A main determinant of inflorescence architecture is the site where floral meristems are initiated. We show that in wild-type Petunia bifurcation of the inflorescence meristem yields two meristems of approximately equal size. One terminates into a floral meristem and the other maintains its inflorescence identity. By random transposon mutagenesis we have generated two mutants in which the architecture of the inflorescence is altered. In the extra petals (exp) mutant the inflorescence terminates with the formation of a single terminal flower. Phenotypic analysis showed that exp is required for the bifurcation of inflorescence meristems. In contrast, the aberrant leaf and flower (alf) mutant is affected in the specification of floral meristem identity while the branching pattern of the inflorescence remains unaltered. A weak alf allele was identified that, after bifurcation of the inflorescence meristem, yields a ‘floral’ meristem with partial inflorescence characteristics. By analysing independent transposon dTph1 insertion alleles we show that the alf locus encodes the Petunia FLORICAULA/LEAFY homolog. In situ hybridisation shows that alf is expressed in the floral meristem and also in the vegetative meristem. Differences and similarities between these Petunia mutants and mutations affecting inflorescence architecture in other species will be discussed.

In the plant kingdom, a wide variety of inflorescence structures exist that can be classified according to their overall architecture (Coen and Nugent, 1994). These different inflorescence architectures are a consequence of differences in the behaviour of meristematic cells. For example, Arabidopsis and Antirrhinum develop racemose inflorescences with a single main axis, where the apex remains meristematic and (multiple) flowers are generated from meristems appearing on the flanks of the apex. Other species, such as tulip, develop a determinate inflorescence where the complete apex transforms into a single terminal flower and meristematic activity is lost. In cymose inflorescences the apex also transforms into a terminal flower, but growth of the inflorescence continues from a new, secondary meristem that forms in the axil of the flower. Thus, multiple terminal flowers are generated on a single inflorescence. Species belonging to the Solanaceae, like Petunia, have been classified as cymose (Child, 1979; Weberling, 1989).

Floral meristem identity genes are likely to play a key role in determining inflorescence architecture, because their expression domain determines where flowers are formed. The two highly homologous genes floricaula (flo) and leafy (lfy) specify floral meristem identity in the racemose inflorescences of Antirrhinum and Arabidopsis, respectively (Coen et al., 1990; Weigel et al., 1992). Expression of flo and lfy is restricted to floral meristems that initiate on the flanks of the inflorescence meristem (Coen et al., 1990; Weigel et al., 1992). At later stages of flower development flo and lfy mRNAs still accumulate in floral organs. The observation that lfy encodes a nuclear protein that can bind DNA suggests that LFY is a transcription factor that activates a yet unknown set of target genes (Weigel, 1995). By analysing periclinal chimeras, that express flo in only one of the three meristematic layers, it was shown that flo acts in a cell non-autonomous manner (Carpenter and Coen, 1995; Hantke et al., 1995). The expression of the downstream floral organ identity genes, deficiens and plena is activated in all three cell layers in these flo chimeras. Whether this is the consequence of diffusion of the FLO protein itself, or via signalling by another diffusable factor remains to be resolved.

While flo and lfy are not expressed in vegetative meristems, the homologous genes from pea, tobacco and Impatiens are expressed in both floral meristems and leaf primordia (Hofer et al., 1997; Kelly et al., 1995; Poteau et al., 1997). The finding that NFL, the tobacco homolog of FLO/LFY is expressed at the site where leaf primordia are initiated suggests a more general role for flo/lfy and homologs from other species in specifying determinacy of lateral organs and meristems (Kelly et al., 1995). However, the function of nfl during development of the simple tobacco leaf remains unknown, because no mutant is available. In pea the unifoliata (uni) gene encodes the FLO/LFY homolog. Similar to the flo and lfy mutants of Antirrhinum and Arabidopsis, the flowers of uni mutants are converted into shoots. In addition, the compound pea leaf with tendrils is converted into a single lamina in uni mutants. This phenotype was explained by assuming a role for uni in maintaining a transient phase of indeterminacy of lateral derivatives formed by the meristem (Hofer et al., 1997). Therefore, the function of flo/lfy and their homologous genes in other plant species seems distinct from direct determination of floral meristem identity.

Modification of the expression pattern of lfy can change an indeterminate inflorescence structure into a determinate floral meristem (Weigel and Nilsson, 1995). The constitutive expression of lfy in Arabidopsis leads to precocious formation of terminal flowers. Transformation of the same lfy construct in aspen leads to flowering within five months while wild-type aspen flowers after 8-20 years. Also in the centroradialus (cen) mutant of Antirrhinum and the terminal flower (tfl) mutant of Arabidopsis ectopic expression of lfy and flo in the apex leads to the formation of a terminal flower (Bradley et al., 1996, 1997). This shows that the activity of LFY or FLO is sufficient to convert the indeterminate inflorescence into a determinate floral meristem.

We have started the genetic analysis of cymose inflorescence development in Petunia. By random transposon mutagenesis we have isolated mutants in which inflorescence development is altered. Among those, mutants with terminal flowers, altered inflorescence or floral meristem organisation and aberrant meristem identity were found. Here we present the characterisation of two of those mutants. In the extra petals (exp) mutant the inflorescence terminates by the formation of a single flower. We show that exp is required for bifurcation of the inflorescence meristem. In the aberrant leaf and flower (alf) mutant floral meristem identity is altered. The alf mutant has phenotypic characteristics highly similar to those of flo and lfy mutants. We describe the cloning and molecular characterisation of the Petunia homolog of flo/lfy and show that it is present at the alf locus. Like the pea gene uni, alf is expressed in both the floral meristem and the vegetative meristem, but no alterations are observed during vegetative development of alf mutants.

Plant material

The alf-S3018 (Doodeman et al., 1984) and exp-W2115 (this work) mutants arose spontaneously in the progeny of the Petunia line W138. The alf-G5509 and alf-T2009 alleles were derived from a plant heterozygous for the alf-S3018 and a wild-type allele. The alf+-X2011 revertant allele was isolated by cross-fertilisation of flowers on a revertant branch of an alf-S3018/alf-G5509 plant. The alf-W2167 dTph1 insertion allele was identified among 2000 Petunia plants by PCR with primers flo1 (5’-GGAATTCCCCATGGACCCAGAGGCTTTC-3’) and the out1 primer (5’-dGGGAATTCGCTCCGCCCCTG-3’) complementary to the terminal inverted repeat of dTph1 as described previously (Koes et al., 1995). The alf-X2586 mutant arose spontaneously among W138 progeny (Angenent et al., personal communication). Allelism was confirmed by crossing heterozygous plants harbouring different alf alleles.

