Broad-spectrum Delta-BA.2 tandem-fused heterodimer mRNA vaccine delivered by lipopolyplex

Since the beginning of the coronavirus disease 2019 (COVID-19) pandemic, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of COVID-19, continues to mutate and generates new variants with increasingly severe immune escape, urging the upgrade of COVID-19 vaccines. Here, based on a similar dimeric RBD design as our previous ZF2001 vaccine, we developed a novel broad-spectrum COVID-19 mRNA vaccine, SWIM516, with chimeric Delta-BA.2 RBD dimer delivered by lipopolyplex (LPP). Unlike the popular lipid nanoparticle (LNP), this LPP-delivered mRNA expresses only in the injection site, which avoids potential toxicity to the liver. We demonstrated the broad-spectrum humoral and cellular immunogenicity of this vaccine to Delta and Omicron sub-variants in naïve mice and as booster shots. When challenged with Delta or Omicron live virus, vaccinated human angiotensin-converting enzyme (hACE2) transgenic mice and rhesus macaques were both protected, displaying significantly reduced viral loads and markedly relieved pathological damages. We believe the SWIM516 vaccine qualifies as a candidate for the next-generation broad-spectrum COVID-19 vaccine.


Introduction
The coronavirus disease 2019 (COVID-19) pandemic has led to over 7 million deaths and countless economic losses worldwide (https://www.who.int/).Building herd immunity through vaccination is the most efficient method for preventing the spread of viruses [1][2][3].However, over three years since the beginning of the pandemic, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) continues to mutate, bringing new variants with increasing capabilities of immune escape, such as the variants of concern (VOCs) and variants of interest (VOIs) defined by the World Health Organization [4][5][6][7].As a result, COVID-19 vaccines based on immunogens from the SARS-CoV-2 prototype (PT) have gradually become ineffective against the prevalent Omicron sub-variants [8][9][10], which requires the development of a new generation of broad-spectrum vaccines to protect against the current and future SARS-CoV-2 variants.To date, several vaccines targeting Omicron sub-variants have been approved, including bivalent mRNA vaccines and XBB.1.5monovalent vaccines by Moderna and Pfizer/BioNTech, an mRNA vaccine SY6006 by CSPC Pharmaceutical Group and two multivalent protein subunit vaccines SCTV01E (Shenzhou Cell) and Coviccine (WestVac Biopharma).These vaccines, containing immunogens from BA.1, BA.5 or XBB.1.5,have displayed potent efficacy against Omicron sub-variants as boosters [11][12][13][14][15].
Previously, our group developed a universal tandem dimeric RBD design of a vaccine against betacoronavirus, which led to the world's first COVID-19 protein subunit vaccine, ZF2001, that has been approved in China, Uzbekistan, Indonesia, and Columbia [16][17][18].Recently, we demonstrated that as with the immunogen of protein subunit and mRNA vaccines, a chimeric RBD dimer design with a Delta RBD and an Omicron (BA.1)RBD could elicit broad-spectrum immunogenicity against multiple SARS-CoV-2 variants in naïve mice and mice prime-vaccinated with inactivated vaccines [19][20][21].However, with the shift of circulating Omicron sub-variants, the RBDs of immunogens should be updated accordingly.
The delivery system is crucial for mRNA vaccines, as it casts a great impact on the location and efficiency of in vivo mRNA expression [22][23][24][25].To date, the most popular delivery system for mRNA vaccines is lipid nanoparticle (LNP) which has been used in Moderna's mRNA-1273 and BioNTech's BNT162b2 [26,27].However, the currently approved LNP formula tends to deliver mRNA into the liver, which poses the potential risk of liver toxicity to vaccinees [24].Alternative to LNP, a core-shell structured lipopolyplex (LPP) is also an effective way to deliver mRNA vaccine [28,29].Compared with LNP, LPP mostly delivers mRNA to be expressed at the injection site after intermuscular inoculation instead of liver cells and displays adjuvant activity by activating the Toll-like receptor signaling pathway in dendritic cells, which improves the safety and immunogenicity of LPP-delivered vaccine simultaneously [28].
In the present study, we provide a novel mRNA vaccine (named SWIM516) with a Delta-BA.2chimeric RBD dimer delivered by LPP (Fig 1A).We evaluated the broad-spectrum immunogenicity of the SWIM516 vaccine to SARS-CoV-2 PT, Delta and Omicron sub-variants in naïve mice and mice prime-vaccinated with inactivated or protein subunit vaccine.We also demonstrated the protection against live viruses in mice (Delta and BA.1) and rhesus macaques (BA.1) by SWIM516.This vaccine demonstrated a great potential to protect against naked mRNA and the LPP-encapsulated mRNA vaccine could express the Delta-BA.2RBD dimer as a single band at its theoretical size in a dose-dependent manner, indicating the integrity of the immunogen translated in vivo (Fig 1E).The expression location of the LPP-delivered SWIM516 vaccine in mice was visualized by in vivo imaging of the bioluminescence of mRNA-expressed luciferase.Our results indicated that, unlike LNP, LPP-delivered mRNA expression was concentrated in the muscle instead of partially expressing in the liver, and the expression can last approximately 48 hours (Fig 1F ).

