key: cord-0984670-p0778euv
authors: Wu, Jun-Jun; Chen, Yong-Xiang; Li, Yan-Mei
title: Adopting STING agonist cyclic dinucleotides as a potential adjuvant for SARS-CoV-2 vaccine
date: 2020-07-24
journal: bioRxiv
DOI: 10.1101/2020.07.24.217570
sha: 25efe58dcbdcb65569c7679d1a5ded4f7a8b6441
doc_id: 984670
cord_uid: p0778euv

A novel STING agonist CDGSF unilaterally modified with phosphorothioate and fluorine was synthesized. CDGSF displayed better STING activity over dithio CDG. Immunization of SARS-CoV-2 Spike protein with CDGSF as an adjuvant elicited an exceptional high antibody titer and a robust T cell response, which were better than the group using aluminium hydroxide as a adjuvant. These results highlighted the adjuvant potential of STING agonist in SARS-CoV-2 vaccine preparation for the first time.

protein. And the degradation of CDNs by phosphoesterase also reduces their stimulating efficiency. To overcome the drawbacks, we decided to modify the CDNs structure with fluorine and phosphorothioate.

Here, we chose CDG as a model CDNs. To improve CDG stability, we have synthesized a novel CDG analogue CDG SF for the first time which replaced one 2'-OH with F atom and one phosphate diesters with phosphorothioate diesters at the same side (Fig. 1b) . CDG SF showed enhanced activity compared with phosphorothioated CDG (dithio CDG, Fig. 1b) . Besides, recombinant S protein immunization with CDG SF as an adjuvant elicited a robust humoral (IgG) and cellular (T cell) response, which was better than aluminium hydroxide group. 

The CDG analogue CDG SF was synthesized based on Jones and co-workers' one flask solution-phase strategy. [14] [15] [16] After a few attempts, some conditions and raw materials had been adjusted as shown in synthetic route (Scheme 1). The commercial available 2'-F guanosine phosphoramidite, D3, was adopted as the second portion to couple with deprotected guanosine D2. To obtain the mono-phosphorothioate modification, sulfurization with 3-((dimethylaminomethylidene)-amino)-3H-1,2,4-dithiazole-5-thione (DDTT) was conducted after coupling. Following cyclization with 2-chloro-5,5-dimethyl-1,3,2dioxaphosphorinane-2-oxide (DMOCP), the compound was oxidized with iodine to form the second phosphodiester (D6). Finally, the target molecule CDG SF was acquired through deprotection and crystallization, with an overall yield of 40%. 

After finishing the synthesis, we were curious about the activity of CDG SF which had not been studied before. We adopted J774A.1 cells (a STING-expressing mouse macrophage) as a model cell line, since the macrophage is a critical immune cell for the antitumor and adjuvant effect of CDNs. Considering that phosphorothioated CDNs is the preferential form for clinical researches, 17 we adopted dithio CDG (Fig. 1b) as a control group. Because of poor penetrability mentioned above, we added the agonists to cells with or without the pre-treatment of a transfection reagent (see methods for details). After incubation, the cells were labelled with anti-CD86 antibody for flow cytometry analysis. As illustrated in Fig. 2 , both of CDG SF and dithio CDG with transfection remarkably upregulated the expression level of activation marker co-stimulatory molecule CD86. More importantly, CDG SF induced higher CD86 level compared to dithio CDG either with or without the transfection. The better performance of CDG SF could be owed to the fluorine modification which brought CDG SF with improved liposolubility and stability. 18 These results indicated that CDG SF was capable of stimulating STING pathway efficiently and might be a better clinical choice relative to dithio CDG. Besides, the poor results of dithio CDG and CDG SF without transfection further highlighted the importance and urgency of developing the delivery method (Fig. 2) . 

