Introduction

Infection with the zoonotic severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes the coronavirus disease 2019 (COVID-19)1,2. Due to the multi-organ tropism of SARS-CoV-2, COVID-19 manifestations are often systemic and characterized by a broad severity spectrum with high morbidity and an elevated risk of mortality in distinct patient groups1,3. After the global spread of SARS-CoV-2 in 2020, the rapid development of several novel vaccine platforms within one year was key to mitigate the pandemic4,5,6,7,8. This unprecedented achievement was possible due to prior knowledge from the development and preclinical studies of vaccine candidates against SARS-CoV and Middle Eastern respiratory syndrome coronavirus (MERS-CoV) that identified the spike protein of human coronaviruses as the cardinal antigenic target to generate broad neutralizing B and T cell responses4,9. In August 2023, the two licensed mRNA-based COVID-19 vaccines BNT162b2 (BioNTech/Pfizer)10 and mRNA-1273 (Moderna)11, both of which encode a modified full-length SARS-CoV-2 S1 spike protein designed to stabilize the prefusion conformation, account for 90% of administered doses in the European Union and the United States12. Large clinical trials and real-world data clearly show that both vaccines are extremely safe and provide high protection against symptomatic and severe infection by eliciting neutralizing B and T cell responses including immunological memory that are also effective against different emerging variants of concern in single-dose or (heterologous) booster settings13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28. Notably, COVID-19 vaccination has similar risk and safety profiles in immunocompromised individuals29, patients with cancer30,31,32,33 or during pregnancy34.

Although vaccines are highly successful in reducing morbidity and mortality of acute COVID-19, their efficacy in preventing long-term consequences of a SARS-CoV-2 infection is less clear. Around 10-15% of COVID-19 patients have persisting health impairments beyond four weeks of symptom onset that are heterologous in their expression and can last for months with significant impairments for the quality of life35,36,37,38,39. For earlier variants of SARS-CoV-2 studies suggested that the likelihood of develo** symptoms after infection (post-COVID-19 condition; PCC) is less frequent in individuals with pre-infection vaccination compared to those without40,92. Nevertheless, this requires further investigation given the lack of longitudinal symptom reporting or plasma sampling in our study.

Finally, it is important to state that the data presented in this manuscript does not provide any hints that post-infection vaccination might exacerbate PCC-like symptoms or adversely impact individuals with earlier infection. We also did not detect any PCC-like signature in uninfected individuals after initial mRNA-based vaccination, neither did we notice any general imprint that could be interpreted as priming for potential PCC development after booster vaccination. Our data suggests that vaccination in PCC patients may play a role in ameliorating post-infection gut-related symptoms via IL-1 family cytokines. While we did not observe any vaccination-associated adverse events, these can affect a very small number of vaccinees and require larger studies. It is nevertheless important to acknowledge virus-induced and vaccine-induced pathology as distinct entities to understand their underpinnings and develop targeted treatments.

Together, our study corroborates the safety and efficiency of mRNA-based COVID-19 vaccines in SARS-CoV-2 naïve individuals or after infection. In addition, we provide biomarker-based data that suggests a benefit of post-infection vaccination for patients with PCC who suffer from gastrointestinal sequelae. This finding invites further exploration into the intricate interplay between vaccination, cytokine modulation, and gastrointestinal health.

