key: cord-0919214-8m17jj7d authors: Tong, Dali; Zhang, Mei; Yang, Yunru; Xia, Han; Tong, Haiyang; Zhang, Huajun; Zeng, Weihong; Liu, Muziying; Wu, Yan; Ma, Huan; Hu, Xue; Liu, Weiyong; Cai, Yuan; Yao, Yanfeng; Yao, Yichuan; Liu, Kunpeng; Shan, Shifang; Li, Yajuan; Gao, Ge; Guo, Weiwei; Peng, Yun; Chen, Shaohong; Rao, Juhong; Zhao, Jiaxuan; Min, Juan; Zhu, Qingjun; Zheng, Yanmin; Liu, Lianxin; Shan, Chao; Zhong, Kai; Qiu, Zilong; Jin, Tengchuan; Chiu, Sandra; Yuan, Zhiming; Xue, Tian title: Single-dose AAV-based vaccine induces a high level of neutralizing antibodies against SARS-CoV-2 in rhesus macaques date: 2022-05-09 journal: bioRxiv DOI: 10.1101/2021.05.19.444881 sha: 4d6fa156a35d0602281847edf5b1d50a0d32baf4 doc_id: 919214 cord_uid: 8m17jj7d Coronavirus disease 2019 (COVID-19), which is triggered by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, continues to threaten global public health. Developing a vaccine that only requires single immunization but provides long-term protection for the prevention and control of COVID-19 is important. Here, we developed an adeno-associated virus (AAV)-based vaccine expressing a stable receptor-binding domain (SRBD) protein. The vaccine requires only a single shot but provides effective neutralizing antibodies (NAbs) over 598 days in rhesus macaques (Macaca mulatta). Importantly, our results showed that the NAbs were kept in high level and long lasting against authentic wild-type SARS-CoV-2, Beta, Delta and Omicron variants using plaque reduction neutralization test. Of note, although we detected pre-existing AAV2/9 antibodies before immunization, the vaccine still induced high and effective NAbs against COVID-19 in rhesus macaques. AAV-SRBD immune serum also efficiently inhibited the binding of ACE2 with RBD in the SARS-CoV-2 B.1.1.7 (Alpha), B.1.351 (Beta), P.1/P.2 (Gamma), B.1.617.2 (Delta), B.1.617.1/3(Kappa), and C.37 (Lambda) variants. Thus, these data suggest that the vaccine has great potential to prevent the spread of SARS-CoV-2. 1 that continues to pose a serious global public health emergency. The disease shows a 2 high infection rate, long incubation period, and rapidly emerging variants, which have 3 led to its rapid spread worldwide 1 . Many vaccines have been developed for the control 4 of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus 5 responsible for COVID-19, including vaccines based on messenger RNA (mRNA) 2, 3 , 6 viral vectors 4, 5 , recombinant proteins 6 , and inactivated SARS- 8, 9 . Indeed, 7 several of these vaccines have been shown to protect population from SARS-CoV-2 8 infection. However, most vaccines lack long-term protection efficacy 10 , and most of 9 them require two or three injections to induce neutralizing antibodies (NAbs). Therefore, 10 developing a vaccine that only requires single-dose immunization and provides long-11 term NAbs would be optimal for combating Adeno-associated virus (AAV) is a single-stranded DNA parvovirus widely used 13 for gene therapy and vaccines 11, 12, 13 . AAV vector-mediated gene therapy products have AdVs, AAVs exhibit longer lasting gene expression and lower immune response. Thus, 25 we applied AAV vectors in the current study to develop a long-term expression vaccine 26 for the prevention of The SARS-CoV-2 spike protein mediates the binding of the virus to the human 28 angiotensin converting enzyme 2 (ACE2) receptor for entry into target cells 15 . As such, 29 it is the main antigen target for vaccines. Based on the SARS-CoV-2 spike protein 1 structure (PDB: 6VXX), we found that the receptor-binding domain (RBD) was not as 2 stable as the domain that spanned the spike protein from Q321 to S591, with the C and 3 N tail forming a stabilizing beta-sheet (Figure 1a) , hereafter termed SRBD. Thermal 4 stability analysis also showed that the SRBD protein (56.88 ± 0.45 °C) was more 5 thermostable than RBD (52.28 ± 0.77 °C) (Figure 1b) . The AAV2/9 serotype was 6 chosen as the vaccine carrier due to its high transduction efficiency in muscles. To 7 assess the immunogenicity of the designed vaccines, we injected AAV-SRBD vaccines 8 intramuscularly into both C57BL/6J and NIH mice at a dose of 1 × 10 11 virus genomes 9 (vg)/mouse, respectively (Figure S1a-b). AAV-CAG-GFP (1 × 10 11 vg/mouse) was used 10 as a control. Results showed that the AAV-SRBD could express well in the muscle of 11 mice ( Figure 1c ). Moreover, AAV-SRBD resulted in high antibody titers in both NIH 12 and