SUMMARY
The Delta variant is now the most dominant and virulent SARS-CoV-2 variant of concern (VOC). In this study, we investigated several virological features of Delta spike protein (SPDelta), including protein maturation and its impact on viral entry of cell-free pseudotyped virus, cell-cell fusion ability and its induction of inflammatory cytokine production in human macrophages and dendritic cells. The results showed that SPΔCDelta exhibited enhanced S1/S2 cleavage in cells and pseudotyped virus-like particles (PVLPs). We further showed that SPΔCDelta elevated pseudovirus infection in human lung cell lines and mediated significantly enhanced syncytia formation. Furthermore, we revealed that SPΔCDelta-PVLPs had stronger effects on stimulating NF-κB and AP-1 signaling in human monocytic THP1 cells and induced significantly higher levels of pro-inflammatory cytokine, such as TNF-α, IL-1β and IL-6, released from human macrophages and dendritic cells. Overall, these studies provide evidence to support the important role of SPΔCDelta during virus infection, transmission and pathogenesis.
INTRODUCTION
The emergence of the highly pathogenic coronavirus disease 2019 (COVID-19) has been a major concern and threat to public health for two years. As of early November 2021, approximately 248 million COVID-19 cases and more than 5 million deaths were reported globally (WHO, 2021). COVID-19 is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which is a member of betacoronaviridae with a single-stranded 30 kb positive-sense RNA genome encoding 29 proteins (Srivastava et al., 2021). Within 2 years, multiple variants of SARS-CoV-2 have emerged (Galloway et al., 2021; Greaney et al., 2021; Nonaka et al., 2021; Paiva et al., 2021; Resende et al., 2021; Santos and Passos, 2021; Tegally et al., 2020; Volz et al., 2021). Delta variant, which was first detected in India and derived from the Pango lineage B.1.617.2, is the most dominant variant of concern (VOC) and has accounted for approximately 99% of new cases of coronavirus worldwide (Banu et al., 2020; Ranjan et al., 2021; Sahoo et al., 2021; Worldometer, 2021). Previous studies suggested that Delta variant infection has a shorter incubation period but a greater viral load (> 1,000 times) than earlier variants (Li et al., 2021; Reardon, 2021). Further, the patients contracted with Delta variant have higher hospitalization rate and more severe outcomes (Twohig et al., 2021). Unfortunately, current vaccines only provide partial protection against the infection of Delta variant, because these vaccines are designed based on the original Wuhan-Hu-1 sequence (Mlcochova et al., 2021). Therefore, it is important to understand the molecular mechanisms of the increased transmissibility and immune evasion of these SARS-CoV-2 variants to facilitate the development of vaccines and therapeutic drugs against COVID-19.
VOCs are mainly classified based on the mutations on their spike protein (SP). The SP of coronavirus is responsible for viral attachment and entry to the host cells (Huang et al., 2020b). The matured SP is cleaved to generate S1 and S2 subunits at specific cleavage sites. The S1 subunit (aa 14-685) is responsible for receptor binding through its receptor-binding domain (RBD). The S2 subunit (aa 686-1273) mediates membrane fusion to facilitate cell entry (Bertram et al., 2013; Hoffmann et al., 2020b; Peacock et al., 2021b). Mutations in SP have resulted in the high rates of transmission and replication of various variants (Zhang et al., 2021a; Zhou et al., 2021) and the immune evasion from antibody neutralization of various variants (McCallum et al.; Mlcochova et al., 2021; Zhang et al.). The SP of the Delta variant has eight mutations compared with the original virus, including T19R, Δ156– Δ157, and R158G in the N-terminus Domain (NTD), D614G, L452R and T478K at the RBD, P681R close to the furin cleavage site, and D950N at the S2 region (Cherian et al., 2021; Planas et al., 2021; Zhang et al., 2021a). It has been demonstrated that the mutations at the NTD of Delta SP alter the antigenic surface near the NTD-1 epitope, thus leading to the lack of binding affinity with the NTD neutralizing antibodies (Zhang et al., 2021a).
Furthermore, the P681R mutation closed to the furin cleavage has been demonstrated to aid the pathogenicity of the virus (Cherian et al., 2021; Liu et al., 2021; Saito et al., 2021). The newly identified furin cleavage site (681-PRRAR↓SV-687) at the S1/S2 site is reported to be critical for the pathogenesis of SARS-CoV-2 in mouse models, and also be responsible for the cell-cell fusion, which is absent in other group-2B coronaviruses (Coutard et al., 2020; Johnson et al., 2020; Xia et al., 2020). It was shown that P681R mutation at this cleavage site endow Delta variant with special features that facilitate the spike protein cleavage and viral fusogenicity (Peacock et al., 2021b; Saito et al., 2021). Study found that a chimeric Delta SARS-CoV-2 bearing the Alpha-SP replicated less efficiently than the wild-type Delta variant, and the reversion of Delta P681R mutation to wild-type P681 attenuated Delta variant replication as well (Liu et al., 2021). These observations suggested that the P618R mutation that occurs in Delta variants contributes immensely to the high replication and transmissibility rate. We are therefore interested in further investigating how the P618R mutation impacts the high replication rate of Delta variants.
