Discussion
Since the outbreak of the COVID-19 pandemic, numerous single herbs and herbal formulations as well as single compounds isolated from the herbal extracts have been reported to exhibit significant anti-SARS-CoV-2 effects through the inhibition of SARS-CoV-2 life cycle by targeting viral proteins such as 3CLpro, PLpro, RNA-dependent RNA polymerase (RdRp), and S protein, and cellular proteins such as ACE2, cathepsin L, and TMPRSS2 (Adeleye et al., 2022; Benarba & Pandiella, 2020; Lee, Park & Cho, 2022). Saponins including triterpenoid saponins derived from various medicinal plants have also been investigated for their therapeutic potential against SARS-CoV-2 (Ebob, Babiaka & Ntie-Kang, 2021). But, most are based on in silico studies (Diniz, Perez-Castillo, Elshabrawy, Filho & de Sousa, 2021; Falade, Adelusi, Adedotun, Abdul-Hammed, Lawal & Agboluaje, 2021; Sinha et al., 2021; Yang et al., 2020) and it is only recently that their inhibitory activities have been demonstrated using pseudotyped or authentic SARS-CoV-2. Licorice-saponin A3, a triterpenoid saponin isolated from licorice, was reported to inhibit SARS-CoV-2 by targeting the NSP7, a core component of RdRp (Yi et al., 2022). Hui and co-workers very recently showed that 3-O-β-chacotriosyl ursolic acid saponins they synthesized displays anti-SARS-CoV-2 activity by binding to S protein (Li et al., 2022). However, to our knowledge, there have been few studies verifying the role of natural triterpenoid saponins in preventing SARS-CoV-2 infection at the stage of membrane fusion. We previously reported that PD, a triterpenoid saponin derived from PG, inhibits SARS-CoV-2 infection and proposed the blockade of membrane fusion as a possible mechanism of action by demonstrating its ability to inhibit a well-known membrane fusion event of spontaneous inhibitory postsynaptic currents in acute brain slices (Kim et al., 2021). Here, we discovered that LA, a triterpenoid saponin isolated from CL and has a very similar chemical composition to PD, exhibits inhibitory activity against SARS-CoV-2. To provide direct evidence that LA blocks the membrane fusion mediated by S protein of SARS-CoV-2, we established cell fusion assay systems based on S protein-ACE2 interaction. The results from these assays revealed that LA indeed hinders the S-mediated membrane fusion event, thereby preventing SARS-CoV-2 infection.
Cholesterol is an essential component of cell membranes involved in virus-host cell fusion and cell-cell fusion for SARS-CoV-2 entry and pathological syncytia formation (Barrantes, 2022; Sanders et al., 2021). Thus, cholesterol-targeting therapeutics have been proposed for the treatment of COVID-19 (Palacios-Rápalo et al., 2021). The role of membrane cholesterol in SARS-CoV-2 infection was highlighted by recent studies showing that 25-hydroxycholesterol inhibits SARS-CoV-2 entry by blocking membrane fusion through depleting accessible cholesterol from plasma membrane (Wang et al., 2020; Zang et al., 2020; Zu et al., 2020). It was also reported that exogenously added 27-hydroxycholesterol accumulates in the plasma membrane lipid rafts, leading to inhibition of SARS-CoV-2 entry (Marcello et al., 2020; Palacios-Rápalo et al., 2021). Together, these studies suggest that transient changes in cholesterol distribution in plasma membranes can exert inhibitory effects on membrane fusion and viral entry. In this regard, we investigated the effects of LA on membrane cholesterol and observed that upon treatment with LA, cholesterol levels at the plasma membrane were significantly increased, implying that LA blocks S-mediated membrane fusion by altering cholesterol distribution in plasma membranes.
In this study, we also explained the structure-activity relationship by comparing the antiviral activity of three different triterpenoid saponins isolated from CL such as LA, LB, and LD. These compounds share structure similarities consisting of 30-carbon triterpene backbone with a single sugar at position C3 and an oligosaccharide chain attached at position C28, but the number of sugar units and the presence or absence of a branched sugar in the oligosaccharide chain are different each other (Figure 1c). Both pSARS-CoV-2 entry assay and cell-to-cell fusion assay identically demonstrated that LA consisting of linearly linked four sugar residues exhibited the most potent antiviral activity compared to LB and LD that are composed of a branched and/or a shorter chain of sugar residues, indicating that both length and linearity of the conjugated sugar residues of triterpenoid saponins are critical for such inhibitory actions. Previously, our molecular modeling studies predicted that triterpene backbone of PD are embedded in the lipid bilayer membrane with an oligosaccharide chain sticking out of the membrane, highlighting the importance of glycosylated group attached at C28 position in the antiviral activity. Thus, the present study corroborates our previous results, further supporting the idea that the linear sugar residues of triterpenoid saponins create a protrusion from the cell membrane and play critical roles in hindering membrane fusion events. Intriguingly, the potency of LA, LB, and LD against SARS-CoV-2 is well correlated with their ability to increase membrane cholesterol levels (Figure 1d, 3d, and S3), raising an interesting possibility that the attached oligosaccharide chain of LA might also be responsible for the alteration of membrane cholesterol. The more detailed studies of the relationships of structure and mechanism of action of triterpenoid saponins including LA are needed to address this hypothesis.
