HR121 targeting HR2 domain in S2 subunit of spike protein can serve as a broad-spectrum SARS-CoV-2 inhibitor via intranasal administration

Abstract

The continuously emerging SARS-CoV-2 variants pose a great challenge to the efficacy of current drugs, this necessitates the development of broad-spectrum antiviral drugs. In the previous study, we designed a recombinant protein, heptad repeat (HR) 121, as a variant-proof vaccine. Here, we found it can act as a fusion inhibitor and demonstrated broadly neutralizing activities against SARS-CoV-2 and its main variants. Structure analysis suggested that HR121 targets the HR2 domain in SARS-CoV-2 spike (S) 2 subunit to block virus-cell fusion. Functional experiments demonstrated that HR121 can bind HR2 at serological-pH and endosomal-pH, highlighting its inhibition capacity when SARS-CoV-2 enters via either cellular membrane fusion or endosomal route. Importantly, HR121 can effectively inhibit SARS-CoV-2 and Omicron variant pseudoviruses entering the cells, as well as block authentic SARS-CoV-2 and Omicron BA.2 replications in human pulmonary alveolar epithelial cells. After intranasal administration to Syrian golden hamsters, it can protect hamsters from SARS-CoV-2 and Omicron BA.2 infection. Together, our results suggest that HR121 is a potent drug candidate with broadly neutralizing activities against SARS-CoV-2 and its variants.

Graphical abstract

HR121 can block the binding of HR2 and HR1 during membrane fusion, thus inhibiting the replication of SARS-CoV-2 and protecting hamsters from infection via intranasal administration.

1. Introduction

In late December 2019, a new respiratory virus emerged. Because of the same receptor and similar sequence with SARS-CoV, this virus was named as SARS-CoV-21, 2, 3. Its pandemic has posed a serious threat to global public health⁴. The continued spread of SARS-CoV-2 worldwide has resulted in the emergence of many variants of interest (VOIs) and concern (VOCs), such as Omicron and its sub-lineages, in which the S protein contains more than 30 mutations that significantly increase resistance to neutralizing antibodies evoked by current COVID-19 vaccines5, 6, 7. These alarming facts call for the development of broad-spectrum inhibitors to block infection by original SARS-CoV-2 as well as its emerging variants.

Like other coronaviruses, SARS-CoV-2 enters host cells using its spike protein. The S protein consists of S1 and S2 subunits. S1 subunit contains a receptor binding domain (RBD) that is responsible for binding to cell surface receptor ACE2⁸. Hence, the viral S1 subunit, especially the RBD, is exposed and vulnerable to host immunity. Indeed, under the pressure of host immunity, most mutations in the emerging variants are found to be located within this region. Many vaccines and drugs have been designed targeting the RBD domain. Unfortunately, these mutations may play important roles in increasing viral fitness, and equally were shown to reduce the neutralizing efficacy of RBD-specific antibodies⁶. The S2 subunit mediates the membrane fusion between host cell and virus, it includes N-terminal fusion peptide (FP), heptad repeat (HR)1, HR2, transmembrane domains and cytoplasmic tail⁹. During the process of fusion, FP was inserted into the cellular membrane, resulting in conformational transition of the S2 subunit, which changes into a fusion intermediate. In the fusion intermediate, three HR1 segments self-assemble into a trimer coil, and three HR2 segments are packed back into the surface groove of the HR1 inner core thus generating a six-helical bundle (6-HB) structure, driving the close contact and fusion between cellular and viral membranes¹⁰. Compared with S1 subunit, the HR1 and HR2 sequences of S2 subunit are more conservative in coronaviruses, especially the sequence in HR2¹¹. Therefore, the fusion intermediate provides important targets for the design of broad-spectrum fusion inhibitors. The peptides derived from HR2 can bind to the HR1 trimer inner core, and showed antiviral activities against coronaviruses, such as SARS-CoV and MERS-CoV¹². Among them, a pan-coronavirus fusion inhibitor EK1 presented potent inhibitory activities against a variety of human coronaviruses, including SARS-CoV, MERS-CoV, and HCoV-NL63^11,12. In addition, a series of fusion inhibitory peptides derived from HR2 of SARS-CoV-2, such as HR2P, EK1C4, were also designed^3,11. These peptides all target the HR1 domain of S2 subunit, and to date, there are no HR1-derived peptides have been reported to target the more conserved HR2 domain and showed antiviral activities, which may mainly because the peptides derived from HR1 exhibited a strong tendency to aggregate with poor solubilities⁷.

In the previous study, we have designed a recombinant protein vaccine HR121, which evoked potent cross-neutralizing antibodies (nAbs) against SARS-CoV-2 and its multiple variants. HR121 forms a stable dimer with highly solubility and can interact with the HR2 domain of the S2 subunit¹³. In the present study, we found that HR121 could compete with HR1, prevent HR1 binding with HR2, thus inhibiting the formation of the 6-HB between HR1 and HR2, and blocking S protein-mediated membrane fusion. HR121 demonstrated broad-spectrum inhibitory activities against 13 pseudo-SARS-CoV-2 viruses, including Delta, Omicron and its current sub-lineages. It also inhibited SARS-CoV-2 prototype strain or Omicron BA.2 variant replication in several host cells. Prior intranasal application of HR121 protected Syrian golden hamster from both SARS-CoV-2 prototype strain and Omicron BA.2 variant infection. Structural study of HR121 in complex with HR2 not only implied the mechanism of HR121 against infection by SARS-CoV-2, but also explained why HR121 could maintain its pan-inhibitory activity. These results suggest that HR121 is a promising fusion inhibitor and can be developed as a potential prophylactic and therapeutical reagent against SARS-CoV-2 and its multiple variants.