Like other high pathogenic viruses (influenza H5N1, SARS-CoV-1 and MERS-CoV), SARS-CoV-2 infection also induced excessive inflammatory response with the release of a large amount of pro-inflammatory cytokines (cytokine storm) that may result in Acute Respiratory Distress Syndrome (ARDS) and multiorgan damage (Zhu and al., 2020). Clinical studies showed that the high mortality of COVID-19 is related to cytokine release syndrome (CRS) in a subgroup of severe patients (Huang et al., 2020a), characterized by elevated levels of certain cytokines including IL-6, TNF-α, IL-8, IL-1β, IL-10, MCP-1 and IP-10 (Hadjadj et al., 2020; Huang et al., 2020a; Mehta et al., 2020; Xu et al., 2020; Yang et al., 2020). The observed cytokine production induced by SARS-CoV-2 infection or spike protein expression has been linked with nuclear factor kappa B (NF-κB) and and activator protein-1(AP-1) signaling pathways that can induce the expression of a variety of proinflammatory cytokine genes (Neufeldt et al., 2020b; Zhu et al., 2021). Like other RNA virus, SARS-CoV-2 was found activate NF-κB and AP-1 transcription factors following the sensing of viral RNAs or proteins by different pathogen pattern recognition receptors (PRRs) and associated signalling cascades, including RLRs and TLRs (Pantazi et al., 2021; Zhu et al., 2021). In addition, angiotensin II type 1 receptor (AT1)-MAPK signalling (Patra et al., 2020) and the cGAS-STING signalling (Neufeldt et al., 2020a) have been identified responsible for the activation of NF-κB and the elevated expression of IL-6 in the SARS-CoV-2 infected or SP expressing cells. However, whether Delta variant or its SP may initiate stronger cytokine storm in patients that leading to more severe illness than other strains still required more investigation.
The current study aims to characterize the cleavage/maturation of various SPs of SARS-CoV-2, especially the Delta variant SP, and their effects on virus infection, cell-cell fusion, cytokine production and related signaling pathways. By using a SARS-CoV-2 SP pseudotyped lentiviral vector or viral-like particles (PVLPs), we demonstrated that SPDelta enhanced S1/S2 cleavage, accelerated pseudovirus infection and promoted cell-cell fusion. We also showed that SPDelta strongly activates NF-κB and AP-1 signaling in THP1 cells. Furthermore, we observed that SPΔCDelta-PVLP stimulation promoted the production of several pro-inflammatory cytokines by human macrophages and dendritic cells.
RESULTS
SARS-CoV-2 Delta SP exhibited enhanced cleavage and maturation in cells and in the pseudotyped virus
To investigate the functional role of SARS-CoV-2 Delta SP, we first synthesized cDNA encoding SARS-CoV-2 Delta SP, as described in CDC’s SARS-CoV-2 Variant Classifications and Definitions (CDC, 2021) (Fig. 1A) and inserted cDNA into a pCAGG-expressing plasmid, as described in the Materials and Methods. Previously described pCAGG-SPΔCWT- and pCAGG-SPΔCG614-expressing plasmids (Ao et al., 2021a) were also used in this study. Meanwhile, we constructed a pCAGG-SPΔCDelta-PD in which arginine (R) at the amino acid position of 681 and asparagine (N) at 950 in SPΔCDelta were reverted to the original proline (P) and aspartic acid (D) to test the effect of these amino acids on the function of SPΔCDelta (Fig. 1A). To enhance the transportation of SP to the cell surface and increase the virus incorporation of SP, the cDNA encoding the C-terminal 17 aa in SARS-CoV-2 SP was deleted in all pCAGG-SPΔC plasmids (Ao et al., 2021a).
To examine the expression of various SPs in the cells and their incorporation in the SPΔC-pseudotyped viral particles (PVPs), each SPΔC-expressing plasmid was cotransfected with a multiple-gene deleted HIV-based vector encoding a Gaussia luciferase gene (ΔRI/ΔEnv/Gluc) and a packaging plasmid (pCMVΔR8.2) in HEK293T cells, as described previously (Ao et al., 2021a). After 48 hrs of transfection, the expression of each SPΔC in the transfected cells was analyzed by an indirect immunofluorescence (IF) assay using human SARS-CoV-2 S-NTD antibodies. The results revealed that all SPΔCs were well expressed in the transfected cells or cell surface (Fig. 1B). Meanwhile, the transfected cells and PVPs were lysed and processed with WB with anti-SP/RBD or anti-S2 antibodies, respectively (Fig. 1C, top and middle panels). Interestingly, the data clearly showed that in the cells, SPΔCDelta was more efficiently processed from the S precursor into S1 and S2 than SPΔCWT and SPΔCG614 (Fig. 1C, compare Lane 3 to Lanes 1 and 2). In PVPs, the majority of SPΔCDelta presented as a mature form (S1 and S2) compared to SPΔCWT and SPΔCG614 (Fig. 1C, compare Lane 7 to Lanes 5 and 6), indicating that SPΔCDelta undergoes a more efficient maturation process. Surprisingly, S1 of SPΔCDelta but not S2 appeared to migrate faster than S1 of SPΔCWT and SPΔCG614 (Fig. 1C, top panel). The possible mechanism for this behavior is currently unknown.
