Acquisition of a multibasic cleavage site does not increase MERS-CoV entry into Calu-3 human lung cells

ABSTRACT

Human-to-human transmission of the highly pathogenic Middle East respiratory syndrome coronavirus (MERS-CoV) is currently inefficient. However, there is concern that the virus might mutate and thereby increase its transmissibility and thus pandemic potential. The pandemic SARS-CoV-2 depends on a highly cleavable furin motif at the S1/S2 site of the viral spike (S) protein for efficient lung cell entry, transmission, and pathogenicity. Here, by employing pseudotyped particles, we investigated whether augmented cleavage at the S1/S2 site also increases MERS-CoV entry into Calu-3 human lung cells. We report that polymorphism T746K at the S1/S2 cleavage site or optimization of the furin motif increases S protein cleavage but not lung cell entry. These findings suggest that, unlike what has been reported for SARS-CoV-2, a highly cleavable S1/S2 site might not augment MERS-CoV infectivity for human lung cells.

IMPORTANCE

The highly cleavable furin motif in the spike protein is required for robust lung cell entry, transmission, and pathogenicity of SARS-CoV-2. In contrast, it is unknown whether optimization of the furin motif in the spike protein of the pre-pandemic MERS-CoV increases lung cell entry and allows for robust human–human transmission. The present study indicates that this might not be the case. Thus, neither a naturally occurring polymorphism that increased MERS-CoV spike protein cleavage nor artificial optimization of the cleavage site allowed for increased spike-protein-driven entry into Calu-3 human lung cells.

INTRODUCTION

The Middle East respiratory syndrome coronavirus (MERS-CoV) is an emerging betacoronavirus (1). The virus is endemic to the Middle East but has been introduced into a total of 27 countries, mainly via infected travelers (2). MERS-CoV naturally infects dromedary camels and can induce common cold-like symptoms in these animals (3–6). Contact with infected camels or consumption of contaminated camel products can result in zoonotic transmission of the virus to humans, and afflicted patients frequently develop fatal disease (1). As of November 2022, 2,600 laboratory-confirmed infections have been reported, of which 48% presented with severe disease or died (2). At present, human-to-human transmission of MERS-CoV is inefficient, with a reproduction number (R0) below 1, and has mainly been observed in hospitals (2, 7–9). However, there is concern that the virus might acquire mutations that augment inter-human transmissibility and increase the risk of pandemic spread. Therefore, surveillance of circulating MERS-CoV for the emergence of novel viral variants is an important task.

The viral spike (S) protein facilitates MERS-CoV entry into target cells. This process depends on S protein interactions with the cellular receptor dipeptidyl peptidase 4 (DPP4) (10) and cleavage of the S protein by cellular proteases. First, the S protein is cleaved by the proprotein convertase furin at the S1/S2 site in infected cells (11–13). Pre-cleavage at the S1/S2 site increases susceptibility to a second cleavage at the S2′ site (14, 15). This second cleavage event occurs during viral entry into target cells, is carried out by the transmembrane serine protease TMPRSS2, and is essential for infectious viral entry into lung cells (16–19). SARS-CoV-2 employs a similar mechanism for S protein activation (20). However, unlike MERS-CoV spike protein (MERS-S), which contains a minimal furin recognition motif at the S1/S2 site (RSVR), the S1/S2 site of the SARS-CoV-2 S protein (SARS-2-S) harbors an additional arginine residue (RRAR), which is essential for efficient furin cleavage, lung cell entry, transmission, and pathogenesis (21–23).

Here, we investigated whether naturally occurring polymorphisms at the S1/S2 site of MERS-S or artificial optimization of the minimal furin recognition motif impact S protein cleavage and lung cell entry.

RESULTS

The naturally occurring polymorphism T746K increases MERS-S cleavage

We screened a public database (https://www.ncbi.nlm.nih.gov/genome/viruses/variation/) for patient-derived MERS-S sequences with polymorphisms in the vicinity of the S1/S2 motif. We identified three S protein amino acid sequences that harbored a polymorphism, T746K, located two positions upstream of the S1/S2 cleavage motif (Fig. 1A; Table 1). We investigated the impact of this polymorphism on S protein cleavage and host cell entry since it is known that basic residues within or adjacent to furin cleavage sites can increase cleavage efficiency (24). For this, the mutation was introduced into the S protein of the reference virus MERS-CoV EMC. Subsequently, S protein expression and function were analyzed using rhabdoviral particles pseudotyped with S protein (VSV_pp), which faithfully model key aspects of MERS-CoV entry into cells (12, 16).

