Single intranasal injection of live attenuated parainfluenza virus vector SARS-CoV-2 vaccine protects hamsters | NASA

2021-12-08 11:06:35 By : Mr. jeff wang

View all hidden authors and organizations

Edited by Peter Palese, Microbiology, Icahn School of Medicine, Mount Sinai, NY, NY; received May 26, 2021; received November 6, 2021

Although SARS-CoV-2 infection in children is usually mild, it is associated with a large number of morbidities and contributes to the dynamics of transmission. There is no SARS-CoV-2 vaccine for young children. The bovine/human parainfluenza virus 3 (B/HPIV3) vector used for intranasal immunization in children was previously evaluated in a phase 1/2 study and is well tolerated in children as young as 2 months old. This manuscript describes a B/HPIV3 vector that expresses a stable pre-fusion version of SARS-CoV-2 S protein (S-2P), and shows that a single intranasal dose is highly immunogenic and works well in hamster models , The most powerful SARS-CoV-2 challenge model available. Based on these results, B/HPIV3/S-2P is a promising candidate vaccine for clinical evaluation as a pediatric vaccine for intranasal immunity against HPIV3 and SARS-CoV-2.

All age groups need a single-dose vaccine that can limit the replication of SARS-CoV-2 in the respiratory tract, which can help control COVID-19. We have developed an intranasal live vector vaccine for infants and children against COVID-19, which is based on a replication-competent chimeric bovine/human parainfluenza virus expressing native (S) or stable before fusion (S-2P) SARS Type 3 (B/HPIV3) -CoV-2 S spike protein, the main protective and neutralizing antigen of SARS-CoV-2. B/HPIV3/S and B/HPIV3/S-2P replicate and stably express SARS-CoV-2 S as efficiently as B/HPIV3 in vitro. Pre-fusion stably increases the S expression of B/HPIV3 in vitro. In hamsters, compared with B/HPIV3/S serum SARS-CoV-2 neutralizing antibody (12 times higher), serum IgA and SARS IgG, a single intranasal dose of B/HPIV3/S-2P induced drops The degree is significantly higher than B/HPIV3/S. CoV-2 S protein (5 times and 13 times), and receptor binding domain IgG (10 times). The antibody showed broad neutralizing activity against SARS-CoV-2 of the A, B.1.1.7 and B.1.351 lineages. 4 weeks after immunization, hamsters received 104.5 50% tissue culture infection dose (TCID50) of SARS-CoV-2 intranasal challenge. In hamsters immunized with the B/HPIV3 empty vector, the average titer of SARS-CoV-2 replication was 106.6 TCID50/g in the lungs and 107 TCID50/g in the nasal tissues, and resulted in moderate weight loss. In hamsters immunized with B/HPIV3/S, the SARS-CoV-2 attack virus was reduced by 20 times in the nasal tissues and was undetectable in the lungs. In hamsters immunized with B/HPIV3/S-2P, no infectious attack virus was detected in the nose and lungs; B/HPIV3/S and B/HPIV3/S-2P were completely prevented after SARS-CoV-2 attack Weight loss. B/HPIV3/S-2P is a promising candidate vaccine that can protect infants and young children from HPIV3 and SARS-CoV-2.

The β-coronavirus Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) emerged in 2019 and spread rapidly across the world (1). In the first year of the pandemic, more than 105 million infections and 2.3 million deaths were reported globally, including more than 27 million cases and 500,000 deaths in the United States (https://covid19.who.int/). Vaccines are being deployed rapidly to control persistent infections and the emergence of worrying mutations (2), accompanied by increased virulence and changes in antigenicity.

SARS-CoV-2 is mainly infected and spread through the respiratory tract (3, 4), and the mucosal surface of the respiratory tract is the main site of infection. COVID-19 is a disease caused by SARS-CoV-2, which is characterized by upper and lower respiratory tract symptoms, fever, chills, body aches and fatigue, and in some cases gastrointestinal tract and other tissue involvement Other symptoms (5, 6).

SARS-CoV-2 infection is triggered by a glycoprotein on the spike (S) surface, which is the main target of SARS-CoV-2 neutralizing antibodies. S protein is a trimeric class I fusion glycoprotein. Each protobody consists of two functionally different subunits, S1 and S2, which are connected by a furin cleavage site; S2 contains an additional proteolytic cleavage site S2'. S2/S2' cleavage is mediated by the transmembrane protease Serine 2 (TMPRSS2) (1, 7⇓ –9). The S1 subunit contains the receptor binding domain (RBD). The S2 subunit contains membrane fusion mechanisms, including hydrophobic fusion peptides and α-helical heptad repeats (7, 9).

The binding of S RBD to its receptor, human angiotensin converting enzyme 2 triggers a change from the closed and metastable pre-fusion conformation to the open and stable post-fusion form, driving membrane fusion to allow the virus to enter (1). The stability of S protein in its natural pre-fusion state should retain antibody epitopes, including the immunodominant sites of RBD, which are required to elicit high-quality neutralizing antibody responses (9⇓ ⇓ ⇓ –13). Therefore, the stable version of S protein before fusion is the best vaccine immunogen (13⇓ –15).

The SARS-CoV-2 vaccine is available, but it is currently restricted to people 12 years of age or older. They are injected intramuscularly and will not directly stimulate the mucosal immunity of the respiratory tract, which is the main site of SARS-CoV-2 infection and shedding. Although the main burden of COVID-19 disease is adults, infections and diseases also occur in infants and young children, leading to the spread of the virus. Therefore, the development of a safe and effective pediatric COVID-19 vaccine is essential for global control of COVID-19. The ideal vaccine should be effective in a single dose, induce long-lasting and extensive systemic immunity, and completely block the respiratory mucosal immunity of T cells and B cells from SARS-CoV-2 infection and spread.

Here, we describe a carrier SARS-CoV-2 vaccine candidate for intranasal immunization of infants and young children. The vaccine is based on an attenuated, replication-capable type 3 parainfluenza virus (PIV3) vector, called B/HPIV3 (16), which expresses the SARS-CoV-2 S protein. B/HPIV3 is composed of Kansas cattle PIV3 (BPIV3) strains, in which BPIV3 hemagglutinin-neuraminidase (HN) and fusion (F) glycoproteins (two PIV3 neutralizing antigens) have been those of human PIV3 strain JS Replacing (16, 17). The BPIV3 framework provides a host range limit for human replication and serves as the basis for a strong and stable attenuation (17, 18). B/HPIV3 was originally developed as a live vaccine candidate for HPIV3 and is well tolerated in young children (17). In addition, B/HPIV3 has been used to express the F glycoprotein of another human respiratory pathogen, Human Respiratory Syncytial Virus (HRSV), as a bivalent HPIV3/HRSV vaccine candidate. This candidate vaccine is well tolerated in children over 2 months (18) (Clinicaltrials.gov NCT00686075), and the optimized version is in further clinical development as a pediatric vaccine (19, 20). In this study, we used B/HPIV3 to express the wild-type (S) or stable before fusion (S-2P) version of the SARS-CoV-2 S protein to create candidate vaccines B/HPIV3/S and B/ HPIV3/S- 2P. These were evaluated as attenuated SARS-CoV-2 intranasal vaccine candidates in vitro and in hamsters.

B/HPIV3 is composed of BPIV3, in which the F and HN genes are replaced by HPIV3 (16) genes (Figure 1A). We codon optimized the 1,273-aa S ORF derived from the first available SARS-CoV-2 genome sequence (GenBank MN908947) (21) for human expression and placed it at the start of the PIV3 gene (transcription start Start) and end (transcription termination and polyadenylation) signals direct their expression as separate mRNAs through the PIV3 transcription mechanism (Figure 1A). We have generated a second version of the gene (S-2P), which contains two proline substitutions before stable fusion at amino acid positions K986P and V987P of S, and a furin cleavage site (residue) at S1/S2. The four amino acids in base 682 to V987P) replace 685; RRAR-to-GSAS), which ablate S1/S2 cleavage (9). We inserted each of the two S gene constructs into the full-length B/HPIV3 cDNA between the N and P genes (Figure 1A). In previous studies, this made possible the efficient and stable expression of heterologous genes with minimal impact on B/HPIV3 vector replication (22). We use the obtained cDNA to recover the recombinant B/HPIV3/S and B/HPIV3/S-2P viruses by reverse genetics, as described previously (23). We propagate the virus stock in Vero cells, which are used to make clinical grade materials for human drug delivery. Sequencing of the full-length viral genome confirmed the absence of detectable mutations.

