key: cord-0802478-dh1z0h7v
authors: Zhang, Na-Na; Li, Xiao-Feng; Deng, Yong-Qiang; Zhao, Hui; Huang, Yi-Jiao; Yang, Guan; Huang, Wei-Jin; Gao, Peng; Zhou, Chao; Zhang, Rong-Rong; Guo, Yan; Sun, Shi-Hui; Fan, Hang; Zu, Shu-Long; Chen, Qi; He, Qi; Cao, Tian-Shu; Huang, Xing-Yao; Qiu, Hong-Ying; Nie, Jian-Hui; Jiang, Yuhang; Yan, Hua-Yuan; Ye, Qing; Zhong, Xia; Xue, Xia-Lin; Zha, Zhen-Yu; Zhou, Dongsheng; Yang, Xiao; Wang, You-Chun; Ying, Bo; Qin, Cheng-Feng
title: A thermostable mRNA vaccine against COVID-19
date: 2020-07-23
journal: Cell
DOI: 10.1016/j.cell.2020.07.024
sha: c42841d269b8e0319cd69015d10c18bd0a8f9ab9
doc_id: 802478
cord_uid: dh1z0h7v

Summary There has been an urgent need of vaccines against coronavirus disease 2019 (COVID-19) due to the ongoing SARS-CoV-2 pandemic. Among all approaches, messenger RNA (mRNA) -based vaccine has emerged as a rapid and versatile platform to quickly respond to such a challenge. Here, we developed a lipid-nanoparticle-encapsulated mRNA (mRNA-LNP) encoding the receptor binding domain (RBD) of SARS-CoV-2 as a vaccine candidate (termed ARCoV). Intramuscular immunization of ARCoV mRNA-LNPs elicited robust neutralizing antibodies against SARS-CoV-2 as well as Th1-biased cellular response in mice and non-human primates. Two doses of ARCoV immunization in mice conferred complete protection against the challenge of a SARS-CoV-2 mouse adapted strain. Additionally, ARCoV was manufactured in liquid formulation and can be stored at room temperature for at least one week. This novel COVID-19 mRNA vaccine, ARCoV, is currently being evaluated in phase 1 clinical trials.

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), a novel human coronavirus closely related to SARS-CoV Zhou et al., 2020; Zhu et al., 2020b) , has spread throughout the world and caused global public health crises. The clinical manifestations caused by SARS-CoV-2 ranged from non-symptomatic infection, mild flu-like symptoms to pneumonia, severe acute respiratory distress syndrome and even deaths Wang et al., 2020) . To date, the Coronavirus Disease 2019 has resulted in more than 3.5 million cases with over 250,000 mortalities (World Health Organization). There is no effective treatment yet available. Therefore, the development of a safe and effective vaccine against COVID-19 is urgently needed.

SARS-CoV-2, together with the other two highly pathogenic human coronaviruses, SARS-CoV and MERS-CoV, belongs to the genus Betacoronavirus of the family Coronavirdae.

Coronaviruses are enveloped positive-sense, single-stranded RNA viruses, and the virion is composed of a helical capsid formed by nucleocapsid (N) proteins bound to the RNA genome and an envelope made up of membrane (M) and envelope (E) proteins, coated with "crown"-like trimeric spike (S) protein. Like other human coronaviruses, the full-length S protein of SARS-CoV-2 consists of S1 and S2 subunits. First, the S protein mediates viral entry into host cells by binding to its receptor angiotensin-converting enzyme 2 (ACE2) through the receptor-binding domain (RBD) at the C-terminal of S1 subunit, which subsequently causes the fusion between the viral envelope and the host cell membrane through the S2 subunit (Hoffmann et al., 2020) .

The full-length S protein, S1 and RBD, are all capable of inducing highly potent neutralizing antibodies and T cell mediated immunity, therefore have been widely selected as promising targets for coronavirus vaccine development (Amanat and Krammer, 2020) . Some recent studies also demonstrated immunization with recombinant RBD of SARS-CoV-2 induced high titers of neutralizing antibodies in the absence of antibody dependent enhancement (ADE) of infection (Quinlan et al., 2020; Tai et al., 2020) . The structure of SARS-CoV-2 RBD alone, RBD-ACE2 and RBD-monoclonal antibody complexes have been resolved in record time at high resolution (Lan et al., 2020; Shang et al., 2020; Walls et al., 2020) , which further improved our understanding of this vaccine target.

Messenger RNA (mRNA) -based therapy has recently emerged as an effective platform for treatment of infectious diseases and cancer (Jackson et al., 2020; Mascola and Fauci, 2020) . In the past few years, with the technological advances in mRNA modification and delivery tools (Ickenstein and Garidel, 2019; Maruggi et al., 2019; Pardi et al., 2020) , mRNA vaccine field is developing extremely rapidly in both basic and clinical research. Preclinical studies have demonstrated that mRNA-based vaccines induced potent and broadly protective immune responses against various pathogens in small and large animals with an acceptable safety profile (Maruggi et al., 2019) . To date, clinical trials for mRNA vaccines against viral diseases including Zika, Ebola, Influenza, Rabies, and Cytomegalovirus infections have been carried out in many countries (Alameh et al., 2020) . One of the key advantages of mRNA vaccine platform is its capability of scalable production within very short period of time, which makes it very attractive in responding to the outbreak of pandemic. The mRNA manufacturing avoids the lengthy process of cell culture and purification procedures and the stringent biosafety measures for traditional virus vaccine production. Clinical scale mRNA vaccine can be rapidly designed and manufactured within weeks once viral antigen sequence becomes available. In March 2020, it took only 42 days for Moderna's mRNA-1273 to enter phase I clinical trial as the very first mRNA vaccine against COVID-19 in the United States (NCT04283461). There are several other SARS-CoV-2 mRNA vaccine candidates currently under development worldwide, which further proves the high potential of mRNA vaccine platform. However, none of these mRNA vaccines in the clinical stage have been evaluated in animal models; the mechanism of mRNA vaccine against COVID-19 is unclear and their effectiveness is yet to be proven (Jiang, 2020) . In the present study, we demonstrated the immunogenicity and protection of a novel mRNA vaccine candidate (termed ARCoV) against SARS-CoV-2 in animal models, which supports further clinical development in humans.

Lipid nanoparticles (LNPs) represent one of the most appealing and commonly used mRNA delivery tools (Ickenstein and Garidel, 2019) . Here, we have developed a vaccine platform based on modified mRNA encapsulated in LNPs for in vivo delivery. The RBD of SARS-CoV-2 (aa 319-541) was chosen as the target antigen for the mRNA coding sequence (Figure S1 ), as shown in Figure 1A . Transfection of the RBD-encoding mRNA in multiple cell lines (HeLa, Huh7, HEK293T and Vero) resulted in high expression of recombinant RBD in culture supernatants ( Figure 1B ) with up to 917.4 ng/mL of RBD in the mRNA transfected HEK293F cells ( Figure   S2A ). RBD protein expressed from mRNA retained high affinity to recombinant human ACE2 demonstrated by kinetic analysis using ForteBio Octet (Figure 1C) , and functionally inhibited the entry of a vesicular stomatitis virus (VSV) -based pseudovirus expressing the SARS-CoV-2 S protein (Nie et al., 2020) in Huh7 cells ( Figure 1D ). Immunostaining further demonstrated that this RBD protein can be recognized by a panel of monoclonal antibodies (mAbs) against SARS-CoV-2 RBD ( Figure S2B ) as well as convalescent sera from three COVID-19 patients ( Figure   1E ).

