells seeded in 24-well plates (5 × 10⁵ cells/well) for the experiments with treatments and immunofluorescence assays, and six-well plates (1 × 10⁶ cells/well) for Western blots. A multiplicity of infection (MOI) of 0.3 was used for all experiments.

To do the FLAG‐tagged protein expression, a modified pcDNA 3.1 (+) (Thermo Fisher Scientific) was generated by cloning the FLAG peptide coding sequence upstream of the multiple cloning sites, using the NheI and BamHI restriction sites, and generating the pcDNA‐FLAG vector (Oliveira et al., 2013). The full‐length of M, N, and E genes were codon-optimized, synthesized (Geneart -Thermo Fisher Scientific), and cloned into pcDNA‐FLAG, generating plasmid pFLAG‐M, pFLAG-N, and pFLAG-E. The pFLAG‐green fluorescent protein (GFP) (Amaral et al., 2016) was used as the control plasmid for the expression of a non-related protein in the immunoprecipitation assays.

VeroE6 and HEK293T cells were seeded for 24 hours before transfection. Transfection was performed with Lipofectamine (Thermo Scientific - #20071882) and PLUS reagents (Thermo Scientific - #15338100). The protocol of plasmids’ transfection is described by (Amaral et al., 2016). For immunofluorescence, the cells were seeded and transfected in 24-well plates; for western blotting in 6-well plates; and for immunoprecipitation in 100 mm plates.

For anti-FLAG immunoprecipitation, HEK293T cells cultivated in 100 mm diameter dishes expressing the FLAG-tagged GFP, E, M, and N proteins were washed twice with PBS and harvested by pipetting up and down in PBS after 48 hours. Cells were resuspended in 500 µL of lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100) containing protease inhibitor cocktail (Roche). Protein lysates were incubated and shaken on ice for 15 min and centrifuged at 12,000 × g for 10 min at 4°C. Supernatants were collected and protein quantification was performed using a BCA protein assay kit (Thermo Scientific). A total of 2,000 µg of protein extract was used to perform the immunoprecipitation, so the samples were diluted with lysis buffer without inhibitors and incubated overnight at 4°C with 30 µL of anti-FLAG agarose-coupled beads (#A2220, Sigma-Aldrich) under mild agitation. Subsequently, the beads were washed five times with 600 µL of ice-cold TBS (50 mM Tris-HCl, pH 7.5, 150 mM NaCl) and eluted with 300 ng/L of FLAG peptide (#F4799, Sigma-Aldrich) for four hours under moderate agitation (Pavan et al., 2016). Supernatants were collected and stored at -20°C for immunoblotting analysis.

For reverse immunoprecipitation, in HEK293T cells, the same protocol of protein lysis and quantification were performed. A total of 300 µg of protein extract was used to perform the immunoprecipitation. The samples were diluted with lysis buffer without inhibitors and incubated overnight at 4°C with anti-PCNA antibody (1:500, #2586, Cell Signaling) under mild agitation. Subsequently, 30 µL of protein A/G Agarose (#20421, Thermo Scientific) was added to each sample, followed by mild agitation for two hours. Samples were washed five times with a wash buffer (25 mM Tris-HCl, pH 7.5, 150 mM NaCl) and the elution was performed with Laemmli Buffer with SDS 1×. Supernatants were collected and stored at -20°C for immunoblotting analysis.