2. Materials and methods

2.1. Ethics statements

The study was approved by the internal review board of Kunming Institute of Zoology, Chinese Academy of Sciences (approval number: SMKX-2021-01-006). All methods were carried out in accordance with relevant guidelines and regulations.

2.2. Cell and viruses

SARS-CoV-2 prototype strain [Accession No.: NMDCN0000HUI in the China National Microbiology Data Center (NMDC)] was provided by Guangdong Center for Disease Control and Prevention (Guangzhou, China)¹⁴. BA.2 variant was isolated from a patient in Yunnan Provincial Infectious Disease Hospital, China (No. SUB1663744906574 in NMDC)¹³. The viruses were propagated and titrated in African green monkey kidney epithelial cells (Vero-E6) (ATCC, No. 1586). VSV-G pseudovirus (G∗ ΔG-VSV-Rluc) was provided by Professor Geng-Fu Xiao, Wuhan Institute of Virology, Chinese Academy of Sciences. Human alveolar epithelial cells (HPAEpiCs) were purchased from ScienCell Research Laboratory (San Diego, CA, USA), and 293T-ACE2 cells were purchased from Yesean company (Shanghai, China) and all the cells were cultured in DMEM medium supplemented with 10% fetal bovine serum (Gibco, USA).

2.3. Plasmid construction and protein purification

The construction of the expression vectors and the purification of the proteins were performed as described previously¹³. Briefly, to express HR121, HR1 and HR2 proteins, the genes encoding these proteins were cloned into E. coli expression vector pET-30a. within EcoRI and XhoI enzyme sites. Thus, these expressed proteins were generated with the redundant 52 amino acid residues containing N-terminal 6-his tags.

2.4. Protein structure simulation and prediction

The crystal structure of HR121 was acquired as previously described¹³, and the crystal structure of HR2 was acquired from PDB database with the ID number 6LVN. HR2 were paired to the ligand-binding region of HR121 using the ZDOCK 3.0.2 program. The interaction of HR121/HR2 protein complex was analyzed using discovery studio software. All the figures were generated by PyMOL program.

2.5. Binding assay

In the binding assay, HR121 protein (1 μg/mL) was coated on a 96-well polystyrene plate in coating buffer (15 mmol/L Na₂CO₃, 35 mmol/L NaHCO₃, pH 9.6) at 4 °C overnight. After removing the coating buffer, the plates were washed 3 times with PBS containing 0.05% Tween-20 (PBST) and blocked with blocking buffer (PBS containing 5% BSA) at 37 °C for 2 h. Then the plate was washed 3 times. Serially diluted HRP labeled HR2 were added to 96 well polystyrene plates coated with HR121 and incubated at 37 °C for 1 h. After washing again, the OPD matrix was added to each well. The reaction was stopped with 2 mmol/L H₂SO₄ and the optical density (OD) value of the hole was read on an ELISA reader at 490 nm/630 nm. In a separate binding test, HR2 protein in the coating buffer was added to 96 well polystyrene plate overnight at 4 °C. After removing the coating buffer, the plates were washed 3 times with PBST and blocked with the blocking buffer at 37 °C for 2 h. Then the plate was washed 3 times. Serially diluted HRP labeled HR121 was added to the coated 96-well polystyrene plate and incubated at 37 °C for 1 h. After washing again, the OPD matrix was added to each well. The reaction was stopped with 2 mmol/L H₂SO₄ and the OD value of the hole was read on an ELISA reader at 490 nm/630 nm.

2.6. Competitive ELISA

Competitive ELISA assay was performed as described previously¹³. Briefly, HR1 protein (1 μg/mL) in coating buffer at 4 °C on a 96-well polystyrene plate overnight. After removing the coating buffer, the plates were washed 3 times with PBST and blocked with blocking buffer at 37 °C for 2 h. The gradient-diluted HR121 protein was premixed with HRP-labeled HR2 protein at 37 °C for 1 h. The mixture was then added to a 96-well plate coated with HR1 protein and incubated at 37 °C for 1 h. After washing again, the OPD matrix was added to each well. The reaction was stopped with 2 mmol/L H₂SO₄ and the OD value of the hole was read on an ELISA reader at 490 nm/630 nm.

2.7. Native polyacrylamide gel electrophoresis (N-PAGE)

Continuously diluted HR121 (2-fold dilution, maximum final concentration of 7 μmol/L) was incubated with equal volume of SARS-CoV-2 HR2 (10 μmol/L) at 37 °C for 0.5 h, then HR1 was added and incubated at 37 °C for 0.5 h. The mixture was loaded on a tris-glycine gel (6%) for electrophoresis for 2 h and observed with Coomassie blue staining.

2.8. Cell–cell fusion assay

HR121 mediated inhibitory activity against SARS-CoV-2 S mediated cell–cell fusion was evaluated as described elsewhere⁷. Briefly, pcDNA3.1 vector encoding SARS-CoV-2 S protein and pHIV-EGFP vector encoding eGFP (Addgene #21373, USA) were cotransfected into 293T cells as effector cells. After 24 h, fresh 293T/SARS-CoV-2-S/EGFP cells were mixed with an equal volume of three times continuously diluted HR121 at concentrations ranging from 6000 to 8 nmol/L and cultured at 37 °C for 2 h. After incubation, the mixture of effector cells and proteins was gently transferred to the 96-well plate inoculated with 293T-ACE2 cells expressing ACE2 receptor. After co-cultured at 37 °C for 2 h, the percentage of fusion cells was calculated under the fluorescence microscope, and the inhibition concentration of 50% was calculated.