The B/HPIV3 vector expresses the wild-type and the stable version of SARS-CoV-2 S protein before fusion, where the S1/S2 cleavage site is ablated. (A) The B/HPIV3 genome map of the SARS-CoV-2 S gene has been added. The BPIV3 gene is blue, the HPIV3 gene is red, and the S gene is yellow. Each gene, including the SARS-CoV-2 S gene, starts and ends with the PIV3 gene start (GS) and gene end (GE) transcription signals (light gray and dark gray bars, respectively). The S gene encodes the wild-type (S) or the stable uncut (S-2P) version of the S protein (9) before fusion, and the AscI restriction site is inserted to place it in the B/HPIV3 N and P genes. Two stable proline substitutions (amino acids 986/7) and four amino acid substitutions to eliminate furin cleavage sites (RRAR to GSAS, amino acids 682 to 685) in the S-2P protein are shown. (B) The stability of SARS-CoV-2 S expression was analyzed by double staining immunoplaque assay. The virus stock solution was titrated on Vero cells and analyzed by double staining immunoplaque assay basically as described in (20), using goat hyperimmune antiserum against recombinantly expressed secreted S-2P protein and against HPIV3 virus particles Rabbit hyperimmune antiserum. HPIV3- and SARS-CoV-2 S-specific staining were red and green false colors, respectively; double staining was yellow. The percentage of yellow patches (±SD) is shown at the bottom. (C) Multi-cycle replication of B/HPIV3 vector on Vero cells. Vero cells in a six-well plate were infected three times with the specified virus at an MOI of 0.01 PFU per cell, and incubated at 32°C for a total of 7 days. Every 24 hours, aliquots of the medium are collected and quickly frozen for subsequent immune plaque titration of Vero cells.

We use polyclonal hyperimmune antiserum to determine virus titer and evaluate the stability of S and S-2P expression, one for PIV3 virus particles and the other for recombinantly expressed secreted S-2P protein. From four (B/HPIV3/S) or eight (B/HPIV3/S-2P) independently recovered and cultivated stocks, 99.4 ± 1.3 and 94.9 ± 3.4% plaques of PIV3 and SARS-CoV- 2 Staining, respectively (Figure 1B, yellow), indicates the stable expression of SARS-CoV-2 S protein. B/HPIV3/S and B/HPIV3/S-2P are effective in multi-cycle replication in Vero cells and are generally similar to B/HPIV3, indicating that the presence of 3.8-kb S or S-2P inserts is not detectable It hinders the replication of the B/HPIV3 vector (Figure 1C).

In order to characterize the expression of SARS-CoV-2 S and B/HPIV3 protein in vitro, we used B/HPIV3 (negative control), B/HPIV3/S or B/HPIV3/S to infect Vero and human lung epithelial A549 cells-2P, The multiplicity of infection (MOI) is 1 plaque forming unit (PFU) per cell. Cell lysates were prepared 48 hours after infection, and Western blot analysis was performed by SDS/PAGE (under reducing and denaturing conditions) and using hyperimmune antisera against secreted SARS-CoV-2 S-2P or PIV3 antigen (Figure 2A ). For these two recombinant viruses, we detected S protein in the lysate as a band consistent with the migration of the uncut S0 precursor protein (Figure 2A, lanes 3, 4, 7 and 8, and SI appendix, Figure S1 ). In B/HPIV3/S-infected A549 and Vero cells (Figure 2A, lanes 3 and 7, and SI appendix, Figure S1), there are other smaller products, consistent with the size of the lysates S1 and S2. The absence of these smaller bands in B/HPIV3/S-2P-infected A549 and Vero cells confirms the absence of proteolytic cleavage of the S-2P protein, and the furin cleavage site has been ablated (Figure 2A, lane 4 and 8 and SI appendix, Figure S1). It is worth noting that the Western blot staining intensity of S protein bands in the two cell lines was stably increased before fusion (Figure 2A, comparing B/HPIV3/S and B/HPIV3/S-2P, lanes 3 to 4 and 7 8; and SI appendix, Figure S1).

Infected cell lysates and purified viral proteins from viral particles. (A) Vero or A549 cells in a six-well plate were infected with B/HPIV3, B/HPIV3/S, and B/HPIV3/S-2P at an MOI of 1 PFU per cell, and incubated at 32°C for 48 hours. Cell lysates were prepared, denatured and analyzed by Western blot. SARS-CoV-2 S protein is detected by goat hyperimmune serum against S protein. BPIV3 N and P proteins (approximate molecular weights: ~60 kDa and ~75 kDa) were detected by rabbit hyperimmune serum against sucrose-purified HPIV3; HPIV3 HN protein (~65 kDa) was detected by rabbit hyperimmunization against HPIV3 HN-derived peptides Serum detected HPIV3 F protein (F0: ~70 kDa; F1: ~48 kDa) through rabbit hyperimmune serum to detect recombinant and purified F extracellular domain, and then immunostained with an infrared fluorophore-labeled secondary antibody and infrared imaging. Use Image Studio software (LiCor) to acquire and analyze images. Immunostaining of GAPDH is shown as loading control. (B) The relative expression of N, S, P, HN and F proteins of B/HPIV3/S and B/HPIV3/S-2P in Vero cells. To obtain these data, three additional repeat infections were performed. Vero cell lysates are evaluated side-by-side in Western blots, as described in A (see SI appendix, Figure S1 of Western blot images). In the same experiment, protein expression was normalized to GAPDH and expressed as a fold change compared to B/HPIV3. The expression of SARS-CoV-2 S protein is related to B/HPIV3/S. (C and D) Silver staining (C) and Western blot analysis (D) of sucrose-purified B/HPIV3, B/HPIV3/S and B/HPIV3/S-2P. Vero-grown virus is purified by 30%/60% discontinuous sucrose gradient centrifugation and gently precipitated by centrifugation to remove sucrose. 1 μg protein per lane was used for SDS/PAGE, and the gel was silver stained (C) and western blot analysis (D).

The quantitative comparison of the protein expression of B/HPIV3/S and B/HPIV3/S-2P in Vero cells from three independent experiments showed that the stabilization before fusion increased the level of SARS-CoV-2 S protein expressed by the vector by about 1.6 Times. Figure 2B and SI appendix, Figure S1A). We also studied how the insertion of the S gene cassette between the BPIV3 N and P genes affects the expression of the backbone vector genes. Quantitative analysis showed that, compared with B/HPIV3, the expression levels of upstream N genes of B/HPIV3/S and B/HPIV3/S-2P increased slightly, while downstream vector genes (BPIV3 P; HPIV3 F and HN) decreased by 40 % To 90% (Figure 2A and B and SI appendix, Figure S1).

In order to evaluate whether SARS-CoV-2 S or S-2P protein may be incorporated into B/HPIV3 particles, we purified the virus cultured in Vero cells by sucrose gradient ultracentrifugation, and analyzed the protein by gel electrophoresis, silver staining and western blotting composition. Figure 2 C and D). In the silver stained gel, a polymer band consistent with SARS-CoV-2 S0 can be seen in the B/HPIV3/S-2P preparation, but not in the B/HPIV3 or B/HPIV3/S preparation. Immunostaining identified this band as S0, and also showed a small amount of S0 in the B/HPIV3/S preparation (Figure 2D), indicating that the stable version of S protein before fusion was more abundantly incorporated into B/HPIV3 carrier particles than wild-type S Lots of protein.

