Twenty-one adult volunteers (mean age, 36 years old [range 27 to 61]) who were scheduled to receive the CoronaVac vaccine were recruited from professionals from the area of clinical laboratories for COVID detection at the Universidad de Santiago de Chile during March 2021. The sample size was calculated according to the associated hypothesis using the G test. Since a 60% protection has been described for this vaccine, a 60% induction of memory at 90 days post-vaccination was expected. This analysis showed that the minimum sample size was 15 volunteers. Because usually, volunteers declined their participation without completing the studies, about 30 volunteers were recruited. Inclusion criteria included the absence of suspected/diagnosed case of COVID-19 and testing negative for COVID-19 for at least six previous PCR analyses in monthly screening tests. In addition, immunological diseases or any associated chronic disease with no treatment were considered exclusion criteria. Blood samples were obtained before the vaccination process. The Scientific Ethics Committee approved the study and informed consent, Universidad de Santiago de Chile (172/2021). In addition, written informed consent was obtained from each enrolled participant. Paired samples were analyzed using Friedman and Quade test with multiple comparison post-test. Unpaired data analysis was performed using Mann–Whitney test. All statistical analyses were accomplished with GraphPad Prism version 8.01 (GraphPad Software Inc, La Jolla, CA). The results were presented as median. p < 0.05 was considered statistically significant.

Blood was collected in BD Vacutainer tubes (REF367281, Becton, Dickinson and Company, USA) by health-trained personnel. The samples obtained (15 ml at each time point) were coded, kept cold, and processed to obtain plasma and peripheral blood mononuclear cells (PBMCs). One milliliter of blood was aliquoted to obtain blood plasma by centrifugation. Then, 300 µl was transferred to cryopreservation tubes and frozen at −80°C until use. White cells were isolated by Ficoll-Hypaque gradient centrifugation (Gibco BRL lymphoprep).

The sequence coding for the spike protein modified by Amanat et al. (GenBank: MN908947.3) was commercially synthesized and cloned into the pcDNA3.1 plasmid (GenScript, Piscataway, NJ, USA). The transfection was performed using the ExpiCHO Expression System. Briefly, the ExpiCHO-S cells were cultured in a vented flask with ExpiCHO Expression medium (Thermo Fisher, Cat. A2910001) in a humidified 8% CO₂ incubator at 37°C at 400g and passaged every 3–4 days at 200,000 cells/ml, until transfection day. The day before transfection, the cells were split at 3 × 10⁶ cells/ml. After 24 h, the cells were diluted at 6 × 10⁶ cells/ml with a fresh medium. The ExpiCHO transfections were performed using the ExpiCHO Expression System Kit (Thermo Fisher, Cat. A29133), according to the manufacturer’s protocol, and cultures were then temperature shifted to 32°C and cultured for 5 to 7 days. Then, cell suspension was centrifuged at 3,220g for 20 min at 4°C to pellet cells. The supernatant was collected and centrifuged at 5,000g for 30 min, the pH was adjusted to 8.0, and inhibitor proteases were added. The supernatant was dialyzed in native dialysis buffer (50 mM Tris–HCl, pH 8.0, 250 mM NaCl) overnight at 4°C. The purification was performed by exclusion chromatography in Ni-NTA (Gold Bio, Cat H-350-100). Briefly, columns were equilibrated using native binding buffer (50 mM Tris–HCl, pH 8.0, 250 mM NaCl, 10 mM Imidazole). The supernatant was loaded onto the equilibrated columns and incubated for 2 h at 4°C in an orbital shaker. Following incubation, the supernatant was collected, and the columns were washed twice with native wash buffer (50 mM Tris–HCl, pH 8.0, 250 mM NaCl, 20 mM Imidazole). Next, the protein was eluted in the elution buffer (50 mM Tris–HCl, pH 8.0, 250 mM NaCl, 250 mM Imidazole) and concentrated in an Amicon Ultra-15 centrifugal filter unit (50,000 NMVL) at 3,220g at 4°C for 15 min. The buffer was discarded, and the protein was washed twice using storage buffer (20 mM Tris–HCl, pH 8.0, 150 mM NaCl). Proteins were analyzed by reducing SDS-PAGE, dyeing with Sypro Ruby (Thermo Fisher, Cat. S12001), and quantified using ImageJ software. Finally, the Spike protein was aliquoted and stored at −80°C, until use.

