Isolation and analysis of SaMADS C, the APETALA 1 cDNA homolog from mustard

Gene activation cascade triggered by a single photoperiodic cycle inducing flowering in Sinapis alba

Molecular genetic analyses in Arabidopsis disclosed a genetic pathway whereby flowering is induced by the photoperiod. This cascade is examined here within the time course of floral transition in the long-day (LD) plant Sinapis alba induced by a single photoperiodic cycle. In addition to previously available sequences, the cloning of CONSTANS (SaCO) and FLOWERING LOCUS T (SaFT) homologues allowed expression analyses to be performed to follow the flowering process step by step. A diurnal rhythm in SaCO expression in the leaves was observed and transcripts of SaFT were detected when light was given in phase with SaCO kinetics only. This occurred when day length was extended or when a short day was shifted towards a 'photophile phase'. The steady-state level of SaFT transcripts in the various physiological situations examined was found to correlate like a rheostat with floral induction strength. Kinetics of SaFT activation were also consistent with previous estimations of translocation of florigen out of leaves, which could actually occur after the inductive cycle. In response to one 22-h LD, initiation of floral meristems by the shoot apical meristem (SAM) started about 2 days after activation of SaFT and was marked by expression of APETALA1 (SaAP1). Meanwhile, LEAFY (SaLFY) was first up-regulated in leaf primordia and in the SAM. FRUITFULL (SaFUL) was later activated in the whole SAM but excluded from floral meristems. These patterns are integrated with previous observations concerning upregulation of SUPPRESSOR OF OVEREXPRESSION OF CO1 (SaSOC1) to provide a temporal and spatial map of floral transition in Sinapis.

MADS-box gene evolution-structure and transcription patterns

This study presents a phylogenetic analysis of 198 MADS-box genes based on 420 parsimony-informative characters. The analysis includes only MIKC genes; therefore several genes from gymnosperms and pteridophytes are excluded. The strict consensus tree identifies all major monophyletic groups known from earlier analyses, and all major monophyletic groups are further supported by a common gene structure in exons 1-6 and by conserved C-terminal motifs. Transcription patterns are mapped on the tree to obtain an overview of MIKC gene transcription. Genes that are transcribed only in vegetative organs are located in the basal part of the tree, whereas genes involved in flower development have evolved later. As the universality of the ABC model has recently been questioned, special account is paid to the expression of A-, B-, and C-class genes. Mapping of transcription patterns on the phylogeny shows all three classes of MADS-box genes to be transcribed in the stamens and carpels. Thus the analysis does not support the ABC model as formulated at present.

Analysis of PEAM4, the pea AP1 functional homologue, supports a model for AP1-like genes controlling both floral meristem and floral organ identity in different plant species

APETALA1 (AP1) and its homologue SQUAMOSA (SQUA) are key regulatory genes specifying floral meristem identity in the model plants Arabidopsis and Antirrhinum. Despite many similarities in their sequence, expression and functions, only AP1 appears to have the additional role of specifying sepal and petal identity. No true AP1/SQUA-functional homologues from any other plant species have been functionally studied in detail, therefore the question of how the different functions of AP1-like genes are conserved between species has not been addressed. We have isolated and characterized PEAM4, the AP1/SQUA-functional homologue from pea, a plant with a different floral morphology and inflorescence architecture to that of Arabidopsis or Antirrhinum. PEAM4 encodes for a polypeptide 76% identical to AP1, but lacks the C-terminal prenylation motif, common to AP1 and SQUA, that has been suggested to control the activity of AP1. Nevertheless, constitutive expression of PEAM4 caused early flowering in tobacco and Arabidopsis. In Arabidopsis, PEAM4 also caused inflorescence-to-flower transformations similar to constitutive AP1 expression, and was able to rescue the floral organ defects of the strong ap1-1 mutant. Our results suggest that the control of both floral meristem and floral organ identity by AP1 is not restricted to Arabidopsis, but is extended to species with diverse floral morphologies, such as pea.

FPF1 modulates the competence to flowering in Arabidopsis

During the transition to flowing the FPF1 gene is expressed in the peripheral zone of apical meristems and in floral meristems of Arabidopsis. Constitutive expression of FPF1 causes early flowering in Arabidopsis under both long-day and short-day conditions and leads to a shortened juvenile phase as measured by the trichome distribution on the abaxial leaf surface. In the classical late flowering mutants, overexpression of FPF1 compensates partially for the late flowering phenotype, indicating that FPF1 acts downstream or in a parallel pathway to the mutated genes. The co-overexpression of 35S::AP1 with 35S::FPF1 leads to a synergistic effect on the shortening of the time to flowering under short-day conditions. The co-overexpression of 35S::FPF1 and 35S::LFY, however, shows only an additive reduction of flowering time and the conversion of nearly every shoot meristem, except the inflorescence meristem, to a floral meristem under the same light conditions. In addition, the constitutive expression of FPF1 attenuates the severe lfy-1 phenotype under short days and phenocopies to a great extent the lfy-1 mutant grown under long-day conditions. Thus, we assume that FPF1 modulates the competence to flowering of apical meristems.