The mRNA-LNP formulations were prepared using a modified procedure as previously described for siRNA (Semple et al., 2010) , followed by tangential flow filtration and purification before being filled into sterile glass vials ( Figure S3 ). The characterization of representative batches of mRNA-LNPs were shown in Table S1 . The final stock of SARS-CoV-2 RBD encoding mRNA-LNPs (termed ARCoV) manufactured under GMP conditions showed an average particle size of 88.85 nm ( Figure 1F ) with >95% encapsulation. Cryo-TEM analysis showed the ARCoV particles exhibit homogenous morphologies of solid spheres that lack aqueous core (Figure 1G ), which demonstrates a key difference between RNA-loaded LNPs and conventional liposomes.

Next, we evaluated the in vivo delivery capability of ARCoV in mice. To visualize the tissue distribution of our mRNA-LNP formulations, a firefly luciferase (FLuc) reporter encoding mRNA-LNP were prepared using the same procedure as ARCoV (Table S1) , and subjected to Bioluminescence Imaging (BLI) analysis upon different immunization routes. Following intramuscular (i.m.) injection, robust expression of FLuc were seen in the upper abdomen as well as the injection site of BALB/c mice at 6 h post injection (Figure 2A) ; subcutaneous (s.c) injection also led to robust FLuc expression in upper abdomen, while no signal was detected in mice receiving intranasal (i.n.) inoculation. Specially, real-time monitoring of the intramuscular immunized mice showed that photon flux peaked at 12 h post injection, faded to undetectable level at 48 h post injection at both upper abdomen and the injection site. Further ex vivo imaging analysis of the ARCoV immunized BALB/c mice showed that the liver represented the most abundant RBD expressing tissue, and slight luminescent signal was also detected in spleen and muscle tissues ( Figure 2B) . Additionally, following intravenous (i.v.) administration of ARCoV mRNA-LNP at 1 mg/kg, the expression of RBD was readily detectable by ELISA at 6 h post injection with the average concentration of 450.6 ng/mL in ICR mice sera ( Figure 2C) . Furthermore, to identify the primary cell types in which the encapsulated mRNAs were translated into antigen, muscle at the injection site and liver were collected from the ARCoV mRNA-LNP and placebo LNP immunized mice and subjected to multiplex immunofluorescent staining of SARS-CoV-2 RBD and different immune cell markers. Robust expression of SARS-CoV-2 RBD was detected in the muscle samples collected from the intramuscular injection site of the ARCoV inoculated mice, which were mostly colocalized with the CD11b-positive monocytes as well as the CD163-positive macrophages and CD103-positive dendritic cells in muscle samples from the vaccine immunized mice ( Figure 2D ). As expected, no RBD expression was seen in muscle tissue from the placebo LNP treated mice, and LNPs also stimulated massive infiltration of monocytes and macrophages, which functioned as adjuvants as previously described (Maugeri et al., 2019) . Furthermore, colocalization of SARS-CoV-2 RBD and CD11b-positive monocytes was also detected within the intramuscular lymph nodes from the ARCoV mRNA-LNP inoculated mice ( Figure S4 ). Abundant SARS-CoV-2 RBD were also detected in the liver from the ARCoV immunized mice (Figure 2E ), which agreed with the in vivo and ex vivo luciferase expression profile (Figure 2A-B) . Further analyses also showed that SARS-CoV-2 RBD fluorescent signal primarily overlapped with the glutamine synthetasepositive pericentral hepatocytes surrounding the CD31-positive central vein (CV), Arg1-positive hepatocytes as well as CD163-positive liver macrophages ( Figure 2E ). These results highlighted the capability of our mRNA-LNP formulations to deliver mRNA in vivo and recruit antigen presenting cells to process the antigens expressed.

We have also determined the immunogenicity and efficacy of ARCoV mRNA-LNP in animals. Initially, groups of immunocompetent female BALB/c mice were immunized with a single dose of ARCoV mRNA-LNP via intramuscular administration, empty LNPs were used as placebo. Following immunization, no local inflammation response in the injection site or other adverse effects were observed during the observation period. A single immunization with ARCoV mRNA-LNP (2 and 30 µg) induced the production of SARS-CoV-2 RBD specific IgG antibodies ( Figure S5A ) and neutralizing antibodies with NT 50 approached ~1/278 and ~1/559 at 28 days post immunization ( Figure S5B) , which was lower than the neutralizing antibody levels in the convalescent sera from selected COVID-19 patients (Ni et al., 2020) . Next, groups of mice were immunized with 2 or 10 µg of ARCoV mRNA-LNPs, and boosted with the same dose at day 14, and sera were collected at 7, 14, 21 and 28 days post initial vaccination and subjected to antibody assays ( Figure 3A) . Remarkably, a second immunization with either 2 or 10 µg of ARCoV mRNA-LNPs resulted in a rapid elevation in both IgG and neutralizing antibodies in mice ( Figure 3B-D) , whereas no SARS-CoV-2 specific IgG and neutralizing antibody was detected in sera from mice vaccinated with empty LNPs. At 28 days post initial immunization, the NT 50 titers in mice immunized with 2 or 10 µg of ARCoV mRNA-LNPs approached ~1/2,540 and ~1/7,079, respectively ( Figure 3C ); and the PRNT 50 reached ~1/2,194 and ~1/5,704, respectively ( Figure 3D ). Recent genome surveillance has recorded novel epidemic SARS-CoV-2 strains with specific mutations in the S protein, which was potentially associated with virus transmission and pathogenesis (Becerra-Flores and Cardozo, 2020; Korber et al., 2020) . Thus, we further evaluated whether mouse sera post ARCoV mRNA-LNP vaccination could cross neutralize different epidemic strains of SARS-CoV-2 ( Table S2 ). All three epidemic strains used in this study shared the same RBD sequence as our mRNA vaccine, while 5N and V34 contained a unique D614G and A653V substitution in the S protein, respectively ( Figure   S1 ). As expected, sera from all ARCoV vaccinated mice showed similar neutralizing capability against all three SARS-CoV-2 epidemic strains, and there was no significant difference in PRNT 50 titers ( Figure 3E) . Together, our results demonstrated that two doses of immunization of ARCoV vaccine have induced high levels of antibodies with broad neutralizing capabilities against SARS-CoV-2 in mice.

We further studied whether SARS-CoV-2 specific T cell immune response was elicited by two doses of immunization of ARCoV mRNA-LNPs in mice via intramuscular administration.

Flow cytometry results showed a significant increase in viral specific CD4 + and CD8 + effector memory T cells (Tem) in splenocytes from ARCoV vaccinated mice in comparison with the placebo LNPs ( Figure 4A ) upon stimulation with peptide pools covering SARS-CoV-2 RBD (Table S3) . Furthermore, Enzyme-linked immunosorbent spot (ELISPOT) assay showed that secretion of IFN-γ, TNF-α, IL-2 in splenocytes from mRNA-LNP immunized mice was significantly higher than those received the placebo vaccination ( Figure 4B) . Meanwhile, there was no significant difference in IL-4 and IL-6 secretion between the ARCoV immunized animals and the placebo immunized ones ( Figure 4C ). Our results demonstrated that mRNA-LNP vaccine successfully induced a Th1-biased, SARS-CoV-specific cellular immune response.