Analyses were performed in plasma samples obtained from the volunteers, before vaccination and 28 days after the second immunization. The cytokines TNF-α, IL-12, IL-6, IL-10, IL-1β, and IL-8 were quantified using the Cytometric Bead Array (CBA) System (Becton Dickinson, BD-USA). Data acquisition was achieved by flow cytometry using a FacsCanto II flow cytometer (BD-USA) and analyzed in the BD FCAP Array software, where values were expressed in pg/ml for each cytokine.

Total antibody levels were determined using a previously established protocol (https://doi.org/10.1038/s41591-020-0913-5). Briefly, ninety-six-well plates were coated at 4°C with 50 ng/well of the recombinant SARS-COV-2 S protein S1 (Biolegend 792906) in PBS. After overnight incubation, the coating solution was removed and 100 μl/well of PBS, containing 0.1% Tween-20 (PBST) and 3% non-fat milk, were added to the plates. Plates were incubated for 1 h at room temperature. The serum samples were previously heated at 56°C for 1 h, and serial dilutions of serum were prepared in 1% non-fat milk prepared in PBST. One hundred microliters of each serial dilution was added, and the plates were incubated for 2 h at room temperature. The plates were washed three times using 0.1% PBST and then 100 μl of a 1:10,000 dilution of donkey anti-human IgG–horseradish peroxidase (HRP) conjugated antibody (Clone Poly24109, Biolegend 410902) was added to each well for 1 h. Plates were again washed with 0.1% PBST and then dried. Next, 100 μl of TMB (3,3′,5,5′-Tetramethylbenzidine BD OptEIA) solution was added to each well. This substrate was left on the plates for 10 min and then the reaction was stopped by the addition of 50 μl per well of 1 M Phosphoric acid. The optical density at 490 nm was measured using an Emax (Molecular Devices, USA) plate reader. Results are calculated as the area under the curve of four-serial dilutions of patient plasma obtained by the ELISA technique.

Rapid tests (OnSite Covid-19 IgG/IgM Rapid Test-CTK Biotech, Inc, USA) were used according to manufacturer’s instructions to detect antibodies before vaccination, and at 28, 48, 78, 108, and 138 dpi.

Pseudotyped viral particles were produced by transient transfection of HEK293T cells using polyethylenimine (PEI) and plasmids pNL4.3-ΔEnv-Firefly and pCMV14–3X-Flag–SARS-CoV-2 SΔ19C in a 1:1 ratio as we recently described (8). Cell culture media was replaced and collected at 16 h and 48 h post-transfection, respectively, and centrifugated to 850g for 5 min at room temperature. Viral particles were diluted with 50% in fetal bovine serum (Sigma-Aldrich) and stored at −80°C. Viral stock was quantified with the HIV-1 Gag p24 Quantikine ELISA kit (R&D Systems) and titrated by serial dilutions. Neutralization assays were performed as we previously reported (8). Briefly, inactivated serum samples were diluted in DMEM with 10% fetal bovine serum and incubated with 3 to 5 ng of p24 of HIV-1-based SARS-CoV-2 pseudotyped particles. After 1 h incubation at 37°C, 1 × 10⁴ HEK-ACE2 cells were added to each well. HEK293T cells incubated with the pseudotyped virus were used as a negative control. Cells were lysed 48 h later, and firefly luciferase activity was measured using the Luciferase Assay Reagent (Promega) in a Glomax 96 Microplate luminometer (Promega). The percentage of neutralization for each dilution was calculated, and neutralization titers, defined as 50% inhibitory dose (ID50), were estimated as previously described (8). Briefly, ID50 for each sample was obtained using a four-parameter non-linear regression. Pass/fail quality control criteria for the assay were as follows (1): the average RLUs of pseudotype control wells are ≥10 times the average RLUs of negative control (2); the % coefficient of variation (CV) of RLUs for triplicate wells control is ≤30% (3); the % CV of duplicate is ≤30% for sample dilutions that yield at least 40% neutralization; and (4) positive control neutralization curve crosses the 50% neutralization cutoff 0 to 1 times. All statistical analyses were performed using GraphPad Prism version 8.0.1 software.

The specific activation of T lymphocytes was evaluated by IFN-γ and Granzyme B kit (R&D system). PBMCs were isolated by using Ficoll-Hypaque gradient centrifugation (Lymphoprep Gibco BRL) and leucocyte cell numbers were quantified using Trypan Blue. Anti–IFN-γ and Anti-Human Granzyme B antibody-coated wells (R&D Systems) were seeded with 300,000 leucocyte cells per well. Cells were stimulated in duplicate with 10 µM SARS-CoV-2 spike structural protein (1.3 μg/ml). OKT3 and CD28 were used as positive control of stimulation (3 and 5 μg/ml, respectively). Culture medium was the negative control. PBMCs were cultured for 48 h before enzymatic revelation of IFN-γ and Human Granzyme B (R&D Systems). Spots were counted by using an ELISPOT reader (A.EL.VIS ELISPOT Scanner) and results are expressed as the mean number of spot-forming units/3 × 10⁶ peripheral blood mononuclear cells after subtraction of the background value.