In order to evaluate the replication and immunogenicity of vaccine candidates in susceptible animal models, we used 5 log10 PFU of B/HPIV3/S or B/HPIV3/S-2P vaccine candidates or B/HPIV3 empty carrier to nasal 30 hamsters. Internal immunity (Experiment 1) (Figure 3A). On the 3rd and 5th days after vaccination, 8 hamsters in each group were killed to assess vector replication in the respiratory tract: turbinates and lungs were collected from 6 animals, tissue homogenates were prepared and analyzed by immune plaque assay ( Figure 3B and C), and the lungs were collected from the remaining two animals and analyzed by immunohistochemistry (IHC) (Figure 3D).

Replication and immunogenicity in hamster models. (A) In the experiment. In Figure 1, a group of 30 6-week-old golden Syrian hamsters were inoculated intranasally with 5 log10 PFU of the designated virus. On the 3rd and 5th days, 6 animals per group were killed every day, and the virus titers in the turbinates (B) and lungs (C) were determined by double staining immunoplaque assay. The individual animal titer is shown by the symbol, and the group average is shown directly below the dotted line; the maximum average peak titer of each group (regardless of the number of days) is shown in bold. The limit of detection (LOD) is 50 PFU/g tissue. In C, the average percentage of double-stained plaques is shown directly above the x-axis, indicating the stability of the S expression of the B/HPIV3 vector during replication in vivo. (D) On the 3rd and 5th days, lung tissues (n = 2 animals per group) were obtained and analyzed by IHC. The serial sections were immunostained for HPIV3 and SARS-CoV-2 antigens using hyperimmune antisera against HPIV3 virus particles and secreted S-2P protein, respectively. A representative image from day 5 is shown. The areas of bronchial epithelial cells that are positive for HPIV3 and SARS-CoV-S are marked with red and blue arrows, respectively (magnification 20 times). (Scale bar, 50 µm.) (EK) Collect serum on day 28 and evaluate serum antibody titers (n = 14 animals per group) to determine the 50% SARS-CoV-2 neutralization titer (ND50) on Vero E6 cell secretion against isolate WA1/2020 (lineage A), USA/CA_CDC_5574/2020 (lineage B.1.1.7) and USA/MD-HP01542/2021 (lineage B.1.351) (EG) or IgG ELISA titer E6 cell secretion S-2P protein (H) or S protein fragment (amino acids 328 to 531) containing SARS-CoV-2 RBD (I), or IgA titer 2P protein of S-2P protein in secreted form, by DELFIA-TRF (J) Determination. (K) The serum was also analyzed to determine 60% PRNT60 and B/HPIV3. Right above the x-axis is the average log10 antibody titer; for EG, the natural number (parentheses) for mutual neutralization titers is also provided. The asterisk indicates the significance of the difference between the groups. ns, meaningless.

As previously reported (19), B/HPIV3 replicated well in the turbinate and lung (6.3 and 5.4 log10 PFU/g on day 3, respectively) (Figure 3 B and C), and decreased by 10 times in the turbinate on day 5. The turbinate titers of B/HPIV3/S and B/HPIV3/S-2P on day 3 were 10 and 100 times lower than that of empty vector, respectively, but surprisingly, the increase on day 5 was consistent with delayed replication in the upper respiratory tract Because of the presence of the insert. Importantly, the average peak nasal titers of all three viruses irrespective of the study day were similar.

In the lungs, B/HPIV3 replication remained high (5.4 log10 PFU/g) for two days (Figure 3C). Similar to the findings of the turbinate, the average titers of B/HPIV3/S and B/HPIV3/S-2P (5.0 log10 and 4.4 log10 PFU/g) on ​​day 3 were lower than the average titers of B/HPIV3 empty carrier, despite the fact that the lungs The difference between B/HPIV3 and B/HPIV3/S did not reach statistical significance. By day 5, the titer of B/HPIV3/S was about 10 times higher than that of the empty vector on any day, indicating that the wild-type S protein contributes to the replication of the vector in the lung. The peak titer of B/HPIV3/S-2P in the lungs was also slightly higher than that of B/HPIV3, but this was not statistically significant.

In order to determine the stability of B/HPIV3 vector in vivo S and S2-P protein expression, we analyzed lung samples by double staining immunoplaque assay; in the lung samples obtained on day 3 after infection, 99.5% and 98.4% of B /HPIV3/S and B/HPIV3/S-2P plaques stably expressed S protein, 99.4% and 97.9% of B/HPIV3/S and B/HPIV3/S-2P plaques obtained on day 5 expressed S protein (Figure 3C, bottom image). Therefore, the vector expression of the two forms of S protein is stably maintained in vivo. We failed to detect infectious B/HPIV3, B/HPIV3/S and B/HPIV3/S-2P in homogenized brain, kidney, liver, spleen or small intestine, indicating that S protein expression did not widen B/ The tropism of HPIV3 carrier hamsters.

We analyzed the antigen expression in the lungs of two immunized animals in each group by IHC on the 3rd and 5th days after immunization; a representative IHC image is shown in Figure 3D. We mainly detected the B/HPIV3 antigen in the columnar epithelial cells of the inner wall of the bronchioles, as shown in the tissues of the B/HPIV3, B/HPIV3/S and B/HPIV3/S-2P immunized animals obtained on the 5th day (Figure 3D , The red arrow above). In animals immunized with B/HPIV3/S and B/HPIV3/S-2P, SARS-CoV-2 S antigen was also detected in the bronchioles (Figure 3D, lower blue arrow). Overall, the B/HPIV3 and SARS-CoV-2 S immunostaining patterns did not differ between animals immunized with B/HPIV3/S and B/HPIV3/S-2P.

In a small supplementary experiment (Experiment S1) (SI Appendix, Figure S2A), we vaccinated each group of six hamsters with B/HPIV3, B/HPIV3/S or B/HPIV3/S-2P, as described above, Evaluation of vaccine replication at a later time point. On day 7, residual B/HPIV3 could be detected in the nasal tissue of only one animal. In the nose tissues (three and two animals) and lungs (two and four animals) of some animals, low titers of residual B/HPIV3/S or B/HPIV3/S-2P (SI Appendix, Figure S2 B and C)). From day 0 to day 7 after vaccination, the weight gain of animals in all groups was similar to that of mock immunized control animals, indicating that hamsters have a good tolerance to the virus (SI appendix, Figure S2D).

In summary, these data show that after intranasal immunization of hamsters, B/HPIV3/S and B/HPIV3/S-2P vectors effectively infect and express SARS-CoV-2 S protein in bronchial epithelial cells. There is no significant difference between them. Tissue distribution between the pre-fusion stable version of B/HPIV3 expressing wild-type and S protein with ablation S1/S2 cleavage site.

We next evaluated the serum antibody response of the remaining animals from the experiment 28 days after intranasal immunization. 1 (n = 14 animals per group). We used SARS-CoV-2 strain WA1/2020 (representative of SARS-CoV-2 lineage A with S amino acid sequence) to measure the SARS-CoV-2 neutralizing antibody titers by 50% neutralization dose (ND50) determination The expression of B/HPIV3/S is the same (Figure 3E). As expected, no SARS-CoV-2 neutralizing antibodies were detected in animals immunized with the B/HPIV3 empty vector. B/HPIV3/S induces a very low response to serum SARS-CoV-2 neutralizing antibodies (geometric mean reciprocal ND50 titer: 0.86 log10, [1:7.2]), while B/HPIV3/S-2P induces approximately 12 times SARS -Higher titers of CoV-2 neutralizing antibodies (geometric mean reciprocal ND50 titer: 1.95 log10 [1:89.1]) (Figure 3E]. We also evaluated the neutralizing ability of serum antibodies induced by the B/HPIV3 carrier. The SARS-CoV-2 variant of concern. The isolate USA/CA_CDC_5574/2020 (UK variant, carrying N501Y, A570D, D614G, P681H, T716I, S982A) of the B.1.1.7 lineage was evaluated in the neutralization test. The D1118H characteristic mutation in the serum S protein of hamsters), and the USA/MD-HP01542/2021 (South African variant of the B.1.351 lineage, carrying characteristic mutations K417N, E484K, N501Y, D614G and A701V in the S) (24, 25 ).