Evaluation of the immunogenicity of the SWIM516 vaccine
To examine the immunogenicity of the SWIM516 vaccine, we immunized groups of C57BL/6 mice (N = 10) twice (10 μg/dose) at a 14-day interval and collected serum samples as well as splenocytes 14 days after the second dose ( day  SARS-CoV-2 variants, including the Omicron sub-variants BA.5 and BF.7 [30].Notably, although these serum samples contained significantly higher titers of antibodies specific to the RBDs of the PT compared to those targeting the Omicron sub-variants, they demonstrated superior efficacy in neutralizing the pseudotyped viruses of Omicron sub-variants, especially BA.2 and its derivatives.(Fig 2B and 2C).
In addition to humoral immunogenicity, the cellular immunogenicity of the SWIM516 vaccine was also evaluated with splenocytes.After re-stimulation with peptide pools of different RBDs (PT, Delta, BA.1, or BA.2), enzyme-linked immune absorbent spot (ELISpot) assay showed that, with SWIM516 vaccination, the number of IFN-γ + spots in all four groups dramatically increased to a higher level (Fig 2D).Interestingly, the T cell activation levels after stimulation by different peptide pools were almost identical.Using in silico analysis, we predicted multiple conserved peptides in the RBDs of PT, Delta, BA.1 and BA.2 that could have high binding affinity with MHC class I (S2 Fig) , indicating SWIM516 could elicit a broadspectrum T cell response to multiple SARS-CoV-2 variants.The intracellular cytokine staining (ICS) assay also consistently demonstrated significantly increased percentages of IFN-γ + and TNF-α + populations in both CD8 + and CD4 + cells after re-stimulation with the BA.2 RBD peptide pool (Fig 2E).These data verified the cellular immunogenicity of the SWIM516 vaccine in response to both BA.1 and BA.2.
Notably, LPP-delivered mRNA vaccine exhibited comparable humoral and cellular immunogenicity as LNP-delivered mRNA vaccines.By delivering a mRNA vaccine using LNP or LPP in a side-by-side comparison experiment, neutralization and ELISpot assays demonstrated similar levels of NAbs and similar numbers of IFN-γ + spots, respectively (S1 Fig).

