A systems immunology study comparing innate and adaptive immune responses in adults to COVID-19 mRNA (BNT162b2/mRNA-1273) and adenovirus vectored vaccines (ChAdOx1-S) after the first, second and third doses (preprint)

We longitudinally profiled immune responses in 102 adults who received BNT162b2 (Pfizer-BioNTech) or ChAdOx1-S (Oxford-AstraZeneca) as their primary vaccinations. Bloods were collected pre-vaccination, 1-7 days after the 1st, 2nd and 3rd doses (BNT162b2 or mRNA-1273) to assess innate and early adaptive responses, and ~28 days after the 2nd and 3rd doses to assess immunogenicity. Using a multi-omics approach including RNAseq, cytokine multiplex assay, proteomics, lipidomics, and flow cytometry we identified key differences in the immune responses induced by the ChAdOx1-S and BNT162b2 vaccines that were correlated with subsequent antigen-specific antibody and T cell responses or vaccine reactogenicity. We observed that vaccination with ChAdOx1-S but not BNT162b2 induced a memory-like response after the first dose, which was correlated with the expression of several proteins involved in complement and coagulation. The COVID-19 Vaccine Immune Responses Study (COVIRS) thus represents a major resource to understand the immunogenicity and reactogenicity of these COVID-19 vaccines.

Persistence of a SARS-CoV-2 variant with a frameshifting deletion for the duration of a major outbreak (preprint)

Australia experienced widespread COVID-19 outbreaks from infection with the SARS-CoV-2 Delta variant between June 2021 and February 2022. Whole-genome sequencing of virus from an early case revealed a sub-consensus level of sequencing reads supporting a 17-nucleotide frameshift-inducing deletion in ORF7a that truncated the peptide sequence. The variant rapidly became represented at the consensus level (Delta-ORF7a{Delta}17del) in most of the outbreak cases in Australia. Retrospective analysis of ORF7a deletions in all GISAID clade GK Delta genomes showed that of 4,018,216 genomes, 134,751 (~3.35%) possessed a deletion in ORF7a, with the ORF7a{Delta}17del mutation by far the most common. Approximately 99.05% of Delta-ORF7a{Delta}17del genomes on GISAID originated from the Australian Delta outbreak, and comprised 87% of genomes in the outbreak. In vitro comparison of lineages in cell culture showed a significantly greater proportion of cells were infected with Delta-ORF7a{Delta}17del than with a contemporaneous Delta variant without ORF7a{Delta}17del (Delta-ORF7aintact), and the proportion was also measurably higher than an early SARS-CoV-2 strain (A.2.2). These results showed that Delta-ORF7a{Delta}17del potentially has a slight growth advantage compared to Delta-ORF7aintact. Delta-ORF7a{Delta}17del viruses still produced ORF7a protein, but significantly less than A.2.2, in a different cellular distribution with a more diffuse expression throughout the cytoplasm of infected cells. These data suggest that the proliferation of Delta-ORF7a{Delta}17del genomes during the Australian Delta outbreak was likely not a result of an intrinsic benefit of the ORF7a{Delta}17del mutation, but rather a chance founder effect. Nonetheless, the abundance of different ORF7a deletions in genomes worldwide suggests these have some benefit to virus transmission. IMPORTANCE Deletions in the ORF7a region of SARS-CoV-2 have been noted since early in the COVID-19 pandemic, but are generally reported as transient mutations that are quickly lost in the population. Consequently, ORF7a deletions are considered disadvantageous to the virus through possible loss-of-function effects. In constrast to these earlier reports, we present the first report of a SARS-CoV-2 variant with an ORF7a deletion that dominated for the entirety of a protracted outbreak, and found no associated fitness disadvantage or advantage in cell culture. The relatively common rise and fall of ORF7a deletion variants over time likely represent chance founder events followed by proliferation until a more fit variant(s) is introduced to the population. Our global clade-level survey of ORF7a deletions will be a useful resource for future studies into this gene region.