To further evaluate the in vivo protection efficacy, we employed a newly developed SARS-CoV-2 mouse adapted strain challenge model . Upon intranasal challenge with mouse adapted strain MASCp6, immunocompetent BALB/c mice sustained robust viral replication in lung and trachea, resulting in moderate pneumonia and inflammatory responses.

Deep sequencing has revealed that MASCp6 contained a unique N501Y substitution at the S protein ( Figure S1 ). To exclude the potential impact of N501Y mutation on antibody response, we also compared the neutralizing antibody titers of mouse sera from ARCoV immunized mice against MASCp6 and wild type SARS-CoV-2 strain 131, respectively. As shown in Figure S6 , there was no significant difference between the PRNT 50 against the two strains. Then, mice that received two doses immunization of ARCoV mRNA-LNPs at 2 or 10 µg were intranasal (i.n.) challenged with 6,000 PFU of SARS-CoV-2 MASCp6 at 40 days after initial vaccination ( Figure 3A) . At day 5 after challenge, mice were euthanized, lung and trachea were analyzed for viral RNA loads, and lung sections were subjected to immunostaining with in situ hybridization (ISH) and histopathological assays. Consistent with their high neutralizing antibody titers, all mice immunized with 2 or 10 µg of ARCoV mRNA-LNPs showed full protection against SARS-CoV-2 infection, no measurable viral RNA were detected in both lungs ( Figure 5A ) and trachea ( Figure 5B) , whereas high level of viral RNAs was detected in lung and trachea (~10 9 and 10 7 RNA copies equivalents per gram, respectively) from mice in placebo group. Immunostaining assay showed abundant SARS-CoV-2 protein expression, mainly along the airway, in lung section from mice receiving placebo LNPs inoculation, while few positive cells were detected in the lung from the ARCoV vaccinated mice ( Figure 5C) . Similarly, ISH assay by RNA scope also detected SARS-CoV-2 specific RNAs in the placebo mice, while no viral RNAs in lung sections from all ARCoV vaccinated animals ( Figure 5D ). More importantly, mice vaccinated with empty LNPs developed typical lung lesions characterized by denatured epithelial tissues, thickened alveolar septa, and activated inflammatory cell infiltration, while no such pathological changes were seen in lung sections from all ARCoV immunized animals ( Figure 5E ). These results demonstrated that two doses ARCoV vaccination completely prevent SARS-CoV-2 replication in lower respiratory tract and protected mice from lung lesions.

We also compared the serum neutralizing antibodies titers of the ARCoV vaccinated mice pre-and post-SARS-CoV-2 challenge. Strikingly, all animals receiving 2 or 10 µg doses of ARCoV mRNA-LNPs showed no significant increase in neutralizing titers after challenge with SARS-CoV-2 ( Figure 6A ). The in vivo protection of a single immunization of ARCoV mRNA-LNPs was also evaluated using the MASCp6 model. We next evaluated the immunogenicity of the ARCoV vaccine in cynomolgus monkeys (Macaca fascicularis), a non-human primate model that was susceptible to SARS-CoV-2 infection (Lu et al., 2020; Rockx et al., 2020) . Two groups of macaques (n=10/each group) were immunized with 100 or 1000 µg of ARCoV mRNA-LNPs via intramuscular administration and boosted with the same dose at 14 days post initial immunization ( Table S4) . The same number of monkeys (n=10) were vaccinated with PBS as placebo ( Figure 7A ). SARS-CoV-2-specific IgG antibodies were readily induced at day 14 post initial immunization, and the boost immunization resulted in a notable increase of IgG titers to ~1/5,210 and ~1/22,085 at day 28 post initial immunization ( Figure 7B ). Fifty percent of animals received high dose ARCoV immunization developed low level neutralizing antibodies at day 14 post initial immunization, while the boost immunization resulted in a notable increase in NT 50 to ~1/699, ~1/6482 in monkeys vaccinated with either low or high dose ARCoV, respectively ( Figure 7C ).

Additionally, there is no significant difference in the serum neutralizing titers between male and female macaques ( Figure S7 ). Meanwhile, IFN-γ ELISPOT assays showed SARS-CoV-2 RBD specific T cell response were stimulated in PBMCs from monkeys vaccinated with either low or high dose of ARCoV at day 5 post boost immunization, but not from animals receiving placebo ( Figure 7D ); There was no significant difference in IL-4 + /CD4 + cell response to SARS-CoV-2 RBD between ARCoV and placebo treated animals ( Figure 7E ), suggesting the induction of Th1-biased cellular immune response by ARCoV immunization.

Finally, as cold chain transportation is not available in many COVID-19 epidemic areas, a vaccine that can be stored at room temperature will be highly desirable. The thermal stability of the ARCoV mRNA-LNP vaccine was evaluated using the FLuc reporter mRNA-LNP formulation. After storage for 1, 4 and 7 days at different temperatures, the FLuc reporter mRNA-LNP was intramuscularly administered into BALB/c mice and mice were subjected to in vivo BLI imaging 6 h later. As shown in Figure S8A , there was no reduction in FLuc expression between all groups, indicating that our mRNA-LNP formulation is stable at 4°C and 25°C for at least 7 days. Storage at 37°C for 7 days only resulted in a ~13% reduction in relative photon flux ( Figure S8B ). These results indicated the high thermostability of ARCoV vaccine.

In the present study, we reported the immunogenicity and efficacy of a novel COVID-19 mRNA vaccine candidate in various animal models. A single dose or two doses immunization of ARCoV elicited robust antibody and T cell responses in mice and non-human primates against multiple epidemic SARS-CoV-2 strains. The NT 50 in sera from non-human primates receiving low dose (100 µg) ARCoV immunization was comparable to those from the convalescent sera from 20 COVID-19 patients ( Figure S9 ), while high dose (1000 µg) ARCoV immunization induced much higher titers of neutralizing antibodies compared with the convalescent sera (Ni et al., 2020) . It has been reported that two or three doses of inactivated SARS-CoV-2 virus vaccine could induce neutralizing antibodies at levels of ~1/50, which provided full protection against SARS-CoV-2 in rhesus macaques . A recent report showed that two doses immunization with a DNA vaccine candidate elicited mean neutralization titers between ~1/70 and ~1/170 in rhesus macaques . In the aforementioned studies, all animals exhibited anamnestic antibody responses following challenge, suggesting that vaccine protection was probably not sterilizing, which was consistent to relative lower neutralizing antibody titers.

In our study, a single dose ARCoV immunization induced anamnestic antibody response, while animals receiving two doses of ARCoV immunization didn't showed enhancement in neutralizing antibody titers upon challenge, suggesting sterilizing immunity may have been induced in mice ( Figure 6A ).