2.9. Inhibition of pseudotyped SARS-CoV-2 infection

To measure neutralize the activity of HR121 against various SARS-CoV-2 variant pseudoviruses (Supporting Information Table S1)¹³, 293T-ACE2 cells (1 × 10⁴/well) were inoculated in a 96-well plate. The next day, HR121 was three-fold serial diluted in another 96-well plate in a volume of 60 μL. Then, 60 μL SARS-CoV-2 pseudovirus [multiplicity of infection (M.O.I.) = 0.1] was added to diluted HR121 and incubated at 37 °C for 1 h. Thereafter, 100 μL mixture and 293T-ACE2 cells were cultured at 37 °C for 24 h, Renilla luciferase activity was measured using Renilla luciferase assay kit (Promega, Madison, WI, USA) and 50% inhibitor concentration (IC₅₀) value of HR121 was calculated.

2.10. Inhibition of authentic SARS-CoV-2 infection

HPAEpiC cells (1.6 × 10⁵/well) were inoculated into a 48-well plate. The next day, HR121 was placed in another 48-well plate with two-fold serial dilution at a volume of 100 μL. Then, 100 μL SARS-CoV-2 WT or BA.2 (M.O.I. = 1) was added to diluted HR121 and incubated at 37 °C for 1 h. Thereafter, the culture medium in HPAEpiC cells was removed and replaced with a mixture of virus and HR121. After incubation at 37 °C for another 1 h, the mixture was removed and washed 3 times with PBS. Then, fresh medium containing the same diluted concentration of HR121 was added. The cells were cultured at 37 °C for 48 h, then the supernatant was collected for viral RNA extraction (Roche Diagnostics, Mannheim, Germany), and the viral load was analyzed and the IC₅₀ value was calculated.

Vero-E6 cells (4 × 10⁴/well) were inoculated into a 96-well plate. The next day, HR121 protein was continuously diluted in another 96-well plate, and virus (M.O.I. = 0.1) was added and incubated at 37 °C for 1 h. Then the culture medium in the cells was removed and the virus–protein mixture was added. After 72 h, the cell activities were determined using Enhanced Cell Counting Kit-8 (Beyotime, China).

SARS-CoV-2 authentic virus neutralization assay for mice or hamster bronchoalveolar lavage (BAL) fluid was performed using the same method.

2.11. Cytotoxicity assay

Cells were added to 96-well plates with gradient dilution HR121 and incubated at 37 °C, 5% CO₂ for 48 h before adding 20 μL MTT (Sigma–Aldrich, USA). After 4 h of incubation, 100 μL of culture supernatant was removed and 100 μL of 12% sodium dodecyl sulfate (SDS)–50% N,N-dimethylformamide (Sigma–Aldrich, USA) was added. The plates were incubated at 37 °C overnight. OD was measured at 570 and 630 nm and 50% cytotoxic concentration (CC₅₀) was calculated.

2.12. Viral load measurement

RNA was extracted from lung tissue using Trizol (Takara, Japan), and RNA concentration was measured using nanodrop 2000 (Thermo Fisher Scientific, USA). RNA was extracted from the culture supernatant with Roche kit, and virus genomic RNA was measured by real-time qPCR using Q5 real-time PCR system (Life Technologies, USA). The primers used to detect gRNA were from the N gene of SARS-CoV-2, including forward: 5′-GGGGAACTTCTCCTGCTAGAAT-3′; reverse: 5′-CAGACATTTTGCTCTCAAGCTG-3′; and probe: FAM-TTGCTGCTGCTTGACAGATT-TAMRA-3′.

2.13. Animal infection assay

To test the protective effect of HR121 on SARS-CoV-2 infected male Syrian golden hamsters, the hamsters at 8 weeks old (purchased from Charles River Company, Beijing, China) were used in SARS-CoV-2 prototype strain challenge, and HR121 was dissolved in PBS. SARS-CoV-2 prototype strain was intranasally injected into hamsters at a dose of 10⁴ 50% tissue culture infectious dose (TCID₅₀) in a volume of 50 μL. Hamsters were administered with 0.5 mg/kg HR121 by nasal route at 0.5 h (n = 8), 2 h (n = 8) before infection or 0.5 h (n = 8), 2 h post-infection (n = 8) in a same volume of 50 μL. Meanwhile, the hamsters administered with PBS in the same way at 0.5 h (n = 8), 2 h (n = 8) before infection or 0.5 h (n = 8), 2 h post-infection (n = 8) were set as controls.

In the same manner, another group of hamsters was used for Omicron BA.2 variant challenge and HR121 protection examination. Briefly, Omicron BA.2 was intranasally injected into hamsters at a dose of 10³ TCID₅₀. Hamsters (n = 10) were administered with 0.5 mg/kg HR121 by nasal route at 0.5 h before infection. Hamsters (n = 10) administered with PBS were set as controls. 72 h after infection, the hamsters were euthanized and dissected, the viral load in lung tissue was measured, and lung pathology was detected.

In the multiple administration of HR121 experiment, a total of 32 hamsters were used. Hamsters (n = 16) were administered with 0.5 mg/kg HR121 by nasal route every 7 days. Seven days after the third administration, hamsters (n = 8) were administered with 0.5 mg/kg HR121 by nasal route at 0.5 h before Omicron BA.2 infection and other hamsters (n = 8) were administered with PBS before infection. Meanwhile, hamsters (n = 8) were administered with 0.5 mg/kg HR121 by nasal route at 0.5 h before infection, and hamsters (n = 8) administered with PBS before infection were set as positive and negative drug controls, respectively. Seventy-two hours after infection, the hamsters were euthanized and dissected, the viral load in lung tissue was measured, and lung pathology was detected.