To obtain the alf/exp double mutant alf-G5509/+ plants were crossed to exp-W2115/exp-W2115 plants. Six of the F1 plants were self-pollinated to obtain F2 progenies and four of these segregated for alf mutants. Because the exp-W2115 is unstable, presumably transposon tagged, five out of six progenies segregated for exp mutants. The alf/exp double mutant phenotype was only seen in those Fr2 families segregating for both exp and alf.

Scanning electron microscopy

SEM was performed as described previously (Souer et al., 1996).

Isolation and sequence analysis of the alf cDNA and genomic fragments

The flop cDNA was isolated by screening a Petunia W115 petal cDNA library with the leafy cDNA under low stringency conditions. Fragments of this clone were subcloned in pBluescript-SK plasmids (Stratagene) and sequenced.

For PCR amplification of insertion and footprint alleles the following primers were used:

flo1, 5’-GGAATTCCCCATGGACCCAGAGGCTTTC-3’

flo3, 5’-GGAATTCTGAGGAACCAGTGCAGCAG-3’

flo4, 5’-CGGGATCCTTAGTAGGGCATTTTTCACCAC-3’

flo5, 5’-CGGGATCCGGAGCTTTGGTGGGCACATAC-3’

flo6, 5’-GCTCTAGATGAACAATGCAGGGATTTC-3’

Amplification products were cloned in pBluescript-KS or pGEM plasmid vectors and sequenced. Sequencing was performed using asymmetric PCR with fluorescent M13 primers employing an Applied Biosystems DNA sequencer model 370A.

Expression analysis

Total RNA extractions and RT-PCR reactions were performed as described previously (Souer et al., 1996), except that the primers used were oligo d(T)17 for first strand synthesis and flo3 and flo5 for PCR amplification. To obtain a quantitative response, alf transcripts were amplified in 25 PCR cycles, while for gapdh transcripts only 18 cycles were used.

In situ hybridisations were performed as described by Cañas et al. (1994). An alf-specific digoxigenin-labelled RNA probe was obtained by T7 polymerase-driven in vitro transcription from the full length cDNA in pBluescript-SK according to the instructions of Boehringer Mannheim. RNA transcripts were partly hydrolysed for 45 minutes in 60 mM Na2CO3, 40 mM NaHCO3.

Development of the wild-type Petunia inflorescence

Members of the Solanaceae, like Petunia are classified as having cymose inflorescences that terminate in a flower. Growth continues from a sympodial meristem in the axis of this flower (Child, 1979; Napoli and Ruehle, 1996; Weberling, 1989). Development of the Petunia inflorescence is very similar in diverse inbred lines, although there are slight differences in the extent of axillary meristem dormancy and internode elongation. The Petunia wild-type inflorescence contains two leaf-like structures at each node, termed bracts (Figs 1 and 2B). At each node one flower develops in the axil of one bract. In the axil of the other bract, an inflorescence shoot develops that will reiterate this branching pattern (Figs 1 and 2A,B). When the flower and the inflorescence have grown out, axillary meristems develop in the axils of both bracts. These meristems are initially vegetative, forming a few leaves, before they transform into an inflorescence meristem that develops in a similar manner as the main inflorescence (Figs 1 and 2B). The extent of outgrowth of these axillary meristems depends on their position in the plant. At the acropetal nodes (near the apex) both meristems remain in a dormant state. At the most basipetal nodes (more distal from the apex) the axillary meristem in the axil of flower and bract will sprout, while the one in the axil of bract and inflorescence usually remains in a dormant state, unless the apex has been removed.

Fig. 1.

Diagram showing a wild-type Petunia plant. The position of flowers is shown by closed circles, and the apical inflorescence meristem by a closed triangle. Leaves and bracts are indicated by large and small grey ovals respectively. Vegetative axillary meristems are shown by open triangles; smaller size indicates stronger dormancy.

Fig. 1.

Diagram showing a wild-type Petunia plant. The position of flowers is shown by closed circles, and the apical inflorescence meristem by a closed triangle. Leaves and bracts are indicated by large and small grey ovals respectively. Vegetative axillary meristems are shown by open triangles; smaller size indicates stronger dormancy.

Fig. 2.

Phenotype of wild-type, alf, exp and exp/alfplants. (A) Wild-type (left) and alf-G5509 (right) plants. (B) Wild-type inflorescence. All flowers are subtended by bracts, but some have been removed to show the branching pattern. (C) alf-G5509 inflorescence. Some bracts have been removed to show the branching pattern. The arrow indicates the first bifurcation of the inflorescence. (D) Inflorescence of a plant carrying the weak alf-T2009 allele. Some bracts have been removed. (E) Detail of floral structures found at inflorescence branches of alf-T2009 homozygous plants. (F) Wild-type Petunia flower. (G) exp flower. The flower has six instead of five petals. (H) Wild-type (left) and exp (right) plants. The exp inflorescence has terminated in a single flower. Note that the flower lacks a pedicel. (I) exp/alf (left) and alf (right) plants. The exp/alf inflorescence is unbranched but indeterminate. Some leaves have been removed. ax, axillary branches; br, bract; cal, carpeloid organs; if, inflorescence shoot; le, leaf; pet, petaloid organ.

Fig. 2.

Phenotype of wild-type, alf, exp and exp/alfplants. (A) Wild-type (left) and alf-G5509 (right) plants. (B) Wild-type inflorescence. All flowers are subtended by bracts, but some have been removed to show the branching pattern. (C) alf-G5509 inflorescence. Some bracts have been removed to show the branching pattern. The arrow indicates the first bifurcation of the inflorescence. (D) Inflorescence of a plant carrying the weak alf-T2009 allele. Some bracts have been removed. (E) Detail of floral structures found at inflorescence branches of alf-T2009 homozygous plants. (F) Wild-type Petunia flower. (G) exp flower. The flower has six instead of five petals. (H) Wild-type (left) and exp (right) plants. The exp inflorescence has terminated in a single flower. Note that the flower lacks a pedicel. (I) exp/alf (left) and alf (right) plants. The exp/alf inflorescence is unbranched but indeterminate. Some leaves have been removed. ax, axillary branches; br, bract; cal, carpeloid organs; if, inflorescence shoot; le, leaf; pet, petaloid organ.