Evaluation of the immunogenicity of SWIM516 as boosters
Based on the evaluation in naïve mice, we further designed a prime-boost experiment to assess the efficacy of SWM516 vaccines as booster shots following the inactivated vaccines (IV) or protein subunit vaccines (PV), which reflects the vaccination status in China and many other countries (Fig 3A).Groups of C57BL/6 mice (N = 8) were prime-vaccinated with two or three doses of IV (0.65 U or 2.6 U/dose) or PV (2.5 μg or 10 μg/dose) at a 14-day interval, and boosted with the SWIM516 mRNA vaccine (RV, 2.5 μg or 10 μg/dose) 14 days later (Fig 3B and 3C).IV or PV were also vaccinated in parallel as controls (Fig 3C).Serum samples and splenocytes were collected 14 days post the last dose (day 49 or day 63) and subjected to neutralization assays and ELISpot assays, respectively.
Neutralization assay results showed that the levels of NAbs against Delta or BA.2 pseudotyped viruses in groups vaccinated with SWIM516 as the booster shot were higher than that in groups with IV or PV boosters (Fig 3D).Compared with three doses of IV (2.6 U/dose, GMT:73), a third dose of SWIM516 (10 μg, GMT:1464), following two doses of IV, elicited 20-folds higher BA.2-specific NAbs.Similarly, a third dose of SWIM516 (10 μg, GMT:260), following two doses of PV, lead to an 18.2-fold increase of BA.2-specific NAbs compared with three doses of PV (10 μg/dose, GMT:4741) (Fig 3D).Notably, the SWIM516-elicited NAbs demonstrated a clear dose-dependent trend in the three-dose groups, as the high-dose groups (10 μg SWIM516) displayed significantly higher Delta-or BA.2-specific NAbs than the lowdose groups (2.5 μg SWIM516) (Fig 3D).Interestingly, a fourth dose of low-dose SWIM516 (2.5 μg) elicited higher levels of NAbs than a fourth dose of IV or PV, but a high-dose SWIM516 (10 μg) as the fourth dose resulted in significantly reduced NAbs levels against both Delta and BA.2 compared with the low-dose group (Fig 3D).This result is consistent with our prior observation that exceedingly high-dose vaccination can lead to reduced level of NAbs [16], possibly due to immune tolerance generated by repeated vaccination [31].The mechanism of such a phenomenon deserves further study.Additionally, the cellular immunogenicity of SWIM516 as booster shots was also evaluated by ELISpot assays.Similar to NAbs levels, the numbers of IFN-γ + spots after re-stimulation of Delta or BA.2 RBD peptide pools were also significantly increased by switching the third or fourth dose from IV or PV to SWIM516 (Fig 3E).These data demonstrated that SWIM516 could be a better choice as booster shots than the PT-based inactivated vaccine or protein subunit vaccine.

Evaluation of the protective efficacy of SWIM516 in mice and rhesus macaques
To further evaluate the protective efficacy of the SWIM516 vaccine, we challenged immunized mice with the SARS-CoV-2 virus using previously established mouse models transiently or stably expressing human angiotensin-converting enzyme (hACE2) [32,33].For the transient hACE2 mice model, we immunized C57BL/6 mice twice (10 μg/dose) on days 0 and 14, transduced them with recombinant type 5 adenovirus expressing hACE2 and challenged them with 1×10 5 TCID 50 of BA.1 or Delta variant via the intranasal route.Lung, trachea, and turbinal tissues were harvested three days post-challenge (Fig 4A).High levels of both Delta and BA.1 virus were detected in all tissues of unvaccinated mice.In comparison, the vaccinated group showed four and two orders of magnitude lower levels of Delta and BA.1 in lung tissues, respectively (Fig 4B).Notably, in tracheal tissues, vaccination also reduced the BA.1 viral load by approximately 1000-fold, which is a significant finding as Omicron was reported to primarily replicate in the respiratory tract [34].
For hACE2 transgenic mice, two different doses of vaccine were injected to investigate whether the protection against BA.1 or Delta was dose dependent.Mice immunized twice with 3 μg or 10 μg per dose were challenged intranasally with 1×10 5 TCID 50 of BA.1 or Delta variant.Lung and tracheal tissues were collected five days post-challenge (Fig 4C).Titration of viral loads showed that, after challenging by either Delta or BA.1, vaccination led to a two to four orders of magnitude decrease in viral load in both lung and tracheal tissues, but a higher dose of vaccination did not lead to significantly decreased viral load compared to the low dose group, indicating that low dose could provide saturated protection (Fig 4C).In addition, the pathological analysis also displayed markedly relieved pathological damage and inflammation caused by SARS-CoV-2 infection in vaccinated mice, despite existing signs of cellular infiltration (Fig 4D and S1 Table).Together, these results verified the protection against both Delta and Omicron variants by the SWIM516 vaccine.

