Multi-omics analysis of human population variation in immune function and in vivo response to BCG vaccination

Immune responses are tightly regulated, yet highly variable between individuals. To investigate human population variation of trained immunity, we immunized healthy individuals with Bacillus Calmette-Guérin (BCG). This live attenuated vaccine induces not only an adaptive immune response against tuberculosis, but also triggers innate immune activation and memory. We established personal immune profiles and chromatin accessibility maps over a time course of BCG vaccination in 323 individuals. This large resource uncovered genetic and epigenetic predictors of baseline immunity and BCG vaccine response. We found that BCG vaccination enhances the innate immune response only in individuals with dormant immune states at baseline, suggesting that exogeneous induction of trained immunity is not a universal booster of innate immunity, but specifically elevates weak innate immune responses. This study advances our understanding of BCG’s heterologous immune-stimulatory effects and trained immunity in humans. Moreover, our results highlight the value of epigenetic cell states as an “endophenotype” that connects immune function with genotype and the environment. The 300BCG cohort consists of 323 healthy adult individuals (184 females and 139 males) and is part of the Human Functional Genomics Project (HFGP). Study participants were included from April 2017 until June 2018 at the Radboud University Medical Center, the Netherlands. They received a standard dose of 0.1 ml BCG (Bulgaria strain, Intervax) intradermally. Blood was drawn immediately before BCG vaccination (d0), and 14 days (d14) as well as 90 days (d90) after vaccination. Exclusion criteria were any vaccination within three months prior to d0, acute (febrile) illness four weeks before inclusion, a medical history of immunodeficiency or any other chronic disease, use of systemic medication other than oral contraceptives or acetaminophen, use of antibiotics three months before inclusion, ancestry other than Western European, previous BCG vaccination, history of tuberculosis, previous contact with tuberculosis patients, being born in a tuberculosis endemic country, pregnancy, or breastfeeding. During the study, individuals were excluded upon medication use other than oral contraceptives or acetaminophen, or illness at time of study visit. Applying these criteria resulted in 273 individuals with measurements for d0, 253 individuals for d14, and 254 individuals for d90, respectively. >>>Submitter declares that the raw data will be deposited in EGA due to patient privacy concerns<<<