To explore the adjuvant effect of CDG SF on SARS-CoV-2 vaccine, we chose recombinant Spike S1+S2 extracellular domain (ECD) glycoprotein (sequence: YP_009724390.1) as a model antigen. Besides, Spike (S) protein plus aluminium hydroxide (Alhydrogel® adjuvant 2%, 100 μg per mouse) was adopted as a control group. Immunizations on Babl/c mice were performed for three times biweekly (Fig. 3a) . One week after last injection, sera and spleen samples were harvested. SARS-CoV-2 S protein-specific T cell response was assessed through IFN-γ enzyme-linked immunospot assay (IFN-γ ELISPOT, see methods for details). Splenocytes were stimulated with 50 μg/mL S protein for 36 h before forming IFN-γ spot. As shown in Fig. 3b , the number of IFN-γ-secreting S protein-specific T cells in CDG SF group was much higher than Alum and S protein group. As expected, Alum adjuvant contributed the limited promotion to the cellular response of S protein. Besides, the spot numbers were corroborated by mouse spleen weight of each group (Fig. 3c ). This result proved that CDG SF could be an excellent adjuvant to significantly improve the SARS-CoV-2-specific T cell responses, prior to Alum. 

As to humoral response, we adopted enzyme linked immuno-sorbent assay (ELISA) to analyse the S protein-specific IgG titers with sera from immunized mice. The plate was precoated with 0.1 μg/well S protein. As illustrated in Fig. 4a , S protein plus CDG SF immunization elicited an exceptionally high SARS-CoV-2-specific IgG titer (endpoint titer up to 819,200), close to Qin's Inactivated SARS-CoV-2 vaccine and higher than IgG titers (about 20,000) in recovered COVID-19 patients. 19 While, Alum group did not exhibit the adjuvant effect on S protein compared with S protein alone, suggesting the need of higher aluminium hydroxide dose. The titers data demonstrated that CDG SF could also notably enhance S protein-specific humoral response.

To further compare the relative intensity of cellular and humoral response induced by CDG SF immunization, we detected antibody isotypes distribution in sera (Fig. 4b, dilution  1:12800 ). In general, the level of IgG1 isotype is related to humoral response (Th2) and the generation of IgG2a reflects the T cell activation (IFN-γ, Th1) . In anti-sera of CDG SF group, the amount of IgG1 was relatively more than IgG2a, IgG2b and IgM, along with minimal levels of IgG3 and IgA. The level ratio (2:1) of IgG1 and Ig2a revealed that the combination of S protein and CDG SF generated a balanced cellular and humoral (Th1/Th2) response. In Alum group, the higher level of IgM over IgG1 and IgG2a indicated an inefficient immune activation. The isotype results again highlighted the adjuvant potential of CDG SF in SARS-CoV-2 vaccine preparation. 

In brief, we have designed and synthesized a novel CDN CDG SF which exhibited better STING activity over dithio CDG. Besides, we proved for the first time that CDG SF could act as an excellent adjuvant to notably improve the S protein-specific T cell and IgG titer level of SARS-CoV-2 vaccine, overcoming the drawbacks of aluminium hydroxide. These results also suggest that STING agonists might provide a great and general adjuvant choice for multiple kinds of SARS-CoV-2 vaccines including inactivated virus vaccine, recombinant RBD protein vaccine, peptide vaccine and DNA/RNA vaccine. 

The CDG SF was synthesized based on the one-flask synthesis strategy by Jones et al with some adjustments. 14 Evaluation of macrophage activation in vitro using J774A.1 cell line J774A.1 cells (mouse monocyte macrophage cell line) were cultured in DMEM (dulbecco's modified eagle medium) containing 10% fetal bovine serum (FBS) at 37℃, 5% CO 2 . After harvest, the cells were planted on 24-well culture plates with a density of 5×10 5 cells/well and cultured overnight. Then, 10 μM of the compounds were added respectively and incubated for 18 h. For the transfection group, Lipofectamine® 3000 was used as a transfection reagent and conducted according to manufacturer's protocol. Samples were mixed with DMEM as a A solution (150μL). Lipofectamine® 3000 (4.5 μL) was mixed with DMEM as a B solution (150 μL). 5 min later, solution A was added to the solution B. Waiting for 15 min, the mixture was added dropwise to 24-well plate at a final concentration of 10 μM and incubated for 18h. Then, cells were harvested and strained with mouse anti-CD86phycoerythrin antibodies (BD Pharmingen, dilution 1/200) at ice for 1h. After washing, the cells were analyzed on BD Calibur flow cytometry.

