Machine Learning-Based Single Cell and Integrative Analysis Reveals That Baseline mDC Predisposition Correlates With Hepatitis B Vaccine Antibody Response.

Tobias R Kollmann,Scott J Tebbutt,Rym Ben-Othman,Michael S Kobor,Amy Huei-Yi Lee,Nicole Gladish,Mark Novotny,Richard H Scheuermann,Travis M Blimkie,Manish Sadarangani,Wayne C Koff,Alba Llibre,Darragh Duffy,Bing Cai,Robert E Hancock,Casey P Shannon,Jamie C Ye,Yun Zhang,Brian D Aevermann

doi:10.3389/fimmu.2021.690470

Abstract

Vaccination to prevent infectious disease is one of the most successful public health interventions ever developed. And yet, variability in individual vaccine effectiveness suggests that a better mechanistic understanding of vaccine-induced immune responses could improve vaccine design and efficacy. We have previously shown that protective antibody levels could be elicited in a subset of recipients with only a single dose of the hepatitis B virus (HBV) vaccine and that a wide range of antibody levels were elicited after three doses. The immune mechanisms responsible for this vaccine response variability is unclear. Using single cell RNA sequencing of sorted innate immune cell subsets, we identified two distinct myeloid dendritic cell subsets (NDRG1-expressing mDC2 and CDKN1C-expressing mDC4), the ratio of which at baseline (pre-vaccination) correlated with the immune response to a single dose of HBV vaccine. Our results suggest that the participants in our vaccine study were in one of two different dendritic cell dispositional states at baseline – an NDRG2-mDC2 state in which the vaccine elicited an antibody response after a single immunization or a CDKN1C-mDC4 state in which the vaccine required two or three doses for induction of antibody responses. To explore this correlation further, genes expressed in these mDC subsets were used for feature selection prior to the construction of predictive models using supervised canonical correlation machine learning. The resulting models showed an improved correlation with serum antibody titers in response to full vaccination. Taken together, these results suggest that the propensity of circulating dendritic cells toward either activation or suppression, their “dispositional endotype” at pre-vaccination baseline, could dictate response to vaccination.

Highlights

Vaccination as a general strategy to prevent infectious disease has been one of the most effective public health measures since its conceptualization and implementation by Edward Jenner in the 18th century, and has resulted in the complete eradication of smallpox, the near elimination of polio, and a dramatic reduction in the incidences of measles, mumps and other common diseases
Single monocytes (MON), myeloid dendritic cells, plasmacytoid dendritic cells, and natural killer cells (NK) were sorted into microtiter plate wells and single cell cDNAs showing positive ACTB expression by Quantitative PCR (qPCR) analyzed by scRNAseq and Uniform Manifold Approximation and Projection (UMAP) embedding of the gene expression data
In a relatively small cohort of HBV vaccine recipients, we identified two distinct myeloid dendritic cells (mDCs) subsets using single cell transcriptomics analysis, the ratio of which at baseline correlated with vaccine response to a single dose of the HBV vaccine

Summary

Introduction

Vaccination as a general strategy to prevent infectious disease has been one of the most effective public health measures since its conceptualization and implementation by Edward Jenner in the 18th century, and has resulted in the complete eradication of smallpox, the near elimination of polio, and a dramatic reduction in the incidences of measles, mumps and other common diseases In contrast to these successes, several notable failures in the development of effective vaccines against other common infectious diseases, including AIDS, tuberculosis, and malaria suggest that the current empirical approach to vaccine design is not effective in eliciting protective immunity in many cases [1]. The application of largescale multi-omics assessment of large vaccination cohorts is costprohibitive, raising the question of whether advanced computational and machine learning methods may allow for the discovery of predictive mechanistic signatures in studies with smaller sample sizes [13]

Methods

Results

Conclusion