HEK 293T cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) with high glucose (Sigma D6429) supplemented with 5% heat-inactivated foetal bovine serum (iFBS) and 1x penicillin-streptomycin (P/S; Sigma P4333). VeroE6 cells were cultured in DMEM (Sigma D6429) supplemented with 10 mM HEPES, 1x MEM non-essential amino acid solution (Sigma M7145), 10% iFBS and 1x P/S.

Plasmodium falciparum blood stage cultures (3D7) were maintained in RPMI 1640 (Gibco, ref. 51800-035) supplemented with 25 mM HEPES, 2.1 g/mL NaHCO₃, 5% AlbuMAX II (Gibco), 0.37 mM hypoxhantine and 5% human erythrocytes. Cultures were incubated at 37°C under 1% CO₂, 3% O₂ and 96% N₂ gas mixture. Parasite development was monitored by Giemsa staining, and cultures were maintained as previously described (24).

Human whole blood of healthy donors was obtained from the Andalusian Public Health System Biobank. Blood was centrifuged at 2,000 rpm for 10 min, plasma and buffy coat were removed and the red blood cell pellet was washed three times in RPMI 1640 supplemented with 25 mM HEPES. Finally, the packed red blood cell pellet was resuspended with an equal volume of complete RPMI 1640 medium (cRPMI), this 50% haematocrit medium was stored at 4°C and used within three weeks for parasite culture and SARS-CoV-2 infection experiments.

pNL4-3.Luc.R-E- (NIH-AIDS reagent program, catalog n° 3418) is an HIV-1-luciferase reporter vector. It contains the luciferase gene flanked by the long-terminal regions (LTRs) for integration into the host genome, gag, pol, tat and rev genes. However, it is defective for nef, env and vpr genes, only allowing a single round of replication. To produce infectious pseudovirions, it needs to be co-transfected with a plasmid expressing an envelope protein: the vesicular stomatitis virus glycoprotein (VSV-G) envelope expressing plasmid (pMD2 VSV-G, Addgene, catalog n° 12259) as a positive control; or the pcDNA3.1 SARS-CoV-2 S prot-HA, a plasmid expressing the codon-optimized SARS-CoV-2 spike protein from isolate Wuhan-Hu-1 with the HA-tag. An empty pcDNA3.1(+) (Invitrogen) plasmid was used as a negative control.

Pseudovirus were produced by co-transfection of HEK 293T cells (plated in 6-well plates) with pNL4-3.Luc.R-E- (800 ng) and an envelope plasmid (800 ng): pMD2 VSV-G, pcDNA3.1 SARS-CoV-2 S prot-HA or pcDNA3.1(+) for no-envelope control. Transfection mixtures were made using 7.5 µL of TransIT^®-LT1 Transfection Reagent (Mirus) and 250 µL of Opti-MEM I media (Gibco). Mock transfection was performed without plasmids as an additional control. Medium was changed at 24h and pseudovirus-containing supernatants were harvested at 48 h post-transfection, centrifuged at 500 x g for 5 min to remove cellular debris, aliquoted and used immediately or stored at −80°C for later use.

Transient expression of the human receptors ACE2 or CD147 and the protease TMPRSS2 was performed by reverse transfection the plasmids pcDNA3.1-ACE2, pcDNA3.1-CD147 and pcDNA3.1-TMPRSS2 into HEK 293T cells. 24 hours prior to transfection, cells were plated in a T75 flask to be 70-80% confluent the following day. Next day, cells were trypsinized, counted and resuspended at a density of 1.6 x 10⁵ cells/mL. The transfection master mix was prepared with 5 µg of the plasmid of interest, 45 µL of TransIT^®-LT1 Transfection Reagent (Mirus) and 1.5 mL of Opti-MEM media (Gibco), the mixture was incubated 15 min and then 11.04 mL of cell suspension was added to the mix. Cells were plated in quadruplicate in a 96-well plate (100 µL/well) and incubated for 24h.

