Tbx5 and Tbx20 act synergistically to control vertebrate heart morphogenesis

Abstract

Members of the T-box family of proteins play a fundamental role in patterning the developing vertebrate heart; however, the precise cellular requirements for any one family member and the mechanism by which individual T-box genes function remains largely unknown. In this study, we have investigated the cellular and molecular relationship between two T-box genes, Tbx5 and Tbx20. We demonstrate that blocking Tbx5 or Tbx20 produces phenotypes that display a high degree of similarity, as judged by overall gross morphology, molecular marker analysis and cardiac physiology, implying that the two genes are required for and have non-redundant functions in early heart development. In addition, we demonstrate that although co-expressed, Tbx5 and Tbx20 are not dependent on the expression of one another, but rather have a synergistic role during early heart development. Consistent with this proposal, we show that TBX5 and TBX20 can physically interact and map the interaction domains, and we show a cellular interaction for the two proteins in cardiac development, thus providing the first evidence for direct interaction between members of the T-box gene family.

Fig. 1

Tbx5 and Tbx20 morpholinos block translation of their respective target proteins. (A) TBX5MO and TBX20MO positions relative to Tbx5 and Tbx20 cDNA. (B) Inhibition of TBX5-V5 translation in vitro by TBX5MO. TBX20MO and ControlMO serve as controls. Each reaction contains 1 μg of Tbx5-V5 circular plasmid along with the indicated amounts of MO. (C) Inhibition of TBX20-V5 translation in vitro by TBX20MO. TBX5MO and ControlMO serve as controls. Each reaction contains 1 μg of Tbx20-V5 circular plasmid along with the indicated amounts of MO. (D) Inhibition of TBX5-V5 translation by TBX5MO in animal caps. TBX20MO and ControlMO serve as controls. Probed with anti-V5 and re-probed with anti-PTP1D/SHP2 as a loading control. Embryos injected with 2 ng mRNA and the indicated amounts of MO. (E) Inhibition of TBX20-V5 translation by TBX20MO in animal caps. TBX20MO and ControlMO serve as controls. Probed with anti-V5 and re-probed with anti-PTP1D/SHP2 as a loading control. Embryos injected with 2 ng mRNA and the indicated amounts of MO.

Fig. 2

Tbx5 and Tbx20 are required for proper cardiogenesis. (A–F) Morpholino-injected tadpoles at the indicated stages. Control morphant embryos (A,D), TBX5 morphant embryos (B,E) and TBX20 morphant embryos (C,F). Arrows indicate the heart region, arrowheads indicate the eye. (G) Chart displaying the percentage of morphants surviving and displaying cardiac abnormalities, as scored by the presence of an unlooped heart tube, a reduction in cardiac mass and the presence of a pericardial edema.

Fig. 3

TBX5 and TBX20 morphants fail to undergo looping and chamber formation and display reduced cardiac cell numbers. Cryosections of TBX5 and TBX20 morphant hearts taken at the anterior (outflow), middle (ventricular) and posterior (atrial) regions. (A–D) ControlMO, (E–H) TBX5MO and (I–L) TBX20MO. Sections stained for tropomyosin (red), DAPI (blue) and cardiac actin (green). (D,H,L) Same sections as B, F and J stained with cardiac actin. In the looped control heart, the middle ventricular section also contains the atrium. (M) Mean number of cells per heart obtained by cell counts of heart tissue in serial sections derived from a minimum of three embryos.

Fig. 4

Cardiac specification is unaltered in TBX5 and TBX20 morphants. Whole-mount in situ with Nkx2.5 on stage matched (A,D,G,J) ControlMO-, (B,E,H,K) TBX5MO- or (C,F,I,L) TBX20MO-derived embryos.

Fig. 5

TBX5 and TBX20 morphants display dramatic morphological defects. Whole-mount in situ hybridization of cleared stage 36 embryos. (A–C) ANF whole-mount in situ hybridization. (D–F) XTnIc whole-mount in situ hybridization. (A,D) ControlMO. (B,E) TBX5MO and (C,F) TBX20MO.

Fig. 6

Tbx5 and Tbx20 are localized to the nucleus and can activate transcription on the Nppa/ANF promoter. (A) Schematic depicting the amino acid positions of the T-box domains of Tbx5 and Tbx20. (B) Schematic of Rat Nppa/ANF-luciferase reporter construct showing T-box-binding site consensus sequences and their relative position within the promoter relative to translation start site. (C–H) Transfected cells were stained with anti-V5 (C,F; Cy2, green) for TBX5-V5 and TBX20-V5, and with anti-phosphotyrosine (D,G; Cy3, red) to visualize cytoplasmic compartment. Overlaid image of Cy2 and Cy3 staining (E,H). (I) Rat ANF-luciferase co-transfected with a constant amount of Tbx5 (100 ng) and increasing amounts of Tbx20 (25, 50, 100 and 500 ng) in 293T cells, and the level of transcriptional activation is expressed as relative luciferase units based on average of three independent experiments performed in triplicate.

Fig. 7

Tbx5 and Tbx20 are not required for the expression of each other. Embryos injected at the one-cell stage with ControlMO, TBX5MO or TBX20MO. (A,C) Whole-mount in situ hybridization showing Tbx5 expression. (B,D) Whole-mount in situ hybridization showing Tbx20 expression.

Fig. 8

TBX5 and TBX20 physically interact. Cell lysates containing GST- and/or HA-tagged proteins were incubated on GST and eluted, and separated by SDS-PAGE. GST proteins were detected using anti-GST antibodies and HA-tagged proteins were detected with anti-HA antibodies. (A) Association of TBX5 with TBX20 is shown by pull-down of HA-TBX20 with GST-TBX5. HA-NKX2-5 serves as positive control. Fifteen percent of output and 7.5% of input was probed. (B,C) Pull-down of full-length HA-TBX20 with a GST-tagged TBX5 deletion series reveals an interaction domain in the N terminus and T-box region of TBX5. Each reaction was probed with anti-HA antibodies (B). Fifteen percent of output and 7.5% of input was probed. (D,E) Pull-down of HA-TBX20 deletion series with full-length GST-TBX5 reveals an interaction domain within the N-terminus and T-box of TBX20. Each reaction was probed with anti-HA antibodies (D). Fifteen percent of output and 7.5% of input was probed, except in the case of ΔN/C, in which the amount of protein probed was only 4% owing to the increase in total amount of ΔN/C protein used in pull-down (see text).

Fig. 9

Tbx5 and Tbx20 synergistically act to regulate cardiac gene expression. (A–L) Embryos injected with the indicated morpholinos at the one-cell stage. (A–D) Nkx2-5 whole-mount in situ hybridization. (E–H) XANF whole-mount in situ hybridization. (I–L) XTnIc whole-mount in situ hybridization. (A,E,I) ControlMO, (B,F,J) TBX5MO injected at suboptimal dose, (C,G,K) TBX20MO injected at suboptimal dose, (D,H,L) TBX5MO and TBX20MO injected in combination at suboptimal doses. All embryo were cleared to reveal heart expression. (M) Statistics for embryos injected with suboptimal doses of TBX5MO and TBX20MO in combination with each other or with ControlMO. Hearts were judged as having defects if they displayed a pericardial edema, an unlooped heart tube or reduction in cardiac mass.