The cleavage of the SARS-CoV-2 SP into S1 and S2 most likely occurs by furin, and the P681R mutation of the SP Delta was suggested to enhance S1/S2 cleavability (Liu et al., 2021; Peacock et al., 2021a). We therefore further tested whether a fusin protease inhibitor, a peptidyl chloromethylketone (CMK), or the reverse change in R681 of SPΔCDelta to P would alter the maturation rate of the S protein. The SPΔCDelta-PD-PVPs or SPΔCDelta-PVPs packaged in the presence or absence of CMK were analyzed by WB with anti-RBD antibody and quantified by densitometry using ImageJ (https://imagej.nih.gov/ij/). The results showed that either CMK treatment or SPΔCDelta-PD clearly negatively impacted the maturation of SPΔCDelta (Fig. 1D).
Delta-SP mediates more efficient pseudovirus infection in a lung epithelial cell line and primary macrophages
To investigate the impact of SPΔCDelta on viral infection, we produced Gluc-expressing Delta-SP-PVPs (Fig. 2A) and tested the infectivity of different pseudoviruses in two human lung epithelial cell lines, Calu-3 and A549 cells. To increase susceptibility to SP-PVP infection, the A549ACE2 cell line was generated by transducting a lentivirus expressing hACE2 and subsequent puromycin selection, as described in the Materials and Methods. hACE2 expression in A549ACE2 cells was verified by WB (Fig. 2B). Then, both cell lines were infected with equivalent amounts (adjusted with p24 values) of the SPΔCWT-, SPΔCG614-, and SPΔCDelta-PVPs for three hours and washed. At 24 and 48 hrs post infection (p.i.), the supernatants were collected, and the infection levels of pseudoviruses were monitored by measuring Gluc activity. The results showed that in both cell lines, the SPΔCDelta-PVPs exhibited the highest infection efficiency, the SPΔCG614-PVPs had a slightly lower infection efficiency than SPΔCDelta, while the SPΔCwt-PVPs showed a significantly lower infection efficiency (Fig. 2C). All of these results indicated that SPΔCDelta-PVPs had a significantly more efficient virus entry step than SPΔCwt.-PVPs in a single cycle replication system.
Next, we also checked the ability of SPΔC-PVPs to infect human differentiated macrophages and dendritic cells. Briefly, human monocyte-derived macrophages (MDMs) or dendritic cells (MDDCs) were infected with equal amounts (adjusted with HIV p24 levels) of SPΔCwt--, SPΔCG614-, and SPΔCDelta-PVPs. At 48 and 72 hrs p.i., the Gluc activity in the supernatant from the infected cell cultures was monitored. The results showed that both human primary cells, especially MDMs, could be infected by SPΔC-PVPs, while SPΔCDelta- and SPΔCG614-PVPs displayed more efficient infection than SPΔCwt-PVPs (Fig. 2D). All of these experimental observations indicate that SPΔCDelta-PVPs have a stronger ability to target MDMs than SPΔCwt-PVPs. The results also suggested that MDDCs can be targeted by SPΔC-PVPs but with less efficiency.
Delta-SP variant enhanced syncytia formation in lung epithelial A549 cells expressing ACE2
Previous studies have shown that SARS-CoV-2 SP is able to possess fusogenic activity and form large multinucleated cells (syncytia formation) (Bussania et al., 2020; Cattin-Ortolá et al., 2020). We then asked whether Delta-SP could possess higher fusogenic activity than other variants. Briefly, 293T cells were transfected with SPΔCWT, SPΔCG614, SPΔCDelta, or SPΔCDeltaPD plasmids by Lipofectamine 2000. At 24 hrs of transfection, we mixed SPΔC-expressing 293T cells with A549ACE2 cells at a ratio of 1:3. At 6 and 30 hrs post transfection, syncytial formation was observed under a microscope, and the results revealed that SPΔCWT and SPΔCG614 induced similar levels and sizes of syncytia. Intriguingly, an increasing number of syncytia formations were observed in the coculture of SPΔCDelta-expressing 293T and A549ACE2 cells (Fig. 3A and B), indicating that SP from the Delta variant has a stronger fusogenic ability. However, SPΔCDeltaPD-expressing 293T/A549ACE2 cell coculture displayed less syncytia formation than SPΔCDelta (Fig. 3 B). A and B), suggesting the importance of P681R for the strong fusogenic activity of SP from the Delta variant.