Fig 1.

Polymorphism T746K enhances S protein cleavage. (A) Schematic illustration of the MERS-CoV S protein and mutations studied. SP, signal peptide; RBD, receptor-binding domain; FP, fusion peptide; HR, heptad repeat; TD, transmembrane domain; S1/S2 and S2′, cleavage sites. (B) T746K enhances MERS-S cleavage. Whole-cell lysates (WCL) of cells expressing the indicated S proteins (or no S protein, control; top left panel) or VSV_pp bearing the indicated S proteins (or no S protein, control; top right panel) were subjected to SDS-PAGE, and S protein cleavage was analyzed using immunoblot (via C-terminal V5-epitope tag). Detection of beta-actin (ACTB) or VSV-M protein served as loading control. The immunoblot data are representative of five (WCL) or six (VSV_pp) independent experiments. For quantification of S protein expression (bottom left panel), total S protein signals (S0 +S2) were normalized against the corresponding ACTB (WCL) or VSV-M (VSV_pp) signals and subsequently compared to EMC (WT) S (set as 1). For quantification of S protein cleavage, total S protein signals (S0 +S2) were set as 100% and the proportion of S2 signals calculated (bottom right panel). Graphs show mean ± SEM of five (WCL) or six (VSV_pp) independent experiments. Statistical significance was assessed by two-tailed, unpaired Student’s t-test with Welch correction (P > 0.05, not significant [ns]; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001). (C) Peptides covering the MERS-S S1/S2 cleavage site with or without mutation T746K (Suc-TL[T/K]PRSVR-AMC), or a furin consensus sequence (Pyr-RTKR-AMC) were incubated with recombinant furin and the mean fluorescence intensity (MFI; excitation wave length: 355 nm; emission wave length: 460 nm) was recorded every 60 s for a total of 1 h, and background signals (addition of water instead of peptide) were subtracted. Graphs show average (mean) data ± SEM of four independent experiments, conducted with technical triplicates.

TABLE 1.

: Data for human patients infected with MERS-CoV variants harboring T746K in their S protein

Sex	Age (years)	Disease	Route of infection	Virus
Male	40	Severe pneumonia	Probably health care acquired	KSA-3929a
Female	64	Pneumonia	Community acquired	KSA-4067b
Female	48	Mild	Household contact	KSA-7179c

^a KSA-3929, GenBank: KT805976.1.

^b KSA-4067, GenBank: KT805975.1.

^c KSA-7179, GenBank: KT806010.1.

Immunoblot analyses of transfected cells expressing S protein and concentrated S protein-bearing particles showed that polymorphism T746K increased S protein cleavage at the S1/S2 site without interfering with S protein expression in cells and S protein incorporation into VSV particles (Fig. 1B, upper panel). This increase in cleavage efficiency was statistically significant (Fig. 1B, lower panel) and in agreement with the increased cleavability predicted by the software ProP 1.0 (25) (Table 2). In keeping with these findings, polymorphism T746K increased furin-mediated cleavage of a peptide mimicking the S1/S2 cleavage site and surrounding sequences (Fig. 1C). The S protein KSA-3929, in which the T746K polymorphism was identified, additionally harbors exchanges E32A and Q1020R (Fig. 1A). However, these exchanges did not further increase S protein cleavage relative to T746K alone (Fig. 1B). Thus, we have identified a naturally occurring polymorphism, T746K, which augments cleavage of the MERS-S protein.

TABLE 2.

