# The 3′ region of the ZPA regulatory sequence (ZRS) is required for activity and contains a critical E-box

## TL;DR

This study identifies the 3′ region of the ZPA regulatory sequence as essential for its activity and highlights the role of a specific E-box in regulating Sonic hedgehog expression during limb development.

## Contribution

The study reveals that the 3′ subdomain of the ZRS, not the central subdomain, is critical for enhancer activity and Hand2 function.

## Key findings

- The 3′ fragment of the ZRS is necessary for activity, while the 5′ and central fragments are not.
- Restoring the 3′ E-box alone is sufficient for robust ZRS activity.
- Hoxd13 sites promote localized activity within the ZPA.

## Abstract

During development, Hand2 and Hoxd13 transcription factors (TFs) regulate Sonic hedgehog (Shh) expression in the zone of polarizing activity (ZPA) in the distal posterior limb mesoderm. The ZPA regulatory sequence (ZRS) is a conserved, limb-specific enhancer that controls Shh expression. The ZRS can be divided into 5′, central, and 3′ subdomains, each with an E-box site that can bind basic helix-loop-helix (bHLH) TFs like Hand2. In addition, two Hoxd13 sites are present in the 5′ and central subdomains. Hand2 purportedly binds the ZRS through the central E-box, and both Hand2 and Hoxd13 have been shown to activate the ZRS in vitro. We hypothesized that the central E-box was required for activity, while the other E-boxes and Hoxd13 sites localize ZRS activity to the distal posterior limb mesoderm.

To identify the functional role of each subdomain, we generated three ZRS fragments (5′, central, and 3′) and combined fragment constructs to test subdomain collective contributions. Additionally, we disrupted the five binding sites, alone or in concert, using site-directed mutagenesis. All ZRS constructs were cloned into a GFP reporter and evaluated in an in vivo chicken limb bioassay. We validated our findings using select ZRS constructs in transgenic mice.

We found that the 3′ fragment was necessary for ZRS activity, while the 5′ and central fragments had no activity alone or when combined. However, combining the 3′ fragment with the 5′ fragment restored robust activity. Further, mutation of all five binding sites markedly reduced ZRS activity. Reinstating each of the Hoxd13 sites restored focal activity, while restoring the 5′ and central E-boxes had little effect. However, the 3′ E-box proved sufficient for robust activity even in the absence of the other four binding sites.

Our data indicate that the ZRS 3′, not the central, subdomain is necessary for activity and contains the 3′ E-box that Hand2 likely uses to induce Shh expression, while the 5′ and central E-boxes appear to be inhibitory. Our data also suggest that the Hoxd13 binding sites promote localized activity within the ZPA.