It is worth noting that the serum neutralizing antibody titer induced by B.1.1.7 isolates by B/HPIV3/S-2P (geometric mean reciprocal ND50 titer: 1.97 log10 [1:93.3]) (Figure 3F) and anti-sera The antibody titer of WA1 is equivalent to /2020 (line A). For the isolate of pedigree B.1.351 (Figure 3G), we observed greater differences in neutralizing titers (geometric mean reciprocal ND50 titer: 1.72 log10 [1:52.2]). Serum antibodies from B/HPIV3/S immunized animals only showed very low neutralizing activity against representatives of these heterologous lineages, which is consistent with the low serum neutralizing antibody titers against WA1/2020 (line A).

We also used purified preparations of the secreted form of S-2P protein (Figure 3H) and S protein fragments (amino acids 319 to 591) with RBD to measure SARS-CoV-2 specific serum IgG by ELISA. Antigen (Figure 3I). Medium serum IgG titers to secreted S-2P protein and RBD were detected in B/HPIV3/S immunized animals, while IgG responses to secreted S-2P (13 times higher) and RBD (10 times higher) were detected Obviously stronger) The antigen is induced by B/HPIV3/S-2P. We also measured the titer of serum IgA to S-2P by dissociation-enhanced lanthanide time-resolved fluorescence immunoassay (DELFIA-TRF) (Figure 3J). Both B/HPIV3/S and B/HPIV3/S-2P induced a strong S-specific IgA response, which was consistent with the IgG response. The IgA titer induced by B/HPIV3/S-2P was significantly higher than that of B/HPIV3/S (5 -fold).

B/HPIV3, B/HPIV3/S and B/HPIV3/S-2P also induced a strong neutralizing antibody response in the 60% plaque reduction test against B/HPIV3 (Figure 3K). The antibody response induced by B/HPIV3/S-2P to the B/HPIV3 carrier was similar to the antibody response induced by the empty B/HPIV3 carrier control, while the antibody response induced by B/HPIV3/S was slightly lower than the control empty vector. These findings support B/HPIV3/S-2P as an excellent candidate vaccine that can induce effective neutralizing antibody responses against SARS-CoV-2 and HPIV3.

In order to evaluate the protection against SARS-CoV-2 challenge in the nose, we used B/HPIV3, B/HPIV3/S and B/HPIV3/S-2P to vaccinate groups of 10 hamsters, as described above (Figure 4A) (Experiment 4A). 2). The serum antibody response (SI Appendix, Figure S3) 27 days after immunization was comparable to that in the experiment. 1 (Figure 3 E, HJ). We used 4.5 log10 50% tissue culture infectious dose (TCID50) of SARS-CoV-2 to challenge hamsters intranasally on the 30th day, and isolated WA1/2020 from the preparations that had undergone deep genome sequencing to confirm their integrity. Monitor the animals for clinical signs and weight loss (Figure 4B).

Protect immune hamsters from SARS-CoV-2 attack. (A) In the experiment. As shown in Figure 2, a group of 10 hamsters were immunized intranasally as shown in Figure 3, and each animal was challenged intranasally with SARS-CoV-2 strain WA1/2020 with 4.5 log10 TCID50 on the 30th day after immunization. (B) Weight change after challenge; the average percentage and SD for each group (n = 5 animals) are shown. The asterisk indicates the significance of the difference between B/HPIV3/S and B/HPIV3/S-2P compared with the B/HPIV3 immunization group, which is determined by two-way ANOVA and Tukey's multiple comparison test. (C and D) Expression of inflammatory cytokines in lung tissue after challenge. On the 3rd and 5th day after challenge, 5 animals in each group were killed and the tissues were collected. Extract total RNA from lung homogenate. The cDNA was synthesized from 350 ng RNA and analyzed by qPCR using a customized 16-gene hamster-specific Taqman array. The qPCR results were analyzed using the comparative threshold cycle (ΔΔCT) method and normalized to β-actin. (C) Relative gene expression of CXCL10 and Mx2, interferon-inducible antiviral response genes, compared with the average expression levels of three unimmunized and unattended controls (dashed lines). (D) The heat map shows the expression of 12 immune response genes on day 3 after SARS-CoV-2 challenge, which is shown as a multiple increase in gene expression relative to the average of the three unimmunized and unattended controls (red) or Reduce (green). (E) SARS-CoV-2 lung viral load after challenge, expressed in log10 genome copies per gram. In order to detect the viruses gN, gE, and sgE that attack the virus WA1/2020, cDNA was synthesized from the total RNA of lung homogenate as described above, and Taqman qPCR was performed (n = 5 animals at each time point). (F and G) SARS-CoV-2 challenge virus titers were measured in homogenates of turbinates (F) and lungs (G) from five animals in each group. (C, E, F, and G) shows the individual titer, average, and SD for each group. The P value is determined in a two-way ANOVA test using Tukey's multiple comparison test. LOD, detection limit; ns, meaningless.

In the first 5 days after SARS-CoV-2 challenge, animals immunized with empty B/HPIV3 vector showed moderate weight loss, which was the only clinical symptom after challenge (an average reduction of 10% on day 5 after challenge), and Use B/HPIV3 HPIV3/S and B/HPIV3/S-2P to generally continue to gain weight. Compared with the B/HPIV3/S-2P immunized animals on day 2 and the B/HPIV3/S immunized animals on day 3, the weight loss of the empty B/HPIV3 carrier immunized group reached a significant level. In order to observe the cytokine response to SARS-CoV-2 at 3 and 5 days after infection, we extracted RNA from the lung homogenates of five animals every day, and determined the expression of key cytokine genes by Taqman assay (Figure 4C and D) ). In Figure 4C, the results of the two most strongly expressed genes after SARS-CoV-2 challenge in B/HPIV3 control immunized animals, namely CXC motif chemokine ligand 10 (CXCL10), are about 50% respectively. The fold and 40-fold increase on the 3rd and 5th day after challenge relative to the unimmunized, unchallenged control and Myxovirus resistance protein 2 (Mx2), and increased by about 40-fold and on the 3rd and 5th days. 20 times. CXCL10 is an interferon (IFN) inducible chemokine and represents a biomarker of the SARS-CoV-2 cytokine storm, providing a correlation with the severity of disease in COVID-19 patients (26). Mx2 is an interferon-induced antiviral response gene. On the 3rd and 5th day after the challenge, the protective effect provided by B/HPIV3/S and B/HPIV3/S-2P immunization was greatly reduced from the induction of CXCL10 and Mx2, which was evident in B/HPIV3/S-2P It is almost undetectable under the circumstances, especially for Mx2 expression.

Expanding on these results, we evaluated a set of 12 immune response genes, including the pro-inflammatory cytokines CC-ligand (CCL) 17, CCL22, interleukin (IL)-12p40, IL-1B, IL-2, and tumor necrosis Factor-α (TNF)-a kind); immunomodulatory factors IL-10, IL-6 and IL-21; anti-inflammatory cytokines IL-13 and IL-4; and IFN-γ (IFN-G) (Figure 4D) ). On the 3rd day after challenge, compared with B/HPIV3/S and B/HPIV3/S-2P immunized animals, the expression level of most genes in the B/HPIV3 vector control immunized animals was higher, indicating that SARS-CoV-2 The challenge induced a strong inflammatory cytokine response in animals immunized with the B/HPIV3 empty vector, but not in animals immunized with B/HPIV3/S and B/HPIV3/S-2P.