Predicted efficiency of spike protein cleavage by proprotein convertases^a

Spike protein	S1/S2 loop sequence	Score
EMC (WT)	LFVEDCKLPLGQSLCALPDTPSTLTPRSVR↓SVPGEMRLASIAFNHPIQVDQL	0.780
EMC (T746K)	LFVEDCKLPLGQSLCALPDTPSTLKPRSVR↓SVPGEMRLASIAFNHPIQVDQL	0.863
KSA-3929	LFVEDCKLPLGQSLCALPDTPSTLKPRSVR↓SVPGEMRLASIAFNHPIQVDQL	0.863
EMC (SSVR)	LFVEDCKLPLGQSLCALPDTPSTLTPSSVR↓SVPGEMRLASIAFNHPIQVDQL	0.388
EMC (S1/S2opt)	LFVEDCKLPLGQSLCALPDTPSTLTPRRKR↓SVPGEMRLASIAFNHPIQVDQL	0.942
EMC (S1/S2SARS2)	LFVEDCKLPIGAGICASYQTQTNSPRRAR↓SVASQSIIAYNHPIQVDQL	0.911
EMC (S1/S2H5)	LFVEDCKLPKYVKSNKLVLATGLRNSPQRERRRKR↓GLFGAIAFNHPIQVDQL	0.881

^a Cleavage scores were predicted by ProP 1.0; Legend: ↓, cleavage site; bold and underlined, core cleavage motif; bold and italics, sequence variation compared to EMC (WT).

T746K does not augment spike protein-driven cell–cell fusion

We next asked whether T746K augments the ability of MERS-S to fuse cells. For analysis of S protein-driven cell–cell fusion, effector 293T cells co-expressing S protein and the β-galactosidase α-fragment were mixed with target 293T cells expressing DDP4 and the Ω-fragment of β-galactosidase followed by measurement of β-galactosidase activity in cell lysates. As a control, we mutated the furin cleavage site in MERS-S from RSVR to SSVR (Fig. 2A) since we and others previously found that this mutation largely abrogates S protein cleavage (14, 26). Indeed, immunoblot analyses revealed that mutant SSVR was not appreciably cleaved (Fig. 2B), in agreement with a low cleavage score calculated by ProP 1.0 (Table 2), and analysis of cell–cell fusion showed that this mutant was largely inactive (Fig. 2C and D). In contrast, EMC (T746K) and KSA-3929 S proteins robustly drove cell–cell fusion, but fusion efficiency was similar to that measured for EMC S protein (Fig. 2C and D). Thus, polymorphism T746K increases MERS-S cleavage but not MERS-S-driven cell–cell fusion.

Fig 2.

Mutation T746K does not augment cell–cell fusion. (A) A MERS-S mutant, in which the S1/S2 site was changed from RSVR to SSVR was included as control in the cell–cell fusion assay, since this mutant is known to exhibit reduced cell–cell fusion. (B) VSV particles bearing no spike, EMC WT, or mutant SSVR were concentrated, and expression of S protein and VSV-M in cell lysates was analyzed using immunoblot. Similar results were obtained in two separate experiments. (C) For analysis of S protein-driven cell-to-cell fusion, effector 293T cells transfected with plasmids encoding the indicated S proteins and β-galactosidase α-fragment were mixed with target 293T cells transfected to express DPP4 and the Ω-fragment of β-galactosidase. β-galactosidase activity in cell lysates was measured at the indicated time points post cell mixing and normalized against the assay background (effector cells expressing no S protein). The mean of three independent experiments with technical quadruplicates is shown; error bars indicate SEM. (D) Cell–cell fusion activity at 24 h post cell mixing is shown. Results obtained for MERS-S WT were set as 1. Statistical significance in panels (C) and (D) was assessed by two-way analysis of variance with Tukey’s multiple comparison test and two-tailed, unpaired Student’s t-test with Welch correction, respectively (P > 0.05, not significant [ns]; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001).

T746K does not augment spike protein-driven entry into Calu-3 human lung cells

To determine whether T746K modulates S protein-driven cell entry, VSV_pp bearing MERS-S were used. Mutation T746K slightly reduced MERS-S-driven entry into Vero (African green monkey kidney) and Calu-3 cells (human lung) and had no significant impact on entry into Huh-7 (human hepatoma) and Caco-2 (human colon) cells (Fig. 3). Furthermore, combination of T746K with exchanges E32A and Q1020R (all found in KSA-3929) did not alter the efficiency of entry into Vero, Huh-7, and Caco-2 cells but moderately augmented entry into Calu-3 cells (Fig. 3). In sum, polymorphism T746K alone is not compatible with robust entry into Calu-3 lung cells, while T746K jointly with E32A and Q1020R increases entry into lung cells.

Fig 3.

