Structural basis for LEAFY floral switch function and similarity with helix-turn-helix proteins

Abstract

The LEAFY (LFY) protein is a key regulator of flower development in angiosperms. Its gradually increased expression governs the sharp floral transition, and LFY subsequently controls the patterning of flower meristems by inducing the expression of floral homeotic genes. Despite a wealth of genetic data, how LFY functions at the molecular level is poorly understood. Here, we report crystal structures for the DNA-binding domain of Arabidopsis thaliana LFY bound to two target promoter elements. LFY adopts a novel seven-helix fold that binds DNA as a cooperative dimer, forming base-specific contacts in both the major and minor grooves. Cooperativity is mediated by two basic residues and plausibly accounts for LFY's effectiveness in triggering sharp developmental transitions. Our structure reveals an unexpected similarity between LFY and helix-turn-helix proteins, including homeodomain proteins known to regulate morphogenesis in higher eukaryotes. The appearance of flowering plants has been linked to the molecular evolution of LFY. Our study provides a unique framework to elucidate the molecular mechanisms underlying floral development and the evolutionary history of flowering plants.

Figure 1

DNA-dependent dimerization of LFY-C. (A) Size-exclusion chromatography. LFY-C (40 μM, black curve), AP1 DNA (10 μM, dark grey curve), LFY-C (40 μM)+AP1 DNA (10 μM, light grey curve) were analysed. LFY-C elution at a volume corresponding to 28 kDa is consistent with the monomer size (25.7 kDa), the DNA duplex elutes earlier than expected at a volume corresponding to 52 kDa because of its elongated shape. The LFY/DNA complex elutes at a volume corresponding to 121 kDa. Molecular weights estimated from the calibration curve (dashed line) are indicated. (B) Molecular mass of LFY-C alone (dashed line) or in combination with AP1 DNA (solid line) determined by multi-angle laser light scattering and refractometry combined with size-exclusion chromatography. Elution profiles were monitored by excess refractive index (left ordinate axis). Dots show the molecular mass distribution (right ordinate axis). Average molecular mass is 64±2 kDa for the LFY-C/DNA complex (65 kDa theoretical size for a dimeric complex) and 35±1 kDa for LFY-C alone (26 kDa theoretical size for LFY-C monomer). (C) Electrophoretic mobility shift assay (EMSA) with 10 nM AP1 DNA and various LFY-C or GFP–LFY-C concentrations. Schematic complexes with LFY-C (filled circle) and GFP–LFY-C (open circle) are depicted.

Figure 2

Sequence alignments. (A) Aligned C-terminal amino-acid sequences of LFY (Arabidopsis thaliana, AAA32826), BgLFY (Brownea grandiceps, AAS79888), FLO (Antirrhinum majus, P23915), NymodLFY (Nymphea odorata, AAF77609), WelLFY (Welwitschia mirabilis, AAF23870), MatstLFY (Matteuccia struthiopteris, AAF77608) and PpLFY1 (Physcomitrella patens, BAD91043). Identical and conservatively substituted residues are depicted on a grey background. Secondary structure elements are indicated. Residues involved in interactions with DNA bases and backbone are labelled with red and blue circles, respectively. Dashed bars indicate disordered regions in the crystal, blue rectangles indicate the residues involved in dimerization. Green triangles indicate the position of Arabidopsis mutations and residues divergent in PpLFY1 are highlighted in pink. (B) Two DNA duplexes containing the LEAFY-binding sites from AP1 and AG promoters present in the LEAFY–DNA complex crystals are depicted. Base pairs related by a dyad (indicated by a black dot) are highlighted in yellow.

Figure 3

Structure of the LFY-C dimer bound to DNA. (A, B) Two orthogonal views of the LFY-C dimer (residues 237–399) bound to DNA. Monomers are coloured in olive and orange with the helix-turn-helix (HTH, helices α2 and α3) motif in red. The DNA duplex is depicted in blue. Figures 3, 4B, 5A and 6 were produced with program Pymol (Delano, 2002). (C) Superposition of the DNA duplex found in the LEAFY–DNA complex (blue) with regular B-form DNA (red).

Figure 4

DNA recognition by LEAFY. (A) Protein–DNA interactions in one AP1 half-site. Dyad-related base pairs 7 and 9 from the other half-site are shown in pink and encircled. Polar and hydrophobic interactions are shown with solid and dashed arrows, respectively. K284 belongs to the other monomer and is depicted in green. The pseudo-dyad coinciding with the crystallographic dyad is depicted in black. (B) Ribbon diagram of one LEAFY monomer bound to its AP1 half-site. The protein is coloured in olive except for the HTH motif shown in red. Polar interactions are indicated by dashed lines. For clarity, only side chains in contact with DNA are shown. (C) Effect of selected mutations on LFY-C DNA-binding affinity to AP1 DNA. EMSAs were performed with wild-type and mutant LFY-C (100–250–750–2000 nM from left to right). Only dimeric complexes are shown except for P308A that gave rise to an unknown higher complex. AP1 m5 mutant DNA contains base pair C:G instead of A:T at position±8 (see Supplementary Table 3 for full DNA sequences). Phenotype of the wild-type Arabidopsis inflorescence (D) and lfy-28 (P308L) mutant inflorescence (F) and flower (E). Scale bar is 1 mm on (D, F) and 0.5 mm on (E).

Figure 5

The LFY-C dimer interface mediates cooperative binding. (A) The dimer interface is viewed perpendicular to the DNA axis. Polar contacts between the two monomers (in orange and olive) are shown with dashed lines. (B) EMSA with increasing concentrations (0, 10, 20, 50, 100, 200, 500, 1000, 2000 and 3000 nM from left to right) of LFY-C wild-type, R390A mutant, H387A mutant, and H387A/R390A double mutant and 50 μM AP1 DNA. Free DNA (F), monomeric (M) and dimeric (D) complexes are indicated. (C) Estimation of dissociation constants for wild-type LFY-C and three mutant versions (H387A, R390A and H387A/R390A). Binding of LFY-C to AP1 DNA was modelled as two equilibrium reactions as detailed in Supplementary data: (1) Binding of a first LFY-C monomer to AP1 DNA, leading to the formation of the monomeric complex (M) and characterized by the K_d1 dissociation constant; (2) binding of a second LFY-C monomer to M, leading to the formation of the dimeric complex (D) and characterized by K_d2. EMSA signals from (B) were quantified and the corresponding experimental values were fitted with theoretical equations describing the two equilibria. The errors and intervals between square brackets indicated correspond to the 95% confidence interval. An elevated K_d1/K_d2 ratio reflects a high level of cooperativity, whereas a ratio of 1 would indicate an absence of binding cooperativity. The single mutations resulted in a weak decrease of cooperativity, whereas the H387A/R390A double mutation strongly decreased the cooperativity.

Figure 6

Comparison of LFY-C with paired and homeodomain DNA binding. (A) Two orthogonal views of LFY-C helices α1–α3 bound to their DNA target site (red) superimposed with the three-helical bundle core of the N-terminal subdomain of the paired domain of Drosophila Prd (blue, PDB-id: 1pdn). (B) Superposition with the homeodomain of Drosophila engrailed bound to DNA (yellow, PDB-id: 1hdd), where the centre of recognition helix α3 inserts into the major groove.