Proteomic Signatures of the Serological Response to Influenza Vaccination in a Large Human Cohort Study

Abstract

The serological response to the influenza virus vaccine is highly heterogeneous for reasons that are not entirely clear. While the impact of demographic factors such as age, body mass index (BMI), sex, prior vaccination and titer levels are known to impact seroconversion, they only explain a fraction of the response. To identify signatures of the vaccine response, we analyzed 273 protein levels from 138 serum samples of influenza vaccine recipients (2019–2020 season). We found that levels of proteins functioning in cholesterol transport were positively associated with seroconversion, likely linking to the known impact of BMI. When adjusting seroconversion for the demographic factors, we identified additional, unexpected signatures: proteins regulating actin cytoskeleton dynamics were significantly elevated in participants with high adjusted seroconversion. Viral strain specific analysis showed that this trend was largely driven by the H3N2 strain. Further, we identified complex associations between adjusted seroconversion and other factors: levels of proteins of the complement system associated positively with adjusted seroconversion in younger participants, while they were associated negatively in the older population. We observed the opposite trends for proteins of high density lipoprotein remodeling, transcription, and hemostasis. In sum, careful integrative modeling can extract new signatures of seroconversion from highly variable data that suggest links between the humoral response as well as immune cell communication and migration.

Longitudinal assessment of human antibody binding to hemagglutinin elicited by split-inactivated influenza vaccination over six consecutive seasons

“…In order to fully understand the host influences on vaccination and the ever-changing influenza variants over time, samples from these cohorts have been analyzed for host RNA gene expression profiles by transcriptomics [ 59 ], host glycomics [ 60 ], proteomics [ 61 ], and metabolomics [ 62 ], as well as DNA methylation and immune senescence [ 63 ]. These data sets have been used to develop predictive models of host responses and vaccine effectiveness [ 64 ].…”

Patient Subtyping Analysis of Baseline Multi-omic Data Reveals Distinct Pre-immune States Predictive of Vaccination Responses

“…The converted files were then analyzed with the directDIA method using default settings. Data preprocessing includes quality control, batch correction, preprocessing by mapDIA, and normalization as described previously [9, 20]. We further processed the data by adjusting for age, sex, BMI, month of vaccination, and prior vaccination status, and then applied Z-transform.…”

“…The details of 225 serum proteome probes preparation from 160 influenza vaccine recipients was previously published [9]. All 225 raw files were first converted to the HTRMS format with the HTRMS converter.…”

“…Recent research has revealed several potential biomarkers associated with influenza vaccine response, including pre-existing immunity, age, sex, and obesity [3][4][5][6], and further studies examined transcriptomics, glycomics, proteomics, and metabolomics data, which showed their potential links with response to the influenza vaccine [7][8][9][10]. The mechanism of vaccine response is complex, and hence not well understood to date.…”