96-well plates seeded with HEK 293T transfected with the receptors or mock transfected (control cells) were infected with 100 µl of pseudovirus-containing supernatants, plates were incubated for 48h and analysed for luciferase activity using the Luciferase Assay System (E1500, Promega), following the manufacturer’s protocol. Cells were rinsed once in 1x PBS and lysed with 40 μL of lysis reagent per well, 20 μL of cell lysate was transferred to a white 96-well plate and 50 µL of Luciferase Assay reagent was added to each well, luminescence was measured in a TECAN infinite 200 instrument with 7.5 sec of readout time.

HEK 293T cells transfected with 2.5, 3.5 and 5 µg of pcDNA3.1-CD147 were lysed 24h and 48h post-transfection with 1x RIPA buffer (abcam) and 1x Complete EDTA free protease inhibitor cocktail (Roche). Cell lysates were quantified by DC protein Assay (Bio-Rad) following manufacturer’s instructions and then mixed with 4x Laemmli’s sample buffer (Bio-Rad) with 10% β-mercaptoethanol. Protein extracts were loaded and run on 10% SDS-PAGE gels and transferred onto a PVDF membrane (Bio-Rad). Membranes were blocked with 5% skimmed milk/TBST 1X (TBS1X-0.1% Tween 20) for 1 h and then incubated overnight at 4°C with primary antibodies; rabbit anti-CD147 (abcam ab108308; 1:1,000 dilution) and mouse anti-GAPDH (Affinity Biosciences, T0004; 1:10,000 dilution) in blocking solution. Following washing, membranes were incubated with the respective secondary antibodies for 1 h at room temperature; goat anti-rabbit IgG (IRDye^® 800CW) (ab216773) and goat anti-mouse IgG (IRDye^® 680RD) (ab216776) diluted 1:20,000. After washing, the fluorescence signal was visualized using an Odyssey CLx Imaging System (LI-COR Biosciences).

SARS-CoV-2/NL/2020 strain passage 3 was obtained from the European Virus Archive GLOBAL (EVA). Viral titre was determined as follows by plaque assay (protocol kindly shared by C. Suñé). Confluent monolayers of VeroE6 cells were grown on 6-well plates and incubated with 300 µL of a 10-fold serial dilution of the virus stock in DMEM supplemented with 0.2% BSA, 10 mM HEPES and 1x MEM, for 1 h at 37°C. After the adsorption hour, cells were overlaid with MEM 1x supplemented with 10% FSB, 2 mM Glutamine, 1x penicillin-streptomycin, plus 0.1% agarose and incubated for 72h at 37°C. Next, cells were fixed with 1% glutaraldehyde for 20 min. Medium was removed by aspiration and cells were washed with distilled water. Lysis plaques were stained with 0.1% crystal violet for 30 min at room temperature and washed with distilled water. Plaques were counted and quantified as plaque-forming units (PFU) per mL. To generate viral stocks, SARS-CoV-2 was propagated on VeroE6 cells in DMEM supplemented with 2% FSB, at MOI of 0.01. After 72h the virus-containing supernatant was collected, centrifuged for 10 min at 2.000 rpm, aliquoted and titrated as described above. Aliquots were stored at −80°C for later use.

Erythrocytes, stored in cRPMI medium at 50% haematocrit (see above), were quantified in a Neubauer chamber. 2 x 10⁶ erythrocytes were incubated in protein low bind tubes with SARS-CoV-2 viral supernatant at a MOI 2, for 1h at 37°C with gentle agitation every 15 min to maintain the erythrocytes in suspension. After centrifugation at 1,500 rpm for 3 min, the supernatant was discarded and the erythrocytes were washed four times with 0.5 mL of phenol free RPMI to remove unbound virus. Erythrocytes were then resuspended at 5 x 10⁴ cells/µL and 500,000 cells were seeded on microscope round cover glasses of 12 mm (VWR) previously treated with 50 uL Concanavalin A (Sigma, 5 mg/mL in water) for 20 min at 37°C in 24-well plates. Erythrocytes were allowed to settle for 20 min at 37°C and unbound cells were rinsed with two PBS washes. Seeded cells were fixed with 4% paraformaldehyde (PFA)/0.0075 glutaraldehyde (GA) in PBS for 20 min at 37°C and analysed by indirect immunofluorescence microscopy.