Mutation T746K does not augment S protein-driven entry into Calu-3 lung cells. The indicated cell lines were inoculated with VSV_pp bearing the indicated S proteins. S protein-driven cell entry was analyzed by quantification of virus-encoded luciferase activity in cell lysates. For normalization, cell entry of particles bearing EMC (WT) S was set as 1. Mean ± SEM of three independent experiments conducted with four technical replicates is shown. Statistical significance was assessed by two-tailed, unpaired Student’s t-test with Welch correction, respectively (P > 0.05, not significant [ns]; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001).

T746K reduces DPP4 binding

To determine why T746K was not compatible with robust cell entry, we investigated whether the mutation interferes with binding to DPP4, the cellular receptor used by MERS-CoV for host cell entry (10). For this, we studied binding of a fusion protein consisting of the DPP4 extracellular domain and the Fc portion of immunoglobulin G to 293T cells expressing MERS-S. A MERS-S variant harboring mutation I529T (Fig. 4A) was included as control since this mutation reduces DPP4 binding but not S protein expression (27). As expected, mutation I529T reduced S protein binding to DPP4. Furthermore, we discovered that also T746K diminished DPP4 binding, although not to the same level as I529T (Fig. 4B). Finally, combination of T746K with exchanges E32A and Q1020R (all found in KSA-3929) resulted in slightly increased DPP4 binding compared to T746K alone (Fig. 4B), although this effect was not statistically significant. These results suggest that, although T746K increases S protein proteolytic processing, it also interferes with S protein binding to the DPP4 receptor.

Fig 4.

Mutation T746K reduces DPP4 binding. (A) Spike scheme and location of mutation I529T. A spike protein harboring mutation I529T, which is known to reduce DPP4 binding, was included as control in the DPP4 binding assay. (B) To measure S protein binding to DPP4, 293T cells were transfected with expression plasmids for the indicated S proteins or empty vector and incubated with DPP4-Fc fusion protein and labeled antibody and subjected to flow cytometric analysis. Left panel: Mean fluorescence intensities were normalized to cells expressing no S protein. The average of three independent experiments conducted with unicates is shown; error bars indicate SEM. Right panel: Data were normalized against binding of MERS-S WT to DPP4, which was set as 1. Error bars indicate the SEM. Statistical significance was assessed by two-tailed, unpaired Student’s t-test with Welch correction (P > 0.05, not significant [ns]; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001).

Optimization of the MERS-S furin motif does not increase S protein-driven entry into Calu-3 human lung cells

Finally, we investigated whether increasing MERS-S cleavage by mutations other than T746K might augment S protein-driven entry into lung cells. For this, we converted the minimal furin cleavage motif at the S1/S2 site into a full furin consensus motif (RRKR, mutant S1/S2_OPT) or replaced the complete S1/S2 cleavage loop by the corresponding sequences found in the H5N1 influenza A virus hemagglutinin (mutant S1/S2_H5) or SARS-CoV-2 S protein (S1/S2_SARS2), respectively (Fig. 5A). These changes were predicted by ProP 1.0 to increase S protein cleavage efficiency compared to wild-type MERS-S (Table 2). Mutant S1/S2_OPT was robustly expressed and incorporated into particles, and exhibited augmented cleavage, with only cleaved S protein being detected in particle preparations (Fig. 5B and C). In contrast, mutants S1/S2_SARS2 and S1/S2_H5 were not cleaved and not incorporated into particles despite appreciable expression in cells, suggesting that the heterologous cleavage loops might have interfered with proper folding or intracellular trafficking of the MERS-S protein (Fig. 5B and C). Further, mutant S1/S2_OPT, but not S1/S2_SARS2 and S1/S2_H5, bound to DPP4 with similar efficiency compared to EMC S protein (Fig. 5D). Finally, mutant S1/S2_OPT mediated reduced Calu-3 lung cell entry compared to EMC S protein, while mutants S1/S2_SARS2 and S1/S2_H5 were unable to mediate entry (Fig. 5E), consistent with lack of particle incorporation. Thus, increasing cleavability of MERS-S may not increase MERS-CoV entry into lung cells.

Fig 5.