There are many advantages of membrane fusion blockers over other viral entry inhibitors. Because most reported mutations of SARS-CoV-2 variants are clustered near the RBD of S protein, compounds that bind to ancestral S protein and block its interaction with ACE2 might exhibit reduced inhibitory activity against SARS-CoV-2 variants as monoclonal antibodies targeting the RBD of the S protein of ancestral virus showed reduced neutralizing activity against new variants of SARS-CoV-2 (Bekliz et al., 2022; Takashita et al., 2022). Moreover, SARS-CoV-2 entry inhibitors that specifically act on the interaction between S protein and ACE2 do not provide cross-reactivity to other coronaviruses that recognize different cellular receptors, such as MERS-CoV that uses the receptor dipeptidyl peptidase 4 (DPP4) (Mou, Raj, van Kuppeveld, Rottier, Haagmans & Bosch, 2013) and novel enveloped viruses that might utilize a different cellular receptor to enter host cells. In contrast, viral membrane fusion blockers have a broader application since the membrane fusion process is shared in all enveloped viruses including SARS-CoV-2 and its variants. Supporting this notion, we here demonstrated that LA effectively inhibits the infection of the ancestral SARS-CoV-2 and its variants including Omicron with similar IC50 values ranging from 2.23 to 3.37 μM. Another advantage is that, since the membrane fusion is a common event in the virial entry pathways, fusion blockers can prevent two main SARS-CoV-2 entry routes with a similar efficiency. In agreement with this hypothesis, we found that LA inhibits the cell surface and endosomal pathways of SARS-CoV-2 entry with similar IC50 values in experiments using pseudotyped and authentic viruses (Figure 2f).
The S-mediated membrane fusion is critical, not only for SARS-CoV-2 entry into host cells but also for syncytia formation. The important roles of syncytia formation in SARS-CoV-2 Infection include evasion from neutralizing antibodies and viral spread by cell-to-cell transmission (Rajah, Bernier, Buchrieser & Schwartz, 2021; Zeng et al., 2022). Moreover, syncytia can lead to cell death via apoptosis or pyroptosis, releasing virus particles to re-infect neighboring cells and triggering an inflammatory response (Sanders et al., 2021; Santana et al., 2021). Here, we observed that LA effectively inhibits S protein-mediated syncytia formation and cellular apoptosis, indicating that LA is able to block SARS-CoV-2 pathological effects. Overall, LA is a natural viral fusion blocker that effectively prevents SARS-CoV-2 and all newly emerging variants from infecting host cells and syncytia formation by hindering the S-mediated membrane fusion. We propose here that LA can be a broad-spectrum antiviral drug not only against SARS-CoV-2 but also against other novel enveloped viral pathogens that might arise in the future.
Author contributions: Conceptualization, T.Y.K., D.S.J., S.K., and C.J.L.; Experimental execution, T.Y.K., S.J., M.K., Y.E.D, and S.-R.S. Methodology and formal analysis, T.Y.K., S.J.; writing-original draft preparation, T.Y.K and S.J.; writing-review and editing, T.Y.K., D.S.J., S.K., and C.J.L. All authors have read and agreed to the published version of the manuscript.
Funding: This work was supported by the Institute for Basic Science (IBS), Center for Cognition and Sociality (IBS-R001-D2) to C.J.L., National Research Foundation of Korea (NRF) with grants funded by the Korean government (MSIT) (NRF-2017M3A9G6068245 to S.K. and NRF-2019R1A2C1083945 to D.S.J.).
Data Availability Statement: Materials are available upon a reasonable request from the corresponding author
Acknowledgement: We appreciate Mr. Chan-yong Kwak for providing the root of Codonopsis lanceolata , which motivated the present study. We also thank Haejin Jung and Ph.D. Taek Seung Kim, senior engineers at the Research Solution Center (RSC) in IBS, for performing the flow cytometry analysis and imaging cell-to-cell fusion using Leica DMi8 microscope, respectively.
Conflict of interest s: The authors declare no conflict of interests.