2.14. Collection of hamster and mouse BAL fluid

Three hamsters were administered with 0.5 mg/kg HR121 by nasal route every 7 days. Two groups of mice, four mice each, were administered 0.5 mg/kg HR121 by nasal route every 14 and 21 days, respectively. Seven days after the third administration, each hamster or mouse was anesthetized and fixed, lying supine on an operating table. After disinfecting the neck and chest skin with 75% alcohol, the neck skin was cut and the sternum was removed, exposing the trachea. A small oblique incision was made at the proximal end of the trachea, where a medical indwelling needle was inserted approximately 1 cm deep, and the incision was ligated and secured using sutures. Next, cold PBS was injected into the lungs, and 2 mL aspirated solution was centrifuged at 800×g for 5 min to remove cell precipitates and the supernatant was collected and frozen at −80 °C for further analyses.

2.15. Detection of viral antigen-specific IgG and IgA in mice BAL fluid samples by ELISA assay

HR121 protein (1 μg/mL) was coated with coating buffer (15 mmol/L Na₂CO₃, 35 mmol/L NaHCO₃, pH 9.6) on a 96-well polystyrene plate at 4 °C overnight. After removing the coating buffer, the plates were washed 3 times with PBST and blocked with blocking buffer (PBS containing 5% BSA) at 37 °C for 2 h. Then the plate was washed 3 times and BAL fluid samples were added. After incubating at 37 °C for 2 h, the plate was washed and anti-mouse IgG or anti-mouse IgA antibody linked with HRP was added. After incubation at 37 °C for another 1 h, the plate was washed again and the OPD matrix was added to each well. The reaction was stopped with 2 mmol/L H₂SO₄ and the OD value of the hole was read on an ELISA reader at 490 nm/630 nm.

2.16. Histological examination

The tissue was fixed in 4% paraformaldehyde, dehydrated with graded ethanol, and embedded in paraffin. Then paraffin sections were dewaxed, rehydrated, and stained with hematoxylin&eosin.

2.17. Statistical analysis

All statistical analyses were performed using GraphPad Prism 9.0.0. P values were labeled in the figures. Mann–Whitney test was used to compare the difference of viral load in lung tissue between each group. P value below 0.05 was considered significant.

3. Results and discussions

3.1. HR121 can bind with HR2 with high affinity

In the previous study, using the full amino acid sequence of the HR fragment of SARS-CoV-2 reference strain (NC_045512.2), we linked HR1, HR2 and HR1 together to construct the recombinant protein HR121¹³. The purified HR121 protein with his tags showed a single band in SDS-PAGE, which was consistent with its theoretical molecular weight of 28 kDa. Western blotting analysis confirmed its expression (Fig. 1A). In addition, the structure of the recombinant protein was analyzed, X-ray crystal diffraction showed that the recombinant protein formed a stable non-symmetric 6-HB structure which consisted of 4 parallel HR1s and 2 antiparallel HR2s, resembling the conformation of fusion intermediate conformation during SARS-CoV-2 infection¹³.

Figure 1.

HR121 can bind to HR2. (A) The expressed and purified HR121 was verified by Western blotting and SDS-PAGE analysis. (B) The predicated model of HR121/HR2 protein complex. HR121 colored in purple and HR2 in yellow. (C) The hydrophobic groove between HR121 and HR2. The hydrophobic surface of HR121 is presented. HR2 protein is shown in cartoon. (D) The detailed interactions between HR121 and HR2. Amino acids involved in the hydrogen bond are indicated and shown in enlarged figure. HR121 colored in purple and HR2 in yellow. The binding of HR2 to HR121 (E) and the binding of HR121 to HR2 (F) were determined by ELISA. In (E) and (F), data are presented as mean ± SD, n = 3; EC₅₀ means 50% effective concentration.

Our previous studies have shown that HR121 can selectively bind to HR2, but not HR1¹³. Here, we analyzed the protein complex structure of HR121/HR2 using discovery studio software and found that HR121 showed a six-α-helical structure, and it can interact with HR2 to form a coiled-coil complex (Fig. 1B). HR121 displayed an extended hydrophobic groove via two of its four central HR1 helices, which bind to HR2 (Fig. 1C). The residues involving the interaction included N122, L125, Q129, A133, K136, A137, K140, T144, K150, Q323, S327, A330, S331, Q337, D338, and N341 in the hydrophobic groove of HR121, and I1, N5, V8, V9, Q12, I15, D16, L18, N19, A22, L25, L29 and Q33 in HR2 (Fig. 1D). We also observed a series of hydrogen-bond interactions between HR121 and HR2. These include residues Q129, K140, Q323 and S327 from HR121, and corresponding the HR2 amino acids A22, Q12, N19 and D16 (Fig. 1D). Their binding sites are like those observed in the 6-HB structures formed by HR1 and HR2 during SARS-CoV-2 infection, and provide structure base for HR121 binding to HR2, thus inhibiting virus infection.

To examine it, we first assessed the affinity between HR2 and HR121 using a two-way ELISA⁷. When diluted HR2 was added to the microplate coated with HR121, it can bind to HR121 in a dose-dependent manner, the mean 50% effective concentration (EC₅₀) of HR2 was 21.63 nmol/L (Fig. 1E). Conversely, while diluted HR121 was added to the microplate coated with HR2, HR121 also showed a high-avidity binding to HR2, the EC₅₀ value of HR121 was 0.1641 nmol/L (Fig. 1F). These findings demonstrate that the affinity between HR121 and HR2 is specific, analogous to the binding of HR1 and HR2 that forms the 6-HB structure.