Optimization of the S1/S2 site increases S protein cleavage but not lung cell entry. (A) Spike scheme and location of mutations. The minimal furin motif at the S1/S2 site in MERS-S was either optimized (mutant S1/S2_OPT), or the complete S1/S2 loop was exchanged against the corresponding sequences found in the hemagglutinin of H5N1 influenza A virus (mutant S1/S2_H5) or the S protein of SARS-CoV-2 (mutant S1/S2_SARS2). (B) Optimization of the furin motif at the S1/S2 site of MERS-S increased S protein cleavage. Whole-cell lysates (WCL) of cells expressing the indicated S proteins (or no S protein, control; left panel), or vesicular stomatitis virus pseudovirus particles (VSV_pp) bearing the indicated S proteins (or no S protein, control; right panel) were subjected to SDS-PAGE. and S protein cleavage was analyzed using immunoblot (via C-terminal V5-epitope tag). Detection of beta-actin (ACTB) or VSV-M protein served as loading control. Representative immunoblot data of three independent experiments are shown. Bands corresponding to uncleaved (S0) and cleaved (S2) MERS-S are indicated. (C) For quantification of S protein expression (left panel), total S protein signals (S0 +S2) were normalized against the corresponding ACTB (WCL) or VSV-M (VSV_pp) signals and subsequently compared against EMC (WT) S (set as 1). For quantification of S protein cleavage, total S protein signals (S0 +S2) were set as 100% and the proportion of S2 signals calculated (right panel). Graphs show the mean of three independent experiments. Error bars indicate SEM. (D) S protein binding to DPP4 was analyzed as described for panel B of Fig. 4. Left panel: The average of four independent experiments conducted with unicates is shown, error bars indicate SEM. Right panel: Data were normalized against binding of MERS-S WT to DPP4, which was set as 1. Error bars indicate the SEM. (E) The indicated cell lines were inoculated with VSV_pp bearing the indicated S proteins. S protein-driven cell entry was analyzed by quantification of virus-encoded luciferase activity in cell lysates. For normalization, cell entry of particles bearing EMC (WT) S was set as 1. Mean ± SEM of three independent experiments conducted with four technical replicates are shown. Statistical significance in panels (C)–(E) was assessed by two-tailed, unpaired Student’s t-test with Welch correction (P > 0.05, not significant [ns]; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001).

DISCUSSION

Since the identification of MERS-CoV in a patient with a novel lung disease more than 10 years ago (28), MERS cases are continually detected (29), due to the ongoing transmission of the virus from dromedary camels to humans. Increased replicative fitness of MERS-CoV of the recombinant lineage 5 has been reported (30), and the adaptation of the virus to efficient human–human transmission is a concern. The cleavage of the viral S protein by host cell proteases is a determinant of coronavirus zoonotic potential (31) as well as transmissibility and pathogenicity (21, 22). The results of the present study suggest that the MERS-S S1/S2 site is cleaved by furin to an extent that is sufficient for robust entry into human lung cells since polymorphism T746K or optimization of the furin motif increased S protein cleavage but not Calu-3 lung cell entry.

Increasing cleavage efficiency may be disadvantageous for the virus because it can come at the cost of reduced DPP4 binding, as demonstrated for mutant T746K, although we cannot exclude that increased dissociation of the S1 from the S2 subunit contributed to the diminished DPP4 binding of this mutant. Mutant S1/S2_OPT also exhibited reduced DPP4 binding, but this effect was minor and not statistically significant, suggesting the factors other than diminished DPP4 binding might contribute to the reduced efficiency of cell entry mediated by the S1/S2_OPT mutant. The reduced DPP4 binding associated with T746K was at least partially reversed by mutations E32A and Q1020R, and it is noteworthy that the S protein harboring all three mutations showed augmented cleavage and facilitated Calu-3 lung cell entry with higher efficiency than the wild-type S protein. Thus, it is conceivable that compensatory mutations in the S protein might allow the virus to gain an advantage from mutations increasing S protein cleavage.

Our results require confirmation with authentic MERS-CoV, primary lung cells, and animal models, and it needs to be taken into account that emergence of viruses with the T746K exchange was only observed in three patients. Nevertheless, our data suggest that augmenting S1/S2 cleavage by mutations within or adjacent to the S1/S2 site might not allow for increased MERS-CoV spread in and between individuals, at least not in the absence of additional mutations.