SARS-CoV-2 vaccines immunization 6-8 week old Babl/c mice (4-5 mice per group, female) were separately subcutaneously vaccinated with SARS-CoV-2 S protein 5 μg/mouse, CDG SF 20 μg/mouse, Alhydrogel® adjuvant 2% 100 μg/mouse. SARS-CoV-2 Spike protein (S1+S2 ECD, gene: YP_009724390.1) was purchased from Sino Biological Inc. Alhydrogel® adjuvant 2% was purchased from InvivoGen. Immunizations were conducted for three times biweekly. Antisera and spleens were collected one week after the last administration. Mice used in the experiments were raised in Animal Facility of Center of Biomedical Analysis in Tsinghua University and treated in compliance with the animal ethics guidelines. The animal protocol (approval number: 16-LYM2) was approved by Institutional Animal Care and Use Committee (IACUC) of Tsinghua University. Animal Facility of Center of Biomedical Analysis in Tsinghua University has been authenticated by Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC).

IFN-γ ELISPOT Kit was purchased from Dakewe Biotech Co., Ltd. Spleens from immunized mice were grinded and filtered through a 40 μm cell strainer. And red blood cells (RBCs) were lysed using lysis buffer for RBC. Splenocytes were counted and added to 96well kit at 1000,000 amount/ 100 μL per well. Then SARS-CoV-2 S protein was added to the well (final concentration: 50 μg/mL) and cells were stimulated for 36 h. The spot forming procedure was performed according to the Kit instruction.

SARS-CoV-2-specific antibody titers 1 μg/mL SARS-CoV-2 S protein in coating buffer (0.1M NaHCO 3 solution, pH=9.6) was added to high-binding 96-well ELISA plate (Costar 3590, 100 μL/well). After incubation for 12h at 4℃ and washing with PBST solution (0.05% Tween in PBS buffer), the wells were blocked by 0.25% gelatin PBS solution for 3h at room temperature. After washing with PBS and PBST, the diluted antisera (1:200) was added to each well (100 μL per well) and incubated for 1.5h at 37℃. After washing again, diluted rabbit anti-mouse IgG-Peroxidase antibodies (1/2000 dilution, Sigma) were added to each well (100 μL per well) and incubated for 1h at 37℃. After washing and spin-drying, 3,3',5,5'-Tetramethylbenzidine (TMB) was added to plate (200 μL per well) and incubated for 4 min and then stopped by 2 M H 2 SO 4 (50 μL per well). Optical absorption was measured at wavelength of 450nm and antibody titer was defined as highest dilution yielding an optical absorption of 0.1 or greater over that of negative control antisera.

SARS-CoV-2-specific antibody isotypes 96-well ELISA plate was coated with SARS-CoV-2 S protein according to the procedure described above. The antisera were diluted to 1:12800 and added to each well, incubating for 1.5h at 37℃. After washing with PBS and PBST, isotype antibodies IgG1, IgG2a, IgG2b, IgG3, IgA and IgM (anti-mouse antibodies from goat, Sigma) were diluted to 1:1000 and added to each well (100μL per well). After incubation for 1.5 h at 37℃ and washing, 200μL TMB substrate described above was added to plate, incubated for 4 min and stopped by 2 M H 2 SO 4 (50 μL per well). Optical absorption was also measured at wavelength of 450 nm.

STING: infection, inflammation and cancer

STING is an endoplasmic reticulum adaptor that facilitates innate immune signalling

Type I interferon as a stimulus for cross-priming

Agonists and inhibitors of the STING pathway: Potential agents for immunotherapy

Cyclic dinucleotides and the innate immune response

Research and Development on Therapeutic Agents and Vaccines for COVID-19 and Related Human Coronavirus Diseases

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

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

SARS-CoV-2 Vaccines: Status Report

Advances in aluminum hydroxide-based adjuvant research and its mechanism

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

Targets of T Cell Responses to SARS-CoV-2 Coronavirus in Humans with COVID-19 Disease and Unexposed Individuals

The trinity of COVID-19: immunity, inflammation and intervention

The Chemistry of the Noncanonical Cyclic Dinucleotide 2'3'-cGAMP and Its Analogs

One-flask syntheses of c-di-GMP and the [Rp,Rp] and [Rp,Sp] thiophosphate analogues

Targeting STING with cyclic di-GMP greatly augmented immune responses of glycopeptide cancer vaccines

Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity

Fluorinated nucleosides as an important class of anticancer and antiviral agents

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