3.2. HR121 can block the formation of 6-HB in an acid insensitive manner

Next, we examined whether HR121 can competitively bind HR2 and block the formation of 6-HB between HR1 and HR2. As previously reported⁷, we premixed HR121 at different concentrations with HR2 at 37 °C for 30 min, then added HR1 (the same amount with HR2) into the mixture for 30 min and applied them to native electrophoresis. Consistent with previous reports⁷, the free HR1 with net positive charge migrated upwards, while the free HR2 with net negative charge migrated downwards. The bands of density HR1/HR2 complex (6-HB structure) increased with the decrease of HR121 concentration (Fig. 2A). In addition, we also conducted a competitive ELISA in a similar way with those of native electrophoresis. It was found that the HR121 interacted with HR2 to block the binding of HR2 and HR1, with IC₅₀ of 44.43 nmol/L, while the IC₅₀ of free HR1 blocking the formation 6-HB was 47.75 nmol/L (Fig. 2B). As SARS-CoV-2 can enter the cells through receptor-mediated endocytosis, and during this process, the virus particle undergoes several changes in pH reduction, ranging from 7.5 to 4.5, the low pH of the endosome may interfere with the binding between the HR2 of SARS-CoV-2 and HR121, thus affecting the effect of HR121 when virus enters via the endosomal route. We therefore studied the effect of pH on the role of HR121 in blocking the formation of 6-HB structure. The results show that HR121 could block the formation of 6-HB structure at pH 4.5–6.0, and the effect was not attenuated compared with in neutral pH. The IC₅₀ were 20.06, 31.44, 32.32, and 7.321 nmol/L respectively (Fig. 2C). These results indicate that HR121 can effectively compete with HR1 to bind HR2 and block the formation of 6-HB fusion core in an acid insensitive manner.

Figure 2.

HR121 can block the formation of six-helical bundle (6-HB) structure between HR1 and HR2 in an acid-insensitive manner. (A) HR121 mediated disruption of 6-HB formation between HR1 and HR2. The intensity of HR1/HR2 complex band decreased with the increasing concentration of HR121 (n = 3). (B) HR121 mediated inhibition of the formation of 6-HB between HR1 and HR2 as shown by ELISA (n = 3). (C) Effect of different pH on HR121 blocking the formation of 6-HB structure (n = 3). Data are presented as mean ± SD. IC₅₀ means 50% inhibitory concentration.

3.3. HR121 inhibited the membrane fusion mediated by spike protein and prototype SARS-CoV-2 pseudovirus entering the host cells

Next, we evaluated the inhibitory activity of HR121 against SARS-CoV-2 Spike-mediated cell–cell fusion. As shown in Fig. 3A and B, HR121 potently inhibited cell–cell fusion in dose-dependent manner with an average IC₅₀ value of 97.53 nmol/L. Subsequently, we examined the inhibitory activity of HR121 against SARS-CoV-2 pseudovirus entering the host cells. As shown in Fig. 3C, HR121 effectively inhibited SARS-CoV-2 pseudovirus entering in 293T-ACE2 cells with an average IC₅₀ value of 436.3 nmol/L. The inhibitory activity of HR121 against SARS-CoV-2 Spike-mediated cell–cell fusion and pseudovirus infection was comparable to those of the previously reported coronavirus inhibitors EK1 and 5-helix^7,12.

Figure 3.

HR121 can inhibit cell–cell fusion mediated by S protein and has inhibitory activity against SARS-CoV-2 pseudovirus. (A) Fluorescent photograph of SARS-CoV-2 S mediated cell–cell fusion in the presence of HR121. (B) Statistical analysis of HR121 inhibitory activity of cell fusion. (C) The inhibitory activity of HR121 on SARS-CoV-2 pseudoviruses. In (B) and (C), data are presented as mean ± SD, n = 3; IC₅₀ means 50% inhibitory concentration.

3.4. HR121 exhibited potent inhibitory activity against authentic SARS-CoV-2 prototype strain

To examine the inhibitory activity of HR121 against the authentic SARS-CoV-2, the HPAEpiC and Vero-E6 cells were infected with SARS-CoV-2 prototype strain, and viral replication were determined by the cell activity of the Vero-E6 cells and the viral particle shedding in the supernatant of the HPAEpiCs respectively. The results show that HR121 significantly inhibited SARS-CoV-2 replication in Vero-E6 (Fig. 4A) and HPAEpiC cells (Fig. 4B) with IC₅₀ of 2.281 μmol/L and 434.8 nmol/L, respectively. To exclude the effect of HR121 on cellular activity, we also assessed the toxicities of HR121 on Vero-E6, HPAEpiC and 293T-ACE2 cells (Fig. 4C–E). As expected, HR121 had no detectable cytotoxicities to all three cells even at the highest concentration of 20 μmol/L. These results indicate that HR121 can effectively inhibit SARS-CoV-2 replication in both HPAEpiC and Vero-E6 cells.

Figure 4.

HR121 mediated inhibition of the authentic SARS-CoV-2 prototype strain and the cytotoxicity in various cells. (A) HR121 mediated inhibition of SARS-CoV-2 replication in Vero-E6 cells (n = 3). (B) HR121 mediated inhibition of SARS-CoV-2 replication in human alveolar epithelial cells (HPAEpiCs) (n = 3). (C–E) The cytotoxicity of HR121 in Vero-E6, HPAEpiC and 293T-ACE2 cells (n = 3). Data are presented as mean ± SD. IC₅₀ means 50% inhibitory concentration.

