Inferring tunicate relationships and the evolution of the tunicate Hox cluster with the genome of Corella inflata

Tunicates, the closest living relatives of the vertebrates, have served as a foundational model of early embryonic development for decades. Comparative studies of tunicate phylogeny and genome evolution provide a critical framework for analyzing chordate diversification and the emergence of the vertebrates. Towards this goal, we sequenced the genome of Corella inflata (Ascidiacea, Phlebobranchia), so named for the capacity to brood self-fertilized embryos in a modified, “inflated” atrial chamber. Combining the new genome sequence for Co. inflata with publicly available tunicate data, we estimated a tunicate species phylogeny, reconstruct the ancestral Hox gene cluster at important nodes in the tunicate tree, and compare patterns of gene loss between Co. inflata and Ciona robusta, the prevailing tunicate model species. Our maximum-likelihood and Bayesian trees estimated from a concatenated 210-gene matrix were largely concordant and showed that Aplousobranchia was nested within a paraphyletic Phlebobranchia. We show that this relationship is not an artifact due to compositional heterogeneity, as had been suggested by previous studies. The Co. inflata genome helps increase the resolution of the ancestral Hox clusters of key tunicate nodes, therefore expanding our understanding of the evolution of this cluster and its potential impact on tunicate morphological diversity. Our analyses of other gene families reveal that several cardiovascular associated genes (e.g., BMP10, SCL2A12, and PDE2a) absent from Ci. robusta are present in Co. inflata. Taken together, our results help clarify tunicate relationships and the genomic content of key ancestral nodes within this phylogeny, providing critical insights into tunicate evolution.