Next, we used the N and E gene-specific assays to detect SARS-CoV-2 genomic RNA and mRNA, and the assays to specifically detect subgenomic E mRNA (sgE), to evaluate the attack virus in lung homogenate by Taqman assay Load, indicating SARS-CoV-2 mRNA synthesis (Figure 4D). In the B/HPIV3 control immunized animals, we detected very high N and E genomic RNA loads, and the average copy numbers on the 3rd and 5th days after challenge were approximately 11.2 log10 and 10.4 log10 per gram, and high sgE copy number (The average is 9.6 and 8.8 log10 servings per gram on Day 3 and Day 5). In contrast, in the lung homogenates of the B/HPIV3/S immunized animals, on the 3rd and 5th days after challenge, the average of genome N and E RNA loads was about 400 to 4 times lower than that of the B/HPIV3 control immunized animals. 4,000 times. On the 3rd and 5th days after the attack, the average sgE load was reduced by about 400 times and 1,000 times. In the lung homogenates of B/HPIV3/S-2P immunized animals, the N and E genomic RNA loads are about 4 to 5 log10 lower than those of the B/HPIV3 empty vector immunized animals, and the sgE load is slightly higher than the detection limit in this assay Among them, only two of the five animals (day 3) and one (day 5) could detect sgE in the lung homogenate.

Finally, we measured the SARS-CoV-2 titers in lung and turbinate tissue homogenates obtained on days 3 and 5 after challenge (Figure 4 F and G). In the turbinates, animals immunized with empty vector had high average titers of 7.0 log10 and 5.0 log10 TCID50/g attacking SARS-CoV-2 tissue on day 3 and day 5, respectively (Figure 4F). In animals immunized with B/HPIV3/S, the titer of challenge virus in the turbinate was about 20 times lower on day 3 and was undetectable on day 5. Importantly, in animals immunized with B/HPIV3/S-2P, no challenge virus was detected in the turbinate on both days. No virus was detected in the lungs of animals immunized with B/HPIV3/S or B/HPIV3/S-2P on the 3rd or 5th day, while the average titer of SARS-CoV was 6.6 log10 TCID50/g and 4.5 log10 TCID50/g. 2 In animals immunized with empty carriers (Figure 4G).

In summary, our data firstly shows that B/HPIV3 expressing SARS-CoV-2 S has a highly protective effect against SARS-CoV-2 attack, and secondly, it shows that the stabilization of S protein before fusion greatly enhances immunogenicity and protective efficacy. Make this vector an ideal candidate for the treatment of SARS-CoV-2. Human clinical trials.

Although SARS-CoV-2 infections in children are usually mild, they are still associated with a large disease burden, and SARS-CoV-2 infections in children contribute to the dynamics of transmission (27⇓ ⇓ –30). In addition to systemic reactions, a pediatric vaccine that induces a strong immune response in the upper and lower respiratory tracts may strongly restrict SARS-CoV-2 in its main infection and shedding sites, which should strengthen protection and limit family and community transmission.

Here, we studied B/HPIV3 as a candidate intranasal SARS-CoV-2 vaccine carrier. The B/HPIV3 carrier is well tolerated and immunogenic in HPIV3 seronegative infants (17). B/HPIV3 expressing HRSV F protein is also well tolerated in children and 2-month-old infants (17, 18), and further versions are in clinical development as a bivalent HRSV/HPIV3 vaccine (15, 16 , 23, 30). Therefore, the B/HPIV3 vector has been well characterized in previous pediatric clinical studies. We inserted two versions of SARS-CoV-2 S protein into B/HPIV3: the wild-type S protein and the uncut S protein that was stable before fusion.

This work is based on our experience of using B/HPIV3 or HPIV3 to express HRSV F protein (19, 20, 22, 31, 32), and it determines: 1) Insert the extraneous protein between the N and P genes of B/HPIV3 The source gene HPIV3 optimizes the insertion expression and genetic stability (22); 2) restores the wild-type assignment at amino acids 263 and 370 in the HPIV3 HN protein (I263T and T370P), removes the previously inserted amino acid substitutions as markers and increases the backbone of the PIV3 vector Genetic stability (32); and 3) Moderately improve the immunogenicity and protective efficacy of HRSV F protein through codon optimization, and stabilize F pre-fusion conformation and modify F protein to efficiently package into B/HPIV3 carrier particles (19 , 20, 31). In the case of SARS-CoV-2 S protein, we increased the stability of the pre-fusion conformation, as demonstrated by others (9, 11, 33) before, by including two can effectively stabilize the β-coronavirus S cell Stable proline mutant proteins in the outer domain, including SARS-CoV-1 and -2 and Middle East respiratory syndrome coronavirus proteins (11, 33). In addition, we ablated the S1/S2 protease cleavage site, as in the structural study of the β-coronavirus S protein before fusion (9, 11, 33).

For vaccine development, most pre-fusion stable versions of SARS-CoV-2 S protein, including the protein encoded by the currently used mRNA vaccine (mRNA1273), contain two stable 2P mutations (amino acids 986 and 987), namely wild-type The furin cleavage site is involved in the genetic fusion of the extracellular domain of the S protein with the C-terminal trimerization motif (9, 10). In contrast, the pre-fusion stable S protein encoded by B/HPIV3/S-2P contains the 2P stable version of the full-length SARS-CoV-2 S protein, with ablated S1/S2 furin cleavage sites, and Including the complete cytoplasm and transmembrane domain. We chose to ablate the furin cleavage site based on two factors. First, although furin cleavage is not necessary for SARS-CoV-2 S fusion activation, S1/S2 cleavage makes the S2' cleavage site available for TMPRSS2 cleavage, thereby promoting S2 cleavage and fusion activation (8). Therefore, removing the furin cleavage site can best stabilize the pre-fusion conformation of the S protein (13, 34). Secondly, changing the RRAR cleavage site to the S protein version of GSAS does not induce cell-cell fusion, which indicates that the multi-base cleavage site is essential for syncytium formation (35); therefore, an alternative ablation from RRAR to GSAS The furin cleavage site provides additional protection by preventing the SARS-CoV-2 S protein from mediating fusion in the live virus vector. Importantly, for B/HPIV3/S and B/HPIV/S-2P, we did not detect any vector replication outside the respiratory tract in the hamster model, which indicates that the tropism of the B/HPIV3 vector is consistent in either case. No change.

In order to evaluate the effect of 2P mutation and ablation of the S1/S2 cleavage site on the expression and immunogenicity of the full-length S protein in the background, we used B/HPIV3/S as a control to express unmodified wild-type S protein. When we compared B/HPIV3/S-2P and B/HPIV3/S, we found that the in vitro expression of the stable non-cleaved S-2P version before fusion was increased. Since the antigen has been denatured and reduced before analysis, and the conformational epitope has been eliminated, the quantitative difference detected by Western blot should reflect the difference in protein expression, rather than the difference in antibody reactivity with S-2P compared to S. Therefore, the increased expression may be due to the decreased cellular degradation of conformationally stable S in the early secretory pathway or lysosome, which is an interesting issue for future S protein biogenesis research.

Pre-fusion stability and lack of lysis are associated with significantly better immunogenicity in the hamster model: Compared with B/HPIV3/S, B/HPIV3/S-2P replicates to similar or lower titers in the hamster respiratory tract, while Induces significantly higher IgA and IgG titers of stable S (5 times and 13 times) and RBD (10 times) before serum fusion, and higher levels of neutralizing serum antibodies against SARS-CoV-2 (9 times) Times) Titer-CoV-2 isolate WA1/2020, representative of the A series, its S amino acid sequence is the same as the amino acid sequence expressed by B/HPIV3/S. It is worth noting that the serum antibodies induced by the stable pre-fusion version expressed by B/HPIV3/S-2P neutralized isolates of the B.1.1.7 (UK) and B.1.351 (South Africa) lineages, representing two The main focus is on variants.