3.5. HR121 displayed potent antiviral activity against multiple SARS-CoV-2 variants

The S proteins of VOCs and VOIs with multiple mutations have shown varying degrees of resistance to neutralizing antibodies, which results in reduced neutralizing efficacies of vaccines and sera from COVID-19 convalescents15, 16, 17, 18, 19, thus enhancing viral infectivity and transmission. Here, we examined the inhibitory activity of HR121 against previous VOCs, VOIs and current Omicron sub-lineages and the mutation sites are in Table S1. The results show that HR121 exhibited potent inhibitory activity against D614G, Alpha, Beta, Gamma, Delta, Lambda, Mu, Omicron sub-lineages BA.1, BA.2, BA.4/5, XBB and XBB.1.5 with IC₅₀ values of 230.2, 138.1, 113.5, 87.4, 679.9, 127.9, 173.9, 620.6, 377.3, 1026.0, 1833.0, and 1523.0 nmol/L, respectively (Fig. 5A–L).

Figure 5.

Inhibitory effect of HR121 to pseudo-typed SARS-CoV-2 variants. (A–J) Inhibitory activity of HR121 on D614G (A), Alpha (B), Beta (C), Gamma (D), Delta (E), Lambda (F), Mu (G), Omicron sub-lineages BA.1 (H), BA.2 (I), BA.4/5 (J), XBB (K), and XBB.1.5 (L) pseudoviruses. Three independent experiments were performed (n = 3). Data are presented as mean ± SD. IC₅₀ means 50% inhibitory concentration.

Considering the toxicity of HR121 was not detected in the cells, we infer that the cytotoxicity of HR121 is negligible and does not contribute to its inhibitory activities to viral replication (Table 1). These results suggest that HR121 is a potent inhibitor against the multiple emerging SARS-CoV-2 variants.

Table 1.

Inhibitory activity of HR121 against SARS-CoV-2 pseudovirus infection.

Pseudovirus	IC50 (nmol/L)a	CC50 (μmol/L)a	Selectivity index (SI)
WTa	436.3	>20	>45.8
D614G	230.2	>20	>86.9
Alpha (B.1.1.7)	138.1	>20	>144.8
Beta (B.1.351)	113.5	>20	>176.2
Gamma (P.1)	87.35	>20	>22.9
Delta (B.1.617.2)	679.9	>20	>29.4
Lambda (C.37)	127.9	>20	>156.4
Mu (B.1.621)	173.9	>20	>115
Omicron BA.1	620.6	>20	>32.2
Omicron BA.2	377.3	>20	>53
Omicron BA.4/5	1026	>20	>19.4

^aIC₅₀, the 50% inhibitory concentration of HR121 on HEK293T-ACE2 cells. CC₅₀, the 50% cytotoxic concentration of HR121 on HPAEpiC, Vero-E6, and HEK293T-ACE2 cells. WT, SARS-CoV-2 reference strain (NC_045512.2).

3.6. Intranasally applied HR121 showed strong protection of Syrian golden hamsters against SARS-CoV-2 infection

SARS-CoV-2 spreads through the respiratory tract and results rapid transmission among members of a population. Previous reports showed that 5-helix could significantly inhibit several VOCs and VOIs, but its protective role was not evaluated in vivo⁷. Here, we firstly investigated the potential prophylactic role of HR121 by prior intranasal administration of HR121 to the SARS-CoV-2 infected Syrian golden hamsters. We pre-treated two groups of Syrian golden hamsters with HR121 via nasal administration at a single dose of 0.5 mg/kg, 0.5 h (pre-0.5) or 2 h (Pre-2) before SARS-CoV-2 challenges. The viral challenges were carried out at 10⁴ TCID₅₀ through the same nasal route (Fig. 6A). HR121 with a single dose of 0.5 mg/kg exhibited promising prophylactic effects in both pre-0.5 h and pre-2 h groups. Nearly no detectable viral load (Fig. 6B) and no significant pathological changes in the lungs of HR121-pre-treated hamsters were observed. In stark contrast, the control group pre-treated with PBS (vehicle group) showed a typical feature of COVID-19 in the lung tissues, with a high viral load of 10⁴ copies/μg RNAs (median) (Fig. 6B), and typical histopathological changes of viral interstitial pneumonia, including thickened alveolar walls, and infiltration of lymphocytes and macrophages and unrecognizable architecture (Fig. 6C).

Figure 6.

The prophylactic and therapeutic effect of HR121 on hamsters infected with SARS-CoV-2. HR121 was dissolved in PBS. (A) SARS-CoV-2 was intranasally injected into hamsters at a dose of 10⁴ 50% tissue culture infectious dose (TCID₅₀). 0.5 mg/kg HR121 was administered by nasal route 0.5 h and 2 h before infection. For the control group, the same volume of PBS was given through the nose. Seventy-two hours after infection, hamsters were euthanized and dissected, the viral load in lung tissue was measured (B) and lung pathology was evaluated (C). (D) SARS-CoV-2 was intranasally injected into hamsters at a dose of 10⁴ TCID₅₀. 0.5 mg/kg HR121 was administered by nasal route 0.5 and 2 h post-infection. For the control group, the same volume of PBS was given. Seventy-two hours after infection, hamsters were euthanized and dissected, the viral load in lung tissue was measured (E), and lung pathology was evaluated (F). In (B) and (E), data are presented as median ± interquartile range (n = 8). In (C) and (F), representative pathologic figures were selected randomly. The thickened alveolar wall (blue arrow), lymphocyte infiltration (green arrow), macrophages and unrecognizable architecture (red circle) were labeled. i.n. means intranasal administration.

Then we tested the therapeutic effect of HR121 0.5 h (post-0.5 group) and 2 h (post-2 group) after SARS-CoV-2 infection (Fig. 6D). High viral titer of 10⁵ copies/μg RNA (median) was detected in control group, while only low level of viral titer (approximate 10–20 copies/μg RNAs) was detected in the lung tissue of the post-0.5 h and post-2 h groups (Fig. 6E). In addition, in post-0.5 h and post-2 h groups treated with HR121, little histopathological changes in the lung tissues were observed. However, the lung tissues of hamsters in control group showed obvious histopathological changes, similar to the counterparts observed in pre-treated controls (Fig. 6F).