Further challenge experiments with a SARS-CoV-2 mouse-adapted strain MASCp6 showed that two doses of immunization of ARCoV completely blocked viral replication in lung and trachea, and prevented the pulmonary pathology in mice (Figure 5) . Although it still needs to be further validated in clinical settings, our results revealed that neutralizing antibody titer level in mice correlates well with protection against SARS-CoV-2 challenge. Regression analysis revealed a cut-off value of ~1:1009 neutralizing antibody titers (PRNT 50 ) as full protection of SARS-CoV-2 lung infection. To our knowledge, this is the first protection correlate identified in mouse model. Considering the limited resource of non-human primates and strict requirement of biosafety facility for SARS-CoV-2 challenge experiment, this protection correlate in mouse model represents a simple and useful benchmark for efficacy tests, which will greatly facilitate and accelerate COVID-19 vaccine development. Due to biosafety facility limitation, we are not able to obtain the protection efficacy data in non-human primates at present. However, based on the comparison of neutralizing antibody levels in macaques vaccinated with inactivated or DNA COVID-19 vaccine candidates Yu et al., 2020) , a protective immunity can be expected in most macaques immunized with two doses of ARCoV. Especially, ARCoV is highly immunogenic in both male and female macaques ( Figure S7 ). Of particular note, although one of ten vaccinated macaques in each group failed to produce detectable (1:30) neutralizing antibodies, IFN-γ ELISPOT assay showed virus specific IFN-γ secretion was detected in all vaccinated macaques (Table S4 ).

An ideal COVID-19 vaccine is supposed to avoid the induction of non-neutralizing antibody and Th2-biased cellular immune response for safety concern (Graham, 2020) . Not like the mRNA-1273 vaccine from Moderna (Corbett et al., 2020) , our ARCoV vaccine chose the RBD as antigen target. Compared with the full length S protein, RBD antigen may induce fewer non-neutralizing antibodies, lowering the risk of potential ADE of SARS-CoV-2 infection, as similar phenomenon has been observed during other coronavirus infection (Olsen et al., 1992) . Recent in vitro study has also suggested that antibodies targeting SARS-CoV-2 RBD at various concentrations did not induce ADE infection (Quinlan et al., 2020) . Considering S-specific IgG antibodies has been suggested to cause acute pulmonary injury in vaccine-challenge animal models of SARS-CoV (Liu et al., 2019) , although the exact S epitopes account for the lung pathology remains to be determined, the use of RBD may minimize this risk. Additionally, vaccine-associated enhanced respiratory disease has been linked to Th2-biased CD4 + T cell responses (Ruckwardt et al., 2019) . As expected, our mRNA-based vaccine induced a Th1-prone T cell immune response to SARS-CoV-2 RBD in mice and macaques (Figures 4 and 7) . Similar results were also reported in the DNA and adenovirus-vectored COVID-19 vaccine candidates (van Doremalen et al., 2020; Yu et al., 2020; Zhu et al., 2020a) . In our mouse challenge experiments, we did not observe enhanced viral replication or clinical disease in all vaccinated animals, even those receiving a single dose of ARCoV vaccination.

We also characterized the in vitro and in vivo expression pattern of our mRNA-LNP formulation. Upon intramuscular injection, robust protein expression was readily detected in the muscle tissue at the injection site, and the most predominant expression was seen in the liver (Figure 2A-B) , which was similar to the results from other LNP formulations (Bahl et al., 2017; Pardi et al., 2015) . Most importantly, multiplex immune co-staining assay showed that robust expression of SARS-CoV-2 RBD was observed in multiple antigen presenting cells, including monocytes, macrophages, and DC cells, in muscle, liver as well as lymph nodes from the ARCoV vaccinated mice (Figure 2D-E) . A recent study has shown that a yellow fever mRNA vaccine delivered by LNP mainly expressed in the injection site as well as the draining lymph nodes in cynomolgus macaques (Lindsay et al., 2019) . Further biodistribution profile of ARCoV in cynomolgus monkeys is being tested in a GLP lab.

To date, limited results have been reported regarding the safety and stability profile of LNPbased mRNA vaccines (Jackson et al., 2020; Maruggi et al., 2019; Stitz et al., 2017) . Our data from cynomolgus monkeys showed 100 µg of ARCoV is sufficient to induce high level neutralizing antibodies, and 1000 µg of ARCoV didn't cause obvious adverse effects, highlighting the safety profile of our mRNA LNP formulation. Extrapolation of dose from animals to humans remains a huge challenge which requires careful consideration between safety and efficacy data. These preclinical data from mouse and non-human primate have provided critical reference for the starting dose of ARCoV in human trials. Lastly, the accessibility and scalability of COVID-19 vaccines is a major challenge to expediting the delivery and massive immunization worldwide, therefore a ready-to-use and thermostable vaccine is highly preferred.

The final ARCoV mRNA-LNP vaccine is manufactured in a liquid formulation without the need of thawing or reconstitution before injection, and a single dose vaccine was prepared in a prefilled syringe for quick self-administration. Stability test results showed our formulation maintained their in vivo delivery efficiency at 4°C and 25°C for at least one week (Figure S8 ), long-term stability of ARCoV vaccine is currently under evaluation. Additionally, ARCoV is administrated with the most commonly used intramuscular vaccination route for human use.

These unique features of ARCoV make it a promising COVID-19 vaccine candidate with universal availability and global accessibility.

In summary, we reported a thermostable mRNA vaccine candidate against SARS-CoV-2 and provided the first line evidence of the immunogenicity and efficacy in multiple animal models.

Although two mRNA vaccine candidates from Moderna and BioNTech/Pfizer are being tested in humans prior to our results, there has no report that an mRNA vaccine can protect animals from SARS-CoV-2 infection, nor the immune correlate of protection. During the revision of our manuscript, the immunogenicity and protection efficacy of mRNA-1273 in mice were also reported (Corbett et al., 2020) . The robust protection observed in both studies highlights the power of mRNA vaccine platform and pave the path forward for a successful COVID-19 vaccine in the near future. Our ARCoV mRNA vaccine was approved for phase I clinical trial (ChiCTR2000034112) on June 19th, 2020.

The challenge experiments in our study were based on a mouse-adapted strain of SARS-CoV-2, further challenge experiments with wild-type SARS-CoV-2 strain in transgenic ACE2 mice or non-human primates will provide more data about the protective efficacy. Another limitation of our study is that the duration of neutralizing antibody induced by ARCoV is yet to be determined.

Experience from other human coronaviruses has indicated the possibility of re-infection due to waning of antibody response (Callow et al., 1990; Wu et al., 2007) . Future studies are needed to evaluate the long-term immune response in animal models and the effectiveness of ARCoV in humans. Additionally, long-term stability assays with clinical grade ARCoV vaccine is being (C) Expression of mRNA-encoded RBD in mice. The serum concentration of RBD were measured by ELISA at 6 hours post inoculation. Data are shown as mean ± SEM and analyzed using unpaired t-test. **** P < 0.0001.

