Selection of distinct Hox-Extradenticle interaction modes fine-tunes Hox protein activity

Abstract

Hox genes encode transcription factors widely used for diversifying animal body plans in development and evolution. To achieve functional specificity, Hox proteins associate with PBC class proteins, Pre-B cell leukemia homeobox (Pbx) in vertebrates, and Extradenticle (Exd) in Drosophila, and were thought to use a unique hexapeptide-dependent generic mode of interaction. Recent findings, however, revealed the existence of an alternative, UbdA-dependent paralog-specific interaction mode providing diversity in Hox-PBC interactions. In this study, we investigated the basis for the selection of one of these two Hox-PBC interaction modes. Using naturally occurring variations and mutations in the Drosophila Ultrabithorax protein, we found that the linker region, a short domain separating the hexapeptide from the homeodomain, promotes an interaction mediated by the UbdA domain in a context-dependent manner. While using a UbdA-dependent interaction for the repression of the limb-promoting gene Distalless, interaction with Exd during segment-identity specification still relies on the hexapeptide motif. We further show that distinctly assembled Hox-PBC complexes display subtle but distinct repressive activities. These findings identify Hox-PBC interaction as a template for subtle regulation of Hox protein activity that may have played a major role in the diversification of Hox protein function in development and evolution.

Fig. 1.

Distinct modes of Exd interaction in arthropod Ubx proteins. (A) Alignment of Ubx peptides (HX to UbdA sequences) from C. salei, S. maritima, A. franciscana, and D. melanogaster UbxIa and IVa isoforms. (B) EMSA of Af-Ubx, Stm-Ubx, Cs-Ubx, and Dm-UbxIa variants, wild type or mutated in the HX or UbdA motifs on the DllR probe. Lanes 1 and 2: DllR probe alone and in the presence of a fixed amount of Exd and Hth; lanes 3–26: fixed amount of Exd, Hth, and a fixed, identical amount of Af-Ubx, Stm-Ubx, or Cs-Ubx variants. Bars above the gel indicate proteins present in the experiment. (C) (Left) Repressive potentials in Drosophila of Cs-Ubx variants. The arm-Gal4–driven ubiquitous expressions of the variants (red) repress the Dll reporter gene (green) to different extents. (Right) Quantifications of Dll reporter repression. The error bars represent SDs; the number above the bracket indicates the P value.

Fig. 2.

Drosophila UbxIa and -IVa isoforms repress Dll using different Exd interaction modes. (A) EMSA of Dm-UbxIVa variants, wild type or mutated, in the HX or UbdA motifs on the DllR probe. Lanes 1 and 2: DllR probe alone and in the presence of fixed amount of Exd and Hth. Lanes 3–8: fixed amount of Exd, Hth, and fixed, identical amount of Dm-UbxIVa variants. Bars above the gel indicate proteins present in the experiment. (B) Repressive potentials of Dm-UbxIa and Dm-UbxIVa variants. (Left) armadillo(arm)-Gal4–driven ubiquitous expression of the variants (red) represses the Dll reporter gene (green) to different extents. (Right) Quantifications of Dll reporter repression. The error bars represent SDs; numbers above the brackets indicate P values.

Fig. 3.

Evolutionary LR extension promotes UbdA-mediated Exd interaction by Ubx for Dll repression. (A) Schematic representation of Cs-Ubx variants, with Dm-Ubx LR insertion (yellow) and motif mutations indicated by black crosses. (B) EMSA of Cs-Ubx variants on the DllR probe. Lane 1: DllR probe alone; lanes 2–12: fixed amount of Exd and Hth and increasing identical amounts of Cs-Ubx variants, as indicated. Bars above the gel indicate the amount of protein present in the experiment. (C) (Left) Repressive potentials in Drosophila of Cs-Ubx(DmLR) variants. arm-Gal4–driven ubiquitous expression of the variants (red) represses the Dll reporter gene (green) to different extents. (Right) Quantifications of Dll reporter repression. The error bars represent SDs; the number above the bracket indicates the P value.

Fig. 4.

Conserved role of the evolutionary LR extension in AbdA. (A) Alignment of AbdA peptides (HX to UbdA sequences) from Cupiennius salei (Cs) Procambarus clarkii (Pc), Drosophila melanogaster (Dm), and the modified Cs-AbdA(DmLR), Cs-AbdA(PcLR), and Cs-AbdA(LRala) proteins. (B) EMSA of Cs-AbdA variants on the DllR probe. Lanes 1 and 2: DllR probe alone and in the presence of fixed amount of Exd and Hth; lanes 3–26: fixed amount of Exd and Hth and fixed, identical amount of Cs-AbdA variants. Bars above the gel indicate the amount of protein present in the experiment. (C) (Upper) Repressive potentials in Drosophila of Cs-AbdA variants. arm-Gal4–driven ubiquitous expression of the variants (red) represses the Dll reporter gene (green) to different extents. (Lower) Quantifications of Dll reporter repression. The error bars represent SDs, and numbers above the brackets indicate P values.

Fig. 5.

Motifs’ requirements for segment-identity specification by Dm-UbxIa. (A) Anterior regions of wild-type and of ubiquitously expressing Dm-UbxIa, Dm-UbxIa^HX, or Dm-UbxIa^UbdA first-instar larvae. Arrows point toward “beards,” small denticles in the naked region of T1 and T1-like transformed segments (T1′). (B) Distribution of Scr transcripts (green) in wild-type and ubiquitously expressing Dm-UbxIa or Dm-UbxIa^UbdA embryos (red). No ectopic Scr transcripts are detected in T2 and T3 segments.