Classification and phylogeny of the MADS-box multigene family suggest defined roles of MADS-box gene subfamilies in the morphological evolution of eukaryotes

The MADS-box encodes a novel type of DNA-binding domain found so far in a diverse group of transcription factors from yeast, animals, and seed plants. Here, our first aim was to evaluate the primary structure of the MADS-box. Compilation of the 107 currently available MADS-domain sequences resulted in a signature which can strictly discriminate between genes possessing or lacking a MADS-domain and allowed a classification of MADS-domain proteins into several distinct subfamilies. A comprehensive phylogenetic analysis of known eukaryotic MADS-box genes, which is the first comprising animal as well as fungal and plant homologs, showed that the vast majority of subfamily members appear on distinct subtrees of phylogenetic trees, suggesting that subfamilies represent monophyletic gene clades and providing the proposed classification scheme with a sound evolutionary basis. A reconstruction of the history of the MADS-box gene subfamilies based on the taxonomic distribution of contemporary subfamily members revealed that each subfamily comprises highly conserved putative orthologs and recent paralogs. Some subfamilies must be very old (1,000 MY or more), while others are more recent. In general, subfamily members tend to share highly similar sequences, expression patterns, and related functions. The defined species distribution, specific function, and strong evolutionary conservation of the members of most subfamilies suggest that the establishment of different subfamilies was followed by rapid fixation and was thus highly advantageous during eukaryotic evolution. These gene subfamilies may have been essential prerequisites for the establishment of several complex eukaryotic body structures, such as muscles in animals and certain reproductive structures in higher plants, and of some signal transduction pathways. Phylogenetic trees indicate that after establishment of different subfamilies, additional gene duplications led to a further increase in the number of MADS-box genes. However, several molecular mechanisms of MADS-box gene diversification were used to a quite different extent during animal and plant evolution. Known plant MADS-domain sequences diverged much faster than those of animals, and gene duplication and sequence diversification were extensively used for the creation of new genes during plant evolution, resulting in a relatively large number of interacting genes. In contrast, the available data on animal genes suggest that increase in gene number was only moderate in the lineage leading to mammals, but in the case of MEF2-like gene products, heterodimerization between different splice variants may have increased the combinatorial possibilities of interactions considerably. These observations demonstrate that in metazoan and plant evolution, increased combinatorial possibilities of MADS-box gene product interactions correlated with the evolution of increasingly complex body plans.

Classification and phylogeny of the MADS-box multigene family suggest defined roles of MADS-box gene subfamilies in the morphological evolution of eukaryotes

The MADS-box encodes a novel type of DNA-binding domain found so far in a diverse group of transcription factors from yeast, animals, and seed plants. Here, our first aim was to evaluate the primary structure of the MADS-box. Compilation of the 107 currently available MADS-domain sequences resulted in a signature which can strictly discriminate between genes possessing or lacking a MADS-domain and allowed a classification of MADS-domain proteins into several distinct subfamilies. A comprehensive phylogenetic analysis of known eukaryotic MADS-box genes, which is the first comprising animal as well as fungal and plant homologs, showed that the vast majority of subfamily members appear on distinct subtrees of phylogenetic trees, suggesting that subfamilies represent monophyletic gene clades and providing the proposed classification scheme with a sound evolutionary basis. A reconstruction of the history of the MADS-box gene subfamilies based on the taxonomic distribution of contemporary subfamily members revealed that each subfamily comprises highly conserved putative orthologs and recent paralogs. Some subfamilies must be very old (1,000 MY or more), while others are more recent. In general, subfamily members tend to share highly similar sequences, expression patterns, and related functions. The defined species distribution, specific function, and strong evolutionary conservation of the members of most subfamilies suggest that the establishment of different subfamilies was followed by rapid fixation and was thus highly advantageous during eukaryotic evolution. These gene subfamilies may have been essential prerequisites for the establishment of several complex eukaryotic body structures, such as muscles in animals and certain reproductive structures in higher plants, and of some signal transduction pathways. Phylogenetic trees indicate that after establishment of different subfamilies, additional gene duplications led to a further increase in the number of MADS-box genes. However, several molecular mechanisms of MADS-box gene diversification were used to a quite different extent during animal and plant evolution. Known plant MADS-domain sequences diverged much faster than those of animals, and gene duplication and sequence diversification were extensively used for the creation of new genes during plant evolution, resulting in a relatively large number of interacting genes. In contrast, the available data on animal genes suggest that increase in gene number was only moderate in the lineage leading to mammals, but in the case of MEF2-like gene products, heterodimerization between different splice variants may have increased the combinatorial possibilities of interactions considerably. These observations demonstrate that in metazoan and plant evolution, increased combinatorial possibilities of MADS-box gene product interactions correlated with the evolution of increasingly complex body plans.