To better understand the formation of floral meristems we analysed the wild-type inflorescence by scanning electron microscopy (SEM). The inflorescence meristem simultaneously generates two bracts before a bifurcation of the central dome yields two halves (Fig. 3A,B). One half develops as a determinate floral meristem that soon after the bifurcation starts to generate sepals, the first floral organs. The other half remains meristematic and will continue with a new division, perpendicular to the last division, to form two new bracts and a new floral meristem. Therefore, the Petunia inflorescence does not completely meet the definition of cymose development as the inflorescence meristem does not terminate, but rather splits into two new meristems of which one is the determinate floral meristem. For this reason, we prefer to use the term inflorescence meristem instead of sympodial meristem since the meristem is formed by bifurcation and does not arise as an axillary structure. During the early stages of flower development shown in Fig. 3A,B, the axillary meristem that develops in the axils of older flowers is not visible yet, but will appear later.

Fig. 3.

SEM analysis of wild-type, alf, exp and exp/alf inflorescences.(A) Wild-type inflorescence. The inflorescence has generated a floral meristem (right) on which the first three sepal primordia have arisen. On the inflorescence meristem (left) the bract primordia have just initiated perpendicular to the bracts formed on an earlier node. (B) Wild-type inflorescence slightly later than in A. Bifurcation of the inflorescence has occurred. The floral meristem has initiated all floral organ primordia. Sepals that enclose the oldest flower at this stage were removed. (C) alf-G5509 inflorescence. The bifurcation of the alf inflorescence meristem is similar to that in wild-type inflorescences. However, both meristems behave as inflorescence meristems as shown by the repetitive formation of bracts and new bifurcations of each meristem. (D) alf-T2009 inflorescence. After bifurcation of the inflorescence meristem the ‘floral’ meristem continues to form bracts but secondary bifurcation does not occur. This meristem soon terminates with the development of carpels (right). (E) Young exp flower. Extra petal and stamen organs develop on the floral meristem (compare with wild-type flower in panel B) Some sepals were removed. (F) exp flower slightly later in development. All sepals have been removed. (G) Side-view of an exp branch. A single terminal flower has developed at the apex. The young meristem on the right is an axillary meristem that has developed in a leaf axil. (H) Side-view of an exp/alf inflorescence. The inflorescence meristem keeps developing bract primordia but does not bifurcate. In the axils of older bracts, axillary meristems have developed. ax, meristem in axil of bract; axl, meristem in axil of leaf; br, bract; ca, carpel; cal, carpel-like; fm, floral meristem; im, inflorescence meristem; le, leaf; pe, petal; se, sepal; st, stamen. The scale bar equals 100 µm in all panels.

Fig. 3.

SEM analysis of wild-type, alf, exp and exp/alf inflorescences.(A) Wild-type inflorescence. The inflorescence has generated a floral meristem (right) on which the first three sepal primordia have arisen. On the inflorescence meristem (left) the bract primordia have just initiated perpendicular to the bracts formed on an earlier node. (B) Wild-type inflorescence slightly later than in A. Bifurcation of the inflorescence has occurred. The floral meristem has initiated all floral organ primordia. Sepals that enclose the oldest flower at this stage were removed. (C) alf-G5509 inflorescence. The bifurcation of the alf inflorescence meristem is similar to that in wild-type inflorescences. However, both meristems behave as inflorescence meristems as shown by the repetitive formation of bracts and new bifurcations of each meristem. (D) alf-T2009 inflorescence. After bifurcation of the inflorescence meristem the ‘floral’ meristem continues to form bracts but secondary bifurcation does not occur. This meristem soon terminates with the development of carpels (right). (E) Young exp flower. Extra petal and stamen organs develop on the floral meristem (compare with wild-type flower in panel B) Some sepals were removed. (F) exp flower slightly later in development. All sepals have been removed. (G) Side-view of an exp branch. A single terminal flower has developed at the apex. The young meristem on the right is an axillary meristem that has developed in a leaf axil. (H) Side-view of an exp/alf inflorescence. The inflorescence meristem keeps developing bract primordia but does not bifurcate. In the axils of older bracts, axillary meristems have developed. ax, meristem in axil of bract; axl, meristem in axil of leaf; br, bract; ca, carpel; cal, carpel-like; fm, floral meristem; im, inflorescence meristem; le, leaf; pe, petal; se, sepal; st, stamen. The scale bar equals 100 µm in all panels.

The alf gene is required for floral meristem identity

Among progeny of the Petunia hybrida line W138, an unstable mutant was identified that exhibited a bushy phenotype and only rarely generated heavily malformed floral organs (Fig. 2A, right; Doodeman et al., 1984; Gerats et al., 1988; Gerats, 1991). Because of the proliferation of green leaf-like structures the mutant was called aberrant leaf and flower (alf). Contrary to what the name suggests, we could not detect differences between alf and wild-type plants during their vegetative phase. The first moment that differences become evident is after transition of the vegetative shoot meristem to an inflorescence meristem. While the wild-type inflorescence generates bracts and flowers (Fig. 2B), the alf inflorescence solely forms branches which develop bracts (Fig. 2C). After having formed several branches subtended by bracts, alf inflorescences occasionally terminate by the formation of floral organs, most often carpels. The formation of axillary meristems in the axils of bracts subtending inflorescence branches appears normal in alf plants.

To further examine these abnormalities we analysed alf−?inflorescences by SEM. Fig. 3C shows that after the formation of bracts, a bifurcation of the alf inflorescence meristem yields two meristems, which is similar to the wild-type inflorescence at this stage of development. However, in contrast to wild-type meristems, both meristems behave as inflorescence meristems as they will generate bracts on their flanks and bifurcate again to form new inflorescence meristems (Fig. 3C). Occasionally, alf inflorescence meristems repeat the formation of bracts without secondary bifurcations (e.g. Fig. 3C, left meristem). These observations show that the alf mutant is affected in the transition from inflorescence meristem identity to floral meristem identity.

One progeny obtained by self-fertilisation of a +/alf-S3018 plant segregated for mutants with a weaker alf phenotype (alf-T2009, Fig. 2D). Upon crossing, alf-T2009 and alf-G5509 indeed appeared to be allelic (not shown). All alf-T2009 inflorescences generate meristems which terminate in floral structures. On the periphery these flowers first develop a number of bracts or sepals after which a central carpel is generated that is usually malformed. In between this central carpel and bracts, chimeric organs develop which can contain sepal, petal, stamen and carpel tissue (Fig. 2E). SEM analysis revealed that after bifurcation of the alf-T2009 inflorescence meristem, the floral meristem has partial characteristics of an inflorescence. This meristem repeats the formation of bracts as inferred from their position on the meristem (Fig. 3D). In contrast to the situation in alf-G5509 inflorescences, in alf-T2009 inflorescences no further bifurcation occurs. After having generated several bracts, the central portion of the meristem develops carpels as can be seen by the typical donut structure of the two carpel primordia (Fig. 3D). Therefore, in the alf-T2009 mutant the initial specification of floral meristem identity is blocked as inferred by the repeated formation of bract primordia, but the meristem acquires in time more floral characteristics causing it to terminate with the formation of carpels. This phenotype, together with the observation that in the allelism test alf-T2009/alf-G5509 heterozygous plants could be recognised as having an intermediate phenotype, shows that alf-T2009 is a weak alf allele.