Most importantly, after high-dose intranasal SARS-CoV-2 attack, we did not detect infectious SARS-CoV-2 attack virus in the respiratory tissues of hamsters immunized with B/HPIV3/S-2P. However, there is protection in the upper respiratory tract at least on the 3rd day after challenge. Animals immunized with B/HPIV3/S are not fully immunized. Although the unstable version of S protein expressed by B/HPIV3/S cannot completely protect animals from attacking virus infection, it greatly reduces the replication of attacking virus in intensity and duration, and prevents weight loss and inflammation after hamster infection. Pulmonary induction of cytokines. Challenges highlight the overall effectiveness of the B/HPIV3 carrier platform.

Unexpectedly, the S-2P version, instead of the wild-type S version, was packaged into B/HPIV3 carrier particles. Why pre-fusion stabilization and ablation of the furin cleavage site lead to incorporation is unclear. In the case of the HRSV F protein, the unmodified wild-type protein is not significantly packaged into the carrier particles, and the transmembrane and cytoplasmic tail domains of the carrier F protein need to be replaced with the transmembrane and cytoplasmic tail domains of the carrier F protein. For HRSV F, packaging into carrier particles leads to a substantial increase in the quantity and neutralizing ability of immune-induced serum antibodies. This effect is similar in quality and magnitude to the effect of stabilizing HRSV F in the pre-fusion conformation (19, 31). Similarly, compared with wild-type S protein, packaging of S-2P protein into B/HPIV3 particles may help, in addition to its stability before fusion, but also higher immunogenicity.

During the SARS-CoV-1 outbreak in 2002/2003, we developed an experimental B/HPIV3 vaccine that expressed the full-length wild-type S protein of SARS-CoV-1 (23, 36), and was used clinically. It was evaluated in the previous study. A single intranasal injection of the vaccine can induce high titers of serum SARS-CoV-1 neutralizing antibodies in hamsters and African green monkeys, and limits SARS-CoV-1 infection and upper and lower respiratory tracts in both models The shedding of (23, 36). In the hamster model, the protection of the upper respiratory tract from attacking virus replication is incomplete, but it is intact in the lungs. In the African green monkey model, we could not detect the infectious attack virus at any location, which indicates that intranasal immunization with B/HPIV3 expressing SARS-CoV-1 S protein can provide complete protection against SARS-CoV-1 attack . The kinetics of SARS-CoV-2 infection and disease are different from those of SARS-CoV-1: In the case of SARS-CoV-2, the infection is often an asymptomatic stage, during which the infected person is highly contagious Sex (37, 38). Therefore, a vaccine that can induce effective respiratory tract immunity and limit respiratory tract shedding is particularly useful for SARS-CoV-2.

Live intranasal virus vaccines usually mimic natural infections and quickly induce local respiratory and systemic immunity after a single dose of vaccine. Since it is technically challenging to reliably determine hamster mucosal antibody titer, we measured the serum IgA titer of SARS-CoV-2 S protein by a highly sensitive immunoassay, indicating a single intranasal dose of B/HPIV3 The /S virus can induce an effective IgA response. Live attenuated vaccines and live vector vaccines usually induce a broad and balanced innate, antibody, and cellular response, including CD8+ and CD4+ T cell responses and resident lung T cells (39). In the case of HRSV, due to this broad and balanced response, live attenuated vaccines and live vector vaccines do not have the Th2-mediated enhancement disease associated with inactivated HRSV vaccines (40⇓ –42). Similarly, the replicated SARS coronavirus vaccine is unlikely to cause Th2 lung immunopathology, which has been observed in animal models after being immunized with the previous inactivated coronavirus vaccine candidate (43⇓ –45). This situation. In the future research of rhesus monkeys, the mucosal and systemic immune response after intranasal immunization with B/HPIV3 vector expressing SARS-CoV-2 S protein will be more extensively characterized.

The known favorable clinical safety features of the B/HPIV3 vector will accelerate the evaluation of derivatives. The vector is easy to operate through reverse genetics, and has favorable replication and stability characteristics, which is conducive to manufacturing and distribution. Live attenuated intranasal vaccines are easy to administer and are usually effective in a single dose. Intranasal vector vaccines may also be effective in primary-boost programs using mRNA vaccines or other vaccines administered through different routes. Based on these characteristics and potential applications as well as very promising results in hamster attack models, B/HPIV3/S-2P is being advanced to phase 1 pediatric clinical studies, and it is expected that SARS-CoV-2 and HPIV3 will be used in infants and young children. However, in older children and adults, immunity to HPIV3 is common and may limit the replication and immunogenicity of the B/HPIV3 vaccine vector. Based on similar candidates of viral vectors that do not normally infect humans, and therefore have no pre-existing immunity in humans, it is being developed as an intranasal vaccine for non-pediatric populations.

Human lung epithelial A549 cells (ATCC CCL-185) were grown in F12 medium (ATCC) containing 5% FBS. LLC-MK2 rhesus monkey kidney cells, African green monkey Vero cells (ATCC CCL-81) or Vero E6 cells (ATCC CRL-1586) were grown in OptiMEM (Thermo Fisher) containing 5% FBS. The Vero E6 cell line is a subclone of Vero cells and can effectively replicate SARS-CoV-2. Vero E6 cells are used in the SARS-CoV-2 neutralization test and titrate the SARS-CoV-2 attack virus.

The Sleeping Beauty Transposase System was used to generate Vero E6 cells stably expressing human TMPRSS2. This system requires two components, a transposon and a transposase vector (46). To generate the TMPRRS2 transposon, PCR was performed from Addgene plasmid #53887 [a gift from Roger Reeves, Johns Hopkins University School of Medicine, Baltimore, Maryland; http://addgene.org/53887; RRID:Addgene_53887 (47)] . The TMPRRS2 PCR fragment was cloned into the Sleeping Beauty transposon plasmid pSBbi-BH (from Eric Kowarz of Goethe University Frankfurt, Germany [Addgene Plasmid #60515; http://addgene.org] using PCR primers flanking a unique SfiI restriction recognition site /60515; RRID: Addgene_60515]). pSBbi-BH contains a constitutive bidirectional promoter flanked by SfiI cloning sites of the target gene, blue fluorescent protein and hygromycin resistance genes. After inserting the SfiI site of pSBbi-BH, the sequence of the TMPRSS2 transposon was confirmed by Sanger sequencing. The transposase plasmid pCMV(CAT)T7-SB100, the second component of the system, was a gift from Zsuzsanna Izsvak, Max Delbrück Center for Molecular Medicine in Berlin, Germany (Addgene Plasmid #34879; http://addgene.org/34879; RRID : Addgene_34879). According to the manufacturer's instructions, TMPRSS2 transposon and transposase plasmid were co-transfected into Vero E6 cells at a molar ratio of 10:1 using TransIT-LT1 transfection reagent (Mirus Bio). After transfection, Vero E6 cells were grown for 2 weeks in the presence of 250 μg/mL Hygromycin B Gold (Invivogen). The integration of TMPRSS2 transposon was confirmed by flow cytometry detection of blue fluorescent protein expression. The expression of TMPRSS2 was confirmed by Western blot using TMPRSS2 polyclonal antibody (Millipore Sigma, HPA035787). Vero E6 cells expressing TMPRSS2 were further propagated in Dulbecco's modified Eagle medium containing 10% FBS, 1% l-glutamine, and 250 µL/mL hygromycin B Gold.

SARS-CoV-2 USA-WA1/2020 Challenge Virus (WA1/2020; Lineage A; GenBank MN985325 and GISAID: EPI_ISL_404895; obtained from Natalie Thornburg, Sue Gerber and Sue Tong, Center for Disease Control and Prevention [CDC], Atlanta, GA ) Passage twice on Vero E6 cells. The amino acid sequence of the S protein of WA1/2020 is the same as that expressed by B/HPIV3/S. USA/CA_CDC_5574/2020 isolate (line B.1.1.7, GISAID: EPI_ISL_751801; CDC deposited sequence, isolate obtained from CDC) and USA/MD-HP01542/2021 isolate (line B.1.351, GISAID: EPI_ISL_751801 ; Sequences deposited by Christopher Paul Morris, Chun Huai Luo, Adannaya Amadi, Matthew Schwartz, Nicholas Gallagher, and Heba H. Mostafa, isolates obtained from Andrew Pekosz of Johns Hopkins University in Baltimore, Maryland) stable in Vero E6 cells Express TMPRSS2. The titration of SARS-CoV-2 was performed by measuring TCID50 in Vero E6 cells (48). Illumina sequence analysis confirmed that the SARS-CoV-2 attack virus library and the complete genome sequence of the B.1.1.7 and B.1.351 variants are identical to the consensus sequence, except for the small background of reading (<10%). All about SARS -CoV-2 experiments are conducted in biosafety level 3 containment laboratories approved by the United States Department of Agriculture and CDC.