Longitudinal characterisation of phagocytic and neutralisation functions of anti-Spike antibodies in plasma of patients after SARS-CoV-2 infection. (preprint)

Phagocytic responses by effector cells to antibody or complement-opsonised viruses have been recognized to play a key role in anti-viral immunity. These include antibody dependent cellular phagocytosis mediated via Fc-receptors, phagocytosis mediated by classically activated complement-fixing IgM or IgG1 antibodies and antibody independent phagocytosis mediated via direct opsonisation of viruses by complement products activated via the mannose-binding lectin pathway. Limited data suggest these phagocytic responses by effector cells may contribute to the immunological and inflammatory responses in SARS-CoV-2 infection, however, their development and clinical significance remain to be fully elucidated. In this cohort of 62 patients, acutely ill individuals were shown to mount phagocytic responses to autologous plasma-opsonised SARS-CoV-2 Spike protein-coated microbeads as early as 10 days post symptom onset. Heat inactivation of the plasma prior to use as an opsonin caused 77-95% abrogation of the phagocytic response, and pre-blocking of Fc-receptors on the effector cells showed only 18-60% inhibition. These results suggest that SARS-CoV-2 can provoke early phagocytosis, which is primarily driven by heat labile components, likely activated complements, with variable contribution from anti-Spike antibodies. During convalescence, phagocytic responses correlated significantly with anti-Spike IgG titers. Older patients and patients with severe disease had significantly higher phagocytosis and neutralisation functions when compared to younger patients or patients with asymptomatic, mild, or moderate disease. A longitudinal study of a subset of these patients over 12 months showed preservation of phagocytic and neutralisation functions in all patients, despite a drop in the endpoint antibody titers by more than 90%. Interestingly, surface plasmon resonance showed a significant increase in the affinity of the anti-Spike antibodies over time correlating with the maintenance of both the phagocytic and neutralisation functions suggesting that improvement in the antibody quality over the 12 months contributed to the retention of effector functions.

COVID-19 convalescents exhibit deficient humoral and T cell responses to variant of concern Spike antigens at 12 month post-infection (preprint)

