Homologous proteins occurring through gene duplication may give rise to novel functions through mutations affecting protein sequence or expression. Comparison of such homologues allows insight into how morphological traits evolve. However, it is often unclear which changes are key to determining new functions. To address these ideas, we have studied a system where two homologues have evolved clear and opposite functions in controlling a major developmental switch. In plants, flowering is a major developmental transition that is critical to reproductive success. Arabidopsis phosphatidylethanolamine-binding protein homologues TERMINAL FLOWER 1 (TFL1) and FLOWERING LOCUS T (FT) are key controllers of flowering, determining when and where flowers are made, but as opposing functions: TFL1 is a repressor, FT is an activator. We have uncovered a striking molecular basis for how these homologous proteins have diverged. Although <60% identical, we have shown that swapping a single amino acid is sufficient to convert TFL1 to FT function and vice versa. Therefore, these key residues may have strongly contributed to the selection of these important functions over plant evolution. Further, our results suggest that TFL1 and FT are highly conserved in biochemical function and that they act as repressors or activators of flowering through discrimination of structurally related interactors by a single residue. FLOWERING LOCUS T ͉ phosphatidylethanolamine-binding protein ͉Raf-kinase inhibitor protein ͉ TERMINAL FLOWER 1 N ovel morphologies arise through the evolution of new protein functions. Duplicated genes are a key source of new functions, acquiring mutations that affect expression and͞or protein sequence (1, 2). Studies of large gene families show that different members diverge and participate in different developmental pathways, for example, homeobox and MADS box genes in various species (3-5). However, it is often unclear and difficult to determine which changes in homologues are critical to establish a novel function.The Arabidopsis homologues TERMINAL FLOWER 1 (TFL1) and FLOWERING LOCUS T (FT) provide an excellent model to address this question (6-9). Flowering plant species arose Ͼ 100 million years ago, and FT and TFL1 have been conserved in diverse species, including monocots and eudicots (10-15). Both TFL1 and FT are key controllers of flowering and plant architecture but act in an opposite manner. TFL1 is a repressor, and FT is an activator. Further, gain-of-function studies gave clear and opposite phenotypes in vivo, showing that protein sequence, rather than expression pattern, largely determines the different functions of TFL1 and FT (8,9,16).TFL1 is expressed in the shoot apical meristem (SAM) and represses the transition to flowering; tfl1 mutants flower early (6,7,17,18). TFL1 also maintains indeterminate growth of the SAM by repressing floral meristem identity genes; tfl1 mutants have their SAMs converted into terminal flowers. TFL1 therefore controls plant architecture by determining where flowers are made a...

In Azotobacter vinelandii, activation ofnif gene expression by the transcriptional regulatory enhancer binding protein NIFA is controlled by the sensor protein NIFL in response to changes in levels of oxygen and fixed nitrogen in vivo. NIFL is a novel redox-sensing flavoprotein which is also responsive to adenosine nucleotides in vitro. Inhibition of NIFA activity by NIFL requires stoichiometric amounts of the two proteins, implying that the mechanism of inhibition is by direct protein-protein interaction rather than by catalytic modification of the NIFA protein. The formation of the inhibitory complex between NIFL and NIFA may be regulated by the intracellular ATP/ADP ratio. We show that adenosine nucleotides promote complex formation between purified NIFA and NIFL in vitro, allowing isolation of the NIFL-NIFA complex. The complex can also be isolated from cell extracts containing coexpressed NIFL and NIFA in the presence of MgADP. Removal of the nucleotide causes dissociation of the complex. Experiments with truncated proteins demonstrate that the amino-terminal domain of NIFA and the C-terminal region of NIFL potentiate the ADP-dependent stimulation of NIFL-NIFA complex formation.

The enhancer binding protein NIFA and the sensor protein NIFL from Azotobacter vinelandii comprise an atypical two-component regulatory system in which signal transduction occurs via complex formation between the two proteins rather than by the phosphotransfer mechanism, which is characteristic of orthodox systems. The inhibitory activity of NIFL towards NIFA is stimulated by ADP binding to the C-terminal domain of NIFL, which bears significant homology to the histidine protein kinase transmitter domains. Adenosine nucleotides, particularly MgADP, also stimulate complex formation between NIFL and NIFA in vitro, allowing isolation of the complex by cochromatography. Using limited proteolysis of the purified proteins, we show here that changes in protease sensitivity of the Q linker regions of both NIFA and NIFL occurred when the complex was formed in the presence of MgADP. The N-terminal domain of NIFA adjacent to the Q linker was also protected by NIFL. Experiments with truncated versions of NIFA demonstrate that the central domain of NIFA is sufficient to cause protection of the Q linker of NIFL, although in this case, stable protein complexes are not detectable by cochromatography.In the free-living diazotrophs Klebsiella pneumoniae and Azotobacter vinelandii, activation of expression of genes involved in nitrogen fixation by the enhancer binding protein NIFA is controlled by the sensor protein NIFL in response to changes in levels of oxygen and fixed nitrogen in vivo. The NIFA protein activates transcription at the N -dependent promoters for nitrogen fixation (nif) genes, in combination with N -RNA polymerase holoenzyme. Transcriptional activation by NIFA is repressed by NIFL in response to increases in levels of fixed nitrogen and extracellular oxygen (reviewed in reference 5). The NIFL proteins from both A. vinelandii and K. pneumoniae have been shown to contain flavin adenine dinucleotide (FAD) as the prosthetic group (14,20). For A. vinelandii NIFL, we have shown that the oxidized form of the protein inhibits NIFA activity, but when the flavin moiety is reduced, NIFA activity is unaffected (14). In addition to its ability to act as a redox sensor, NIFL is also responsive to adenosine nucleotides in vitro, the inhibitory activity of the protein being stimulated by the presence of ADP (8). The NIFL protein is comprised of two domains tethered by a Q linker (4, 6,7). Q linkers are short (20 to 30 residues), hydrophilic sequences rich in Gln, Glu, Pro, Arg, and Ser, which serve to tether independently folding domains of some regulatory proteins (24). The N-terminal domain contains the flavin binding site and shows some homology to other oxygen-and redox-sensing proteins (4). The C-terminal domain of A. vinelandii NIFL shows significant homology to the histidine protein kinase transmitter domains, including the conserved histidine residue (4). We have shown previously that this domain binds nucleotides, particularly ADP (22). However, phosphotransfer between NIFL and NIFA has never been demonstrated (2,15,21),...