Discussion
In the past years, the continuously emerging Omicron sub-variants displayed escalated capability of immune escape, which has gradually rendered the widely administered PT-based vaccines ineffective, urging the development of a new generation of broad-spectrum COVID-19 vaccines to protect against a broad range of SARS-CoV-2 variants for the control of COVID-19 pandemic worldwide [4][5][6][8][9][10]35,36].Moreover, as the PT-based inactivated vaccine (e.g.BBIBP-CorV) and protein subunit vaccine (e.g.ZF2001) have been administered in many countries, it is also important to examine the immunogenicity of new COVID-19 vaccines as booster shots following these earlier vaccines, because pre-existing immunity is known to affect vaccinees' immune response to booster vaccines [21,37].Currently, the widely administered mRNA vaccines utilize LNP as the delivery system.However, LNP-encapsulated mRNA vaccines have been reported to express in liver cells, which brings potential side effect caused by liver-toxicity and limits the application of the mRNA vaccines [24].The reason for this effect can partially be attributed to the interaction of LNP with apolipoprotein E (apoE) that enhances LNP entry into primary hepatocytes through the low-density lipoprotein (LDL) receptor (LDLR) [38].To address this problem, the LPPdelivery system provides a solution.With a core-shell structure markedly different from that of LNP, LPP manages to deliver mRNAs to be expressed mostly in the muscle of the injection site with only a small proportion expressing in spleen and lymph node, as demonstrated by in vivo and ex vivo fluorescent images [29,39] (Fig 1F).Consequently, the LPP-encapsulated vaccines may have the advantage to the safety of mRNA vaccine by avoiding liver toxicity.So far, the safety of LPP has been verified by SW-BIC-213, an LPP-delivered mRNA vaccine evaluated in clinical trials and granted EUA in Laos [40][41][42].Moreover, LPP-delivered mRNA vaccine could elicit similar humoral and cellular immunogenicities as LNP-delivered ones in side-byside comparisons, suggesting LPP can be an alternative to LNP (S1 Fig) [24].
Here, we provide a new LPP-delivered COVID-19 mRNA vaccine, SWIM516, with chimeric Delta-BA.2RBD dimer as immunogen based on our dimeric RBD design [16,[19][20][21].The mRNA expression of this vaccine is limited in the injection site, highlighting its potential advantage in safety.The SWIM516 vaccine expands our lineage of tandem repeat RBD-dimer designs that includes the protein subunit vaccine ZF2001 [16][17][18], the chimeric Delta-BA.1 protein subunit vaccine [19] and Delta-BA.1 RBD-dimer mRNA vaccine [20,21].As expected, the SWIM516 vaccine demonstrated broad-spectrum immunogenicity to the SARS-CoV-2 PT, Delta VOCs and a wide range of Omicron sub-variants in mice including BF.7 [30], as well as high protective efficacy in mice and rhesus macaques against the challenge of live SARS-CoV-2 virus.SWIM516 is also effective as a booster shot following PT-based inactivated or protein subunit COVID-19 vaccines, making it suitable for the population in China and many other countries.
Although we were unable to conduct a viral challenge with the BA.2 live virus due to limited resources, we found that the levels of neutralizing antibodies against BA.2 and BA.1 elicited by the SWIM516 vaccine were comparable, suggesting that SWIM516 could provide similar levels of protection against BA.2 and BA.1.In addition, despite markedly relieved pathological damage signs of cellular infiltration can still be observed in the lungs of mice and rhesus macaques vaccinated with SWIM516.
In summary, we developed a new COVID-19 mRNA vaccine with the RBDs of Delta and BA.2 as an immunogen and demonstrated its broad-spectrum immunogenicity to SARS-CoV-2 PT, Delta, and multiple Omicron sub-variants, as well as its protection against live virus in mice (Delta and BA.1) and rhesus macaques (BA.1).This vaccine demonstrated a great potential to protect against a wide range of Omicron sub-variants, including BA.2, and BA.5, BF.7.We believe that it qualifies as a candidate for next-generation broad-spectrum COVID-19 vaccines.