(D) Multiplex immunostaining analysis for the expression of LNP-delivered mRNA in mouse muscle tissues. Female BALB/c mice (n=3) were immunized with 10 µg of ARCoV mRNA-LNP, and empty LNP was used as control (n=3). Muscle tissue at the injection site was collected 6 hours post injection and subjected to multiplex immunofluorescent staining for SARS-CoV-2 RBD (white) as well as other cell markers including Desmin (gold), CD11b (green), CD163 (red) and CD103 (magenta). Magnifications of the areas boxed in white are shown on the right. Arrows indicate double-positively-stained cells. See also Figure S4 . Female BALB/c mice were intramuscularly immunized with 2 µg (n=8), 10 µg (n=8) of ARCoV or Placebo (n = 5), and boosted with an equivalent dose 14 days later. Serum was collected at 7, 14, 21 and 28 days post initial vaccination.

(A) Schematic diagram of immunization, sample collection and challenge schedule.

(B) SARS-CoV-2 specific IgG antibody titer was determined by ELISA.

(C-D) The NT 50 and PRNT 50 was determined using VSV-based pseudovirus and infectious SARS-CoV-2, respectively. The dashed lines indicate the detection limit of the assay. Data are shown as mean ± SEM. Significance was calculated using a two-way ANOVA with multiple comparison tests (n.s., not significant ; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001) (E) Serum cross neutralization against SARS-CoV-2 epidemic strains in ARCoV immunized mice. PRNT against the three SARS-CoV-2 epidemic strains were performed using mouse sera collected at 28 days post initial immunization. Data are analyzed by one-way ANOVA with multiple comparison tests. (n.s., not significant) See also Figure S5 .

(A) SARS-CoV-2 RBD specific CD4 + and CD8 + effector memory T cells (CD44 + CD62L -) in splenocytes were detected by flow cytometry.

(B-C) ELISPOT assay for IFN-γ, TNF-α, IL-2, IL-4 and IL-6 in splenocytes. Data are shown as mean ± SEM. Significance was calculated using unpaired t-test (n.s., not significant; *, P < 0.05; **, P < 0.01)

Forty days after the initial immunization, mice were intranasally inoculated with the mouseadapted SARS-CoV-2 (MASCp6), and the indicated tissues were collected at 5 days post challenge for detection of viral loads and lung pathology. (C) Immunostaining of lung tissues with SARS-CoV-2 S specific mAb. Scale bar: 100 µm.

(D) ISH assay for SARS-CoV-2 RNA. Scale bar: 50 µm. Positive signals are shown in brown.

(E) H&E staining of lung pathology. Scale bar: 100 µm. Representative images from 4 or 5 mice are shown.

See also Figure S5 . (A-B) Paired sera were collected from animals receiving two doses (2 or 10 µg) or single dose (2 or 30 µg) vaccination before (Pre) and 5 days after (Post) SARS-CoV-2 challenge. The NT 50 values were analyzed for differences using paired t-test (n.s., not significant; *, P < 0.05; **, P < 0.01). Three to 6-year-old male or female cynomolgus macaques were immunized intramuscularly with 100 µg (n=10) or 1000 µg (n=10) of ARCoV and boosted with the same dose at 14-day interval. Serum was collected on day 0, 14, 28 post initial immunization and subjected to antibody assays.

(A) Schematic diagram of ARCoV immunization, sample collection and immunological assays.

(B-C) The IgG titers and NT 50 were determined by ELISA and SARS-CoV-2 pseudovirus neutralization assay, respectively. Dotted lines indicate the limits of detection. Data are shown as mean ± SEM. Significance was calculated using two-way ANOVA with multiple comparison tests. (n.s., not significant; *, P < 0.05; ****, P < 0.0001).

(D-E) The production of IFN-γ or IL-4 in PBMCs stimulated by SARS-CoV-2 RBD was measured by ELISPOT assay or flow cytometry. Data are shown as mean ± SEM. Significance was calculated using one-way ANOVA with multiple comparison tests (n.s., not significant ;*, P < 0.05; ****, P < 0.0001).

See also Table S4 . Figure S1 . Amino acid sequence alignment of full S protein of SARS-CoV-2 isolates used in this study. Related to Figures 1 and 3 .

Invariant residues are shown as black dots. RBD sequences are shown in gray. Variant mutations are marked in light red. ARCoV is manufactured through rapid mixing of mRNA in aqueous solution and a mixture of lipids in ethanol. This process yields self-assembled LNPs with mRNA encapsulated inside. Tangential flow filtration was used to remove ethanol and to concentrate the solution. Following the Quality Control (QC) procedure, the final product was filtered into sterilized glass syringes or glass vials. BALB/c mice were intramuscularly immunized with 2 µg (n=7) or 30 µg (n=8) of the ARCoV vaccine or Placebo (n=5). Serum was collected at 14, 28 days post immunization and analyzed by ELISA (A) and pseudovirus neutralization assay (B). Data are shown as mean ± SEM. Significance was calculated using a two-way ANOVA with multiple comparison tests (n.s., not significant; *, P < 0.05; **, P < 0.01; ***, P< 0.001; ****, P < 0.0001).

Six to eight weeks after immunization, all immunized mice were inoculated intranasally with the SARS-CoV-2 mouse-adapted strain MASCp6, and their lungs (C) and trachea (D) were collected for detection of viral RNA loads at 5 days post challenge. Data are shown as mean ± SEM; Significance was calculated using a one-way ANOVA with multiple comparison tests. (**, P < 0.01; ****, P < 0.0001). Ten cynomolgus macaques were immunized intramuscularly with 100 µg or 1000 µg of ARCoV, respectively, and boosted with the same dose at a 14-day interval. The serum neutralizing antibody titers from male and female macaques were calculated respectively. Dotted lines indicate the limits of detection. Significance was calculated using a one-way ANOVA with multiple comparison tests. (**, P < 0.01; ****, P < 0.0001). (A) BLI of FLuc expression in mice .The FLuc encoding mRNA-LNPs were stored at 4°C, 25°C or 37°C for 1, 4, and 7 days before being dosed to BALB/c mice. IVIS imaging was performed 6 hours post inoculation.

(B) Photon flux was quantified from ROI analysis. The data are representative of at least three independent experiments, and error bars indicate the SEM. Significance was calculated using two-way ANOVA with multiple comparison tests. ( n.s., not significant ;***, P < 0.001; ****, P < 0.0001). Figure 6 .

The serum neutralizing antibody titers were calculated from cynomolgus macaques immunized with 100 µg (n=10) and 1000 µg (n=10) ARCoV and COVID-19 patients' convalescent sera (n=20), respectively. Dotted lines indicate the limits of detection. Significance was calculated using a one-way ANOVA with multiple comparison tests. (n.s., not significant ; **, P < 0.01).

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, qincf@bmi.ac.cn (C.F.Q.)

All unique/stable reagents generated in this study are available from the Lead Contact with a completed Material Transfer Agreement.

This study did not generate any unique datasets or code.

All animal studies were performed in strict accordance with the guidelines set by the Chinese Hospital with written informed consent.

African green monkey kidney cell Vero (ATCC, CCL-81), human cervical carcinoma cell HeLa Patient-derived SARS-CoV-2 isolates (listed in Table S2 ) were passaged in Vero cells and the virus stock was aliquoted and titrated to PFU/ml in Vero cells by plaque assay. The mouse adapted SARS-CoV-2 strain MASCp6 and the VSV-based SARS-CoV-2 pseudovirus have been described previously Nie et al., 2020) . All experiments involving infectious SARS-CoV-2 were performed under Biosafety Level 3 facilities in AMMS.