MATERIALS AND METHODS

Plasmids

Expression plasmids pQCXIP-DsRed-hDPP4, pCAGGS-MERS-S EMC (WT), pCAGGS-MERS-S EMC (I529T), pCAGGS-MERS-S EMC (SSVR), pQCXIP-beta-galactosidase alpha fragment, and pQCXIP-beta-galactosidase omega fragment have been described previously (14, 16, 27, 32). Expression plasmids for MERS-S EMC (T746K) and MERS-S KSA-3929 (based on Hu/Riyadh-KSA-3929/2015, GenBank: KT805976.1, harboring S protein mutations E32A, T746K and Q1020R) were generated by overlap-extension PCR with primers harboring the desired mutations and plasmid MERS-S EMC (WT) as template. The resulting PCR products were cloned into the pCAGGS plasmid (33), making use of KpnI and XhoI restriction sites. The same strategy was employed to generate a mutant MERS-S harboring an optimized furin consensus sequence at the S1/S2 cleavage site (mutations S749R and V750K), as well as MERS-S mutants in which the wild-type S1/S2 cleavage loop (730-PLGQSLCALPDTPSTLTPRSVRSVPGEMRLASIAF-763) was replaced by that of the S protein of SARS-CoV-2 (hCoV-19/Wuhan/WH01/2019, GenBank: QIU81765.1, 665-PIGAGICASYQTQTNSPRRARSVASQSIIAY-695) or the hemagglutinin of an avian influenza virus of the subtype H5N1 (319-PKYVKSNKLVLATGLRNSPQRERRRKRGLFGAIA-352, A/duck/Guangxi/3579/2005(H5N1), GenBank: ABJ96784.1), which were all cloned into the pCG1 plasmid (kindly provided by Roberto Cattaneo) using BamHI and SalI restriction sites. Of note, we used untagged MERS-S for pseudovirus entry studies and cell–cell fusion assays, while S proteins with a C-terminal V5-epitope tag were used for experiments addressing S protein cleavage and particle incorporation. The integrity of all PCR-amplified sequences was verified by automated sequence analysis (Microsynth Seqlab).

Cell culture

HEK293T (kidney, human), Vero (kidney, African green monkey) and Huh-7 (liver, human) cells were cultivated in Dulbecco’s modified Eagle’s medium (DMEM; PAN Biotech). Caco-2 cells (colon, human) were grown in minimum essential medium (Thermo Fisher Scientific). Calu-3 cells (lung, human) were grown in DMEM/F-12 medium (Thermo Fisher Scientific). All media were supplemented with 10% fetal bovine serum (FBS, PAN Biotech) and penicillin/streptomycin solution (PAN Biotech). Culture media for Caco-2 and Calu-3 cells further received non-essential amino acids and sodium pyruvate (both Thermo Fisher Scientific). All cells were incubated in a humidified atmosphere at 37°C and 5% CO₂. Transfection of HEK293T cells was performed by calcium-phosphate precipitation.

Prediction of S protein cleavage efficiency

For prediction of the cleavage efficiency of MERS-S by proprotein convertases, the respective S1/S2 loop sequences of wild-type and mutant S proteins were analyzed using the ProP 1.0 online tool (25) (available at: https://services.healthtech.dtu.dk/services/ProP-1.0/), and the resulting cleavage scores are shown in Table 2.

Antibodies

The following antibodies were used as primary antibodies: Anti V5 (Invitrogen), anti β-actin (Sigma-Aldrich), and anti VSV-M (Kerafast). As secondary antibodies, an anti-mouse HRP (horseradish peroxidase, Dianova) and an anti-human AlexaFlour488 (Thermo Fisher Scientific)-conjugated antibody was employed.

Production of VSV pseudoparticles (VSVpp) and transduction of target cells

We employed a previously described protocol for VSV_pp production (14, 27). In brief, HEK293T cells expressing either MERS-S, VSV-G, (positive control) or no viral glycoprotein (empty vector, negative control) following transfection were inoculated with VSV*ΔG-fLuc (34) for 1 h, washed, and further maintained in the presence of neutralizing anti VSV-G antibody I1. At 18 h post inoculation, supernatants containing VSV_pp were collected, cleared from cellular debris by centrifugation, and stored at −80°C until further use. Target cells for transduction experiments were grown in 96-well plates. For transduction, cells were inoculated with the respective VSV_pp and incubated for 18 h before the activity of the virus-encoded firefly luciferase (fLuc) was measured in cell lysates using a commercial kit (PJK) and a plate luminometer (Hidex).