Ethics statement
The animal experiments conducted in the Institute of Microbiology, Chinese Academy of Sciences (IMCAS) were in strict accordance with the recommendations described in the Guide for the Care and Use of Laboratory Animals of IMCAS Ethics Committee (approval number: APIMCAS2022124).The SARS-CoV-2 challenge experiments on mice and macaques were conducted under the approval from the Animal Ethics Committee of the Wuhan Institute of Virology (WIV), Chinese Academy of Sciences (CAS) (approval number: WIVA42202002-01) according to the National Guidelines on Animal Work in China.

Animals
C57BL/6 mice (female, 6-8 weeks) were purchased from Beijing Vital River Animal Technology Co., Ltd (licensed by Charles River) and were housed in a specific-pathogen-free (SPF) mouse facilities in IMCAS with temperature-, humidity-and light cycle-control (20 ± 2˚C; 50 ± 10%; light, 7:00-19:00; dark, 19:00-7:00).Rhesus macaques (9 females and 9 males, 5~8 years-old) were purchased from Hubei Topgene Biotechnology Co., Ltd.All macaques are in good health and are not involved in other experimental procedure.These macaques were housed in the laboratory animal facilities in Wuhan Institute of Virology (WIV), Chinese Academy of Sciences (CAS) for immunization and SRAS-CoV-2 challenge.All macaques were allowed free access to water and standard diet and provided with a 12-hour light and dark cycle (temperature: 21.0-24.3˚C,humidity: 46.5%-100%).SARS-CoV-2 challenge and neutralizing antibody titration assays were performed with approval under Biosafety Level 3 (BSL3) and ABSL3 conditions by the Institutional Biosafety Committee of WIV.

Preparation of mRNA
Messenger RNA of SWIM516 vaccine was transcribed in vitro using linearized plasmid encoding Delta-BA.2RBD dimer as template.The Delta-BA.2RBD dimer was flanked by the 5' and 3' untranslated regions and a 120 nt poly-A tail.In vitro transcription (IVT) and 5'-capping were conducted in one step using one-step mRNA IVT kit (Hongene Biotech, China).Fifty percent of UTP was substituted with 1-methylpseudourine-5'-triphosphate during transcription.Then, mRNA was treated with Antarctic Phosphatase to remove residual 5'-triphosphates (37˚C, 30 min) and purified using column chromatography.Capped mRNA was dissolved in 25 mM sodium acetate (pH 5.2) and stored at -80˚C until use.

Encapsulation of mRNA by LPP or LNP
For LPP encapsulation, a core complex was first prepared by mixing protamine dissolved in 25 mM sodium acetate (pH 5.2) with mRNA diluted in 10 mM citrate buffer (pH 4.0) in a volume ratio of 1:5 (protamine: mRNA), and then incubated for 30 min at room temperature.Next, ionizable lipid, DOPE (Avanti Polar Lipids, USA), cholesterol (A.V.T Pharmaceutical, China) and mPEG-DMG (Avanti Polar Lipids, USA) were dissolved in ethanol using previously optimized molecular ratio [39], and then combined with the core complexes at a ratio of 3:1 using Inano E platform (Micro & Nano Technology Inc, China).For LNP encapsulation, ionizable cationic lipid SM102, DSPC, cholesterol and DMG-PEG2000 purchased from A.V.T (A.V.T Pharmaceutical, China) were mixed at a ratio of 50:10:38.5:1.5 (mol/mol), and then combined with mRNAs diluted in sodium acetate at pH 4.0 using the Inano E platform (Micro & Nano Technology Inc, China).Finally, the LNPs or LPPs were dialyzed against PBS (pH 7.4), concentrated using ultra centrifugal filters (EMD Millipore, USA), filtered through a 0.22-μm filter and stored at 4˚C until further use.