The virus stock of the recombinant B/HPIV3 vector was propagated on Vero cells at 32°C and titrated by double staining immunoplaque assay, basically as described previously (20), using rabbit hyperimmune antiserum anti-sucrose gradient purification HPIV3 virus particles (22) and goat hyperimmune antiserum N25-154 target the secreted form of recombinantly expressed SARS-CoV-2 S protein (amino acids 1 to 1208), containing two proline substitutions (KV to PP, amino acid 986) And 987) and four amino acid substitutions (RRAR to GSAS, amino acids 682 to 685), stabilize the pre-fusion conformation of S and ablate the furin cleavage site between S1 and S2 (9). The plasmid encoding this secreted pre-fusion stable uncut S protein (2019-nCoV S-2P_dFurin_F3CH2S) was developed by Barney Graham and Kizzmekia Corbett, National Institute of Allergy and Infectious Diseases (NIAID) Vaccine Research Center, NIH and University Jason McLellan Provides Texas in Austin, Texas. The plasmid was transfected into Expi293 cells, and the secreted S protein was purified from the tissue culture supernatant by affinity chromatography and size exclusion chromatography and used to immunize goats. To perform double-stained immune plaque assays, the Vero cell monolayer in a 24-well plate was infected with a 10-fold serial dilution of the sample. Infected monolayer cells covered with 0.8% methylcellulose medium, cultured at 32°C for 6 days, fixed with 80% methanol, HPIV3 specific rabbit hyperimmune serum immunostaining to detect B/HPIV3 antigen, goat hyperimmune serum for the above secretion SARS-CoV-2 S detects the co-expression of S protein, followed by the donkey anti-rabbit IRDye680 IgG and donkey anti-goat IRDye800 IgG secondary antibodies coupled with infrared dye. Scan the board with Odyssey infrared imaging system (LiCor). The fluorescent stains of PIV3 protein and SARS-CoV-2 S are shown in green and red, respectively, and provide yellow plaque staining when combined.

The cDNA clone encoding the B/HPIV3 antigenome was previously constructed (49) and was previously modified by two amino acid substitutions (I263T and T370P) in the HN protein, removing the two sequence markers and restoring the complete wild-type sequence (22). The ORF encoding the full-length 1,273 aa wild-type SARS-CoV-2 S protein was codon-optimized for human expression, and a cDNA clone (BioBasic) was synthesized commercially. A second version of the cDNA was synthesized, which encodes a stable version of the S protein before the full-length fusion (called S-2P): this version contains two proline substitutions and four at the furin cleavage site above. Amino acid substitutions for secreted S protein (9). S and S-2P cDNA are the same in other respects. They are designed to have a BPIV3 gene link in front, containing (in 3'to 5'order) gene end (AAGTAAGAAAAA), intergenic (CTT) and gene start (AGGATTAATGGA) motifs, followed by a short sequence (CCTGCAGGATG)) Include the initiating ATG (underscore) in a context that is conducive to translation initiation (Figure 1) (50). The AscI site is located on the flanks of each cDNA, and the synthesized DNA is inserted into a unique AscI site that exists in the downstream non-coding region of the B/HPIV3 N gene in the B/HPIV3 antigenome cDNA (Figure 1). The sequence of the antigenome cDNA carried by the plasmid was fully confirmed by Sanger sequencing, and the plasmid was used to transfect BHK21 cells to clone BSR T7/5 as described above (23) to produce B/HPIV3/S and B/HPIV3/S -2P recombinant virus. The virus stock is grown in Vero cells, and the virus genome purified from the recovered virus is completely sequenced by Sanger sequencing on the overlapping uncloned RT-PCR fragments, confirming that there are no accidental mutations.

Vero cells in a six-well plate were infected with the specified virus in triplicate at an MOI of 0.01 PFU per cell. After the virus is adsorbed, the inoculum is removed, the cells are washed, 3 mL of fresh medium is added to each well, and cultured at 32 ℃ for 7 days. Collect 0.5 mL of medium every 24 hours and quickly freeze, add 0.5 mL of fresh medium to each well. Virus aliquots were titrated together in Vero cells in a 24-well plate by the infrared fluorescence double staining immunoplaque assay described above.

Infect Vero or A549 cells in a 6-well plate with B/HPIV3, B/HPIV3/S, and B/HPIV3/S-2P at an MOI of 1 PFU per cell, and incubate at 32°C for 48 hours. The cells were washed once with cold PBS and then lysed with 300 µL LDS Lysis Buffer (Thermo Fisher Scientific), which contained NuPAGE Reducing Agent (Thermo Fisher Scientific). Cell lysates were passed through QIAshredder (Qiagen), heated at 95°C for 10 minutes, and separated on a 4% to 12% Bis-Tris NuPAGE gel (Thermo Fisher Scientific) in the presence of antioxidants (Thermo Fisher Scientific) , And the separated protein is transferred to the polyvinylidene fluoride membrane. The membrane is blocked with blocking buffer (LiCor) and combined with goat hyperimmune serum against SARS-CoV-2 S, rabbit polyclonal hyperimmune serum against purified HPIV3 (see cells, viruses and reagents above), rabbit against SARS-CoV-2 S HPIV3 polyclonal hyperimmune serum was incubated with HN peptide (YWKHTNHGKDAGNELETC) (22) or the recombinant purified extracellular domain of HPIV3 F protein (32) in blocking buffer at 4 °C overnight. Contains mouse monoclonal antibodies against GAPDH (Sigma) to provide loading controls. Incubate the membrane with infrared dye-labeled secondary antibodies (donkey anti-rabbit IgG IRDye 680, donkey anti-goat IgG IRDye 800 or IRDye 680, and donkey anti-mouse IgG IRDye 800, LiCor). Acquire images and use Image Studio software (LiCor) to quantify the intensity of individual protein bands. The relative abundance of viral proteins was normalized by GAPDH and presented as a fold change compared to the B/HPIV3 vector.

To analyze the protein composition of virus particles, the virus was grown on Vero cells, purified from the supernatant by 30%/60% discontinuous sucrose gradient centrifugation, and gently precipitated by centrifugation to remove sucrose, as previously described (51) . Determine the protein concentration of the purified preparation before adding the lysis buffer (Pierce BCA Protein Assay Kit, Thermo Fisher Scientific), 1 µg protein per lane is used for SDS/PAGE, silver staining (Pierce Silver Staining Kit, Thermo Fisher Scientific) and protein Blotting.

