Identification and characterization of the zebrafish pharyngeal arch-specific enhancer for the basic helix-loop-helix transcription factor Hand2

Abstract

The development of the vertebrate jaw relies on a network of transcription factors that patterns the dorsal-ventral axis of the pharyngeal arches. Recent findings in both mouse and zebrafish illustrate that the basic-helix-loop-helix transcription factor, Hand2, is crucial in this patterning process. While Hand2 has functionally similar roles in these two species, little is known about the regulatory sequences controlling hand2 expression in zebrafish. Using bioinformatics and Tol2-mediated transgenesis, we have generated zebrafish transgenic reporter lines in which either the mouse or zebrafish arch-specific hand2 enhancer direct expression of a fluorescent reporter. We find that both the mouse and zebrafish enhancers drive early reporter expression in a hand2-specific pattern in the ventral pharyngeal arches of zebrafish embryos. These lines provide useful tools to follow ventral arch cells during vertebrate jaw development while also allowing dissection of hand2 transcriptional regulation during this process.

Fig. 1

Identification and cloning of potentialhand2 regulatory regions. A. 12 kb of genomic sequence flanking the zebrafish hand2-coding region compared to Xenopus tropicalis, human and mouse genomes using dcode ECR browser. The zebrafish hand2 coding sequence is shown at the top, with blue indicating exons and yellow indicating 5′ and 3′ untranslated regions (UTRs). The red peaks identify conserved non-coding regions that show sequence similarity of at least 50%. The distance between vertical gray lines is 1 kb. Asterisks indicate the region chosen for further analysis. B. Nucleotide alignment of the mouse (mm9) and zebrafish (danRer5) ECRs identified in (A) illustrates 66.4% identity. C. The 600 bp zebrafish hand2 enhancer was cloned into p5E-FAbas upstream of a β-actin minimal promoter using FseI and AscI. The middle entry clone contains a tag-RFP reporter and the 3′ entry clone contains the SV40 poly-adenylation (pA) signal. These three plasmids were combined in a Multisite Gateway reaction with the destination vector pDestTol2CG2 containing cmlc2:GFP as a transgenesis marker, resulting in Hand2-directed tag-RFP expression. D. The previously described mouse Hand2 arch-specific enhancer was similarly cloned into p5E-FAbas and recombined with a middle entry clone containing EGFP and a 3′ SV40 pA entry clone in the Tol2 destination vector, pDestTol2, resulting in Hand2-directed EGFP expression. Plasmids are not drawn to scale.

Fig. 2

Both zebrafish and mouse enhancers drive transgene expression in a hand2-specific pattern within the pharyngeal arches. Dorsolateral views are shown at 30 and 48 hpf, and lateral views at 72 and 84 hpf and 4 dpf. A–E. Tg(hand2:tag-RFP) embryos were examined for RFP fluorescence between 30 h post fertilization (hpf) and 4 day post-fertilization (dpf). RFP was first detected at 30 hpf in the ventral aspect of pharyngeal arches 1–3 (A). At 48 hpf, RFP was present in the ventral aspects of arches 1–4 (B). At 72 hpf, RFP also appeared ventrally located (C). By 84 hpf (D) and 4 dpf (E), RFP was present in the developing cartilages while also appearing to expand dorsally. F-J. Tg(mHand2:EGFP) embryos were examined at the same stages as the Tg(hand2:tag-RFP) embryos. EGFP activity was observed in a hand2 specific expression pattern in the ventral aspects of arches 1–3 beginning around 30 hpf (F) and persisting in arches 1–4 through 48 hpf (G). EGFP activity continued as chondrogenesis began around 72–84 hpf (H, I) and was still present at 4 dpf (J), though EGFP activity expanded dorsally at both 84 hpf and 4 dpf. K–O. Whole mount in situ hybridization analysis of hand2 expression. Endogenous hand2 expression was present in arches 1–4 at 30 hpf (K) and 48 hpf (L). While endogenous hand2 expression remained at 72 and 84 hpf (M, N), expression began to be restricted by 4 dpf (O). P–R. Compressed confocal z-stacks of Tg(hand2:tag-RFP), Tg(mHand2:EGFP) embryos at 36 hpf in a lateral view, illustrating localization of tag-RFP (P), EGFP (Q) and co-localization of both reporters (R) within the ventral pharyngeal arches.

Fig. 3

The zebrafish hand2 pharyngeal arch-specific enhancer drives lacZ expression in the mouse pharyngeal arches. A. Cloning of the zhand2 arch-specific enhancer into a minimal Hsp68 promoter-lacZ expression vector (Hsp68-lacZ). Plasmid is not drawn to scale. B. Ventral view of a transient transgenic embryo at embryonic day (E) 10.5 showed expression of the zhand2-Hsp68-lacZ transgene in the distal aspects of the mandibular arch (1). β-gal activity was also present in arch 2, but slightly more proximal to that observed in the mandibular arch. C. Transverse sectional analysis of the embryo shown in B illustrated the β-gal activity in the distal mandibular arch mesenchyme. D. Ventral view of endogenous Hand2 expression in control embryos at E10.5, illustrating the distal mesenchyme of the mandibular and second arches. E. Transverse sections through the mandibular arch of the embryos shown in D further illustrated the distal expression.

Fig. 4

The minimal hand2 arch-specific enhancer is regulated by endothelin signaling. A. A 280 bp region of sequence contained within the larger zhand2 enhancer shown in Fig. 1C was cloned into p5E-FAbas and recombined with a middle entry clone containing a mCherry fluorescent reporter into the Tol2 destination vector pDestTol2CG2, resulting in hand2-directed mCherrry expression. Plasmid is not drawn to scale. B–E. In Tg(shand2:mCherry) embryos, mCherry fluorescence was observed in the Hand2 domain beginning around 30 hpf (B) and continuing through arch development (C, D). mCherry fluorescence can still be detected in the ventral cartilages as late as 7 dpf (E). F, G. Injecting 5 ng of an edn1 morpholino led to decreased hand2 expression at 48 hpf (G) compared to wild-type expression (F). H, I. mCherry activity was similarly decreased in Tg(shand2:mCherry) embryos following injection of the edn1 morpholino (I) compared to uninjected Tg(shand2:mCherry) embryos (H).

Fig. 5

Transcriptional regulation of the hand2 arch-specific enhancer. A. The zebrafish enhancer (bottom) is shown, with the location of several potential regulators as predicted by MatInspector marked by colored boxes. The identities of the colored boxes are shown below the enhancer; a similar map of the mouse Hand2 enhancer is shown for comparison (top). Each gray or black bar represents 50 bp. B. A hand2-luciferase vector was transfected into 10T1/2 mouse embryonic fibroblasts along with a dlx5a expression vector. Transfection efficiency was controlled by co-transfection with a renilla reporter construct. Activation of zhand2 by Dlx5a was normalized to the activation obtained when using an empty expression vector (mock). The error is presented as S.E.M * p=0.0026.