In vitro characterization of LPPs
LPPs were diluted into PBS for determining size distribution and zeta potential.Zeta potentials were measured by diluting LPPs into folded capillary zeta cell at pH 4.3 or pH 7.2 and loaded into Zetasizer Pro (Malvern Panalytical, UK).Particle size distribution was measured with dynamic light scattering.For cryo-electron microscopy, LPPs were transferred onto a glowdischarged ultrathin carbon-coated copper grid, blotted for 2 s with filter paper in FEI Vitrobot Mark IV (Thermo Fisher Scientific, USA), followed by quick plugging into liquid ethane.Frozen grids were loaded into a Talos transmission electron microscope (Thermo Fisher Scientific, USA) equipped with a field emission gun operated at 200 kV.Images were recorded on a direct electron detector (ED20).For in vivo imaging, mRNAs encoding luciferase were encapsulated by LNP or LPP and injected in both legs of BALB/c mice (10 μg/mouse).Bioluminescence image of mice was captured using IVIS Spectrum Imaging System (PerkinElmer, USA) at 6, 12, 24 and 48 h post-injection.

mRNA transfection and Western blot
HEK293T cells were transfected with TransIT-mRNA (Mirus Bio, USA).Basically, for naked mRNA, 1 μg was added to 100 μl serum-free Opti-MEM together with TransIT-mRNA reagent (2 μl) and booster reagent (2 μl).The complex was incubated for 3 min before added dropwise to 5 × 10 5 cells cultured in complete medium in 12-well plate.For LPP encapsulated mRNA vaccine, 1 μg was added directly in complete medium in 12-well plate.Supernatants were collected 36 h post transfection and stored at -20˚C until use.For western blot, supernatant samples were combined with loading buffer with dithiothreitol, separated by 10% SDS-PAGE and transferred to Polyvinylidene difluoride (PVDF) membrane using a semi-dry apparatus (WIX Technology, China).Then, the membrane was blocked with 5% non-fat milk diluted in TBS-T buffer, blotted with SARS-CoV-2 Spike/RBD primary antibody (Sino Biological, China) for 1h and goat anti-rabbit IgG-HRP secondary antibody (Easybio, China) for 1 h.Finally, the membrane was developed using Beyotime BeyoECL Plus (Beyotime Biotech, China).

Evaluation of the immunogenicity
All vaccines were immunized by injecting female mice aged 6-8 weeks via the i.m. route.For evaluating the immunogenicity in naïve mice, groups of female C57BL/6 mice (n = 10) were immunized with two doses of mRNA vaccine (10 μg/mouse per dose) or PBS as sham on day 0 and day 14.Blood samples were collected on day 28, splenocytes were prepared on day 28 immediately after sacrifice.For evaluating the immunogenicity of booster shots, groups of female C57BL/6 mice (n = 8) were immunized with three shots (day 0, day 21 and day 35) or four shots (day 0, day 21, day 35 and day 49) of inactivated, protein subunit or mRNA vaccines.The low/high dose per mouse for inactivated, protein subunit and mRNA vaccines are 0.65 U/2.6 U, 2.5 μg/10 μg and 2.5 μg/10 μg, respectively.Blood samples were collected on day 49 (three dose) and day 63 (four dose), splenocytes were prepared at the same day as blood samples immediately after sacrifice.For the side-by-side comparison of LNP and LPP, groups of female BALB/c mice (n = 5) were immunized with two shots (5 μg/mouse per dose) of PT-PT RBD dimer mRNA vaccine with a 14-day interval.Spleens and blood samples were collected on day 21 and day 28, respectively.All blood samples were further centrifuged and the serum in supernatants were stored at -80˚C until use.All splenocytes were homogenized with a tissue grinder in 1 ml of serum-free DMEM, filtered with a 40 μm cell strainer (Corning, USA), followed by lysis of red blood cell with red blood cell lysis buffer (Solarbio Life Science, China).Splenocytes were counted using Celldrop FL automated cell counter (DeNovix, USA) after staining with 0.4% trypan blue solution.Live splenocytes were then immediately used for Intracellular cytokine staining (ICS) assay and Enzyme-linked immunospot (ELISpot) assay.