We evaluated the immune response to the CoronaVac vaccine applied in the Chilean population following an interval of 28 days between the first and second dose schedule. We measured the capability of the vaccine to activate the humoral and cellular responses in volunteers without suspected/diagnosed COVID-19 and with negative PCR over time throughout the whole study. Table 1 shows the demographic and baseline characteristics of the enrolled volunteers. We assessed the presence of anti-SARS-CoV-2 antibodies using rapid detection tests before the first dose of CoronaVac, and 28, 48, 78, 108, and 138 dpi. We found that all volunteers (21 subjects) showed anti-SARS-CoV-2 IgG at 28 dpi, and one (5%) showed specific IgM too. After 108 dpi, one volunteer showed no IgG, and 138 dpi, we found 3 volunteers present no IgG.

The innate immune response is critical for vaccination as it activates the myeloid immune cells to release inflammatory mediators and initiate adaptive immunity the primary and post-boost immunization (9). To assess the pro- and anti-inflammatory plasma cytokine profile, we performed flow cytometry and found that, 28 dpi, only IL-8 increased its level from 14.7 ± 2.6 (on day 0) to 55.9 ± 63.8 pg/ml (on 28 dpi) ( Figure 1 ). No significant changes were observed between 0 and 28 dpi in the levels of tumor necrosis factor (TNF)-α (4.7 ± 9.5 to 4.7 ± 5.1 pg/ml), interleukin (IL)-12 (6.4 ± 20.9 to 4.7 ± 9.8 pg/ml), IL-6 (8.2 ± 7.4 to 8 ± 8.6 pg/ml), IL-10 (6.2 ± 4.3 to 5.4 ± 2.7 pg/ml), and IL-1β (11.7 ± 8.5 to 11.7 ± 18.8 pg/ml) ( Figure 1 ).

We measured the level of the anti-SARS-CoV-2 spike antibodies in 15 volunteers before and after vaccination ( Figure 2A ). We found an increase in the anti-S antibodies in all volunteers after 28 dpi (0 dpi = 24 ± 19, 28 dpi = 759 ± 278); however, after 90 dpi, the anti-S antibody levels declined (90 dpi = 137 ± 92.3), although they were higher than pre-immunization levels ( Figure 2A ). When we analyzed data from volunteers for whom samples were not always collected, the increase in anti-S antibody levels showed the same kinetics ( Figure 2B ). Additionally, we measured SARS-CoV-2 neutralizing antibodies (NAbs) using the HIV-1–SΔ19 pseudotype (8). We found no volunteers with anti-spike neutralizing response before immunization (0 dpi = 4 ± 0.7, Figure 3A ). In contrast, after 28 and 90 dpi, we found anti-S neutralizing antibodies in all volunteers in different levels (28 dpi = 388.1 ± 454.5, 90 dpi = 57.1 ± 67.6, Figures 3B, C ). All 15 volunteers (complete follow-up) show an increase in the IC₅₀ after 28 dpi ( Figure 3D ). Furthermore, after 90 dpi, the neutralizing antibody levels declined; however, some volunteers still showed this type of antibodies. Importantly, these results are consistent with those observed for the volunteers with a not complete follow-up ( Figure 3E ).

To explore the cellular immune responses to SARS-CoV-2 after CoronaVac vaccination, we isolated PBMCs from the 15 volunteers with complete follow-up, and the PBMCs obtained were stimulated in vitro with the SARS-CoV-2 spike (S) protein. We observed a significant increase in the number of IFN-γ spot-forming cells (SFCs) that specifically responded to the recombinant S protein after 28 dpi (0 dpi = 129 ± 402, 28 dpi = 343.5 ± 1,199) ( Figure 4A ). In most of the cases, IFN-γ SFCs continue to increase at day 90 dpi (90 dpi = 490 ± 384.2). Similarly, a significant increase in the number of granzyme SFCs was also observed in all, but one sample, after 28 dpi (0 dpi = 129 ± 402, 28 dpi = 256 ± 152.4) ( Figure 4B ), and most of them increased further at day 90 dpi. Altogether, these results suggest the activation and presence of CD4+ Th and cytotoxic CD8+ T memory cells, all required for an effective viral clearance (10).