C57BL/6J mice (Figure 1d and e). Thus, we used SRBD as the antigen for 13 generating an AAV-based COVID-19 vaccine. To evaluate the tissue-specific 14 expression patterns of AAV2/9, we analyzed the expression of AAV-CAG-GFP in 15 several major mouse organs (Figure S1c-e). Green fluorescent protein (GFP) signaling 16 was found in the injected muscle cells and liver cells of mice, but not in other major 17 organs, i.e., heart, lung, spleen, kidney, and whole brain. Moreover, histological 18 analysis illustrated that no significant pathological changes occurred in the major 19 tissues of AAV-injected mice, e.g., lung, heart, liver, spleen, and kidney, compared with 20 the naïve C57BL/6J mice ( Figure S1f ). These results suggest that the AAV-SRBD 21 vaccine exhibited good immunogenicity and safety in mice. To further examine vaccine safety and efficacy, we tested the AAV-SRBD vaccine 23 in a nonhuman primate (NHP) species. Two groups of rhesus macaques (Macaca 24 mulatta) were used for the study. The first group included three macaques with high-25 dose vaccine (1×10 12 vg/macaque) and was used for long-term monitoring of the NAbs 26 titers. The second group included 13 macaques with different doses. 13 macaques were 27 randomly divided into three groups, then received a single-dose immunization of 28 1×10 12 vg/macaque (high-dose, four macaques), 1×10 11 vg/macaque (middle-dose, 29 6 three macaques), or 1×10 10 vg/macaque (low-dose, three macaques) of AAV-SRBD, 1 respectively. AAV-CAG-GFP (1×10 12 vg/macaque) was used as the control (three 2 macaques) ( Table S1 ). The dosage dependent effect of the vaccine, the body weight, 3 antibody titer, pathological indicators in blood and hepatic function of macaques in the 4 second group were examined until 70 post vaccination (dpv). All intramuscular-injected 5 macaques in the second group showed normal body weight post injection ( Figure S2a ). Blood samples from all macaques were collected to assess the antibody titers ( Figure 7 1f). One potential limitation of AAV vaccine application is that most humans and 8 macaques have experienced wild-type AAV exposure, which can result in pre-existing 9 AAV antibodies and inhibition of AAV transduction in primates 16 . Given this, we 10 estimated the levels of AAV2/9 antibodies in the macaques before and after vaccination. First, we randomly selected seven macaques, including two macaques each in the low-12 and middle-dose groups and three macaques in the high-dose group (group 1) to 13 evaluate their pre-existing levels of AAV2/9 antibodies. All tested macaques were 14 AAV2/9 antibody-positive ( Figure 1g and h), suggesting that AAV2/9 antibodies may 15 commonly exist in this species. Interestingly, the AAV2/9 antibodies levels did not 16 change significantly from 56 to 217 dpv compared with day 0, even in the high-dose 17 AAV-SRBD macaques. These results suggest the pre-existence of AAV2/9 antibodies 18 in macaques before immunization, and that intramuscular injection of the AAV-based 19 vaccine did not boost AAV antibody levels. Even though AAV2/9 antibodies pre-existed in the macaques, the seroconversion 21 rate (antibody titer > 800) reached 100% on 35 dpv in the high-(7/7) and middle-dose 22 (3/3) macaques, but only 33.3% (1/3) at 56 dpv in the low-dose macaques (Figure 1i ). Accordingly, the AAV-SRBD vaccine demonstrated good immunogenicity in the high-24 and middle-dose macaques, but not in the low-dose macaques, as tested by enzyme-25 linked immunosorbent assay (ELISA) (Figure 1i and Figure S2c ) and competitive 26 ELISA ( Figure S2d -g). The SRBD NAbs in the high-and middle-dose macaques 27 effectively inhibited interactions between the RBD and ACE2, and efficacy was better 28 than that of mixed sera from convalescent COVID-19 patients with severe disease (H-29 1 responses in NHPs, and vaccine efficacy appeared to be highly dose-dependent (P = 2 0.0293 in Figure 1i ; P = 0.0221 in Figure S2e ; P = 0.0090 in Figure S2g by one-way 3 ANOVA). To assess the long-term humoral immune response of the AAV vaccine, we 4 also monitored the SRBD antibody levels in the high-dose macaques (group 1) from 5 days 0 to 598 dpv ( Figure 1j ). Results indicated that SRBD antibodies emerged on 21 6 dpv in the high-dose macaques and remained at a high level until 598 dpv, with an 7 average titer higher than found in the H-M 17, 18 . The absorbance at 450nm (A450) values 8 demonstrated that the binding of RBD to SRBD antibodies increased with time but 9 decreased slightly at 364 dpv and 598 dpv (Figure 1j and S2h), as found for the 10 inhibitory ability of NAbs (Figure 1k and S2i). However, the inhibition rate was still 11 higher than that in the H-M samples at 598 dpv. These results indicate that AAV-SRBD 12 triggers a robust and long-lasting humoral response after a single-dose of vaccine, and 13 that pre-existing AAV2/9 antibodies do not interfere with the vaccine immunity. 