Amino acid sequence alignment of full S protein of SARS-CoV-2 isolates was performed using MAFFT (Katoh and Standley, 2013) . Wuhan-Hu-1 (GenBank nos. MN908947.3) was used as the reference strain.

The mRNA was produced in vitro using T7 RNA polymerase-mediated transcription from a linearized DNA template from plasmid ABOP-028 (GENEWIZ), which encodes codonoptimized RBD region of SARS-CoV-2 ( Figure S1 ) and incorporates the 5' and 3' untranslated regions and a poly-A tail. The FLuc-encoding mRNA (FLuc-mRNA) was prepared from plasmid ABOP-010 (GENWIZ) in the same procedure.

Lipid-nanoparticle (LNP) formulations were prepared using a modified procedure of a method previously described for siRNA(Ickenstein and Garidel, 2019) ( Figure S3 ). Briefly, lipids were dissolved in ethanol containing an ionizable lipid, 1, 2-distearoyl-sn-glycero-3-phosphocholine (DSPC), cholesterol and PEG-lipid (with molar ratios of 50:10:38.5:1.5). The lipid mixture was combined with 20 mM citrate buffer (pH4.0) containing mRNA at a ratio of 1:2 through a Tmixer. Formulations were then diafiltrated against 10× volume of PBS (pH7.4) through a tangential-flow filtratio (TFF) membrane with 100 kD molecular weight cut-offs (Sartorius Stedim Biotech), and concentrated to desired concentrations, passed through a 0.22 µm filter, and stored at 2~8 °C until use. All formulations were tested for particle size, distribution, RNA concentration and encapsulation.

ARCoV sample (3 µl) was deposited on a holey carbon grid that was glow-discharged (Quantifoil R1.2/1.3) and vitrificated using a Vitrobot Mark IV (Thermo Fisher Scientific) instrument. Cryo-EM imaging was conducted on a Talos F200C Equipped with a Ceta 4k x 4k camera, operated at 200 kV accelerating voltage.

Size measurements were performed using dynamic light scattering (DLS) on a Malvern Zetasizer Nano-ZS (Malvern). Samples were irradiated with red laser ( λ = 632.8 nm) and scattered light were detected at a backscattering angle of 173°. Results were analyzed to obtain an autocorrelation function using the software (Zetasizer V7.13).

HeLa, HEK293T, Huh7 or Vero cells were seeded in 24-well plates at 200,000 cells/well.

Eighteen hours later, the cells were transfected with RBD or control mRNA (2 µg/ml) using Lipofectamine™ 3000 Transfection Reagent (Thermo Fisher Scientific). Six hours later, the medium was replaced with Opti-MEM™ I Reduced Serum Medium (Thermo Fisher Scientific).

The supernatant was collected at 48 hours after transfection, clarified by centrifugation at 1000 x g, and then mixed with 5×SDS loading buffer (non-reducing). The samples were loaded for SDS-PAGE without heating. The secreted RBD protein was then detected by Western blotting with a monoclonal antibody (mAb) against the SARS-CoV-2 RBD protein (Sino Biological).

A hexa-His tag was added to the C-terminus of signal peptide-RBD to facilitate further purification processes. The optimized RBD gene was cloned into the pcDNA3.1 vector (Beijing BioMed Gene Technology) with Hind III and EcoR I (Thermo Fisher Scientific) restriction sites, resulting in a pcDNA3.1-sp-RBD-His plasmid. 293T cells were seeded in 15 cm dishes at 5,000,000 cells/dish. Eighteen hours later, the cells were transfected with pcDNA3.1-sp-RBD-His (1 µg/ml) using TurboFect™ Transfection Reagent (Thermo Fisher Scientific). Six hours later, the medium was removed and cells were washed with PBS for 3 times, followed by addition of Opti-MEM™ I Reduced Serum Medium. The supernatant was collected per 24 hours for 4 days. The collected supernatant was centrifuged at 1,000 g for 3 minutes before filtration using 0.45 µm Membrane Filter (Millipore), and purified using NI-NTA agarose beads (Qiagen).

The purified protein was concentrated using Pierce™ Protein Concentrator PES, 3K MWCO (Thermo Fisher Scientific).

The real-time RBD-ACE2 binding assay was performed by biolayer interferometry using ForteBio Octet RED96e. Briefly, Streptavidin (SA) Biosensor from ForteBio was used to capture 10 µg/ml biotin-ACE2 (Sino Cell) onto the surface of the SA biosensor. After reaching base line, sensors were subjected to the association step containing 75.6, 30.2, 12.1, 4.84 or 1.94 nM purified RBD-His proteins for 900 seconds and then dissociated for 100 seconds. The K D , K on and K dis were calculated by Data Analysis Octet.

Competitive inhibition assay was performed using SARS-CoV-2 pseudovirus as described previously (Nie et al., 2020) . Briefly, Huh7 cells were seeded in a 96-well plate at 50,000 cells/well for 20 hours. The cells were incubated with 50 µg/ml of BSA, RBD (purified RBD-His protein), or recombinant RBD-His (rRBD, Sino Biological) for 1 hour at 37 °C, followed by treatment with 650 TCID 50 /well of the pseudovirus for 1 hour at 4 °C. Cells were washed with DMEM medium for 3 times and cultured at 37 °C for 22 hours. Luciferase substrate (PerkinElmer) was then added to plates followed by incubation in darkness at room temperature for 2 minutes. The lysate was transferred to white solid 96-well plates for the detection of luminescence using GloMax® 96 Microplate Luminometer (Promega).

HEK293F cells were seeded in a 24-well cell culture plate at 100,000 cells/well in opti-MEM TM I Reduced Serum Medium (Thermo Fisher Scientific). Eighteen hours later, 1 mg of RBDencoding mRNA and equal amount of LNP were transfected into cells using Lipofectamine TM 2000 Reagent (Thermo Fisher Scientific) following the manufacturer's guidelines. The cells were further cultured with 5% CO 2 at 37 °C for 15 hours. Culture media were collected and analyzed by ELISA as described below.

Female ICR mice (4-6-week-old) were purchased from Shanghai SLAC Laboratory Animal Co.,

Ltd. (Shanghai, China). Twenty mice were randomly divided into two groups (n=10/group).

mRNA-LNP or LNP was intravenously administrated at 1 mg/kg into animals. The orbital blood was collected at 6 hours after administration, centrifuged at 5,000 g at 4 °C for 10 minutes. Sera were collected and stored at -80 °C for further test. RBD expression level was determined by ELISA as described below.

Evaluation of RBD expression in vitro and in vivo was performed by ELISA. Briefly, 96-well microtiter plates were coated with 5 µg/ml of human ACE2 (Kactus Biosystems) overnight at 4 °C. The coated plates were washed once with PBS and blocked with 5% BSA at 4 °C for 12

hours. Plates were then washed twice with PBS and incubated with serial dilutions of cell culture media or mouse sera at room temperature for 1 hour, prior to three further washes and subsequent 1 hour incubation with SARS-CoV-2 S rabbit mAb (Sino Biological) as primary antibody at room temperature. After three washes with PBS, plates were incubated with HRPconjugated goat anti-rabbit IgG-Fc antibody as secondary antibody (Sino Biological), followed by incubation with TMB substrate (Solarbio). The absorbance at 450/620 nm was measured and accurate quantification were conducted using SpectraMax iD3 (Molecular Devices).