All animal studies were approved by the NIAID Animal Care and Use Committee. In the Exp. 1, 5- to 6-week-old female golden Syrian hamster (Envigo Laboratories) group (n = 30), pre-screened as HPIV3 seronegative, anesthetized and inoculated intranasally with 100 µL Leibovitz’s L-15 medium (Thermo Fisher Scientific ) B/HPIV3, B/HPIV3/S, or B/HPIV3/S-2P virus containing 5 log10 PFU. On the 3rd and 5th days after vaccination, 6 hamsters in each group were killed by CO2 inhalation, and turbinates, lungs, kidneys, liver, spleen, intestines, brain and blood were collected to assess virus replication. Lung tissue samples for histology were obtained from two other hamsters in each group every day. For the quantification of B/HPIV3 vector replication, the tissue was homogenized in Leibovitz's L-15 medium, and the clarified homogenate was analyzed by the double-stained immunoplaque assay on Vero cells, as described above. On the 28th day after immunization, serum was collected from the remaining 14 animals in each group to evaluate the immunogenicity of the candidate vaccine against SARS-CoV-2 and HPIV3. The B/HPIV3 vector-specific neutralizing antibody was detected by the 60% plaque reduction neutralization test (PRNT60) (20) on Vero cells in a 24-well plate, using the B/HPIV3 version to express from the proximal end of the first promoter The position of the eGFP protein in the genome. In order to determine the response of serum neutralizing antibodies to SARS-CoV-2, a 2-fold dilution of heat-inactivated hamster serum was tested in a micro-neutralization test for the presence of 100 TCID50-replicated antibodies that neutralized SARS-CoV-2 in Vero For E6 cells, there are four wells for each dilution in a 96-well plate. Read the presence of the cytopathic effect of the virus on the 4th day. Calculate the serum dilution (ND50) that completely prevents the cytopathic effect in 50% of the wells using the formula of Reed and Muench (52). Serum IgG antibodies against SARS-CoV-2 were also measured by ELISA, using two different recombinant expression and purified forms of S: one is the secreted form of the above-mentioned S-2P (plasmid was generously provided by Barney Graham, Kizzmekia Corbett and Jason McLellan ), the other is the SARS-CoV-2 S protein fragment (amino acids 328 to 531) containing RBD, obtained from David Veesler through BEI Resources, NIAID, and NIH (53). The RBD fragment was expressed in Expi293 cells from a codon-optimized ORF and purified as described above for the secreted S-2P protein. According to the supplier's protocol, the serum IgA antibody against the secreted form of S-2P is measured by the DELFIA-TRF immunoassay (Perkin-Elmer) enhanced with europium ions.

In the supplementary Exp. S1, group (n = 6) 5 to 6 weeks old golden Syrian hamsters were immunized as described above, or simulated immunization with diluent only. The hamsters were weighed every day from day 0 to day 7 after inoculation. On the 7th day, the hamsters were killed by inhalation of CO2, and the turbinates and lungs were collected to assess virus replication.

At Exp. 2, a group of 6-week-old male golden Syrian hamsters (n = 10) were immunized as described above. On the 30th day after immunization, hamsters were challenged intranasally with 100 µL of 4.5 log10 TCID50 of SARS-CoV-2. Five hamsters in each group were killed by CO2 inhalation on the 3rd and 5th day after challenge, and tissues were collected to evaluate the challenge virus replication (n = 5 per group). The presence of the attacking virus in the clarified tissue homogenate was then assessed by the TCID50 determination of Vero E6 cells.

Follow the manufacturer's instructions to extract total RNA from 0.125 mL lung homogenate (0.1 g/mL) using TRIzol reagent, Phasemaker Tubes Complete System (Thermo Fisher) and PureLink RNA Mini Kit (Thermo Fisher). Total RNA was also extracted from the lung homogenates of three control hamsters (unimmunized and unchallenged) in the same manner. Use High-Capacity RNA-to-cDNA Kit (Thermo Fisher) to synthesize cDNA from 350 ng RNA. The low-density Taqman gene array (Thermo Fisher) is configured to contain TaqMan primers and probes for 15 hamster (Mesocricetus auratus) chemokine and cytokine genes, which were designed based on previous reports (54⇓ –56) . Hamster β-actin is included as a housekeeping gene. Add a mixture of cDNA and 2×Fast Advanced Master Mix (Thermo Fisher) to each filling port of the array card to perform real-time PCR using QuantStudio 7 Pro (Thermo Fisher). The qPCR results were analyzed using the comparative threshold cycle (ΔΔCT) method, which was standardized as β-actin, and expressed as the fold change of the average expression of three uninfected and unattended hamsters. The result in Figure 4D is shown as a heat map using the Gene Expression Similarity Survey Suite (GENESIS program, version 1.8.1, http://genome.tugraz.at).

In order to detect the viral genome N (gN), E (gE) and sgE mRNA of SARS-CoV-2 attack virus WA1/2020 in lung homogenate, cDNA was synthesized from total RNA as described above, and Taqman qPCR for N and E And sgE were performed using the previously described primers and probes (57⇓ –59) and 2x Fast Advanced Master Mix (Thermo Fisher). The determination was performed on the QuantStudio 7 Pro real-time PCR system (Thermo Fisher). Use serially diluted pcDNA3.1 plasmids containing gN, gE, or sgE sequences to generate a standard curve. The sensitivity of the Taqman test is 10 copies, which corresponds to a detection limit of 5 log10 copies per gram of tissue.

Hamster lung tissue samples were fixed in 10% neutral buffered formalin, processed by Leica ASP6025 tissue processor (Leica Biosystems), and embedded in paraffin. Next, 5-μm tissue sections were stained with H&E for routine histopathology. For IHC evaluation, sections are deparaffinized and rehydrated. After epitope retrieval, the sections were labeled 1:1,000 anti-SARS-CoV-2 S (N25-154) with goat hyperimmune serum and 1:500 with rabbit polyclonal anti-HPIV3 serum (22). Perform color development on the Bond RX platform (Leica Biosystems) according to the manufacturer's protocol. Use Bond Polymer Refine Detection Kit (Leica Biosystems) to complete DAB chromogen detection. VisUCyte anti-goat HRP polymer (R&D Systems) replaces the standard Leica anti-rabbit HRP polymer in the kit to bind the SARS-CoV-2 S goat antibody. The slides are finally removed by washing with gradient alcohol and xylene before mounting. The sections were examined by a certified veterinary pathologist using an Olympus BX51 optical microscope, and photomicrographs were taken with an Olympus DP73 camera.

Use Prism 8 (GraphPad Software) to evaluate the significance of the data set using one-way analysis of variance and Tukey's multiple comparison test. Data is only considered significant when P ≤ 0.05.

All research data are included in the main text and SI appendix.

We thank the staff of the National Institute of Allergy and Infectious Diseases (NIAID) Comparative Medicine Branch for their support of animal research; Sonja Gallo for assistance in sequence analysis; Barney Graham and Kizzmekia Corbett, NIAID Vaccine Research Center, and the University of Texas at Austin Jason McLellan of the branch school provided the plasmid 2019-nCoV S-2P_dFurin_F3CH2S, which encodes the stable version of the SARS-CoV-2 S extracellular domain, which has been used to generate ELISA antigens for our research; have helpful discussions and opponents with Peter L. Collins Draft comments. This research was supported by the internal research program of NIAID and NIH.

↵1X.L., CL and YM have made equal contributions to this work.

↵2C.LN, SM and UJB made equal contributions to this work.

Author contributions: XL, CL, YM, H.-SP, CS, LY, INM, SA, CM, CLN, SM and UJB design research; XL, CL, YM, H.-SP, CS, LY, INM, CM and CLN conducted research; RFJ, BAPL, SMB, VJM, JH and JWY contributed new reagents/analysis tools; XL, CL, YM, H.-SP, CS, LY, INM, SA, CM, CLN, SM and UJB analyzed the data; XL, CLN, and UJB wrote this paper.

Competitive interest statement: XL, CL, CLN, SM and UJB are the inventors of provisional patent application number 63/180,534, entitled "Recombinant Chimeric Bovine/Human Parainfluenza Virus 3 Expressing SARS-CoV-2 Spike Protein and Its Use "Provided by the U.S. Department of Health and Human Services.

This article is directly contributed by PNAS.

This article contains online support information https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2109744118/-/DCSupplemental.

This open access article is distributed under Creative Commons Attribution 4.0 (CC BY).

Thank you for your interest in advertising on PNAS.

Note: We only ask you to provide your email address so that the people you recommend the page to know that you want them to see it, and that it is not spam. We do not capture any email addresses.

Feedback privacy/legal

Copyright © 2021 National Academy of Sciences. Online ISSN 1091-6490. PNAS is a partner of CHORUS, COPE, CrossRef, ORCID and Research4Life.