MERS-S expression and incorporation into VSV_pp

HEK293T cells transfected to express MERS-S variants with a C terminal V5-tag were used for the preparation of whole-cell lysates (WCL) or VSV_pp. To investigate incorporation of S proteins into VSV_pp, VSV_pp were concentrated from culture supernatants by high speed centrifugation (25,000 × g, 120 min, 4°C) through a sucrose cushion and lysed. Samples were analyzed by SDS-PAGE under reducing conditions, proteins were transferred onto nitrocellulose membranes (Hartenstein), the membranes blocked with 5% skim milk or bovine serum albumin (BSA), and incubated with primary and secondary antibodies. Western blots were developed using a self-made chemiluminescence reagent in combination with the ChemoCam imaging system and the ChemoStar Professional software (Intas Science Imaging Instruments). Signal intensities were quantified using the program ImageJ (FIJI distribution) (35). To determine particle incorporation of S proteins, the combined signals for uncleaved S protein (S0) and cleaved S protein (S2) were normalized against the VSV-M signal. To determine cleavage efficiency, the combined signal of S0 and S2 was set as 100%, and the relative S0 and S2 signals were calculated.

MERS-S/DPP4 interaction

DPP4 binding to S proteins was determined as previously reported (27). In brief, HEK293T cells expressing the respective S protein were washed and incubated for 1 h in 1% BSA/phosphate-buffered saline (PBS) containing soluble DPP4 equipped with a C-terminal human Fc-tag (solDPP4-Fc, 1:200; ACRO Biosystems). Thereafter, the cells were washed and incubated with AlexaFlour488-conjugated anti-human antibody, again for 1 h. Finally, cells were washed, fixed with 4% paraformaldehyde solution (AppliChem), and analyzed using flow cytometry. Analysis of differences in DPP4 binding by MERS-S EMC WT, MERS-S EMC T746K, and MERS-S KSA-3929 was performed using the LSR II Flow Cytometer and the FACS Diva software (both BD Biosciences). The software FCS Express 4 Flow research (De Novo software) was used for data analysis. In addition, the ID7000 Spectral Cell Analyzer, along with the ID7000 software (Sony Biotechnology, San Jose, CA, USA), was used to analyze differences in DPP4 binding by MERS-S EMC WT, MERS-S EMC S1/S2_OPT, MERS-S EMC S1/S2_SARS2, and MERS-S EMC S1/S2_H5. To determine the unspecific assay background signal (which was later subtracted from all samples), we used MERS-S EMC-expressing cells that were only incubated with AlexaFlour488-conjugated anti-human antibody.

Furin-mediated cleavage of AMC substrate

Recombinant human furin (0.2 µg, R&D Systems) was incubated with 1 nmol of Pyr-RTKR- 7-amino-4-methylcoumarin (AMC), Suc-TLKPRSVR-AMC, or Suc-TLTPRSVR-AMC (customized). Furin-mediated cleavage was determined every minute for 60 min using a Cytation 3 Imaging Reader (355-nm excitation and 460-nm emission). Background was subtracted.

MERS-S-driven cell–cell fusion

MERS-S-driven cell–cell fusion was analyzed employing a modified version of a previously described quantitative fusion assay (32), which is based on complementation of a split β-galactosidase. In brief, 293T effector cells were cotransfected with expression plasmids for MERS-S protein (or empty vector, control) and the β-galactosidase α-fragment, while 293T target cells were cotransfected with expression vector for DPP4 and the β-galactosidase Ω-fragment. At 24 h posttransfection, effector cells were washed with PBS, resuspended in culture medium, and seeded on top of the target cells. Cell–cell fusion was subsequently analyzed at different time points post mixing. For this, β-galactosidase substrate (Gal-Screen, Thermo Fisher Scientific) was added to the cells, and samples were incubated for 90 min in the dark at room temperature before luminescence was recorded using a Hidex Sense plate luminometer (Hidex).

Statistical analysis

If not stated otherwise, two-tailed, unpaired Student’s t-test with Welch correction or two-way analysis of variance with Tukey’s multiple comparison test was performed to assess statistical significance (P > 0.05 ns, not significant; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001).

DATA AVAILABILITY

All data relevant to the present study are shown within the paper.