3.7. HR121 can inhibit Omicron BA.2 replication both in vivo and in vitro

As Omicron sub-lineages are the dominant variants at present, and HR121 has shown potent inhibitory activities against Omicron BA.2 and BA.4/5 in the pseudovirus system, we further tested its neutralization capacity against authentic Omicron BA.2 variant. The results showed that HR121 could greatly inhibit the replication of Omicron BA.2 in HPAEpiC cells with IC₅₀ value of 383.1 nmol/L (Fig. 7A), which was similar to that against the pseudovirus (Fig. 5I).

Figure 7.

The inhibition effect of HR121 on authentic Omicron BA.2 in vitro and vivo. (A) HR121 mediated inhibition of Omicron BA.2 variant replication in human alveolar epithelial cells (HPAEpiCs) (n = 3). (B) Omicron BA.2 was intranasally injected into hamsters at a dose of 10³ 50% tissue culture infectious dose (TCID₅₀). 0.5 mg/kg HR121 was administered by nasal route 0.5 h before infection. HR121 was dissolved in PBS. For the control group, the same volume of PBS was given. Seventy-two hours after infection, hamsters were euthanized and dissected, the viral load in lung tissue was measured (C), and lung pathology was evaluated (D). In (A), data are presented as mean ± SD. In (C), data are presented as median ± interquartile range (n = 10). In (D), representative pathologic figures were selected randomly. The thickened alveolar wall (blue arrow), lymphocyte infiltration (green arrow), macrophages and unrecognizable architecture (red circle) were labelled. i.n. means intranasal administration.

We also examined the potential prophylactic effect of HR121 in Omicron BA.2 infected Syrian golden hamsters. By the same route of intranasal administration, we treated Syrian golden hamsters with a single dose of 0.5 mg/kg HR121 at 0.5 h (pre-0.5) before challenging with Omicron BA.2 (TCID₅₀ = 10³) (Fig. 7B). HR121 showed a promising protective effect against Omicron BA.2 infection. The pulmonary viral loads in HR121-administrated hamsters were nearly undetectable (Fig. 7C), and there were no obvious pathological changes in the lung tissues compared with the control group (Fig. 7D).

3.8. Repeated nasal administration of HR121 did not induce mucosal antibodies to suppress its antiviral activity

Our previous study found that subcutaneous immunization with HR121 protein in Freund's or aluminum adjuvant can induce antibody production¹³. Repeated use of HR121 in the nasal cavity may induce mucosal immunity and produce antibodies to neutralize HR121, weakening its antiviral effect. To investigate this issue, we administered HR121 by nasal route to two groups of hamsters three times at one-week intervals, wherein one group of hamsters were pre-treated with HR121 at 0.5 h before virus infection (group HR121 × 3 + HR121), the other group of hamsters was pre-treated with PBS at 0.5 h before virus infection (group HR121 × 3). Meanwhile, another two group of hamsters pre-treated with HR121 (group HR121) and PBS (group vehicle) at 0.5 h before virus infection were set as positive and negative drug controls, respectively (Fig. 8A). After three repeated intranasal doses of HR121, re-administration HR121 (HR121 × 3 + HR121) could still exhibit a good protective effect in the hamsters, in which a remarkable decrease in the viral loads (Fig. 8B) and few pathological changes in the lung tissues were observed (Fig. 8C), comparable to those in the positive controls (HR121). In contrast, in the hamsters treated with PBS after three doses of HR121 (HR121 × 3), no significant protective role in the viral loads (Fig. 8B) and pathological changes (Fig. 8C) were presented, similar to those in the negative controls (vehicle). Therefore, these data suggest that: 1) the inhibitory effect of HR121 cannot maintain to 7 days after the last intranasal treatment; 2) repeated administration with only HR121 via nasal route was not easy to evoke mucosal antibodies to neutralize the role of HR121. To further confirm it, we collected BAL fluid from another 3 hamsters treated with three doses of HR121 in the same method and detected no obvious neutralizing antibodies against the authentic SARS-CoV-2 Omicron BA.2 variant (Fig. 8D). We also nasally administered HR121 to two groups of BALB/c mice three times at two-week and three-week intervals, respectively, and found that there were no specific anti-HR121 IgAs and IgGs were evoked in their BAL fluid (Fig. 8E). Together, our results indicate that repeated nasal administration of HR121 does not induce mucosal antibodies to suppress its antiviral activity.

Figure 8.

Repeated administration of HR121 did not induce mucosal antibodies. (A) The hamsters were divided into 4 groups (n = 8, in each group). Two groups received nasal administration of HR121 (0.5 mg/kg per dose) three times at 7-day intervals, in which one group received an additional dose of HR121 at 0.5 h, and the other group received PBS before infection. Another two control groups received only one dose of HR121 and PBS at 0.5 h before infection, respectively. Omicron BA.2 variant was used for challenges at a tissue culture infectious dose (TCID₅₀) titer of 2 × 10³. Seventy-two hours after infection, hamsters were euthanized and dissected, the viral load in lung tissue was measured (n = 8, in each group) (B), and a representative pulmonary pathologic figure was presented (C). (D) Neutralization activity of the bronchoalveolar lavage (BAL) fluid from the hamsters administrated with three doses of HR121 (0.5 mg/kg per dose) at 7-day intervals. In vehicle group, n = 1; in HR121 group, n = 3. (E) Anti-HR121 antibody titers in the BAL fluid from the mice treated with three doses of HR121 (0.5 mg/kg per dose) at 14-day and 21-day intervals (n = 4, in each group), respectively. In (B), data are presented as median ± interquartile range. In (D) and (E), data are presented as mean ± SD. In (C), representative pathologic figures were selected randomly. The thickened alveolar wall (blue arrow), lymphocyte infiltration (green arrow), macrophages and unrecognizable architecture (red circle) were labeled. i.n. means intranasal administration.