14 The SRBD NAbs from high dose macaques (group 1) were further measured using Omicron variant was lower than that against wild-type SARS-CoV-2 virus, but all sera 26 tested still kept in high level, showing a PRNT50 from 893 to 11 112 after 182 dpv. These data together illustrated that the AAV-SRBD vaccine induced high and effective 28 NAbs against the wild-type SARS-CoV-2, Beta, Delta and Omicron variants in rhesus 29 8 macaques, and the AAV-SRBD vaccine provided efficient cross neutralization against 1 major SARS-CoV-2 variants. Based on these results, it is reasonable to believe that our 2 vaccine is broad spectrum and could provide protection for future emerging variants. To assess the antigen-specific T cell responses to the AAV-SRBD vaccine, we used 4 the RBD peptide pool to stimulate peripheral blood mononuclear cells (PBMCs) 5 collected from high-dose macaques at 35 dpv. Compared to the bovine serum albumin 6 (BSA) control, the percentages of the CD4 + IFN-γ + , CD4 + IL-2 + , CD4 + IL-4 + , and CD8 + 7 IL-4 + T cells increased under RBD peptide pool stimulation (Figure 1p and Figure S4 ). These results indicate that RBD-specific Th1 (IFN-γ + and IL-2 + ) cell and Th2 (IL-4 + ) 9 cell responses in PBMCs can be activated by stimulation of the RBD peptide pool after 10 vaccination. The toxicity of the SRBD vaccine was further evaluated in rhesus macaques. As 12 of 598 dpv, no deaths, impending deaths, or significant abnormalities in clinical 13 physiology were found in any macaque in group 1. Widely analyzed pathological 14 indicators also showed that lymphocyte subgroup (CD20 + , CD3 + , CD3 + CD4 + , and 15 CD3 + CD8 + ) distribution was normal before and after intramuscular injection ( effectively inhibited interactions between the RBD mutants and ACE2 ( Figure S8 ). 10 Therefore, the SRBD NAbs appear to offer long-term inhibitory activity against SARS- CoV-2 variants. These results indicate that the SRBD vaccine has the potential to block 12 infection from SARS-CoV-2 variants. In conclusion, we developed a single-dose vaccine that can provide long-term 14 protection against SARS-CoV-2. Our results showed that SRBD is more thermostable 15 than RBD. The AAV-SRBD vaccine could induce good seroconversion rate in both NIH 16 and C57BL/6J mice. This vaccine overcomes the multiple injection requirement of 17 current vaccines and provides high-level and long-lasting RBD NAbs. The presence of 18 pre-existing immunity to AAV2/9 here did not restrict the delivery or efficacy of AAV-19 SRBD. We suspect that AAV rapidly enters the cells and AAV antibody titer is relatively 20 low in muscles, allowing AAV-SRBD to overcome the inhibition of AAV antibodies. A 21 potent immune response to AAV-ovalbumin was observed when AAV was administered The AAV vaccines (1 × 10 12 vg/mL) were added to carbon-coated copper grids 3 previously glow-discharged at low air pressure and stained with 2% uranyl acetate for 4 90 s. The EM was operated at an acceleration voltage of 120 kV. Images were recorded 5 using a Tecnai G2 Spirit 120kV EM at 23 000× magnification. To compare the thermal stability of RBD and SRBD, circular dichroism (CD) spectra 24 were acquired on a Chirascan Spectrometer (Applied Photophysics, Leatherhead, UK). Prior to CD measurements, the sample buffers were changed to PBS and the protein 26 concentration was adjusted to 0.5 mg/mL, as determined by its absorbance at 280 nm. For thermal titration, CD spectra were acquired from 20 to 95 °C with temperature steps 2 of 5 °C and wavelengths between 180-260 nm. The CD signals at 222 nm were used to 3 characterize structural changes during thermal titration. The data were fitted by Prism 4 v6 to calculate the Tm values. Vaccine immunogenicity analysis 6 The C57BL/6J and NIH mice (8 weeks old; male and female; 20-25 g body weight for 7 C57 mice and 30-35 g body weight for NIH mice) were randomly divided into five 8 groups (five mice per group). The mice were intramuscularly injected with RBD or 9 SRBD AAV vaccines or the AAV-CAG-GFP control at a dose of 1 × 10 11 vg (20 μL). The macaques were randomly divided into four groups (seven macaques in high-dose 11 group and three macaques in other groups) and intramuscularly injected with 1 mL of 12 SRBD vaccine (1 × 10 12 , 1 × 10 11 , and 1 × 10 10 vg/macaque for high/middle/low dose, 13 respectively) or AAV-CAG-GFP control (1 × 10 12 vg/macaque). Blood was collected 14 from macaques before immunization and at every 7 days before day 42 after injection The neutralizing titer was determined as describe previously with slight modification 26 . 