Immune cell type divergence in a basal chordate

Evolutionary adaptations often occur at the level of cell types and cellular function. Innate immune cells are a promising system for studying cell type evolution, as they are widespread across metazoans, have several conserved functions, and are under selective pressure from pathogens. However, molecular characterizations of invertebrate immune cells are limited, and it remains unclear whether invertebrate immune cell types are homologous to those in vertebrates. Here we use single-cell RNA sequencing, in situ hybridization, and live reporters to define the identity of blood cell states from a basal chordate, Ciona robusta. We find evidence that C. robusta circulating blood contains a differentiation hierarchy that gives rise to at least eight major morphotypes, constituting approximately half of mature blood cell states. The mature cell states include phagocytes, as well as cells variously expressing vanadium-binding proteins, carbonic anhydrases, pattern recognition receptors, cytokines, and complement factors. Despite the expression of homologs to vertebrate immune components, extensive divergence between tunicate and vertebrate immune cells obscures cell state homology. Altogether, this work modernizes blood cell classifications in C. robusta and extends the known repertoire of immune cells within chordates. Overall design: Circulating blood cells were collected from adult Ciona robusta animals, then run directly and analyzed with scRNA-seq. Two libraries were collected in this study, and they were combined with one library from a previous study for analysis (reanalysis of Sample GSM8869531 from Series GSE292926). This combined data (combined_processed_data.h5ad, an H5ad Hierarchical Data Format file) was reanalyzed by removing likley doublets, normalizing across all libraries, identifying of highly variable genes, log normalizing, z-scoring, performing PCA on highly variable genes, batch correcting using BBKNN, Leiden clustering, and UMAP embedding. Additionally, cells from all samples were assigned to SNP profiles using Cellsnp-lite and Vireo. In the combined_processed_data.h5ad, adata.obs['library'] indicates which cell barcodes are associated with each library: 230518_1 and 230518_2 are from this GEO Series (Cr_blood_230518_1 and Cr_blood_230518_2, respectively), and 220428_1 is the reanalyzed Sample GSM8869531 from Series GSE292926. *************************************************************** The table below lists GEO accessions reused/reanalyzed for this study. ***************************************************************