Sterilized glass coverslips were treated with 6M HCL (Synth) and placed in each well on 24-well plates. Vero E6 cells were seeded at a density of 1×10⁵ cells in DMEM with 10% FBS. After the experimental procedures, the wells were washed 1× with PBS, permeabilized with ice-cold methanol 100% for 10 minutes, or fixed with paraformaldehyde 4% (PFA – Sigma Aldrich 158127) for 15 minutes. The cells were then permeabilized with PBS-Tween-20 0.1% (PBS-T - Sigma Aldrich P1379) for 10 minutes, blocked with 1% BSA-Tween-20 0.3% for 30 minutes, and incubated overnight with a solution containing the primary antibody in (1:200) 1% BSA-Tween-20 0.3% at 4°C. After a washing step with PBS (three times for 10 minutes each), a solution containing the secondary antibody (1:250) and PBS-Tween20 0.1% was added to each well for one hour in a dark chamber. The wells were then washed three times with PBS (for 10 minutes each), a solution containing Hoechst (Sigma Aldrich – 861405 - 1:1,000) was added for 10 minutes, followed by three times washing with PBS (for 10 minutes each), the glass coverslips were removed from the wells with the aid of tweezers and added to glass slides with 5 μL of Glycerol (Sigma Aldrich G5516). The primary antibodies used were: PCNA (Cell Signaling #2586), N (Invitrogen #MA5-35943), FLAG (Sigma # F3165). The secondary antibodies used were: Alexa-Fluor-594 Goat anti-Rabbit (Jackson ImmunoResearch #711-585-152), Alexa-Fluor-488 Goat Anti-Mouse (Jackson ImmunoResearch #705-545-003). To quantify the signal on immunofluorescences we used the ImageJ software to transform the images in 8 bit (grayscale), and check-in Set Measurements the options: Area and Integrated density; select an ROI of nuclei area (Hoechst staining) to determinate the fluorescence signal on nuclei; the cytoplasmatic fluorescence signal was obtained by a total cell - nuclei area ROI. Besides that, we computed the signal of the background (region without cells). Finally, the Corrected Cell Total Fluorescence (CTCF) was calculated by multiplication of the integrated density of gray of nuclei or cytoplasm by their respective areas and subtracting the background signal.

Vero E6 cells were seeded at 8×10⁶ in a P100 plate and transfected with pFLAG-GFP or pFLAG-M. After 48 hours, the subcellular fractionation was finalized (Baldwin, 1996). Briefly, the cytoplasmic fraction was first isolated from the nuclear solution using a cytoplasmic buffer (10 mM HEPES; 60 mM KCl; 1 mM EDTA; 0.075% NP-40; 1 mM DTT; protease inhibitor 1×) (pH=7.6) and centrifuged in 1,400×g for 30 minutes. The cytoplasmic supernatants were collected into a new tube. The nuclear pellet was washed with cytoplasmic buffer without NP-40 two times followed by a five-minute centrifugation at 1,000×g. The final pellet was resuspended with nuclear buffer (20 mM TrisCl; 420 mM NaCl; 1.5 mM MgCl₂; 0.2 mM EDTA; 25% glycerol; protease inhibitor 1×) (pH=8.0) and incubated on ice for 10 minutes, and vortexed periodically. Finally, the cytoplasmic and nuclear solutions were centrifuged at 12,000×g for 10 minutes and the supernatants were collected into new tubes. The quantification was performed by BCA.

For PLA protocol, Vero E6 cells were grown on chambered slides at a confluency of 70% and transfected with pFLAG-M. After 24 hours, cells were fixed with ice-cold methanol for 15 min and permeabilized with 0.1% TritonX-100 in PBS. The next steps faithfully followed the manufacturer’s specifications (Duolink^® In Situ – Sigma-Aldrich). Briefly, cells were blocked and stained with primary antibodies (anti-FLAG 1:400 - Sigma F7425 and anti-PCNA 1:400 – Sta Cruz sc-56) and PLA probes (anti-mouse MINUS - DUO92004 and anti-rabbit PLUS - DUO92002). Then, cells were incubated with Ligase for 30 minutes (37°C) and Polymerase for signal amplification for 100 minutes (37°C) (Duolink^® In Situ Orange - DUO92102). For negative controls, cells were stained, missing one of the primary antibodies, or both. After all PLA protocol steps, cells were stained with anti-FLAG 1:400 and anti-rabbit Alexa Fluor 488 1:1,000 (Invitrogen - A-11008) to identify transfected cells. Nuclei were stained with DAPI. After the respective washes, slides were mounted with Fluoromount-G (Invitrogen). The contrast was evenly enhanced in all presented images for better visualization of PLA dots.