To further confirm the strong fusogenic ability of SPΔCDelta, we also generated A549 cells stably expressing SPΔCwt, SPΔCG614 or SPΔCDelta (named A549-SPΔCDelta, A549-SPΔCG614 or A549-SPΔCwt cells) (Fig. 3C). Since A549-SPΔCwt and A549-SPΔCDelta cells displayed similar levels of SPΔC expression based on a WB analysis, we then tested their fusogenic ability by mixing A549-SPΔCDelta or A549-SPΔCwt cells with the A549ACE2 cell line using a similar experimental process as described above. Meanwhile, A549-SPΔCDelta or A549-SPΔCwt cells were cocultured with A549 cells as a control. The results confirmed that the coculture of A549-SPΔCDelta cells and A549ACE2 cells formed large syncytia formation more efficiently than that of A549-SPΔCwt cells (Fig. 3D), confirming the stronger fusogenic activity of the SP of the Delta variant.
Delta variant SP stimulates higher NFκB and AP1 signaling pathway activities
The severity of COVID-19 is highly correlated with dysregulated and excessive release of proinflammatory cytokines (Huang et al., 2020a). Given that the NFκB and AP1 signaling pathways are among the critical pathways responsible for the expression of proinflammatory cytokines and chemokines (Hojyo et al., 2020; Kawasaki and Kawai, 2014), we therefore examined the activities of these two signaling pathways triggered by SPΔC in the monocyte cell line THP1 and THP1-derived macrophages. First, we generated THP1-NF-κB-Luc and THP1-AP-1-Luc sensor cell lines by transducing THP1 cells with a lentiviral vector encoding the luciferase reporter gene driven by NFκB- or AP1-activated transcription response elements (Fig. 4A), as described in the Materials and Methods. To obtain THP1-derived macrophages, THP-1-NF-κB-Luc and THP1-AP-1-Luc sensor cell lines were treated with phorbol 12-myristate 13-acetate (PMA) (100 nM) for 3 days. Additionally, we produced genome-free SPΔC-PVLPs by cotransfecting each SPΔC-expressing plasmid with a packaging plasmid (pCMVΔR8.2) in 293T cells, and the expression of SPΔC in the purified PVLPs was verified by WB with an anti-RBD antibody (Fig. 4B). Then, different THP1 sensor cell lines and THP1-derived macrophages were treated with different SPΔC-PVLPs of same amount (adjusted by p24) for 6 hrs, and the luciferase activity in treated cells was measured by a luciferase assay system (Promega). Interestingly, we found that the NFκB activity induced by SPΔCwt and SPΔCG614-PVLPs was slightly higher than that induced by the VLP control (Gag). However, the SPΔCDelta-treated THP1 cells/macrophages produced significantly higher (3∼7-fold) NFκB activity compared with SPΔCwt or SPΔCG614 (Fig. 4C). Consistent with this finding, SPΔCDelta also triggered higher (∼2-fold) AP1 signaling pathway activities in THP1/macrophages than SPΔCwt or SPΔCG614 (Fig. 4D). These results indicated that SPΔCDelta triggered significantly stronger signals to activate the NFκB and AP1 pathways in the monocyte cell line THP1 and THP1-derived macrophages.
Delta variant SP stimulates higher proinflammatory cytokine production in human macrophages (MDMs) and dendritic cells (MDDCs)
Previous studies have shown that SARS-CoV-2 infection can stimulate the production of immunoregulatory cytokines (IL-6, IL-10) in human monocytes and macrophages (Boumaza et al., 2021). We further investigated whether SP of the Delta variant can induce higher levels of proinflammatory cytokine and chemokine in MDMs and MDDCs. Briefly, human MDMs and MDDCs were treated with the same amount (adjusted by p24) of SPΔC-PVLPs, including SPΔCwt-, SPΔCG614-, SPΔCDelta-PVLPs or control VLPs (Gag-VLPs). After 24 hrs of incubation, the cytokines released in the supernatants were determined by a MSD (Meso Scale Discovery) immunoassay. The results revealed that SPΔCwt-PVLP stimulation did not result in a significant change in cytokine release from MDMs compared to the control VLPs (Fig. 5A). However, in MDMs, SPΔCDelta-PVLPs induced significantly higher levels of several proinflammatory cytokines, such as IFN-γ, TNF-α, IL-1β, and IL-6, while SPΔCG614-PVLPs also induced increase of these cytokines but overall to a less extent, when compared with SPΔCwt-PVLPs (Fig. 5A). For example, the SPΔCDelta-PVLPs elevated TNF-α level 61-fold in comparison with SPΔCwt-PVLPs, contrastingly, SPΔCG614-PVLPs only increased to approximately 33-fold. Nevertheless, all of SPΔC-PVLPs showed no stimulating effects on IL-2 and IL-8 production. Interestingly, SPΔCDelta-PVLPs and SPΔCG614-PVLPs also slightly increased anti-inflammatory cytokines IL-4, IL-10 and IL-13, indicating the negative feedback of inflammation may exist during these stimulations.
