We isolated and characterized two flower-specific genes from petunia. The protein products of these genes, designated floral binding protein 1 (FBP1) and 2 (FBP2), are putative transcription factors with the MADS box DNA binding domain. RNA gel blot analysis showed that the fbp1 gene is exclusively expressed in petals and stamen of petunia flowers. In contrast, the FBP1 protein was only detectable in petals and not in stamens, suggesting post-transcriptional regulation of the fbp1 gene in these tissues. The fbp2 gene is expressed in petals, stamen, carpels, and at a very low level in sepals but not in vegetative tissues. We analyzed the spatial expression of these fbp genes in floral organs of two homeotic flower mutants. In the blind mutant, whose flower limbs are transformed into antheroid structures on top of normal tubes, identical expression levels of both genes were observed in the antheroid structures as in normal anthers. In the homeotic mutant green petals, the petals are replaced by sepaloid organs in which the expression of fbp1 is strongly reduced but not completely abolished. Our results suggest a regulation of the fbp1 gene expression by the green petals (gp) gene. Expression of the fbp2 gene was not affected in the green petals mutant. In contrast to the proposed models describing floral morphogenesis, our data indicated that homeotic genes can be functional in one whorl only.

The petunia MADS box floral binding protein (fbp) gene 1 represents a class B homeotic gene determining the identity of second and third floral whorl organs. Suppression of fbp1, which is highly homologous to the Antirrhinum gene globosa and Arabidopsis gene pistillata, results in the conversion of petals to sepals and stamens to carpels. In contrast to fbp1, the petunia homeotic gene pMADS1, encoding a protein homologous to the Antirrhinum protein DEFICIENS, has been shown to be involved in the formation of petals only. We demonstrated that the induction of fbp1 is established independent of pMADS1, whereas at later developmental stages, fbp1 is up-regulated by pMADS1 in petals. On the other hand, the induction and maintenance of pMADS1 expression are not affected by fbp1. To obtain information about the functional interaction between fbp1 and pMADS1, an fbp1 cosuppression mutant with mild phenotypic alterations was crossed with a green petals mutant in which pMADS1 expression was abolished. Progeny plants, heterozygous for the pMADS1 gene, had flowers with a more pronounced reversion from petals into sepals than was observed for the parent fbp1 mutant. The morphology of the third whorl organs was not changed. These observations, together with expression levels of pMADS1 and fbp1 in mutant flowers, provide evidence for functional control of fbp1 by PMADS1 in vivo.

The petunia mutant green petal (gp, line PLV) shows a homeotic effect in one floral whorl, that is, the conversion of petal to sepal. We demonstrate that this mutant contains a chromosomal deletion, including the petunia MADS box gene pMADS1. Second whorl petal development in this null mutant can be restored with a CaMV 35S-pMADS1 transgene, demonstrating the essential role of pMADS1 in this process. Because gp (PLV) shows only a minor effect on stamen development, the homeotic effects of pMADS1 are different from those of B-type genes in Antirrhinum and Arabidopsis. Two other MADS box genes, pMADS2 and fbp1 (Angenent et al. 1992), require pMADS1 to maintain expression in the second whorl. However, in the absence of pMADS1 these two genes continue to be expressed in the third whorl. The functions assigned to pMADS1 are further supported by experiments in which we phenocopy gp by cosuppression of pMADS1 gene expression. The flowers, obtained through cosuppression and phenotype restoration, display different degrees of sepal to petal conversion. Analysis of these flowers indicate that pMADS1 controls growth under the zone of petal and stamen initiation, which causes the corolla tube and stamen filaments to emerge as a congenitally fused structure.

In both Antirrhinum (Antirrhinum majus) and Arabidopsis (Arabidopsis thaliana), the floral B-function, which specifies petal and stamen development, is embedded in a heterodimer consisting of one DEFICIENS (DEF)/APETALA3 (AP3)-like and one GLOBOSA (GLO)/PISTILLATA (PI)-like MADS box protein. Here, we demonstrate that gene duplications in both the DEF/AP3 and GLO/PI lineages in Petunia hybrida (petunia) have led to a functional diversification of their respective members, which is reflected by partner specificity and whorl-specific functions among these proteins. Previously, it has been shown that mutations in PhDEF (formerly known as GREEN PETALS) only affect petal development. We have isolated insertion alleles for PhGLO1 (FLORAL BINDING PROTEIN1) and PhGLO2 (PETUNIA MADS BOX GENE2) and demonstrate unique and redundant properties of PhDEF, PhGLO1, and PhGLO2. Besides a full homeotic conversion of petals to sepals and of stamens to carpels as observed in phglo1 phglo2 and phdef phglo2 flowers, we found that gene dosage effects for several mutant combinations cause qualitative and quantitative changes in whorl 2 and 3 meristem fate, and we show that the PHDEF/PHGLO1 heterodimer controls the fusion of the stamen filaments with the petal tube. Nevertheless, when the activity of PhDEF, PhGLO1, and PhGLO2 are considered jointly, they basically appear to function as DEF/GLO does in Antirrhinum and to a lesser extent as AP3/PI in Arabidopsis. By contrast, our data suggest that the function of the fourth B-class MADS box member, the paleoAP3-type PETUNIA HYBRIDA TM6 (PhTM6) gene, differs significantly from the known euAP3-type DEF/AP3-like proteins; PhTM6 is mainly expressed in the developing stamens and ovary of wild-type flowers, whereas its expression level is upregulated in whorls 1 and 2 of an A-function floral mutant; PhTM6 is most likely not involved in petal development. The latter is consistent with the hypothesis that the evolutionary origin of the higher eudicot petal structure coincided with the appearance of the euAP3-type MADS box genes.

Basal asterid families, and to a lesser extent the asterids as a whole, are characterized by a high variation in petal and stamen morphology. Moreover, the stamen number, the adnation of stamens to petals, and the degree of sympetaly vary considerably among basal asterid taxa. The B group genes, members of the APETALA3 (AP3) and PISTILLATA (PI) gene lineages, have been shown to specify petal and stamen identities in several core eudicot species. Duplicate genes in these lineages have been shown in some cases to have diversified in their function; for instance in Petunia, a PI paralog is required for the fusion of stamens to the corolla tube, illustrating that such genes belonging to this lineage are not just involved in specifying the identity of the stamens and petals but can also specify novel floral morphologies. This motivated us to study the duplication history of class B genes throughout asterid lineages, which comprise approximately one-third of all flowering plants. The evolutionary history of the PI gene subfamily indicates that the two genes in Petunia result from an ancient duplication event, coinciding with the origin of core asterids. A second duplication event occurred before the speciation of basal asterid Ericales families. These and other duplications in the PI lineage are not correlated with duplications in the AP3 lineage. To understand the molecular evolution of the Ericales PI genes after duplication, we have described their expression patterns using reverse transcription polymerase chain reaction and in situ hybridization, reconstructed how selection shaped their protein sequences and tested their protein interaction specificity with other class B proteins. We find that after duplication, PI paralogs have acquired multiple different expression patterns and negative selective pressure on their codons is relaxed, whereas substitutions in sites putatively involved in protein-protein interactions show positive selection, allowing for a change in the interaction behavior of the PI paralogs after duplication. Together, these observations suggest that the asterids have preferentially recruited PI duplicate genes to diverse and potentially novel roles in asterid flower development.