Search

Significance The protovertebrate Ciona intestinalis is an ideal system to investigate both gene regulatory networks that underlie cell-type specification and how cell types have evolved. In this study, we use single-cell technology, experimental manipulations, and computational analyses to understand the role of the regulatory determinant POU IV—a homolog of Brn3 in vertebrates—in specifying various sensory cell types in Ciona. Surprisingly, the misexpression of POU IV throughout the epidermis led to the formation of hybrid sensory cell types, including those exhibiting properties of both palp sensory cells and bipolar tail neurons. These results demonstrate the interconnectedness of diverse sensory specification networks and give insights into the opportunities and challenges of reprogramming cell types through the targeted misexpression of cellular determinants., The protovertebrate Ciona intestinalis type A (sometimes called Ciona robusta) contains a series of sensory cell types distributed across the head–tail axis of swimming tadpoles. They arise from lateral regions of the neural plate that exhibit properties of vertebrate placodes and neural crest. The sensory determinant POU IV/Brn3 is known to work in concert with regional determinants, such as Foxg and Neurogenin, to produce palp sensory cells (PSCs) and bipolar tail neurons (BTNs), in head and tail regions, respectively. A combination of single-cell RNA-sequencing (scRNA-seq) assays, computational analysis, and experimental manipulations suggests that misexpression of POU IV results in variable transformations of epidermal cells into hybrid sensory cell types, including those exhibiting properties of both PSCs and BTNs. Hybrid properties are due to coexpression of Foxg and Neurogenin that is triggered by an unexpected POU IV feedback loop. Hybrid cells were also found to express a synthetic gene battery that is not coexpressed in any known cell type. We discuss these results with respect to the opportunities and challenges of reprogramming cell types through the targeted misexpression of cellular determinants.

Ascidians possess relatively small and compact genomes. This feature enables us to easily isolate cis-regulatory DNAs of genes of interest. Particularly, cis-regulatory DNAs of genes showing tissue- or cell-type-specific expression are routinely used for the artificial induction of gene expression. This strategy helps us to label cells, tissues, and organs of interest, and to investigate gene functions through overexpression, ectopic expression, and the disruption of functions by dominant-negative forms. Thus, cis-regulatory DNAs provide a powerful tool for tissue-specific genetic manipulation in studies of ascidian development and physiology. This chapter summarizes the types of cis-regulatory DNAs as a genetic manipulation tool, describes the methods used for isolating cis-regulatory DNAs, and provide reported examples of the use of cis-regulatory DNAs as molecular tools for investigating gene functions.

The larval nervous system of the ascidian Ciona intestinalis exhibits an abstract form of the vertebrate nervous system. The Ci-Galphai1 gene, which encodes a G-protein alpha subunit, is specifically expressed in distinct sets of neurons in C. intestinalis larvae, including papillar neurons of the adhesive organ, ocellus photoreceptor cells, and cholinergic and GABAergic neurons in the central nervous system (CNS). A GFP reporter gene driven by the 4.2-kb 5' flanking region of Ci-Galphai1 recapitulated the endogenous gene expression patterns. Comparative genomic analysis of the Galphai1 gene between C. intestinalis and Ciona savignyi identified an 87-bp highly conserved non-coding sequence located between -3176 and -3090 bp upstream of the gene. Deletion of this conserved upstream sequence resulted in the complete loss of reporter expression in the central nervous system, while reporter expression in the adhesive organ and mesenchyme cells remained unaffected. The conserved upstream sequence can activate gene expression from basal promoters in the brain vesicle, although it requires additional cis-regulatory sequences to fully activate the CNS-specific gene expression. These results suggest that different types of central neurons share a common transcriptional activation mechanism that is different from that of papillar neurons.