For co-localization analysis, images were taken with a Zeiss LSM 780 NLO confocal microscope coupled to a HeNe (543 nm), an Argon (488 nm) and a Diode (405 nm) lasers (Core Facility for Scientific Research – University of Sao Paulo (CEFAP-USP)). Images were acquired with an objective α Plan-Apochromat 100x/NA 1.46 in oil immersion. Fluorescent signal was detected on a 32 channel GaAsP QUASAR detector with the following parameters: Alexa Fluor 594 (578 – 692 nm), Alexa Fluor 488 (491 – 587 nm), DAPI (412 – 491 nm). Pinhole was set to 1 airy unit, and z-stacks were taken with intervals of 340 nm. Presented images show a single representative z-stack. Overlap in signal between different channels was measured with Plot Profile plugin on ImageJ. Images were contrast-enhanced for better visualization.

The proteins were collected from Vero E6 and HEK293 cells using a cell lysis buffer (50 mM Tris-Cl pH 7.5; 150 mM NaCl, 1 mM EDTA, 1% Triton x-100, and protease and phosphatase inhibitor cocktail). To obtain the lysates, cells were maintained with lysis buffer for 15 minutes on ice and centrifuged at 12,000 × g for 10 minutes. Samples containing total protein were separated by SDS-PAGE and transferred to nitrocellulose membranes. Membranes were blocked for one hour at room temperature (RT) with 5% non-fat powdered milk dissolved in TBS-Tween-20 (TBS-T) (50 mM Tris-Cl, pH 7.5; 150 mM NaCl; 0.1% Tween-20), incubated overnight (4°C) with primary antibodies and with secondary antibodies against mouse or rabbit IgG conjugated with peroxidase (Amersham) for one hour at RT. The membranes were washed three times with TBS-T, and incubated with the SignalFire ECL Reagent (Cell Signaling Technology) for protein bands visualization. The band’s densitometry was performed using ImageJ software v1.53 (National Institutes of Health). The following antibodies were used: anti-PCNA (#2586); anti-γH2AX (#9718); anti-β-actin (#4967) (Cell Signaling), anti-α-tubulin (CP06) (Calbiochem), anti-N (MA5-35943) (Invitrogen), anti-laminA/C (#A303-430A) (Bethyl) and anti-FLAG (#A2220) (Sigma).

Lung tissue sections, obtained via Minimally Invasive Autopsies, (COVID-19 case and a control lung tissue of a non-smoker subject) were stained in silanized slides (Sigma Chemical Co. St. Louis, Missouri, EUA). Briefly, the histological sections were deparaffinized in xylene, rehydrated through a graded series of ethanol, and kept in a tri-phosphate buffer (TBS) pH 7.4. After deparaffinization, slides were hydrated for five minutes in graded alcohol series (100%, 95%, 70%, and H₂O) and incubated in Tris-EDTA for 50 minutes at 95°C pH 9.0. The blockage of the endogenous peroxidases was done by hydrogen peroxide 10v (3%).

The slides were then incubated with Ab mix (TBS (1%), BSA (4%), and Tween20 (0.02%)) for 20 minutes at RT followed by treatment overnight at 4°C with anti- γ-H2AX (Cell Signaling #9718S) diluted 1:500 in Ab mix or anti-PCNA diluted 1:1000 (DAKO #M0879). The slides were then washed twice in TBS (1%) and finally incubated for one hour at RT with 1:500 goat anti-rabbit/mouse HRP polymer detection kit – ImmunoHistoprobe Plus (ADVANCED-BIOSYSTEMS) for 30 minutes at 37°C). The diaminobenzidine (DAB) was used as a chromogen (Sigma-Aldrich Chemie, Steinheim, Germany). Finally, the counterstaining was done with Harris Hematoxylin (Merck, Darmstadt, Germany), and the slides were mounted with a coverslip in Permount (Fischer #SP15-500). The use of this material for research purposes has been previously approved by the Institutional Ethical Board CAAE #30364720.0.0000.0068).

Lung tissue sections A plaque assay was performed as described in Kashyap et al. (2021). Briefly, after seeding, Vero E6 (8 × 10⁵ cells per well, 6-well plates) were incubated overnight and infected with 200 PFU per well. PCNA-I1 (Cayman #20454) or Verdinexor (Cayman #26171) were added at final concentrations of 0.5 µM or 0.1 µM and 0.1 µM or 1 µM, (Lu and Dong, 2019) in overlays composed of DMEM supplemented with 10% FBS plus carboxymethylcellulose sodium salt (Sigma-Aldrich #C5678) 2%, immediately post-viral adsorption. A MTT cell proliferation assay (Roche #11465007001) was done to verify the toxicity of the drugs ( Supplementary Figure S5 ). After four days, cells were fixed with formaldehyde and stained with crystal violet to count plaques.