extra petals is required for bifurcation of the inflorescence meristem

In a random transposon-mutagenesis experiment we found another mutant with an altered inflorescence. This mutant was initially identified for its increased number of petals and, therefore, named extra petals (exp). While wild-type Petunia flowers have five petals (Fig. 2H), exp flowers contain six full-grown petals (Fig. 2G). Often an additional one or two petals are recognisable that do contain a central vein, but remain very narrow, possibly due to mechanical constraints. About half of the flowers on exp plants contain six mature anthers, while the other half contains five anthers similar to the situation in wild-type flowers. The number of sepals and carpels in exp flowers was similar to the numbers found in wild-type plants. SEM analysis of young exp flowers shows that they contain seven to eight organ primordia in whorl 2 and six to eight primordia in whorl 3, while the number of sepal and carpel primordia in whorls 1 and 4 is similar to wild-type flowers (compare Fig. 3E,F with 3B). To examine if the effect of exp on floral organ number is dependent on their identity, we generated the double mutant with the floral organ identity gene green petals (gp). The gp locus encodes a MADS-box protein and is the ortholog of deficiens from Antirrhinum and apetela-3 of Arabidopsis (van der Krol et al., 1993). Loss of gp function results in the transformation of petals into sepal-like organs. exp/gp flowers consisted of five sepals in whorl 1, seven to eight sepal-like organs in whorl 2, five to six anthers in whorl 3 and two carpels in whorl 4 (not shown), indicating that the exp mutation increases the number of floral organs in whorl 2 regardless of their identity.

Closer examination of exp plants showed that their inflorescence had an unusual architecture. The exp inflorescence (i.e. the structure between the first two bracts) consists of a single terminal flower that almost completely lacks the pedicel (Fig. 2H). Apparently, the formation of this terminal flower is due to the complete transformation of the apical inflorescence meristem into a floral meristem, as (1) no remains of the inflorescence meristem is detectable after this transition and (2) the flower is positioned apically (Fig. 3G). This is consistent with the observation that exp plants loose their apical dominance once a terminal flower is generated, similar to wild-type plants from which the inflorescence apex is manually removed. As a consequence, the dormancy of the vegetative meristems in the axils of leaves is broken and they will generate a series of leaves before they terminate with the formation of a single flower. Thus, exp plants generate multiple inflorescences that each consist of a single flower. To further define the role of exp in inflorescence development we determined the phenotype of the exp/alf double mutant. Fig. 2I shows that exp/alf plants have, like alf mutants, an indeterminate inflorescence that only contains bracts and completely lacks flowers. In addition, the exp/alf double mutant has lost the branching pattern that is typical of alf inflorescences, as it consists of a single branch bearing bracts (compare the two plants in Fig. 2I or Fig. 3H and 3C). In the axils of the older bracts vegetative meristems appear, similar to the situation in wild-type plants.

Taken together these data show that exp and alf function in two distinct processes. exp is required for bifurcation of the inflorescence meristem into two new meristems, but does not seem to have an effect on the identity of these meristems. On the other hand, alf is required to determine the floral identity for one of the two meristems, but does not seem to affect the initiation of new meristems.

Isolation of the flo/lfy homolog of Petunia

The phenotypic defects seen in alf mutant plants resemble those in floricaula and leafy mutants (Coen et al., 1990; Weigel et al., 1992) suggesting that the alf locus may contain the Petunia homolog of lfy and flo. To test this, we isolated the flo/leafy homolog of Petunia (flop, floricaula/leafy ortholog Petunia). The flop cDNA was isolated by screening a petal cDNA library of the Petunia hybrida line W115 with the full length lfy cDNA under low stringency conditions.

By Southern blot analysis under low stringency conditions only a single band could be detected after hybridisation with the flop cDNA (data not shown). Furthermore, upon screening a Petunia inflorescence cDNA library with the flop cDNA, only cDNAs derived from identical transcripts were isolated. Therefore, we presume that the flop gene is single copy in Petunia.

The flop cDNA fragment is 1412 bp long, potentially encoding a protein of 412 amino acids (Fig. 4). The deduced amino acid sequence of the flop gene has 82% identity with FLO, 72% with UNI, 66% with LFY, and 65% with BOFH, the FLO/LFY homolog of Brassica oleracea. FLOP is most homologous to the tobacco NFL1 protein sharing 93% identity (Fig. 5). By PCR and sequence analysis we determined that the positions of the two introns in the flop gene are identical with those in flo, lfy and nfl (Coen et al., 1990; Kelly et al., 1995; Weigel et al., 1992), consistent with the view that they are orthologs.

Fig. 4.

mRNA sequence and structure of the alf gene, the flo/lfy homolog of Petunia. (A) Genomic map of the alf gene. Thick bars represent exon sequences. The position of dTph1 insertions is indicated by open triangles. The arrow indicates the direction of transcription. (B) The deduced amino acid sequence is shown below the nucleotide sequence. The target site duplication sequence of dTph1 insertion alleles is underlined. The position of the introns is indicated by two slashes (\/) after nucleotides 470 and 874, respectively. The GenBank accesion number for the alf cDNA is AF030171. (C) Sequences of footprint alleles created after excision of dTph1 from the insertion allele alf-S3018. The target site duplication is underlined.

Fig. 4.

mRNA sequence and structure of the alf gene, the flo/lfy homolog of Petunia. (A) Genomic map of the alf gene. Thick bars represent exon sequences. The position of dTph1 insertions is indicated by open triangles. The arrow indicates the direction of transcription. (B) The deduced amino acid sequence is shown below the nucleotide sequence. The target site duplication sequence of dTph1 insertion alleles is underlined. The position of the introns is indicated by two slashes (\/) after nucleotides 470 and 874, respectively. The GenBank accesion number for the alf cDNA is AF030171. (C) Sequences of footprint alleles created after excision of dTph1 from the insertion allele alf-S3018. The target site duplication is underlined.

Fig. 5.

Alignment of the deduced amino acid sequence of alf with NFL (Kelly et al., 1995), FLO (Coen et al., 1990), UNI (Hofer et al., 1997), LFY (Weigel et al., 1992) and BOFH (Anthony et al., 1996). Amino acids that are identical in all six sequences are shown in black boxes. In addition, amino acids identical to ALF are shown in grey boxes to reveal the higher similarity of ALF with NFL.