Evaluation of protective efficacy in mice
For transient model, C57BL/6 mice (female, 6-8 weeks, n = 5) were immunized with two doses of mRNA vaccine (10 μg/mouse per dose) or PBS as sham on day 0 and day 14, followed by infection of recombinant type 5 adenovirus expressing hACE2 via i.n.route at 42. Viral challenge with SARS-CoV-2 Delta or BA.1 virus (1×10 5 TCID 5 ) were conducted five days later on day 47.On day 54, lung, trachea and turbinal tissues of each mouse were harvested for examination of viral load.For stable model, human ACE2 transgenic mice (female, 6-8 weeks, n = 5) were immunized with two doses of mRNA vaccine (3 μg or 10 μg/mouse per dose) or PBS as sham on day 0 and day 14, followed by viral challenge with SARS-CoV-2 Delta or BA.1 virus (1×10 5 TCID 5 ) on day 33.Lung and trachea were collected on day 38 for examination of viral load and pathological damages.

Evaluation of protective efficacy in rhesus macaques
Rhesus macaques (n = 6) were immunized with clinical-grade SWIM516 mRNA vaccine (Low dose: 25 μg/animal per dose as; High dose: 45 μg/animal per dose) or Placebo on day 0 and day 21, followed by collection of blood samples on day 28, day 35 and day 42.Viral challenge with SARS-CoV-2 BA.1 virus (1×10 5 TCID 5 ) were conducted on day 42 via endotracheal intubation.Seven days post infection (day 49), animals were euthanized and dissected for pathological examination of lungs.Tissue samples were collected from the left lung (upper lobe, middle lobe and inferior lobe) and the right lung (upper lobe, middle lobe, inferior lobe and accessory lobe) for viral load and histopathology.

Enzyme-linked immunosorbent assay (ELISA)
The recombinant proteins of SARS-CoV-2 PT, Delta, BA.1, BA.2, BA.2.12.1, BA.2.75, BA.3, BA. 5 or BF.7 RBDs were purchased from GenScript.ELISA plates (Corning, USA) were coated overnight with each RBD (2 μg/ml) in 0.05 M carbonate-bicarbonate buffer, pH 9.6.Then, the plates were blocked in 5% non-fat milk diluted in PBS-T buffer.Serum samples were subjected to a three-fold serial dilution starting from 1:20.After adding diluted serum to each well, the plates were incubated for 1 h at 37˚C.Goat anti-mouse IgG-HRP antibody (Santa Cruz, USA) was added to plates as secondary antibody and incubated for 1 h at 37˚C.The plates were developed with 3,3 0 ,5,5 0 -tetramethylbenzidine (TMB) substrate (Promega, USA).After stopping reaction with 2M hydrochloric acid, the absorbances at 450 nm and 630 nm were measured using a microplate reader (Agilent Technologies, USA).Absorbance values were calculated by subtracting the absorbance at 630 nm from that at 450 nm of the same well.Endpoint titers were defined as the highest reciprocal dilution of serum to yield an absorbance greater than 2.1-fold of the background values.Antibody titer below the limit of detection was determined as one third of the detection limit.

Quantitation of viral load in rhesus macaques
Rhesus macaque lungs were collected as described above and fixed in neutral formalin.Lung samples were weighed, added into serum-free DMEM (1,10 W/V) and grinded with tissue homogenizer.Supernatants were isolated by centrifugation (4500 rpm, 30 min).Viral RNA was extracted with QIAamp Viral RNA Mini Kit (Qiagen, USA) as template of RT-PCR.Detection of SARS-CoV-2 was conducted with HiScript II One Step qRT-PCR SYBR Green Kit (Vazyme, China) on the ABI StepOne Real-Time PCR system (Thermo Fisher Scientific, USA) with the following conditions: 50˚C for 3 min and 95˚C for 30 s, 40 cycles of amplification at 95˚C for 10 s and 60˚C for 30 s.The following primers were used: forward 5'-CAATGGTTAAGGCAGG-3'; reverse 5'-CTCAAGGTCTGGATCACG-3'.Viral load (copies/ mL) in each sample was calculated based on standard curve.

Histopathology analysis of lungs
Mice and rhesus macaque lung samples were collected with the method described above.Samples were fixed in neutral formalin and then stained with hematoxylin and eosin for histopathology analysis.Histopathological changes of all collected tissues were graded by a doubleblind evaluation, including inflammation and structural changes.Scores 0, 1, 2 and 3 indicate no, mild, medium and severe histopathological damage, respectively.