Surprisingly, in MDDCs, SPΔCDelta-PVLP stimulation resulted in a significant increase in most pro-inflammation cytokines we tested, including IFN-γ, TNF-α, IL-1β, IL-2, IL-6, IL-8, and IL-12p70 (Fig. 5B). Among them, IL-6, TNF-α and IL-2 were the most increased cytokines (8∼13-fold), followed by IFN-γ and IL-1β (5∼6-fold). The levels of IFN-γ, IL-2, and IL-6 in the supernatans of SPΔCG614-PVLPs treated MDDCs were higher than that of SPΔCWT-PVLPs or control VLPs, also to a less extent. In contrast, IL-10 production appears to be negatively regulated by all SPΔC-PVLPs, including the control VLPs. Altogether, the above results suggested that SPΔCDelta could induce remarkably higher levels of most proinflammatory cytokines tested and small differentce in some anti-inflammatory cytokines than VLP-SPΔCG614 or SPΔCwt in human MDMs and MDDCs.
DISCUSSION
The SARS-CoV-2 Delta variant has higher transmissibility and thus has become the predominant strain worldwide in 2021(Li et al., 2021). It is important to understand the mechanisms of the increased transmissibility and cytokine release triggered by this variant. In this study, we investigated the cleavage and maturation efficiency of the Delta variant Spike protein (SPΔCDelta) during pseudovirus assembly and its impact on cell-free pseudovirus infection and cell-cell fusion activities. The results demonstrated enhanced cleavage and maturation of SPΔCDelta in the produced viral particles. Additionally, the studies clearly showed that SPΔCDelta mediated more efficient pseudovirus infection and mediated a significantly enhanced cell-cell fusion process. Furthermore, our analyses revealed that SPDelta-PVLPs had stronger effects on stimulating NF-κB and AP-1 signaling in THP1 cells and elevated the production of several essential proinflammatory cytokines by human MDMs and MDDCs, compared with SPWT-PVLPs.
SARS-CoV-2 transmission and pathogenesis require the polybasic cleavage site between S1 and S2 in its SP (Johnson et al., 2021; Peacock et al., 2021a). This furin protease cleavable site is critical for the maturation of SARS-CoV-2 and its biological functions (Boson et al., 2021; Hoffmann et al., 2018; Hoffmann et al., 2020a). In an attempt to understand the mechanisms by which the Delta variant is more infectious than other variants, our study revealed that SPΔCDelta was significantly more efficient in the processing of the precursor S into S1 and S2 compared with SPΔCWT and SPΔCG614 in both the cells and PVPs. In the PVPs, the majority of SPΔCDelta was presented as mature forms (S1 and S2), while some portions of the precursor S of SPΔCWT and SPΔCG614 were still present in the PVPs (Fig. 1C, right panel). To further investigate the importance of the furin cleavage site for the enhanced cleavage of the Delta variant, we produced SPΔCDelta-PVPs packaged in the presence or absence of the furin protease inhibitor CMK. As expected, CMK significantly inhibited the cleavage of the spike protein of the Delta variant.
Among the mutations in the Delta variant spike protein, an amino acid proline (P681) at the N-terminus of the polybasic cleavage site (RRAR) was changed to arginine (R), known as the P681R mutation (Saito et al., 2021). The P681R mutation is of great importance because it is part of a proteolytic cleavage site for furin and furin-like proteases. The P618R mutation clearly plays a critical role in the SP of the Delta variant to abrogate host O-glycosylation (Zhang et al., 2021b). To further investigate whether this altered polybasic cleavage site (RRRAR) is required for effective furin cleavage of SARS-CoV-2 SP, we reverted the arginine (R) of 681 and asparagine (N) of 950 back to the original proline (P) and aspartic acid (D) on SP of the Delta variant (SPΔCDelta-PD). The WB results showed that the cleavage of the SPΔCDelta-PD-PVPs was comparable to that of SPΔCWT-, SPΔCG614- or SPΔCDelta-PVPs produced in the presence of CMK. Consistent with other reports (Peacock et al., 2021b; Saito et al., 2021), our results further demonstrated that the P618R mutation in SPΔCDelta is essential for the enhanced furin cleavage of Delta variant SP. Meanwhile, we also observed that S1 of SPΔCDelta appeared to migrate faster than S1 of SPΔCWT and SPΔCG614. Although the mechanism is currently unknown, it could be related to the possible altered glycosylation content of SPΔCDelta because a previous study revealed that P681R could circumvent host O-glycosylation (Zhang et al., 2021b). In addition, multiple amino acid mutations and deletions present in S1 of Delta SP may also partially contribute to this alteration. More detailed studies are required to analyze the possible underlying mechanism(s).