Background The duration and magnitude of SARS-CoV-2 immunity after infection, especially with regard to the emergence of new variants of concern (VoC), remains unclear. Here, immune memory to primary infection and immunity to VoC was assessed in mild-COVID-19 convalescents one year after infection and in the absence of viral re-exposure or COVID-19 vaccination. Methods Serum and PBMC were collected from mild-COVID-19 convalescents at ∼6 and 12 months after a COVID-19 positive PCR (n=43) and from healthy SARS-CoV-2-seronegative controls (n=15-40). Serum titers of RBD and Spike-specific Ig were quantified by ELISA. Virus neutralisation was assessed against homologous, pseudotyped virus and homologous and VoC live viruses. Frequencies of Spike and RBD-specific memory B cells were quantified by flow cytometry. Magnitude of memory T cell responses was quantified and phenotyped by activation-induced marker assay, while T cell functionality was assessed by intracellular cytokine staining using peptides specific to homologous Spike virus antigen and four VoC Spike antigens. Findings At 12 months after mild-COVID-19, >90% of convalescents remained seropositive for RBD-IgG and 88.9% had circulating RBD-specific memory B cells. Despite this, only 51.2% convalescents had serum neutralising activity against homologous live-SARS-CoV-2 virus, which decreased to 44.2% when tested against live B.1.1.7, 4.6% against B.1.351, 11.6% against P.1 and 16.2%, against B.1.617.2 VoC. Spike and non-Spike-specific T cells were detected in >50% of convalescents with frequency values higher for Spike antigen (95% CI, 0.29-0.68% in CD4 + and 0.11-0.35% in CD8 + T cells), compared to non-Spike antigens. Despite the high prevalence and maintenance of Spike-specific T cells in Spike ‘high-responder’ convalescents at 12 months, T cell functionality, measured by cytokine expression after stimulation with Spike epitopes corresponding to VoC was severely affected. Interpretations SARS-CoV-2 immunity is retained in a significant proportion of mild COVID-19 convalescents 12 months post-infection in the absence of re-exposure to the virus. Despite this, changes in the amino acid sequence of the Spike antigen that are present in current VoC result in virus evasion of neutralising antibodies, as well as evasion of functional T cell responses. Funding This work was funded by project grants from The Hospital Research Foundation and Women’s and Children’s Hospital Foundation, Adelaide, Australia. MGM is THRF Early Career Fellow. BGB is THRF Mid-Career Fellow. This project has been supported partly with Federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, under Contract No. 75N93021C00016 to A.S. and Contract No. 75N9301900065 to A.S, D.W. Evidence before this study We regularly searched on PubMed and Google Scholar in June-October 2021 using individual or combinations of the terms “long-term immunity”, “SARS-CoV-2”, “antigenic breadth”, “variant of concern” and “COVID-19”. We found studies that had assessed immune correlates at multipe time points after COVID-19 disease onset in convalescents, but not the antigenic breadth of T cells and antibodies and not in relation to VoC. Other immune studies in virus naive vaccinees, or vaccinated convalescents evaluated VoC-specific immunity, but not in convalescents that have not been vaccinated. In summary, we could not find long-term studies providing and in-depth evaluation of functionality of humoral and cell-mediated immunity, combined with addressing the adaptability of these immune players to VoC. Added value of this study The window of opportunity to conduct studies in COVID-19 convalescents (i.e. natural immunity to SARS-CoV-2) is closing due to mass vaccination programs. Here, in a cohort of unvaccinated mild-COVID-19 convalescents, we conducted a comprehensive, longitudinal, long-term immune study, which included functional assays to assess immune fitness against antigenically different VoC. Importantly, the cohort resided in a SARS-CoV-2-free community for the duration of the study with no subsequent re-exposure or infection. Our findings reveal a deeply weakened humoral response and functional vulnerability of T cell responses to VoC Spike antigens. Implications of all the available evidence This study provides a valuable snapshot of the quality of SARS-CoV-2 natural immunity and its durability in the context of a pandemic in which new variants continuously emerge and challenge pre-existing immune responses in convalescents and vacinees. Our results serve as a warning that delays in vaccination programs could lead to an increase in re-infection rates of COVID-19 convalescents, caused by virus variants that escape humoral and cell-mediated immune responses. Furthermore, they reinforce the potential benefit of booster vaccination that is tuned to the active variants.

'One Year Later' - SARS-CoV-2-Specific Immunity in Mild Cases of COVID-19 (preprint)