Fig. 5.

Alignment of the deduced amino acid sequence of alf with NFL (Kelly et al., 1995), FLO (Coen et al., 1990), UNI (Hofer et al., 1997), LFY (Weigel et al., 1992) and BOFH (Anthony et al., 1996). Amino acids that are identical in all six sequences are shown in black boxes. In addition, amino acids identical to ALF are shown in grey boxes to reveal the higher similarity of ALF with NFL.

The flop cDNA is transcribed from the alf locus

To determine if the alf phenotype was caused by mutation of flop, we performed a PCR reaction with flop specific primers on genomic DNA isolated from a family of plants segregating for the mutable alf-S3018 allele. Compared to wild-type, the amplification products of alf-S3018 contained an insertion of approximately 300 bp. Sequence analysis of this fragment showed that a dTph1 element was inserted in the protein coding sequence, 1 bp upstream the exon2/intron2 boundary at bp 873 (Fig. 4).

To prove that the insertion of this dTph1 element caused the alf phenotype, we screened a Petunia population by PCR for independent insertions in the flop gene (Koes et al., 1995). Among 2000 plants we found one plant carrying a dTph1 insertion in the first exon of flop (alf-W2167, Fig. 4A,B). Selfing of this +/alf-W2167 plant yielded progeny that segregated 3:1 for wild-type and alf phenotypes. A third independent mutant, alf-X2586, was found among W138 progeny (Angenent et al., personal communication). By PCR and sequence analysis we identified a dTph1 element in the second exon of flop (Fig. 4A,B). Crosses of heterozygotes harbouring different alf alleles, segregated 3:1 for wild-type and alf mutants, confirming that the mutants represent alleles of one locus.

The finding that three independent transposon insertions in flop all cause alf phenotypes showed that flop originated from the alf locus. For this reason we will from now on refer to the flop gene as alf. Because the three alf insertion alleles show identical phenotypic alterations we presume that they are all null alleles.

Sequence analysis of alf alleles

To determine the consequence of sequence alterations in the alf gene we sequenced a number of excision alleles originating from alf-S3018. A revertant allele, alf+-X2011, was obtained by cross-fertilisation of (nearly) wild-type flowers that formed on a revertant branch of an alf-S3018/alf-G5509 heterozygote. The resulting progeny segregated 3:1 for wild-type and alf phenotypes indicating that the reversion had occurred in the L2 layer of the parental plant. PCR and sequence analysis of genomic DNA of an alf+-X2011 homozygote showed that the dTph1 element had perfectly excised from the alf gene (Fig. 4C). Sequence analysis of the weak alf-T2009 allele revealed that dTph1 excision had created a 3 bp footprint that leaves the reading frame intact. An extra threonine residue is now present at position 287, apparently resulting in a protein with partial activity (Fig. 4C). One of the selfings derived from a plant heterozygous for alf-S3018 and a wild-type allele yielded a family with alf plants that had lost the dTph1 element in the alf gene as determined by PCR. Sequencing of this new alf-G5509 allele showed that the footprint left by the transposable element shifted the reading frame thereby creating a stop codon at position 290, two amino acids after the original dTph1 insertion site (Fig. 4C). This shows that small changes around amino acid 288 at least partially influence the activity of alf and that the C-terminal part after amino acid 288 is required for alf activity.

Expression pattern of alf

To establish the expression domain of alf, we performed RT-PCR on total RNA extracted from various plant parts. We detected alf transcripts in young flowers, vegetative apices and inflorescence apices (Fig. 6, lane 1, 3 and 5). No expression was found in roots, while only low expression could be detected in leaves. Therefore, alf is expressed in both the inflorescence and the vegetative meristem.

Fig. 6.

RT-PCR analysis of alf RNA in different tissues and different genotypes. (Top) PCR amplification with alf specific primers. (Bottom) PCR amplification with gapdh specific primers. Lane 1, floral buds 1 cm.; lane 2, roots; lane 3, vegetative apices; lane 4, leaves; lane 5, wild-type inflorescences; lane 6, alf-T2009 inflorescences; lane 7, alf-W2167/alf-W2167 inflorescences; lane 8, alf-W2167/Alfr+ inflorescences; lane 9, Alfr+/Alfr+ inflorescences; lane 10, alf-G5509/alf-G5509 inflorescences; lane 11, ten-fold dilution of the RNA used for lane 9; lane 12, 100-fold dilution of the RNA used for lane 9.

Fig. 6.

RT-PCR analysis of alf RNA in different tissues and different genotypes. (Top) PCR amplification with alf specific primers. (Bottom) PCR amplification with gapdh specific primers. Lane 1, floral buds 1 cm.; lane 2, roots; lane 3, vegetative apices; lane 4, leaves; lane 5, wild-type inflorescences; lane 6, alf-T2009 inflorescences; lane 7, alf-W2167/alf-W2167 inflorescences; lane 8, alf-W2167/Alfr+ inflorescences; lane 9, Alfr+/Alfr+ inflorescences; lane 10, alf-G5509/alf-G5509 inflorescences; lane 11, ten-fold dilution of the RNA used for lane 9; lane 12, 100-fold dilution of the RNA used for lane 9.

To determine the spatiotemporal expression pattern of alf, we performed in situ hybridisations at various stages of development (Fig. 7). By analysing serial sections of vegetative meristems we detected alf expression in a ring of cells, a few cells wide, around the central zone (Fig. 7A). This expression domain coincides with the places where leaf primordia are initiated. alf transcripts remain present in young leaf primordia, but disappear when the leaves grow older.

Fig. 7.

In situ localisation of alf mRNA in vegetative meristems and inflorescences of a wild-type Petunia. (A) Young Petunia plantlet. Expression in the vegetative meristem is detected at the site where leaf primordia will appear. (B) Inflorescence meristem. A sector of alf-expressing cells marks the site where the floral meristem will be developed. Strong expression is also detected in the bract primordium (left). The structure overlying the apex is a bract from an earlier node. (C) Young floral meristem. Sepal primordia are formed in which alf is expressed. The zone on the meristem were later petal and stamen primordia will arise also expresses alf. An inflorescence meristem without alf expression is seen on the right in the axil of a bract. (D) Young flower. alf expression is detected in all floral organ primordia. Note that the stamen on the right lacks detectable alf expression. (E) Part of an older flower. alf expression is detected in the developing stigma, the carpel wall and the bottom of the placenta. In petals alf expression can still be detected. an, anther; br, bract; ca, carpel; fm, floral meristem; im, inflorescence meristem; le, leaf; pe, petal; pl, placenta; sam, shoot apical meristem; se, sepal; st, stamen. The scale bar equals 100 µm in all panels.