In this study, we also observed that SPΔCDelta enhanced cell-free pseudovirus infection in A549ACE2+ and Calu-3 cell lines and macrophages, indicating that SPΔCDelta is an important factor contributing to the increased infectiousness of the Delta variant. Additionally, it was observed that the infection mediated by SPΔCDelta-pseudovirus was only slightly higher than that mediated by SPΔCG614-pseudovirus, suggesting that the G614 mutation present in SPDelta may be one of the main driving forces for the increased infectivity of the Delta virus (Daniloski et al., 2020; Zhang et al., 2020). This finding is in agreement with previous studies showing that the D614G mutation in SARS-CoV-2 SP contributes immensely to virus infectivity and replication (Daniloski et al., 2020; Zhang et al., 2020). It should be noted that our results were based on single-cycle SPΔC-pseudovirus replication; thus, the infection advantage of the native SARS-CoV-2 Delta variant needs further investigation.
It is well known that spike protein expressed at the surface of infected cells is sufficient to generate fusion with neighboring cells. Here, we further showed that significantly enhanced syncytia formation was observed when SPΔCDelta-expressing 293T or A549 cells were cocultured with A549ACE2 cells. This observation raised potential importance in terms of the SARS-CoV-2 Delta variant’s virulence since cell-to-cell fusion may provide another efficient method of viral dispersal in the host, thus indicating its stronger transmission among the population, as described previously (Michael Rajah et al., 2021). This finding is in agreement with recent reports that B.1.617.2 SP mediates highly efficient syncytia formation compared with wild-type SP (Michael Rajah et al., 2021; Mlcochova et al., 2021; Planas et al., 2021; Zhang et al., 2021a). Moreover, this efficient cell-to-cell transmission ability of SPΔCDelta may enhance its resistance to host immune responses, such as antibody-mediated neutralization (Mlcochova et al., 2021; Planas et al., 2021). It is worth noting that we did not investigate the impact of TMPRSS2 on the cell-cell fusion process. For SARS-CoV-2, cleavage of S by furin at the S1/S2 site is required for subsequent cleavage by TMPRSS2 at the S2’ site. Previous studies have demonstrated that TMPRSS2 could enhance the infectivity and fusogenic activity of different coronaviruses, including SARS-CoV-2 (Buchrieser et al., 2020; Glowacka et al., 2011; Kleine-Weber et al., 2018; Matsuyama et al., 2010). Future investigations into the role of TMPRSS2 in SP Delta-induced syncytia formation and infection will provide a better understanding of the persistence, dissemination, and immune or inflammatory responses of Delta variants.
The severity of COVID-19 is highly correlated with dysregulated and excessive release of proinflammatory cytokines (Huang et al., 2020a). Hence, we also tested whether macrophages or dendritic cells act as major modulators of the immune response by producing a large amount of cytokines and chemokines to recruit immune cells and presenting antigens to them. The engagement of the spike protein of SARS-CoV-2 with the receptor ACE2 on THP1-derived macrophages is reported to initiate signaling pathways and activate the production of proinflammatory cytokines, including IL-6, TNF-α, and MIP1a (Pantazi et al., 2021). Here, we showed that NFκB and AP1 signaling pathway activities were also enhanced by SPΔCDelta compared with SPΔCWT in THP1 cells and THP1-derived macrophages, suggesting that SPΔCDelta might promote the inflammatory status of these cells. Similarly, the SP of SARS-CoV-1 has been discovered to activate NF-κB and stimulate the release of IL-6 and TNF-α (Wang et al., 2007).
A previous study reported that high plasma levels of TNFα, IL-1, IL-6, IL-8 and other inflammatory mediators were found in severe COVID-19 patients, and the serum IL-6, IL-8 and TNF-α levels were strong and independent predictors of disease progression, severity and death. (Del Valle et al., 2020; Huang et al., 2020a; Santa Cruz et al., 2021). Interestingly, we found that SPΔCDelta significantly enhanced the expression of several proinflammatory cytokines (TNF-α, IL-1β, and IL-6) in both MDMs and MDDCs (Fig. 5). Especially for MDDCs, increased levels of other proinflammatory cytokines, including IFN-γ, IL-2, IL-8, and IL-12p70 were also detected (Fig. 5). However, SPΔCwt only exhibited induction of IFN-γ in MDDCs, but did not show any effect on other cytokine production of MDMs or MDDCs. This is agree with previous study that revealed, upon SARS-CoV-2 infection, neither macrophage, nor dendritic cells produce the pro-inflammatory cytokines (Niles et al., 2021). Mutations in Delta SP seem to be the key points that cause the differential expression of cytokines. Given the fact that IFN-γ, TNF-α, IL-1β and IL-12 are T-helper-1 (Th1) cytokines, it also suggests that the Th1/Th2 balance has further shifted to Th1 dominance after stimulation with SPΔCDelta-PVLP. Along with proinflammatory cytokines, three anti-inflammatory cytokines (IL-4, IL-10 and IL-13) were also increased in SPΔCDelta-PVLP treated macrophages compared with SPΔCWT-PVLP-treated macrophages. Consistently, higher secretion of T-helper-2 (Th2) cytokines such as IL-4 and IL-10 has been reported in ICU patients than in non-ICU patients (Huang et al., 2020a). Their functions are to suppress both inflammation and the TH1 cellular response, indicating that the balances between pro- and anti-inflammation, as well as the balances between TH1 and TH2 cellular responses existing in patients, are important for the clinical outcomes of COVID-19 therapy. However, the lower IL-10 level in all PVLPs treated MDDCs is surprising and the reason of this is unclear. In conclusion, SPΔCDelta-treated macrophages and DCs are in a higher inflammatory state and in a Th1-dominant Th1/Th2 balance.