For detection of in vivo distribution of FLuc mRNA-LNPs, female BALB/c mice aged 6-8 weeks (n=18) were inoculated with 10 µg of the FLuc mRNA-LNP via intramuscular (i.m.), subcutaneous (s.c.) or intranasal (i.n.) routes, respectively. At indicated times post inoculation, animals were injected intraperitoneally (i.p.) with luciferase substrate (Promega). After reaction for 3 minutes, fluorescence signals were collected by IVIS Spectrum instrument (PerkinElmer)

for 60 seconds. For in vitro imaging, female BALB/c mice of 6-8 weeks old (n=2) were intramuscularly inoculated with 10 µg FLuc mRNA-LNP and LNP, respectively. Six hours later, animals were injected intraperitoneally (i.p.) with luciferase substrate (Promega) followed by reaction for 3 minutes. Tissues including brain, heart, liver, spleen, lung, kidney and muscle were collected immediately, and fluorescence signals of each tissue were collected by IVIS imager for 60 seconds. The fluorescence signals in regions of interest (ROIs) were quantified using Living Image 3.0.

FLuc mRNA-LNP was incubated at 4, 25 or 37 °C for 1, 4, and 7 days. Female BALB/c mice aged 6-8 weeks (n=27) were inoculated with 10 µg of the incubated FLuc mRNA-LNP via intramuscular (i.m.) route. Six hours after administrations, animals were injected intraperitoneally (i.p.) with luciferase substrate (Promega), followed by reaction for 3 minutes.

Fluorescence signals were collected by IVIS Spectrum instrument (PerkinElmer) for 60 seconds, and the fluorescence signals in regions of interest (ROIs) were quantified using Living Image 3.0.

The expression of RBD in tissues from ARCoV or placebo vaccinated mice was detected by multiplex immunofluorescent assay. Mouse lung or muscle paraffin sections (4 µM) were deparaffinized in xylene and rehydrated in a series of graded alcohols. Antigen retrievals were performed in citrate buffer (pH6.0) with a microwave (Sharp) for 20 minutes at 95 °C followed by a 20 minutes cool down at room temperature. Multiplex fluorescence labeling was performed using TSA-dendron-fluorophores (NEON 7-color Allround Discovery Kit for FFPE (Histova Biotechnology). Briefly, endogenous peroxidase was quenched in 3% H 2 O 2 for 20 minutes, followed by blocking reagent for 30 minutes at room temperature. Primary antibody was incubated for 2 to 4 hours in a humidified chamber at 37°C, followed by detection using the HRP-conjugated secondary antibody and TSA-dendron-fluorophores. Afterwards, the primary and secondary antibodies were thoroughly eliminated by heating the slides in retrieval/elution buffer (Abcracker®, Histova Biotechnology) for 10 seconds at 95 °C using microwave. In a serial fashion, each antigen was labeled by distinct fluorophores. Multiplex antibody panels applied in this study are: SARS-CoV S1 subunit protein (1:1000, Sino Biological); Glutamine . After all the antibodies were detected sequentially, the slices were imaged using the confocal laser scanning microscopy platform Zeiss LSM880.

For single-dose immunization, groups of 6-8-week-old female BALB/c mice were immunized intramuscularly with ARCoV mRNA-LNP (2 µg, n=7; 30 µg, n=8), or Placebo (n=5) in 50 µl using a 3/10cc 29½G insulin syringe (BD Biosciences). Serum was collected at 1 day before immunization and 14 and 28 days post immunization for detection of SARS-CoV-2 -specific IgG and neutralizing antibody responses as described below. For two-dose immunization, groups of 6-8-week-old female BALB/c mice were immunized intramuscularly with ARCoV mRNA-LNP

(2 µg, n=8; 10 µg, n=5) in 50 µl using a 3/10cc 29½G insulin syringe (BD Biosciences), and boosted with equal dose of ARCoV mRNA-LNP on day 14 post initial immunization. Sera were collected at 1 day before initial immunization and days 7, 14, 21 and 28 after initial immunization for detection of SARS-CoV-2 -specific IgG and neutralizing antibody responses as described below. Spleen tissues were collected at day 28 post initial immunization for evaluation of cellular immune responses by ELISPOT and flow cytometry as described below.

The SARS-CoV-2 challenge model based on the mouse adapted strain MASCp6 has been characterized in detail . BALB/c mice immunized with ARCoV were challenged intranasaly with MASCp6 (6,000 PFU/mouse) at the indicated times. On day 5 post challenge, all animals were sacrificed, and the lung and trachea tissues as well as sera were collected for subsequent antibody detection, viral RNA level determination, histopathology assay, immunofluorescence staining and RNA ISH assay as described below.

A total of 30 adult cynomolgus monkeys (weighing 2.3-4.6 kg) were purchased from Guangzhou Xusheng Biotechnology Co., Ltd (See details in Table S4 ). Animals with similar age and weight were allocated to each group (male/female ratio=1:1). All animals were immunized intramuscularly with 100 µg (n=10) or 1000 µg (n=10) of ARCoV mRNA-LNP, and boosted with the same dose of ARCoV mRNA-LNP on 14 days post initial immunization. Empty LNPs were set as placebo control (n=10). Clinical signs were recorded during a 14-day observation period. Blood was collected before immunization and 14 and 28 days after initial immunization to detect SARS-CoV-2 specific IgG and neutralizing antibodies as described below. PBMCs were collected on day 19 after initial immunization for antigen specific T cell detection. IFN-γ and IL-4 levels were determined by ELISPOT and flow cytometry as previously described (Erhart et al., 2018; Rodriguez-Ruiz et al., 2018) Sera antibody titer evaluation (c) PRNT assay. PRNT was performed as described previously (Li et al., 2018 water. The 50% neutralization titer (PRNT 50 ) was calculated by the method of Spearman-Karber (Hamilton et al., 1977) .

Cellular immune responses in the vaccinated mice were assessed using IFN-γ, TNF-α, IL-2, IL-4, or IL-6 precoated ELISPOT kits (MabTech), according to the manufacturer's protocol.

Briefly, the plates were blocked using RPMI 1640 (Thermo Fisher Scientific) containing 10% FBS and incubated for 30 minutes. Immunized mouse splenocytes were then plated at 300,000 cells/well, with peptide pool for SARS-CoV-2 RBD protein (2 µg/ml of each peptide, see Table   S3 ), Concanavalin A (ConA, Sigma) as positive control or RPMI 1640 media as negative control.

After incubation at 37 °C, 5% CO 2 for 36 hours, plates were washed with wash buffer and biotinylated anti-mouse IFN-γ, TNF-α, IL-2, IL-4 or IL-6 antibody was added to each well followed by incubation for 2 hours at room temperature. Following the addition of AEC substrate solution, the air-dried plates were read using the automated ELISPOT reader AID ELISPOT (AID). The numbers of spot-forming cells (SFC) per 1,000,000 cells were calculated.