Data are presented as mean ± Standard Deviation (SD). Statistical analysis of the data was performed by Student’s t-test or ANOVA. P-values of ≤0.05 were considered statistically significant. All experiments were performed in biological duplicates. The odds ratio was calculated to verify if the translocation of the PCNA to the cytoplasm was dependent on the M transfection. A total of 147 cells were analyzed. A value of odds ratio greater than one indicates that the event observed is dependent on the object analyzed.

In the study by Gordon et al. (2020), the interactome of the structural and non-structural proteins of SARS-CoV-2 was performed in HEK293T cells (Gordon et al., 2020). The analysis of affinity-purification–mass spectrometry (AP-MS)-based proteomics showed that PCNA interacts with E and M SARS-CoV-2 proteins. In our study, we validated this interaction through an anti-FLAG immunoprecipitation assay ( Figure 1A ). FLAG-tagged GFP, E, M, and N proteins were expressed in HEK293T cells and immunoprecipitated with anti-FLAG antibodies. As a result, we identified that PCNA co-immunoprecipitated with the E and M SARS-CoV-2 proteins but not with the N protein, which is expected since PCNA was not identified as an interactor of N protein (Gordon et al., 2020). Anti-FLAG immunoprecipitation confirmed the interaction between M and PCNA in two independent experiments ( Supplementary Figure S1 ). We also confirmed that PCNA and M interact, through a reverse immunoprecipitation assay ( Figure 1B ), in HEK293T cells previously transfected with FLAG-tagged GFP, E, M, and N. We identified that only FLAG-tagged M co-immunoprecipitated with PCNA, revealing a more specific interaction between these proteins. Furthermore, we explored the interaction between FLAG-M and PCNA by proximity ligation assay (PLA). Vero E6 cells expressing FLAG-M were fixed 24 h.p.t. (hours post-transfection) and labeled with primary antibodies against FLAG (rabbit) and PCNA (mouse), followed by PLA probes conjugated to anti-rabbit or anti-mouse. PLA signal is emitted when probes attached to primary antibodies are closer than 40 nm, indicating protein interaction. As negative controls, transfected cells were labeled with only one of the primary antibodies or omitting both. Figure 1C (panels A-C and Supplementary Figure S2 ) shows positive PLA dots in transfected cells, while minimal or no signal is seen in the respective controls ( Figure 1C – panels D-L). Confocal microscopy analysis of the PLA assay also showed that the M-PCNA interaction occurs in the cytoplasm, indicating a possible translocation of the PCNA to the cytoplasm, induced by the M protein ( Supplementary Figure S3 ).

To better understand how the interaction of M-PCNA could be acting on the cells, we conducted immunofluorescence (IF) assays in Vero E6 cells expressing FLAG-M. 24 hours after transfection, cells were stained and analyzed by widefield microscopy ( Figure 2A ). Although PCNA did not entirely co-localize to the structures where M is present, we observed that in transfected cells, PCNA presented a more cytosolic pattern when compared to the non-transfected ones, as indicated by plot profile analysis ( Figure 2A – Arrows 1 and 2). To further investigate this phenomenon, we quantified the fluorescence intensities of nuclear and cytoplasmic PCNA in cells expressing FLAG-M and non-transfected cells. Figure 2B shows that in non-transfected cells, PCNA has a propensity to be nuclear, but under FLAG-M expression, the PCNA is translocated to the cytoplasm. We also evaluated the translocation of PCNA to the cytoplasm by odds ratio. The odds ratio for cytoplasmic PCNA with FLAG-M expression was 6.34 (Confidence Interval 95% =2.83-13.3) compared to non-transfected cells, indicating the correlation of M protein with the translocation of the PCNA to the cytoplasm. In addition, the subcellular fractionation of Vero E6 cells transfected with FLAG-M or FLAG-GFP was carried out ( Figure 2C ). In agreement with IF data, FLAG-M transfected cells showed an evident reduction of PCNA in the nuclear fraction, while it remained nuclear in cells expressing FLAG-GFP ( Figure 2C ). These results indicate that M protein acts on PCNA translocation from the nucleus to the cytoplasm.