Overall, we demonstrated that the SARS-CoV-2 Delta variant spike protein exhibited enhanced cleavage and maturation, which may play an important role in viral infection and cell-cell transmission. Furthermore, we revealed that SPDelta had stronger effects on stimulating NF-κB and AP-1 signaling in monocytes and the release of proinflammatory cytokines from human macrophages and dendritic cells. All of these studies provide strong evidence to support the important role of Delta SP during virus infection, transmission and pathogenesis.
MATERIALS AND METHODS
Plasmid constructs
The SARS-CoV-2 SP protein-expressing plasmids (pCAGGS-nCoVSPΔC and pCAGGS-nCoVSPΔCG614) were described previously (Ao et al., 2021a). The gene encoding SPΔCDelta or SPΔCDelta-PD was synthesized (Genescript) and cloned into the pCAGGS plasmid, and each mutation was confirmed by sequencing. pEF1-SPΔCwt, pEF1-SPΔCG614 or pEF1-SPΔCDelta was constructed by inserting the cDNA encoding SPΔCwt, SPΔCG614 or SPΔCDelta through the BamHI and NheI sites into the pEF1-pcs-puro vector (Ao et al., 2008). The HIV RT/IN/Env trideleted proviral plasmid containing the Gaussia luciferase gene (ΔRI/E/Gluc) and the helper packaging plasmid pCMVΔ8.2 encoding the HIV Gag-Pol plasmids have been described previously (Ao et al., 2016; Zhang et al., 2016).
Cell culture, antibodies and chemicals
Human embryonic kidney cells (HEK293T), human lung (carcinoma) cells (A549), A549ACE2, Calu-3 cells and THP1-sensor cells were cultured in Dulbecco’s modified Eagle’s medium or RPMI 1640 medium supplemented with 10% fetal bovine serum (F.B.S.) and 1% penicillin/streptomycin. To obtain human MDMs or MDDCs, human peripheral blood mononuclear cells (hPBMCs) from healthy donors were collected by sedimentation on a Ficoll (Lymphoprep; Axis-Shield) gradient, adherent to 24-well plates for 2 hrs, and then treated with macrophage colony stimulator (M-CSF) or granulocyte-macrophage-stimulating factor (GM-CSF) and IL-4 (R&D system) for 7 days.
The THP1-NF-κB-Luc and THP1-AP-1-Luc sensor cell lines were described previously (Ao et al., 2021b). To obtain THP1-derived macrophages, THP-1-NF-κB-Luc and THP1-AP-1-Luc sensor cell lines were treated with phorbol 12-myristate 13-acetate (PMA) (200 ng/mL) for 3 days followed by 2 days of rest, as previously described (Starr et al., 2018). A549-expressing human ACE2 (A549ACE2) cells were generated by transducing A549 cells with the ACE2-expressing lentiviral vector (pLenti-C-mGFP-ACE2) (Origene, Cat# PS100093) and then selected with puromycin according to the manufacturer’s procedure.
The rabbit polyclonal antibody against SARS-CoV-2 SP/RBD (Cat# 40592-T62) or human SARS-CoV-2 S-NTD antibody (E-AB-V1030) was obtained from Sino Biological or Elabscience. Mouse monoclonal antibody (1A9) against SARS-CoV-2 SP-S2 (Cat# ab273433) was obtained from Abcam. Anti-HIVp24 monoclonal antibody was described previously (Ao et al., 2007; Qiu et al., 2011). Anti-human ACE2 antibody (sc-390851) was obtained from Santa Cruz Biotechnology Inc. Furin inhibitor I, a peptidyl chloromethylketone (CMK) (Cat# 344930), was obtained from Millipore Sigma.
Virus production and infection experiments
SARS-CoV-2 SPΔC pseudotyped viruses (CoV-2-SPΔC-PVs, CoV-2-SPΔCG614-PVs and CoV-2-SPΔCDelta-PVs) or pseudotyped virus-like particles (VLPs) were produced by transfecting HEK293T cells with pCAGGS-SPΔCWT, pCAGGS-SPΔCG614 or pCAGGS-SPΔCDelta and pCMVΔ8.2 with or without a Gluc-expressing HIV vector ΔRI/E/Gluc (Ao et al., 2021a). After 48 hrs of transfection, cell culture supernatants were collected, and VPs or VLPs were purified from the supernatant by ultracentrifugation (32,000 rpm) for 2 hrs. The pelleted VPs or VLPs were resuspended in RPMI medium, and virus titers were quantified by HIV-1 p24 amounts using an HIV-1 p24 ELISA.
