Synthesis of antigenic peptide pools of SARS-CoV-2 structural and functional proteins
Detailed analyses of the viral genome sequences of SARS-CoV-2 virus with the genomes of SARS virus and MERS virus revealed that the structural proteins Spike (S) and Membrane (M) of the coronavirus have high mutation rate, while the Envelope (E), Nucleocapsid (N) and Protease (P) regions are highly conserved. To identify potential vaccine targets of SARS-CoV-2, we analyzed the genes of all structural proteins of the virus, including S, M, E, N, and the polyprotein cleavage protease (P), and synthesized selected pools of pentadecamer peptides spanning across important functional domains of these polyproteins, including the Receptor Binding Domain (RBD) of S protein, the full-length of E membrane protein, the entire M protein as it is the most abundant protein of coronavirus, and the NBD plus the SR domains including the helix-turn-helix motif of N protein, and the domain III of Mpro (Fig. 1 a & b).
Antigen-specific immune effector cells can be generated through antigen presentation by DCs. The protocol includes in vitro generation of DC, followed by antigen exposure and co-culture with autologous lymphocytes. We isolated peripheral mononuclear cells (PBMCs) from healthy volunteers, and generated mature DCs from monocytes. The specific antigen pools of SARS-CoV-2 were used to pulse DCs and then cocultured with peripheral blood lymphocyts to activate antigen-specirtic T cells as illustrated in Fig. 1c.
Assessment of SARS-CoV-2 antigen-specific precursor frequency in healthy individuals
To investigate the frequency of immune effectors to SARS-CoV-2, we examined antigen-specific T cells in 19 healthy volunteers who have no known prior exposure to the SARS-CoV-2. PBMCs were isolated from the whole blood of donors, and the pooled viral S, M, E, N and P peptides, as well as a control HIV peptide pool, were added to the PBMCs to activate immunogenic response for 17 hours, followed by IFN-γ ELISPOT analysis. The quantification of IFN-g specific spots represented the fold of antigen-specific T cell activation and expansion. Positive response was arbitrarily set at a 1.5-fold increase in the numbers of IFN-γ-secreting T cell spots in the test wells versus the control wells (PBMC alone or treated with HIV peptides).
We observed a diverse range of SARS-CoV-2-specific primary T cell frequencies in the healthy population, as presented in Fig. 2a. The antigen specific response was quantified based on the fold increase of spots with respect to the PBMC alone group (Fig. 2b). We set the scales as low responders if the number was 1.5-2 fold higher than the PBMC background, median responders if it was 2-3 fold higher, and high responders if it was 3 folds or higher. As shown in Fig. 2c, the five mixed peptide pools (SMENP) and the N peptide pool induced the most robust response (16% and 15%, respectively) as compared to the other four individual peptide pools (Fig. 2c). The individuals had different response preference to the various viral antigens, with more than 70% of the tested subjects showing no detectable cellular immune response to the SARS-CoV-2 antigens (no responders). Furthermore, the high responders to the S peptide pool, mainly the RBD domain that is the popular immunogenic target used in the COVID-19 vaccine design, is only 4%, similar to those of the high responders to the M and P peptide pools.
Activation of SARS-CoV-2-specific immune cells in vitro
To investigate whether T cells in different respondrers could be activated by the SARS-CoV-2 viral antigens, we selected donors with different responses in the precursor frequency test to perform an in vitro T cell activation assay. Immature DCs were generated from adherent blood monocytic cells for 5 days in the presence of GM-CSF and IL-4 16. T cells were co-cultured with DCs pulsed with the various SARS-CoV-2 peptide antigens, including pooled SMENP, and the individual viral protein peptide pools, or a negative control HIV peptide pool, for 12 days, followed by ELISPOT analyses. The results showed that the activation potential of the individual antigen-specific T cells correlated with their corresponding T cell precursor frequencies, i.e., the high responders developed the strongest T cell response, and the no responders developed little to no response (Fig. 3a). The high responder group showed enhanced specific cellular immune response by more than 30 fold after 12 days in culture, and the medium responder group and low response group increased about 15 folds and 10 folds, respectively. On the other hand, the expansion of the antigen-specific T cells in the no responder group was relatively low (Fig. 3b). Again, there was a diversity in the individual preference in response to the various viral antigens of SARS-CoV-2, e.g, the high responder had the lowest response to the E antigen, whereas the no response donor #2 had the highest response to the E antigen.
Immune booster to enhance specific anti-viral cellular immunity
Immune booster is part of a standard vaccination protocol. To see if the anti-viral immune response could be enhanced in the no responders by an immune booster application, we re-stimulated the in vitro cultured T cells from the two no responders with DCs pulsed with the same SARS-CoV-2 antigen peptide pools, and extended the culture for 30 days. ELISPOT assay was then performed to measure the specific T cell responses. As compared with the 12 day results, the background value of the non-specific cells decreased significantly (Fig. 4a), and the specific expansion of T cells increased to more than 20 folds (Fig. 4b). The result suggests that individuals with low frequency of immune response to the SARS-CoV-2 virus could benefit from booster vaccination.
Anti-SARS-CoV-2 effector activities of the in vitro DC-SEMNP-activated T cells
Upon TCR engagement and stimulation by antigens in association with MHC molecules, specific immune effector functions can be demonstrated by the activation and release of specific effector molecules such as IFN-g, TNF-α, IL-2 and CD107a 17–19. We examined the SARS-CoV-2 viral antigen-specific T cell response by intracellular staining for TNF-α, IFN-g, IL-2 and CD107a. The generation of a SARS-CoV-2 specific T cell response was determined by comparing T cell stimulation with a control HIV peptide pool. We observed several folds of increases in the SARS-CoV-2 antigen-reactive T cells over the control T cells (Fig. 5), indicating that DCs presenting the SMENP epitopes elicited a strong anti-SARS-CoV-2 T cell response.