Erythrocytes from a P. falciparum culture were quantified in a Neubauer chamber and the parasitemia was determined by Giemsa-stained blood smears. 2 x 10⁶ erythrocytes from a P. falciparum culture at 5.5% parasitaemia were incubated in protein low bind tubes with SARS-CoV-2 at a MOI 2 for 1h at 37°C with gentle agitation every 15 min to maintain cells in suspension. After incubation, erythrocytes were washed, seeded, and fixed on round cover glasses, as described before. Indirect immunofluorescence was performed using an antibody against the SARS-CoV-2 N protein and DAPI staining for detection of the parasite DNA.

P. falciparum culture at 7% parasitaemia was incubated with SARS-CoV-2 at a MOI 2 for 1h at 37°C in a T-25 flask. A control culture was incubated with DMEM, 2% FSB. After incubation, erythrocytes were centrifuged at 1,500 rpm for 5 min, the red blood cell pellet was resuspended in 500 µL of cRPMI and transferred to a protein low-binding Eppendorf. After centrifugation, erythrocytes were resuspended in 50 µL of cRPMI and diluted with 50 µL of cRPMI medium, 5% haematocrit (to dilute the culture below 3.5% parasitemia), the cell suspension was transferred to a well in a 96-well plate and incubated for 24 and 48 hours at 37°C. Culture development was monitored by Giemsa-stained blood smears and quantified by microscopic examination.

Coverslips with fixed erythrocytes were rinsed with PBS twice and permeabilized with 0.1% Triton^® X-100 in PBS for 15 min, after three PBS washes, free aldehyde groups were quenched with 30mM glycine for 5 min followed by PBS washes. Then, cells were blocked with 3% BSA/1x PBS for 45 min. The primary antibody, mouse anti-SARS-CoV-2 N protein (abcam, ab281300), was diluted 1:1,000 in antibody solution (1% BSA/0.5% Tween^® 20/1x PBS) and centrifuged at 13,300 rpm for 1 min to remove aggregates prior to incubation for 2h at room temperature on a horizontal shaker. After three washes with PBS for 10 min, the secondary antibody goat anti-mouse Alexa Fluor 594 (Invitrogen, A11020), was diluted 1:500 and incubated for 45 min. After three washes with PBS for 10 min, the coverslips were mounted on glass slides using VECTASHIELD^® Vibrance™ mounting medium (Vector laboratories). Images were acquired using a fluorescence microscope (DMi8, Leica, Germany), with 100x objective and the LAS X software (Leica). Erythrocytes were visualized by phase contrast (gray). Stacks (0.19-μm z step) acquisition was performed to capture 3D images of the red blood cells. Maximum intensity projections of anti-N and DAPI channels were performed using ImageJ 1.52p software (National Institutes of Health). Quantification of SARS-CoV-2 viral particles bound to the erythrocytes after four washes was performed in maximum intensity projection images. The percentage of erythrocytes with viral particles was calculated counting 2,490 red blood cells and 3,418 red blood cells in the case of the P. falciparum cultures.

All work performed was approved by the local genetic manipulation (GM) safety committee. All work with live SARS-CoV-2 was performed in biosafety laboratory level 3 facilities at the IPBLN, Granada, Spain.

Due to contradictory evidence of the function of CD147 as a truly SARS-CoV-2 receptor (9–11), we first investigated whether the transient overexpression of CD147 in HEK 293T cells could mediate the internalization of SARS-CoV-2 pseudovirions. In order to do this, luciferase reporter pseudovirions based in lentiviral vectors were constructed (25) where the cellular entry will be mediated by the expression of the SARS-CoV-2 spike (S) protein. In the same system, the vesicular stomatitis virus glycoprotein (VSV-G) was used as a positive control of infection due to its broad cell-type tropism. In addition, particles without the S protein were generated as a negative control. Infection was measured by a luciferase activity assay at 48 hours post-infection (h p.i.) in HEK 293T cells expressing the ACE2 or the CD147 receptors. The amount of luciferase activity reflects the expression of integrated proviruses (26). We found that only cells expressing the ACE2 receptor and the TMPRSS2 protease were permissive for the entry of SARS-CoV-2 pseudovirions ( Figure 1 ; Supplementary Figure S1 ). As a positive control, assays were performed with the same amount of VSG-G pseudoviral supernatant, this showed higher luciferase activity in all the assays. In addition, pseudoparticles without an envelope protein were producing a similar level of luciferase activity as the mock supernatant without pseudoparticles.