Fig. 7.

In situ localisation of alf mRNA in vegetative meristems and inflorescences of a wild-type Petunia. (A) Young Petunia plantlet. Expression in the vegetative meristem is detected at the site where leaf primordia will appear. (B) Inflorescence meristem. A sector of alf-expressing cells marks the site where the floral meristem will be developed. Strong expression is also detected in the bract primordium (left). The structure overlying the apex is a bract from an earlier node. (C) Young floral meristem. Sepal primordia are formed in which alf is expressed. The zone on the meristem were later petal and stamen primordia will arise also expresses alf. An inflorescence meristem without alf expression is seen on the right in the axil of a bract. (D) Young flower. alf expression is detected in all floral organ primordia. Note that the stamen on the right lacks detectable alf expression. (E) Part of an older flower. alf expression is detected in the developing stigma, the carpel wall and the bottom of the placenta. In petals alf expression can still be detected. an, anther; br, bract; ca, carpel; fm, floral meristem; im, inflorescence meristem; le, leaf; pe, petal; pl, placenta; sam, shoot apical meristem; se, sepal; st, stamen. The scale bar equals 100 µm in all panels.

In the inflorescence meristem, alf expression is restricted to the site where bracts and the floral meristem will arise (Fig. 7B). The region of the inflorescence meristem that remains meristematic is seen as a zone that lacks alf expression. Slightly later in development alf is expressed inside the bracts and at the peripheral zone of the floral meristem, where sepals will form (not shown). When sepals are formed the expression domain moves towards the centre of the floral meristem to the sites where petal and stamen primordia will arise, but is absent from the centre of the floral meristem (Fig. 7C). Later, signals were detected mainly in petal, stamen and carpel primordia (Fig. 7D). The expression in stamens seems unequally distributed, as some sections lacked signal while in subsequent sections expression could be detected. Around stage five of flower development (Angenent et al., 1995), when the gynoecium is fused at the top and the style and stigma start to develop, alf signal is apparent in petals and carpels (Fig. 7E). In addition, also on the bottom of the placenta strong alf signals can be detected.

In conclusion, alf is transcribed in the inflorescence at the sites where floral meristems and organ primordia are formed and in vegetative meristems at the sites where leaf primordia are initiated.

Transcription of alf is not auto-regulated

In Antirrhinum, FLO was reported to (auto)regulate its own expression (Carpenter et al., 1995). To determine if a similar control exists in Petunia, we analysed the expression of alf by RT-PCR in plants harbouring different alf alleles. The levels of alf transcripts in inflorescences homozygous for the weak alf-T2009 allele, the unstable alf-W2167 allele and the stable recessive alf-G5509 allele are similar to those in homozygous (Alf+/Alf+) or heterozygous (Alf+/alf-W2167) wild-type inflorescences (Fig. 6). This indicates that ALF does not control the accumulation of its own mRNA.

Development of the wild-type Petunia inflorescence

A remarkable transition observed during plant development is the shift from the vegetative to the reproductive phase. This shift is characterised by the development of an inflorescence meristem that generates floral meristems. In racemose inflorescences these floral meristems axillary while in cymose inflorescences the floral meristem develops in a terminal position. In cymes, new (sympodial) meristems are thought to form in the axils of these flowers leading to the typical zig-zag branching pattern of cymose inflorescences. Our results show that this is, at least in Petunia, an incorrect view. In Petunia, floral meristem development involves a bifurcation, not a termination of the inflorescence meristem, yielding two meristems of about the same size. One of these meristems will grow out to form a flower. This pattern by which floral meristems are initiated is highly similar to flower development in tomato, where a bifurcation of inflorescence meristems also occurs (Allen and Sussex, 1996). In pea, flower formation also involves bifurcation of the inflorescence meristem (Tucker, 1989). In Fig. 8, the pattern by which the wild-type Petunia inflorescence develops is drawn schematically as a reiterative cycle.

Fig. 8.

Model explaining the action of alf and exp during inflorescence development in Petunia. The wild-type situation runs through the largest cycle, with the inflorescence meristem (blue) repeating the top to bottom pattern. The inflorescence meristem (blue) develops two bracts (green). exp causes the meristem to split in two halves. The action of alf is restricted to one of these meristems causing it to acquire a determinate floral meristem identity (red). In the absence of exp, bifurcation is omitted, but alf converts the meristem into a single terminal flower. In the absence of alf, bifurcation of the inflorescence meristem still occurs but floral meristem identity is not acquired. When both alf and exp are inactive and both the bifurcation and floral meristem identities are lost, the inflorescence keeps forming bracts without any branching.

Fig. 8.

Model explaining the action of alf and exp during inflorescence development in Petunia. The wild-type situation runs through the largest cycle, with the inflorescence meristem (blue) repeating the top to bottom pattern. The inflorescence meristem (blue) develops two bracts (green). exp causes the meristem to split in two halves. The action of alf is restricted to one of these meristems causing it to acquire a determinate floral meristem identity (red). In the absence of exp, bifurcation is omitted, but alf converts the meristem into a single terminal flower. In the absence of alf, bifurcation of the inflorescence meristem still occurs but floral meristem identity is not acquired. When both alf and exp are inactive and both the bifurcation and floral meristem identities are lost, the inflorescence keeps forming bracts without any branching.

In this paper, we described two genes each of which is required for one of the two major processes that determine the final architecture of the Petunia inflorescence: exp, which is required for bifurcation of the inflorescence meristem, and alf, which is required for floral meristem identity.

alf determines floral meristem identity

By in situ hybridisation we show that the expression of alf in the inflorescence marks the formation of a floral meristem, before the bifurcation of the apex makes it anatomically visible. At this early stage, the zone of the inflorescence that expresses alf comprises a bit more than half of the meristem (Fig. 7B). The bifurcation itself is independent of alf, since it also occurs in alf mutants, thereby maintaining the branching pattern. Another marker for bifurcation of the Petunia inflorescence is the no apical meristem gene (nam) that is expressed in a stripe of cells that separate the inflorescence and floral meristem (Souer et al., 1996). Also in this case the floral meristem appears as a slightly larger zone than the inflorescence meristem. However, when the bifurcation of the inflorescence can be seen microscopically, the two new meristems appear approximately equal in size (Fig. 3A,B). This suggests that the border cells of the expression domain of alf and nam in the inflorescence are not incorporated in the floral meristem or that the inflorescence cells divide faster at this stage of development.