2 Briefly, antibodies were serially diluted with DMEM containing 2.5% FBS, and mixed 3 with equal volume of virus suspension and incubated at 37°C for 1 h. The mixture was 4 added to Vero E6 monolayer cells in 24-well plates and incubated for another 1 h, and 5 the inoculate was replaced with DMEM containing 2.5% FBS and 0.9% 6 carboxymethyl-cellulose. The plates were fixed with 8% paraformaldehyde and stained 7 with 0.5% crystal violet 3 days later. Plaque reduction neutralizing titer was calculated 8 using the "inhibitor vs normalized response (Variable slope)" model in the GraphPad 9 Prism 8.0 software. The Cut-off value is calculated by the negative control (geometric 10 mean + 3 times of geometric standard deviation). Total RNA was extracted from organs with Trizol reagent (Invitrogen, 15596026) and Organs (muscle, liver, heart, lung, spleen, kidney and brain) were sectioned (40-μm 4 thick) for immunohistochemical analysis with a freezing microtome. After washing 5 with PBS three times (5 min each time), the slices were blocked with 3% bovine serum 6 albumin (BSA) and 0.1% Triton-X100 in PBS for 1 h at room temperature. The slices 7 were then stained using 1:1 000 anti-GFP antibody (Earthox, E002030-02) in blocking All data are presented as means ± standard error of the mean (SEM), except for the 3 titers of NAbs, which were quantified using geometric mean + geometric standard 4 deviation. Student's t-test and paired t-test were used to determine the statistical 5 significance of differences between two groups. One-way analysis of variance 6 (ANOVA) and two-way ANOVA were used to determine statistical significance for 7 different dose groups and the curve graphs, respectively. Quantification graphs were 8 analyzed using GraphPad Prism v8 (GraphPad Software). *: P < 0.05; **: P < 0.01; 9 ***: P < 0.001. We thank all colleagues from the National Biosafety Laboratory (Wuhan), Chinese 12 Academy of Sciences, China, for their support during the study. We thank the Center SARS-CoV-2 vaccines in development Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine Efficacy and Safety of the mRNA-1273 SARS-CoV-2 Vaccine Safety and immunogenicity of ChAdOx1 nCoV-19 vaccine administered 10 in a prime-boost regimen in young and old adults (COV002): a single-blind, randomised, 11 controlled, phase 2/3 trial Immunogenicity and safety of a recombinant adenovirus type-5-vectored COVID-14 19 vaccine in healthy adults aged 18 years or older: a randomised, double-blind, placebo-15 controlled, phase 2 trial A vaccine targeting the RBD of the S protein of SARS-CoV-2 induces protective 18 immunity Safety and immunogenicity of an inactivated SARS-CoV-2 vaccine, BBIBP-CorV: a 21 randomised, double-blind, placebo-controlled, phase 1/2 trial provides durable immunogenicity and protection in non-human primates Adeno-associated virus vector as a platform for gene therapy 4 delivery Cell entry mechanisms of SARS-CoV-2 Neutralizing antibodies against adeno-associated virus examined prospectively in 10 pediatric patients with hemophilia Characterization of SARS-CoV-2-specific antibodies in COVID-19 patients reveals 13 highly potent neutralizing IgA Potent Neutralization of SARS-CoV-2 by Hetero-bivalent Alpaca Nanobodies 16 Targeting the Spike Receptor-Binding Domain Intramuscular injection of AAV8 in mice and macaques is associated with 19 substantial hepatic targeting and transgene expression Intramuscular Adeno-Associated Virus-Mediated Expression of (p) Percentage of CD3 + CD4 + , CD3 + CD4 + IFN-γ + , CD3 + CD4 + IL-2 + , CD3 + CD4 + IL-4 + , 6 and CD3 + CD4 + IL-17A + cells in blood of high-dose rhesus macaques activated by BSA 7 or RBD peptide (n = 6 macaques in each group). Values are means ± SEM or geometric mean + geometric standard deviation for 9 antibody titer and PRNT50. *: P < 0.05; **: P < 0.01; ***: P < 0.001.