To measure the infection ability of SARS-CoV-2 SPΔC pseudotyped VPs, equal amounts of each SPΔC-PVs stock (as adjusted by p24 levels) were used to infect A549ACE2, Calu-3 cells, human MDMs or MDDCs. After different time intervals (24, 48 and 72 hrs), the supernatants were collected, and the viral infection levels were monitored by measuring Gaussia luciferase (Gluc) activity. Briefly, 50 µl of coelenterazine substrate (Nanolight Technology) was added to 10 µl of supernatant, mixed well and read in a luminometer (Promega, U.S.A.).
To evaluate the effects of various SCoV-2 SPΔC-VLPs on the NF-κB and AP-1 signaling pathways, the same amount of each SPΔC-pseudotyped VLP stock (10 ng, as adjusted by the p24 levels) was directly added to THP1-NF-κB-Luc or THP1-AP1-Luc sensor cells. After 6 hrs, the cells were collected and subjected to luciferase assay as described previously (Ao et al., 2021b). To test the effect of different SPΔC-VPs on cytokine production in MDMs and MDDCs, the same amount of each SPΔC-VP stock (20 ng, as adjusted by the p24 levels) was added to MDMs and MDDCs, and the supernatants were collected after 24 hrs. The cytokine (IFN-γ, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, TNF-α) levels in the supernatants were measured using the MSD V-PLEX proinflammatory Panel 1 (human) Kit (Mesoscale Discovery, USA, Cat# K15049D-1) following the manufacturer’s procedure.
Generation of different SPΔC-expressing A549 stable cell lines
Production of lentiviral vectors expressing different SPΔC: 293T cells were cotransfected with pEF1-SPΔCwt, pEF1-SPΔCG614 or pEF1-SPΔCDelta with packaging plasmid Δ8.2 and VSV-G expressing plasmid. Forty-eight hours posttransfection, each lentiviral vector particle in the supernatant was collected. Then, each produced lentiviral particle was used to transduce A549 cells, and the transduced cells were selected with puromycin for one week. SPΔCwt/mutant expression in the different transduced A549 cells was evaluated by WB using an anti-RBD antibody.
Immunofluorescence assay
293T cells transfected with various SARS-CoV-2 SPΔC-expressing plasmids were grown on glass coverslips (12 mm2) in a 24-well plate. After 48 hrs, cells on the coverslip were fixed in 4% paraformaldehyde for 5 minutes and permeabilized with 0.2% Triton X-100 in PBS. The cells were then incubated with primary antibodies against the N-terminal domain of SARS-CoV-2 SP followed by the corresponding FITC-conjugated secondary antibodies. The cells were viewed under a computerized Axiovert 200 fluorescence microscope (ZEISS).).
Syncytium formation assay
293T cells were transfected with pCAGGS-SPΔCWT, SPΔCG614, SPΔCDelta or SPΔCDeltaPD plasmids using Lipofectamine 2000. After 24 hrs, the cells were washed, resuspended and mixed with A549ACE2 cells at a 1:3 ratio and plated into 48-well plates. For syncytium formation of the stable cell line, A549-SPΔCWT or A549-SPΔCDelta cells were detached with 0.05% trypsin and mixed with A549 or A549ACE2 cells. At different time points, syncytium formation was observed, counted and imaged by bright-field microscopy (Axiovert 200, ZEISS).
Statistics
Statistical analysis of cytokine levels, including the results of GLuc assay, Luciferase assay, and various cytokine/chemokines assay, were performed using the unpaired t-test (considered significant at P≥0.05) by GraphPad Prism 9 software.
AUTHOR CONTRIBUTIONS
Experimental design, X. Y, Z.A. and M.J.O; Investigation, Z.A., M.J.O., and O.T.A; Writing-Original Draft Preparation, Z.A. and M. J. O. and O.T.A; Review, Z.A. M. J. O. and X.Y. Supervision, X.Y.
DECLARATION OF INTERESTS
The authors declare no competing interests.
ACKNOWLEDGEMENTS
We thank Dr. Darwyn Kobasa for providing the Calu-3 cell lines and technique supports. Titus Olukitibi is a recipient of the University Manitoba Graduate scholarship. This work was supported by Canadian 2019 Novel Coronavirus (COVID-19) Rapid Research Funding (OV5-170710) by Canadian Institute of Health Research (CIHR) and Research Manitoba, and CIHR COVID-19 Variant Supplement grant (VS1-175520) to X-J.Y. This work was also supported by the Manitoba Research Chair Award from the Research Manitoba (RM) to to X-J.Y.