Although the Petunia inflorescence does not meet the definition for cymose development, it also does not develop a straight main axis as in racemose inflorescences (Figs. 1 and 2B). The structure of repetitive bifurcations that makes up the Petunia inflorescence is best seen in alf mutants. Because in wild-type plants the vigorous growth of the inflorescence pushes the flower aside, only a weak zig-zag pattern remains. Expression of lfy and flo in the racemose inflorescences of Arabidopsis and Antirrhinum occurs on the flanks of the inflorescence dome, while alf is expressed in the terminal part of the inflorescence dome (Fig. 7B). Thus, the expression pattern of these three homologs reflects the different behaviour of inflorescence apices that results in different inflorescence architectures. Antirrhinum mutants harbouring the flo-640 frame shift allele contain little or no flo transcripts. Carpenter et al. (1995) suggested that flo may be autoregulated. However, alf, like lfy (Weigel et al., 1992), does not appear to be autoregulated since the alf transcript levels are normal in alf plants. (Fig. 6). Therefore, regulation of alf and flo seems to differ. Alternatively, the transcript derived from the flo-640 allele may be highly unstable due to the sequence alterations caused by the transposon footprint (de Vetten et al., 1997; van Hoof and Green, 1996).

During subsequent stages of flower development alf becomes sequentially active in the organ primordia in all 4 floral whorls, although the expression in stamen primordia is not uniform. The alf expression pattern is very similar to that of lfy. In contrast, flo, is only expressed in whorls 1, 2 and 4, while expression of nfl remains limited to whorls 1 and 2 (Kelly et al., 1995). Since the mutants usually lack floral meristems the role of these genes in floral organ development remains obscure.

alf expression in vegetative meristems

Like homologs isolated from tobacco, pea and Impatiens, alf is also expressed in vegetative meristems, while flo and lfy are inactive there. Recently, the analysis of the pea mutant unifoliata and the expression pattern of the uni gene revealed a role for uni in maintaining cells in a transient phase of indeterminacy (Hofer et al., 1997). Apart from the floral phenotype, which resembles that of flo and lfy mutants, the compound leaf with tendrils is converted into simple single lamina in uni mutants. By in situ hybridisation it was shown that uni is still expressed in relatively old leaf primordia, where it presumably maintains cells in this transient indeterminate state to facilitate the formation of a compound leaf.

Wild-type Petunias develop simple leaves consisting of a single lamina. The expression of alf in leaf primordia is lost rapidly after the outgrowth of the leaf primordia. Therefore, it seems likely that the expression of alf in leaves points towards an old function of alf in leaf development that has been lost during evolution in Petunia. A similar situation might exist in tobacco, where NFL expression was detected in leaf primordia. In tomato, the compound leaf is unaltered in the falsiflora (fa) mutant, although it has been suggested that it encodes a FLO homolog (Allen and Sussex, 1996; Coen and Nugent, 1994). Molecular analysis of the fa mutant and the flo homolog of tomato might clarify this contradiction.

exp and inflorescence development

We have shown that the wild-type inflorescence forms floral meristems by bifurcation of the inflorescence meristem. In the exp mutant each inflorescence branch terminates with the formation of a single flower, thereby converting the normally indeterminate inflorescence into a determinate one. The exp mutant thereby resembles to some extent the phenotype of the cen and tfl mutants of Antirrhinum and Arabidopsis (Bradley et al., 1996, 1997). In the cen mutant about ten flowers are produced before the inflorescence meristem generates a terminal flower. The cen gene is expressed in the sub-apical region of the inflorescence meristem where it represses the activation of flo in the inflorescence meristem. In turn, the expression of cen seems to be dependent on FLO as cen is not expressed in a flo mutant (Bradley et al., 1996). Therefore, cen seems to maintain the inflorescence identity of the apical dome after floral induction by regulating the expression domain of flo. Although the phenotype of the cen/flo mutant has not been reported, one would expect that, given the cascade of regulation of flo and cen and their role in inflorescence development, the flo/cen double mutant would be indistinguishable from the flo single mutant.

Our analysis of the exp/alf double mutant suggests a completely different role for exp than for cen. Similar to the situation in the exp single mutant (Fig. 8, arrow to the right), bifurcation of the inflorescence is lost in the alf/exp double mutant (Fig. 8, smallest cycle). In addition to the loss of bifurcations, the loss of ALF function in an exp background converts the terminal flower into an indeterminate inflorescence meristem. Therefore, it seems that exp is required for bifurcation of the inflorescence meristem alone and has no influence on meristem identity.

The nam gene of Petunia, which is involved in establishing the border of the apical meristem and of floral organ primordia, is expressed in a stripe of cells in between the inflorescence and floral meristem, before bifurcation of the inflorescence meristem has occurred (Souer et al., 1996). In nam mutants, however, no alterations in inflorescence architecture can be detected presumably because of redundancy between nam and nam homologous genes (Souer, 1997). Given the function of nam and exp in separating meristems and organ primordia, nam and exp might function in the same pathway, where exp is a specific separator of inflorescence and floral meristems and nam has a function throughout the plant body.

When bifurcation is lost, as in exp plants, the complete apical dome is converted into a flower. This suggests that alf is expressed throughout the apex in the exp mutant. Since our analysis of exp and the exp/alf double mutant points towards a role for exp in bifurcation of the inflorescence meristem and not in floral meristem identity, this might be an indirect effect of the lack of bifurcation. It could be that alf is still transcribed in part of the apical dome only, but induces floral meristem identity in the whole apex because it acts non-autonomously at the moment bifurcation has been omitted. Preliminary results indicate that the induction of floral organ development in exp floral meristems is unequally distributed on the apex, with organ initiation starting in one part of the meristem first. This part presumably reflects the site in the apex where the alf gene is transcribed. In Antirrhinum, it was shown that FLO is able to induce expression of floral organ identity genes non-autonomously (Carpenter and Coen, 1995; Hantke et al., 1995). Establishing the expression domain of alf in exp plants might clarify this point.

The conversion of the whole apical meristem into a floral meristem in exp mutants causes the formation of extra floral organs in whorl 2 and 3. The formation of extra floral organs in whorl 2 is independent of organ identity as it also occurs in the gp/exp double mutant, where petals are replaced by sepals. Possibly, the formation of extra floral organs is a secondary effect that is caused by extra space due to the conversion of the whole apex into a floral meristem.

The flop cDNA was kindly provided by Prof. N.-H. Chua (Rockefeller University, New York). We thank Saskia Kars for assistance with scanning electron microscopy and Gerco Angenent for providing the alf-X2586 allele. Thanks are also due to Pieter Hoogeveen and Martina Meesters for taking care of the plants, and to Wim Bergenhenegouwen and Fred Schuurhof for photographic work. E. S. is supported by a grant from the Netherlands Technology Foundation (STW), with financial aid from the Netherlands Organization for the Advancement of Research (NWO).

