Abstract
A major goal in biology is to understand the rules by which cis-regulatory sequences control spatially and temporally precise expression patterns. Here we present a systematic dissection of the proximal enhancer for the notochord-specific transcription factor brachyury in the ascidian chordate Ciona. The study uses a quantitative image-based reporter assay that incorporates a dual-reporter strategy to control for variable electroporation efficiency. We identified and mutated multiple predicted transcription factor binding sites of interest based on statistical matches to the JASPAR binding motif database. Most sites (Zic, Ets, FoxA, RBPJ) were selected based on prior knowledge of cell fate specification in both the primary and secondary notochord. We also mutated predicted Brachyury sites to investigate potential autoregulation as well as Fos/Jun (AP1) sites that had very strong matches to JASPAR. Our goal was to quantitatively define the relative importance of these different sites, to explore the importance of predicted high-affinity versus low-affinity motifs, and to attempt to design mutant enhancers that were specifically expressed in only the primary or secondary notochord lineages. We found that the mutation of all predicted high-affinity sites for Zic, FoxA or Ets led to quantifiably distinct effects. The FoxA construct caused a severe loss of reporter expression whereas the Ets construct had little effect. A strong Ets phenotype was only seen when much lower-scoring binding sites were also mutated. This supports the enhancer suboptimization hypothesis proposed by Farley and Levine but suggests that it may only apply to some but not all transcription factor families. We quantified reporter expression separately in the two notochord lineages with the expectation that Ets mutations and RBPJ mutations would have distinct effects given that primary notochord is induced by Ets-mediated FGF signaling whereas secondary notochord is induced by RBPJ/Su(H)-mediated Notch/Delta signaling. We found, however, that ETS mutations affected primary and secondary notochord expression relatively equally and that RBPJ mutations were only moderately more severe in their effect on secondary versus primary notochord. Our results point to the promise of quantitative reporter assays for understanding cis-regulatory logic but also highlight the challenge of arbitrary statistical thresholds for predicting potentially important sites.
Highlights
The small, simple chordate embryo, invariant cleavage patterns, compact genome and ease of transgenesis in Ciona and other ascidians have made them important models for the systematic dissection of developmental Gene Regulatory Networks (GRNs) (Imai et al, 2006; Satoh, 2014; Shi et al, 2005; Kubo et al, 2009; Satou and Peter, 2020)
Various transcription factor binding motifs of interest have been mutated, but these efforts have been split between Halocynthia and Ciona and between the proximal bra enhancer and the more distal “shadow” bra enhancer in Ciona
One somewhat unexpected finding was that our initial ETS mutant construct that destroyed the strongest predicted ETS site did not show a major decrease in expression when assessed in the quantitative dual reporter assay
Summary
The small, simple chordate embryo, invariant cleavage patterns, compact genome and ease of transgenesis in Ciona and other ascidians have made them important models for the systematic dissection of developmental Gene Regulatory Networks (GRNs) (Imai et al, 2006; Satoh, 2014; Shi et al, 2005; Kubo et al, 2009; Satou and Peter, 2020). There are robust methods for identifying transcription factors and signaling molecules of interest in distinct cell types, perturbing their functions, and identifying downstream genes that are differentially expressed in response to those perturbations (Satou et al, 2001a; Satou et al, 2001b; Satou et al, 2003; Imai et al, 2004; Sasaki et al, 2014; Stolfi et al, 2014). As in other model organisms, one of the biggest challenges in GRN analysis is determining whether transcriptional regulatory effects are direct or indirect. The gold standard for determining whether transcriptional interactions are direct or indirect usually involves cis-regulatory analysis (Stolfi and Christiaen, 2012; Irvine, 2013). If the mutation of known or predicted Transcription Factor Binding Sites (TFBS) in an enhancer construct abrogates reporter expression in the expected cell type(s), it provides strong though not unequivocal support for the interaction being direct
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