Methods

Recruitment and sampling of vaccination cohorts

Twenty SARS-CoV-2 naïve healthcare workers from the University Hospital Halle (Saale), Germany, were recruited between December 2020 and January 2021 at the beginning of the German COVID-19 vaccination campaign to study vaccination efficiency after the initial mRNA vaccine rollout. Of these 20 individuals, 19 received the BioNTech-Pfizer Vaccine BNT162b2 (Tozinameran/Comirnaty), one the Moderna mRNA-1273 (Elasomeran/Spikevax) vaccine. As vaccination control, we used a cohort of additional 11 healthcare workers who received the seasonal influenza vaccine (VaxigripTetra 2020/2021) between October and November 2020 at the University Hospital Halle (Saale). Blood sampling of these vaccination cohorts was performed on the day of vaccine administration and four weeks later. The demographic characteristics of both vaccination cohorts are listed in Table 1. To assess the effect of post-infection vaccination on symptoms of post-COVID-19 condition, we used 540 individuals from the cohort study for digital health research in Germany (DigiHero) including 96 individuals without prior SARS-CoV-2 infection. Individuals were recruited between August 2021 and February 2022 via mailed invitation47. Participants completed an online questionnaire focusing on the detection and course of acute COVID-19, its sequelae, and vaccination status. Blood from the DigiHero participants was sampled once at a median of 8 months after the onset of the first COVID-19 symptoms. Demographic characteristics are listed in Table 2. The study was approved by the institutional review board (approval numbers approval number 2020-039 and 2020-076) and conducted in accordance with the ethical principles stated by the Declaration of Helsinki. Informed written consent was obtained from all participants or legal representatives. Plasma was isolated from whole blood via centrifugation of whole blood for 15 minutes at 2000 × g, followed by centrifugation at 12,000 × g for 10 minutes. All plasma samples were stored at - 80 °C before further use.

SARS-CoV-2 and influenza antibody profiling

Relative titers of antibodies targeting the S1 domain of the spike (S) protein and the nucleocapsid protein (NCP) of SARS-CoV-2 were determined using the Anti-SARS-CoV-2-ELISA IgA/IgG and Anti-SARS-CoV-2-NCP-ELISA kits from Euroimmun (Lübeck, Germany). ELISAs were coated with the respective recombinant antigen. To determine the relative titers of IgG class antibodies directed against influenza A and B, we used the Anti-Influenza-A-Virus-ELISA (IgG) and Anti-Influenza-B-Virus-ELISA (IgG) ELISA Kits from Euroimmun. Influenza A ELISA plates were coated with inactivated influenza A strains (Texas, H3N2; Singapore, H1N1; California, H1N1 (Porcine Influenza)) isolated from the allantoic fluid of infected chick embryos. In the case of influenza B, plates were coated with the inactivated virus of the B/Hong Kong/5/72 variant. Assays were performed according to the manufacturer’s instructions. Readouts were performed at 450 nm using a Tecan Spectrophotometer SpectraFluor Plus (Tecan Group Ltd., Männedorf, Switzerland).

Quantification of soluble factors in human plasma

Plasma cytokines were quantified using the LEGENDplex Human B Cell Panel (13-plex) and the Human Anti-Virus Response Panel (13-plex) (BioLegend) according to the manufacturer’s instructions. In addition, plasma levels of IL-5, IL-18, IL-23, IL-33, and CCL2/MCP-1 were quantified using the respective capture beads and corresponding detection antibodies from the LEGENDplex Human Inflammation Panel (Cat. No. 740809) and Human Th Panel (Cat. No. 741027) (BioLegend). Read out of the LEGENDplex assays was performed on a BD FACSCelesta. Concentrations were calculated using the LEGENDplex cloud-based Qognit Data Analysis Software (BioLegend). Heatmap of log-transformed plasma levels were generated with the R package pheatmap using the R version 4.3.1 and RStudio 2023.06.1.

Statistical analysis

Differences in plasma levels of antibodies or cytokines between the two groups were studied using the unpaired two-sided t-test. Comparisons between multiple groups were performed using ordinary one-way ANOVA followed by post-hoc testing (Tukey’s multiple comparisons test). The association between categorial symptom reporting and the number of post-infection vaccinations was tested with the chi-squared test for trend in proportions. All statistical analyses as well as the linear regression and Pearson correlation analyses for antibody plasma levels over time were performed using GraphPad PRISM 9.5.1 (GraphPad Software, La Jolla, CA, USA). Heatmaps were generated with the package pheatmap using R version 4.3.1 and RStudio 2023.06.1. Ranges of p values are indicated with asterisks: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.