Fig 2 .
Fig 2. Evaluation of the immunogenicity of the SWIM516 vaccine.(a) Mice immunization and sample collection schedule for evaluating the immunogenicity of SWIM516 vaccine.(b) Titers of IgG specific to the RBDs of the indicated SARS-CoV-2 variants on day 28.Numbers on top indicate the average IgG titers of each group.Dashed line indicates starting dilution (20 folds).Data are shown as geometric mean titer (GMT) ± 95% confidence interval (CI).(c) NT 50 of neutralizing antibodies against the pseudotyped viruses (pVirus) of indicated SARS-CoV-2 variants.Numbers on top indicate the GMT of each group.Radar chart was drawn based on GMT.Dashed line indicates starting dilution (60 folds).Data are shown as GMT ± 95% CI.(d) ELISpot assay quantifying the IFNγ-secreting splenocytes after re-stimulation by RBD peptide pool of indicated SARS-CoV-2 variants.Data are shown as means ± SD (standard deviation).Numbers on top indicate average SFU/10 5 splenocytes.(e) ICS assays quantifying the proportions of IFNγ-and TNFα-secreting CD8 + and CD4 + T cells stimulated by BA.2 RBD peptide pool.Data are shown as means ± SD.ELISA and pseudovirus neutralization assays were repeated twice.All statistical significances were calculated by Mann-Whitney test (*, p<0.05; **, p<0.01; ***, p<0.001, ****, p<0.0001).https://doi.org/10.1371/journal.ppat.1012116.g002

Fig 3 .
Fig 3. Evaluation of the immunogenicity of SWIM516 as boosters.(a) Mice immunization and sample collection schedule for evaluating the immunogenicity of SWIM516 vaccine as the third or fourth shot.(b) The low and high dosages of inactivated vaccine (IV), protein subunit vaccine (PV) and mRNA vaccine (RV) used in the prime-boost immunization protocols.(c) The prime-boost immunization protocols of three-dose (left) or four-dose (right) groups.Underlined numbers on top indicate group numbers.The numbers in groups names of (d) and (e) correspond with the group numbers in (c), letter L and H indicate low-dose and high-dose groups, respectively.(d) NT 50 of neutralizing antibody against Delta or BA.2 pseudotyped viruses in three-dose (left) or four-dose (right) groups.Numbers on top indicate GMT.Dashed line indicates starting dilution (60 folds).Data are shown as GMT ± 95% CI.Pseudovirus neutralization assays were repeated twice.(e) ELISpot assay quantifying the IFNγ-secreting splenocytes after restimulation by Delta or BA.2 RBD peptide viruses in three-dose (left) or four-dose (right) groups.Data are shown as means ± SD.Numbers on top indicate average SFU/10 5 splenocytes.All statistical significances were calculated by Mann-Whitney test (*, p<0.05; **, p<0.01; ***, p<0.001).https://doi.org/10.1371/journal.ppat.1012116.g003

Fig 5 .
Fig 5. Evaluation of the protective efficacy of SWIM516 in rhesus macaques.(a) Rhesus macaques immunization, challenge and sample collection schedule for evaluating the protective efficacy of SWIM516 vaccine.(b) PRNT 50 of neutralizing antibody against BA.1.Numbers on top indicate GMT, data are shown as GMT ± 95% CI.(c) Diagram of samples collected in different lobes of rhesus macaque lungs after BA.1 challenge.Legend indicated the lobe of lungs (UL: upper left; ML: middle left; LL: lower left; UR: upper right; MR: middle right; LR: lower right).(d) Quantitation of viral load in rhesus macaque lung samples, data are shown as mean ± SD.(e) Pooled analyses of pathological scores for lung tissues of macaques after BA.1 challenge, data are shown as GMT ± 95% CI.Statistical significances were calculated by Mann-Whitney test (*, p<0.05; ****, p<0.00001).(f) Pathological analysis of lung tissues of rhesus macaque after BA.1 challenge.Black bars represent 100 μm.https://doi.org/10.1371/journal.ppat.1012116.g005