To confirm the interaction between M and PCNA, we performed confocal immunofluorescence of FLAG-M transfected cells. Plot profile analysis shows an overlap between FLAG-M and PCNA signals, indicating the proximity of the proteins ( Figure 3 ). Moreover, as mentioned before, the interaction between FLAG-M and PCNA was detected by PLA ( Figure 1C ), and the co-localization of PLA puncta with FLAG-M was shown by confocal microscopy ( Supplementary Figure S3 ). Taken together, these results indicate the ability of M protein in inducing PCNA translocation from the nucleus, and to interact with it in the cytoplasm.

PCNA is one of the essential proteins for DNA replication and DNA repair, and it is also known as a DNA damage marker. To address whether M expression could be related to the induction of DNA damage. We transfected HEK293T and Vero E6 cellswith FLAG-M and looked at γH2AX and PCNA expression levels since both proteins elevated expression is associated with DNA damage. In HEK293T transfected cells, we could observe a slight increase in the levels of PCNA and γH2AX ( Figure 4A ), but no significant difference was observed in the Vero E6 cell line ( Figure 4B ). Because transfection efficiency could be a limiting factor for detecting inconspicuous events like DNA damage, we also explored γH2AX levels in transfected Vero E6 cells by immunofluorescence ( Figure 4C ). Comparing the γH2AX fluorescence intensity in FLAG-M transfected cells with non-transfected cells, we found a significant increase in the phosphorylation level of H2AX, suggesting that M involvement with PCNA may promote DNA damage ( Figure 4D ).

To verify if our findings concerning M protein and the DNA damage markers PCNA and γH2AX were also reproducible during infection, we evaluated the effect of SARS-CoV-2 infection in Vero E6 cells. Immunofluorescence analysis showed PCNA staining in the cytoplasm of infected cells, with PCNA presenting mainly a nuclear pattern on mock control ( Figure 5A ). The results ( Figures 5B, C ) show a higher intensity of PCNA in the cytoplasm of infected cells with a concomitant reduction in its nuclear signal compared to mock-infected cells. In addition, a higher γH2AX fluorescence intensity was observed in infected cells compared to non-infected cells, which confirms that the infection promotes DNA damage ( Figures 5D, E ). We also evaluated PCNA and γH2AX levels by western blotting in HEK293T and Vero E6 infected cells ( Figure 5F ). The effect observed during infection is a higher expression of PCNA and phosphorylation of histone H2AX in both cell lines ( Figures 5G, H ).

We then compared PCNA (top panels) and γH2AX (bottom panels) expression levels in lung sections obtained from control and COVID-19 patients through immunohistochemistry assay ( Supplementary Figure S4 ). Our results indicate high expression of both markers in COVID-19 patient (right panels), suggesting the involvement of both PCNA and γH2AX in SARS-CoV-2 infection in vivo.

Our data indicate that PCNA translocates to the cytoplasm after M protein expression or SARS-CoV2 infection. To analyze if this phenomenon plays a role in the replicative cycle of the virus, two inhibitors of this transport were used to prevent the PCNA migration to the cytoplasm and test their possible anti-viral activity. The inhibitor of PCNA (PCNA I1) prevents the transportation of the PCNA to the cytoplasm by stabilizing its trimer form in the nucleus. Verdinexor is a selective inhibitor of nuclear export protein exportin-1 (XPO1), also called Chromosome Region Maintenance 1(CRM1), and was already described as essential in the PCNA translocation (Bouayad et al., 2012; Perwitasari et al., 2016). Vero E6 cells were tested for PCNAi and Verdinexor toxicity, showing that 10 to 0.1 µM are safe concentrations for both drugs, since they did not reduce cells viability ( Figure S5 ). Vero E6 cells were then infected and submitted to a dose-dependent anti-viral in vitro efficacy assay. The results ( Figure 6 ) indicate that PCNAi 0.5 µM and Verdinexor 0.1 µM have the best effects against SARS-CoV-2, reducing the viral replication roughly by 20% and 15%, respectively, compared to the DMSO.