T cell proliferation in immunized mice were evaluated using a FACSCalibur flow cytometer (BD Biosciences). Briefly, a total of 1,000,000 mouse splenocytes were stimulated with SARS-CoV-2 RBD peptide pool (2 µg/ml of each peptide, see Table S3 ) for 2 hours at 37°C with 5% CO 2 .

Brefeldin A (1 µg/ml, MCE) was then added into splenocytes and incubated for 4 hours.

Following two washes with PBS, splenocytes were permeabilized and stained with fluorescently CoV-P3 (5'-FAM-AGCTGCAGCACCAGCTGTCCA-BHQ1-3'). Viral RNA load was expressed on a log10 scale as viral RNA equivalents per g after comparison with a standard curve produced using serial ten-fold dilutions of SARS-CoV-2 RNA.

For histopathology, lung tissues from mice were fixed in 4% neutral-buffered formaldehyde for 48 hours, embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E).

Images were captured using Olympus BX51 microscope equipped with a DP72 camera. Original magnification was 20×.

For immunostaining, paraffin tissue sections were deparaffinized with xylene, rehydrated through successive bathes of ethanol/water and incubated in 3% H 2 O 2 at room temperature. The sections were then put in 10 mM sodium citrate buffer for 1 hour at 96 °C for antigen retrieval and blocked with BSA at saturation for 20 minutes. Primary antibody against SARS-CoV S protein (Sino Biological) was incubated for 2 hours in a humidified chamber at 37°C, followed by detection using the TSA-dendronfluorophores. Original magnification was 20×.

SARS-CoV-2 genome RNA ISH assay was performed with RNAscope® 2.5 HD Reagent Kit (Advanced Cell Diagnostics) according to the manufacturer's instruction. Briefly, formalin-fixed paraffin-embedded tissue sections of 5 µm were deparaffinized by incubation for 60 minutes at 60 °C. Endogenous peroxidases were quenched with hydrogen peroxide for 10 min at room temperature. Slides were then boiled for 15 minutes in RNAscope Target Retrieval Reagents and incubated for 30 minutes in RNAscope Protease Plus before probe hybridization. Tissues were counterstained with Gill's hematoxylin and visualized with standard bright-field microscopy.

Original magnification was 40×.

All data were analyzed with GraphPad Prism 8.0 software. No statistical methods were used to predetermine sample size, unless indicated. The investigators were not blinded to allocation during experiments and outcome assessment unless indicated (RT-qPCR). Unless specified, data are presented as mean ± SEM in all experiments. Analysis of variance (ANOVA) or t-test was used to determine statistical significance among different groups (*P< 0.05; **P< 0.01; ***P< 0.001; ****P< 0.0001; n.s., not significant).

Messenger RNA-Based Vaccines Against Infectious Diseases

SARS-CoV-2 Vaccines: Status Report

Preclinical and Clinical Demonstration of Immunogenicity by mRNA Vaccines against H10N8 and H7N9 Influenza Viruses

SARS-CoV-2 viral spike G614 mutation exhibits higher case fatality rate

The time course of the immune response to experimental coronavirus infection of man

SARS-CoV-2 mRNA Vaccine Development Enabled by Prototype Pathogen Preparedness. bioRxiv

Immunological analysis of phase II glioblastoma dendritic cell vaccine (Audencel) trial: immune system characteristics influence outcome and Audencel up-regulates Th1-related immunovariables

Development of an inactivated vaccine candidate for SARS-CoV-2

Rapid COVID-19 vaccine development

Rapid adaptation of SARS-CoV-2 in BALB/c mice: Novel mouse model for vaccine efficacy. bioRxiv

Trimmed Spearman-Karber method for estimating median lethal concentrations in toxicity bioassays

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Lipid-based nanoparticle formulations for small molecules and RNA drugs

The promise of mRNA vaccines: a biotech and industrial perspective

Don't rush to deploy COVID-19 vaccines and drugs without sufficient safety guarantees

MAFFT multiple sequence alignment software version 7: improvements in performance and usability

Spike mutation pipeline reveals the emergence of a more transmissible form of SARS-CoV-2. bioRxiv

Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor

Development of a chimeric Zika vaccine using a licensed live-attenuated flavivirus vaccine as backbone

Visualization of early events in mRNA vaccine delivery in non-human primates via PET-CT and near-infrared imaging

Antispike IgG causes severe acute lung injury by skewing macrophage responses during acute SARS-CoV infection

Comparison of SARS-CoV-2 infections among 3 species of non-human primates. bioRxiv

mRNA as a Transformative Technology for Vaccine Development to Control Infectious Diseases

Novel vaccine technologies for the 21st century

Linkage between endosomal escape of LNP-mRNA and loading into EVs for transport to other cells

Detection of SARS-CoV-2-Specific Humoral and Cellular Immunity in COVID-19 Convalescent Individuals

Establishment and validation of a pseudovirus neutralization assay for SARS-CoV-2

Monoclonal antibodies to the spike protein of feline infectious peritonitis virus mediate antibody-dependent enhancement of infection of feline macrophages

Recent advances in mRNA vaccine technology

Expression kinetics of nucleoside-modified mRNA delivered in lipid nanoparticles to mice by various routes

The SARS-CoV-2 receptor-binding domain elicits a potent neutralizing response without antibodydependent enhancement. bioRxiv

Comparative pathogenesis of COVID-19, MERS, and SARS in a nonhuman primate model

Combined immunotherapy encompassing intratumoral poly-ICLC, dendritic-cell vaccination and radiotherapy in advanced cancer patients

Immunological Lessons from Respiratory Syncytial Virus Vaccine Development

Rational design of cationic lipids for siRNA delivery

Structural basis of receptor recognition by SARS-CoV-2

A thermostable messenger RNA based vaccine against rabies

Characterization of the receptor-binding domain (RBD) of 2019 novel coronavirus: implication for development of RBD protein as a viral attachment inhibitor and vaccine

ChAdOx1 nCoV-19 vaccination prevents SARS-CoV-2 pneumonia in rhesus macaques

Structure, Function, and Antigenicity of the SARS-CoV-2 Spike Glycoprotein

Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in

A new coronavirus associated with human respiratory disease in China

Duration of antibody responses after severe acute respiratory syndrome

DNA vaccine protection against SARS-CoV-2 in rhesus macaques

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Safety, tolerability, and immunogenicity of a recombinant adenovirus type-5 vectored COVID-19 vaccine: a dose-escalation, open-label, non-randomised, first-in-human trial

A Novel Coronavirus from Patients with Pneumonia in China

We thank Drs. Chunyun Sun, Jing Ma, and Qiang Gao for critical materials, and helpful discussion. This work was supported by the National Key Research and Development Project of China (2020YFC0842200, 2020YFA0707801, 2016YFD0500304), and the Special Grant from

Development of LNP-encapsulated mRNA vaccine (ARCoV) targeting the RBD of SARS-CoV-2ARCoV induces neutralizing antibodies and T-cell immunity in mice and NHPs.ARCoV vaccination confers full protection against SARS-CoV-2 challenge in mice.

ARCoV is an LNP-encapsulated mRNA vaccine platform that is highly immunogenic and safe in mice and non-human primates, conferring protection against the challenge a SARS-CoV-2 mouse adapted strain.