As SARS-CoV-2 pseudoparticles were not internalized in the CD147-transfected cells, we decided to test whether CD147 was properly expressed. The expression of CD147 in HEK 293T was confirmed by Western blot analysis ( Figure 2 ). As observed in Figure 2A , when CD147 was transiently overexpressed, most of the protein was in the low glycosylated form (LG-CD147, ~32 kDa), and the amount increased with time. We also analysed the endogenous expression of CD147 in HEK 293T and we found that this cell line expressed low but detectable levels of LG-CD147 and high glycosylated CD147 (HG-CD147, ~42 kDa) ( Figure 2B ). Despite the expression of the two CD147 isoforms, we found not significant luciferase activity for SARS-CoV-2 pseudovirions when using non-transfected HEK 293T cells ( Figure 1 ). In addition, the transient expression of ACE2 in HEK-293T was also analysed by immunofluorescence and flow cytometry, we found heterogeneous levels of expression in 10% of the transfected population ( Supplementary Figure S1 ). Although the percentage of ACE2 expression was not high, it was enough to detect a 24.6-fold increase of SARS-CoV-2 pseudovirus entry in HEK 293T cells previously transfected with ACE2 and TMPRSS2 compared to the negative control (No env) in luciferase assays ( Figure 1 ).

Since CD147 alone does not appear to be a functional receptor for SARS-CoV-2 infection, we explored whether SARS-CoV-2 was able to adhere and/or enter to human erythrocytes via an alternative mechanism. Infection assays were performed with the SARS-CoV-2/NL/2020 strain. Erythrocytes were incubated for 1 h with SARS-CoV-2 at a multiplicity of infection (MOI) of 2. After four washes with phenol red-free RPMI to remove unbound virus, erythrocytes were seeded on microscope coverslips, fixed and analysed by indirect immunofluorescence microscopy (IF) using an antibody against the SARS-CoV-2 nucleocapsid (N) protein. Adhesion and entry of SARS-CoV-2 to red blood cells could be observed ( Figure 3 ) but to a lower rate in comparison with VeroE6 infected under the same conditions ( Supplementary Figure S2 ). After four washes, 10.94% of the erythrocytes (standard deviation, SD 1.56%) had viral particles attached. However, virions were not accumulated inside the erythrocytes or the plasma membrane as observed in VeroE6 cells ( Supplementary Figure S2 ).

As P. falciparum remodels the erythrocyte membrane (22, 23), this could render malaria-infected red blood cells (iRBCs) more vulnerable to the adhesion and entry of SARS-CoV-2, potentially, promoting co-infection. To test this hypothesis, a P. falciparum culture at 5.5% parasitaemia was incubated for 1 h with SARS-CoV-2 at MOI:2. After incubation, erythrocytes were washed with phenol red-free RPMI, seeded on coverslips, fixed and analysed by IF using a SARS-CoV-2 N antibody for SARS-CoV-2 detection, and DAPI staining for detection of the parasite DNA. As observed in Figure 4 , after four washes, few viral particles were detected (1 to 5 particles per cell) in a low percentage of the parasitized red blood cells (9.13%; SD 1.82%). In addition, the potential co-infection of SARS-CoV-2 and P. falciparum in red blood cells was mainly observed during the ring stage of the parasite. These results suggest that the presence of P. falciparum inside erythrocytes does not facilitate or increase the SARS-CoV-2 entry to these cells.

We also wanted to test whether the presence of SARS-CoV-2 in malaria-infected erythrocytes could affect the parasite infection process, especially invasion and growth. To test this, we incubated a P. falciparum culture at high parasitaemia with SARS-CoV-2 for 1 h at MOI:2. After incubation, culture cells were washed once with RPMI and further incubated for 24 and 48 h at 37°C in a 96-well plate. After incubation, culture development was monitored by Giemsa-stained blood smears that were quantified by microscopic examination. We found no significant effect in P. falciparum parasitaemia and cell stages between SARS-CoV-2 pre-incubated cultures and control cultures incubated with infection medium without the virus ( Supplementary Table S1 ).