4. Conclusions

The RBD domain in the S1 subunit of coronavirus S protein evokes the main neutralizing antibodies in vivo and serves as the prime target for the development of vaccines and therapeutic antibodies20, 21, 22, 23, 24. Unfortunately, the RBD is variable during virus evolution^6,15. This resulted in the variants of SARS-CoV-2 with enhanced capacities to escape neutralization from most of the antibodies and vaccines, and increased transmissibility and disease severity, thus posing a huge threat to public health. In contrast, the S2 subunit is shaded in the S1 subunit in a naïve state, is not easily recognized by neutralizing antibodies produced in host and shows fewer mutations⁷. Therefore, the HR1 and HR2 domains are the most conserved regions of the coronavirus S protein. Consequently, antiviral drugs targeting these domains may have a high broad spectrum^{3,11,12,25,26}. In the past several decades, the HR1 domain has proved to be an important target for the design of fusion inhibitors27, 28, 29, 30.

However, there was a report demonstrating that HR2 domain was more conserved than HR1 domain, but few peptidic fusion inhibitors were designed to target it³¹. To date, there were only some 5-helical recombinant proteins were reported to target the HR2 domain and inhibit SARS-CoV-2 and its variants in vitro^7,32. Here, we designed a recombinant protein HR121, it targets the conserved HR2 domain in the fusion intermediate conformation of S2 protein. HR121 forms a stable dimer and interacts with the HR2 domain, preventing the formation of 6-HB between HR1 and HR2 in an acid-insensitive manner, thus preventing virus spike-mediated cell–cell fusion. In addition, as expected, HR121 inhibited D614G as well as Alpha, Beta, Gamma, Mu, Delta, and Omicron sub-lineages in vitro. Meanwhile, HR121 efficiently protect Syrian golden hamster from SARS-CoV-2 prototype strain and Omicron BA.2 variant infection. These results indicate that HR121 is a promising prophylactic drug candidate against current circulating and future emerging SARS-CoV-2 variants.

Many fusion inhibitors targeting S2 proteins, such as EK1C, showed better activity than HR121 in inhibiting SARS-CoV-2, but HR121 is easily purified in a large scale, and because of its larger molecular weight and more complicated structure than those of peptides¹³, it is not easy to be degraded by extracellular catabolic enzymes, which promotes the application of HR121. Furthermore, our study demonstrated that intranasal administration of HR121 could efficiently protect Syrian golden hamsters from SARS-CoV-2 prototype strain and Omicron BA.2 variant infection.

Our previous studies have shown that subcutaneous or intramuscular immunization with HR121 protein in Freund's or aluminum adjuvant induces strong neutralizing antibodies that provide effective protection against SARS-CoV-2 infection in hACE2 mice, Syrian golden hamsters and rhesus macaques¹³. Recently, a large number of studies have demonstrated that intranasal delivery of SARS-CoV-2 vaccines with some appropriate adjuvants or viral vectors can promote mucosal immune responses³³. Due to the immunogenicity of HR121, it may induce mucosal antibodies that neutralize its antiviral activity after multiple intranasal administration. In the present study, we found that repeated nasal delivery of HR121 in hamsters and mice induced few mucosal antibodies that can impair its antiviral effect. This result was in line with that of RBD vaccine, intranasal immunization with only RBD protein failed to elicit detectable RBD-specific antibodies in the sera and BAL fluid of the vaccinated mice³⁴. Therefore, repeated administration of HR121 in the nasal cavity is feasible to prevent the emerging SARS-CoV-2 variant infections. In addition, the serum from breakthrough infection patients combined with the fusion inhibitor EK1 could produce a potent synergistic effect against several current SARS-CoV-2 variants in vitro³⁵, thus in parallel, HR121 can be developed as a potential fusion inhibitor in the uninfected and convalescent populations, and combined with some alternative vaccines, such as RBD-based vaccines. However, given that the HR121 protein induces robust antibodies after subcutaneous injection, which may affect its antiviral effect at the mucosal sites, it may not be appropriate to apply HR121 as a fusion inhibitor and a vaccine together. In the following study, based on the structure of HR121, we will continue to optimize its sequence to obtain more derivatives with better antiviral activity.

Acknowledgments

We thank Kunming National High-level Biosafety Research Center for Non-human Primates for providing the experimental platform and professor Songying Ouyang for their work in protein structure analysis. This work was supported in part by grants from the National Natural Science Foundation of China (82151218, 81971548), National Key Research and Development Program of China (2021YFC2301703, 2021YFC2301303, 2022YFC2303700), Yunnan Key Research and Development Program (202103AC100005, 202103AQ100001, 202102AA310055, China) and CAS “Light of West China”.

Author contributions

Wei Pang and Yong-Tang Zheng conceived and designed the study. Ying Lu, Fan Shen, Wen-Qiang He, An-Qi Li, Ming-Hua Li, and Xiao-Li Feng performed the experiments. Ying Lu and Fan Shen analyzed the data. Wei Pang, Ying Lu, and Yong-Tang Zheng prepared the manuscript.

Conflicts of interest

Ying Lu, Fan Shen, Wen-Qiang He, Yong-Tang Zheng, and Wei Pang are listed as inventors on the Chinese patent related “A recombinant fusion protein derived from HR domains in S2 subunit of SARS-CoV-2 and its application (authorization number: ZL202111167024.2)” to this work. Other authors declare that they have no competing interests.

Appendix A. Supplementary data

The following is the Supplementary data to this article.