An integrated single-cell atlas of blood immune cells in aging.

Abstract

Recent advances in single-cell technologies have facilitated studies on age-related alterations in the immune system. However, previous studies have often employed different marker genes to annotate immune cell populations, making it challenging to compare results. In this study, we combined seven single-cell transcriptomic datasets, comprising more than a million cells from one hundred and three donors, to create a unified atlas of human peripheral blood mononuclear cells (PBMC) from both young and old individuals. Using a consistent set of marker genes for immune cell annotation, we standardized the classification of immune cells and assessed their prevalence in both age groups. The integrated dataset revealed several consistent trends related to aging, including a decline in CD8⁺ naive T cells and MAIT cells and an expansion of non-classical monocyte compartments. However, we observed significant variability in other cell types. Our analysis of the long non-coding RNA MALAT1^hi T cell population, previously implicated in age-related T cell exhaustion, showed that this population is highly heterogeneous with a mixture of naïve-like and memory-like cells. Despite substantial variation among the datasets when comparing gene expression between age groups, we identified a high-confidence signature of CD8⁺ naive T cell aging marked by an increased expression of pro-inflammatory genes. In conclusion, our study emphasizes the importance of standardizing existing single-cell datasets to enable the comprehensive examination of age-related cellular changes across multiple datasets.