Background: Duration and quality of immunity to SARS-CoV-2 have significant implications for the management of COVID-19 pandemic. Here, we present a comprehensive set of immunological data from a cohort of individuals (n=43), 12 months after mild COVID-19 disease and in the absence of virus re-exposure.Methods: Serum and PBMC were collected from mild-COVID-19 convalescents 12 months after the COVID-19 positive PCR (n=43) and from healthy SARS-CoV-2-seronegative controls (n=15). Serum titers of SARS-CoV-2-specific immunoglobulins were quantified by ELISA and virus neutralisation activity was assessed using SARS-CoV-2-Spike pseudovirus particles. Frequencies of Spike and RBD-specific memory B cells were quantified by flow cytometry. Magnitude of memory T cell responses was quantified and phenotyped with an activation-induced marker assay.Findings: In the absence of re-exposure to SARS-CoV-2 Spike- and RBD-specific antibodies were present in 90% of COVID-19 convalescents 12 months post-infection. RBD-specific IgG + memory B cells were maintained in 88.9% of patients, while 62% of patients had serum neutralising activity. Functionally mature memory CD4 + and CD8 + T cells were maintained at frequencies previously reported for earlier time points post-COVID-19, indicating substantial maintenance of durable T cell responses. Interpretations: Immunity to SARS-CoV-2 persists for 12 months in mild COVID-19 convalescent patients that retain high Spike-specific antibody titres, virus neutralisation capacity and circulating RBD-specific memory B cells. Significantly, T cell immunity remained stable 12 months post-infection. This study offers vital information on the duration of natural COVID-19 immunity and its potential protective effect against SARS-CoV-2 reinfection and clinical disease, with clear implications for the ongoing management of the global pandemic. Funding Statement: This work was funded by project grants from The Hospital Research Foundation and Women’s and Children’s Foundation, Adelaide, Australia. This work has been supported by NIH contract 75N9301900065 (A.S, D.W).Declaration of Interests: A.S. is currently a consultant for Gritstone, Flow Pharma, Arcturus, Epitogenesis, Oxfordimmunotech, Caprion and Avalia. LJI has filed for patent protection for various aspects of T cell epitope and vaccine design work.Authors PGV, CMH, MGM, AELY, HB, ZAM, ZAD, AA, DA, JG, CF, SO, EMM, DJL, GM, EJG, BAJR, DS, CKL, MRB, DW, RAB, SCB and BGB declare no conflict of interest.Ethics Approval Statement: Study protocols were approved by the Central Adelaide Clinical Human Research Ethics Committee (#13050) and the Women’s and Children’s Health Network Human research ethics (protocol HREC/19/WCHN/65), Adelaide, Australia.

SARS Coronavirus-2 microneutralisation and commercial serological assays correlated closely for some but not all enzyme immunoassays (preprint)

BackgroundSerological testing for SARS-CoV-2 specific antibodies provides important research and diagnostic information relating to COVID-19 prevalence, incidence, and host immune response. A greater understanding of the relationship between functionally neutralising antibodies detected using microneutralisation assays and binding antibodies detected using scalable enzyme immunoassays (EIA) is needed in order to address protective immunity post-infection or vaccination, and assess EIA suitability as a surrogate test for screening of convalescent plasma donors. We assessed whether neutralising antibody titres correlated with signal cut-off ratios in five commercially available EIAs, and one in-house assay based on expressed spike protein targets. MethodsSera from individuals recovered from patients or convalescent plasma donors who reported laboratory-confirmed SARS-CoV-2 infection (n=200), and negative control sera collected prior to the COVID-19 pandemic (n=100) were assessed in parallel. Performance was assessed by calculating EIA sensitivity and specificity with reference to microneutralisation. ResultsNeutralising antibodies were detected in 166 (83%) samples. Compared with this, the most sensitive EIAs were the Cobas Elecsys Anti-SARS-CoV-2 (98%) and Vitros Immunodiagnostic Anti-SARS-CoV-2 (100%), which detect total antibody targeting the N and S1 antigens, respectively. The assay with the best quantitative relationship with microneutralisation was the Euroimmun IgG. ConclusionsThese results suggest the marker used (total Ab vs IgG vs IgA), and the target antigen are important determinants of assay performance. The strong correlation between microneutralisation and some commercially available assays demonstrate their potential for clinical and research use in assessing protection following infection or vaccination, and use as a surrogate test to assess donor suitability for convalescent plasma donation.

Long-term persistence of neutralizing memory B cells in SARS-CoV-2 (preprint)

Considerable concerns relating to the duration of protective immunity against SARS-CoV-2 have been raised, with evidence of antibody titres declining rapidly after infection and reports of reinfection. Here we monitored antibody responses against SARS-CoV-2 receptor binding domain (RBD) for up to six months after infection. While antibody titres were maintained, half of the cohort’s neutralising responses had returned to background. However, encouragingly in a selected subset of 13 participants, 12 had detectable RBD-specific memory B cells and these generally increased out to 6 months. Furthermore, we were able to generate monoclonal antibodies with SARS-CoV-2 neutralising capacity from these memory B cells. Overall our study suggests that the loss of neutralising antibodies in plasma may be countered by the maintenance of neutralising capacity in the memory B cell repertoire.