Allen
,
K. D.
and
Sussex
,
I. M.
(
1996
).
Falsiflora and anantha control early stages of floral meristem development in tomato (Lycopersicon esculentum Mill
.).
Planta
200
,
254
264
.
Angenent
,
G. C.
,
Franken
,
J.
,
Busscher
,
M.
,
van Dijken
,
A.
,
van Went
,
J. L.
,
Dons
,
H. J. M.
and
van Tunen
,
A. J.
(
1995
).
A novel class of MADS box genes is involved in ovule development in Petunia
.
Plant Cell
7
,
1569
1582
.
Anthony
,
R. G.
,
James
,
P. E.
and
Jordan
,
B. R.
(
1996
).
Cauliflower (Brassica oleracea var botrytis L) curd development: The expression of meristem identity genes
.
J. Exp. Bot.
47
,
181
188
.
Bradley
,
D.
,
Carpenter
,
R.
,
Copsey
,
L.
,
Vincent
,
C.
,
Rothstein
,
S.
and
Coen
,
E.
(
1996
).
Control of inflorescence architecture in Antirrhinum
.
Nature
379
,
791
797
.
Bradley
,
D.
,
Ratcliffe
,
O.
,
Vincent
,
C.
,
Carpenter
,
R.
and
Coen
,
E.
(
1997
).
Inflorescence commitment and architecture in Arabidopsis
.
Science
275
,
80
83
.
Cañas
,
L. A.
,
Busscher
,
M.
,
Angenent
,
G. C.
,
Beltrán
,
J.-P.
and
van Tunen
,
A. J.
(
1994
).
Nuclear localization of the Petunia MADS box protein FBP1
.
Plant J.
6
,
597
604
.
Carpenter
,
R.
and
Coen
,
E. S.
(
1995
).
Transposon induced chimeras show that floricaula, a meristem identity gene, acts non-autonomously between cell layers
.
Development
121
,
19
26
.
Carpenter
,
R.
,
Copsey
,
L.
,
Vincent
,
C.
,
Doyle
,
S.
,
Magrath
,
R.
and
Coen
,
E.
(
1995
).
Control of flower development and phyllotaxy by meristem identity genes in Antirrhinum
.
Plant Cell
7
,
2001
2011
.
Child
,
A.
(
1979
). A review of branching patterns in the Solanaceae. In:
The Biology and Taxonomy of the Solanaceae
(ed.
J. G.
Hawkes
,
R. N.
Lester
and
A. D.
Skelding
), pp.
345
356
.
Academic Press
,
London
.
Coen
,
E. S.
,
Roemro
,
J. M.
,
Doyle
,
S.
,
Elliott
,
R.
,
Murphy
,
G.
and
Carpenter
,
R.
(
1990
).
floricaula: A homeotic gene required for flower development in Antirrhinum majus
.
Cell
63
,
1311
1322
.
Coen
,
E. S.
and
Nugent
,
J. M.
(
1994
).
Evolution of flowers and inflorescences
.
Development
Supplement
107
116
.
de Vetten
,
N.
,
Quattrocchio
,
F.
,
Mol
,
J.
and
Koes
,
R.
(
1997
).
The an11 locus controlling flower pigmentation in petunia encodes a novel WD-repeat protein conserved in yeast, plants, and animals
.
Genes Dev.
11
,
1422
1434
.
Doodeman
,
M.
,
Gerats
,
A. G. M.
,
Schram
,
A. W.
,
de Vlaming
,
P.
and
Bianchi
,
F.
(
1984
).
Genetic analysis of instability in Petunia hybrida 2. Unstable mutations at different loci as the result of transpositions of the genetic element inserted at the An1 locus
.
Theor. Appl. Genet.
67
,
357
366
.
Gerats
,
A. G. M.
,
Kaye
,
C.
,
Collins
,
C.
and
Malmberg
,
R. L.
(
1988
).
Polyamine levels in Petunia genotypes with normal and abnormal floral morphologies
.
Plant Physiol.
86
,
390
393
.
Gerats
,
A. G. M.
(
1991
).
Mutants involved in floral and plant development in Petunia
.
Pl. Sci.
80
,
19
25
.
Hantke
,
S. S.
,
Carpenter
,
R.
and
Coen
,
E. S.
(
1995
).
Expression of floricaula in single cell layers of periclinal chimeras activates downstream homeotic genes in all layers of floral meristems
.
Development
121
,
27
35
.
Hofer
,
J.
,
Turner
,
L.
,
Hellens
,
R.
,
Ambrose
,
M.
,
Matthews
,
P.
,
Michael
,
A.
and
Ellis
,
N.
(
1997
).
UNIFOLIATA regulates leaf and flower morphogenesis in pea. Curr. Biol.
7
,
581
587
.
Kelly
,
A. J.
,
Bonnlander
,
M. B.
and
Meeks-Wagner
,
D. R.
(
1995
).
NFL, the tobacco homolog of FLORICAULA and LEAFY, is transcriptionally expressed in both vegetative and floral meristems
.
Plant Cell
7
,
225
234
.
Koes
,
R.
,
Souer
,
E.
,
van Houwelingen
,
A.
,
Mur
,
L.
,
Spelt
,
C.
,
Quattrocchio
,
F.
,
Wing
,
J.
,
Oppedijk
,
B.
,
Ahmed
,
S.
,
Maes
,
T.
,
Gerats
,
T.
,
Hoogeveen
,
P.
,
Meesters
,
M.
,
Kloos
,
D.
and
Mol
,
J. N. M.
(
1995
).
Targeted gene inactivation in Petunia by PCR-based selection of transposon insertion mutants
.
Proc. Natl. Acad. Sci. USA
92
,
8149
8153
.
Napoli
,
C. A.
and
Ruehle
,
J.
(
1996
).
New mutations affecting meristem growth and potential in Petunia hybrida Vilm
.
J. Hered.
87
,
371
377
.
Poteau
,
S.
,
Nichols
,
D.
,
Tooke
,
F.
,
Coen
,
E.
and
Batty
,
N.
(
1997
).
The induction and maintenance of flowering in Impatiens
.
Development
124
,
3343
3351
.
Souer
,
E.
,
van Houwelingen
,
A.
,
Kloos
,
D.
,
Mol
,
J.
and
Koes
,
R.
(
1996
).
The no apical meristem gene of Petunia is required for pattern formation in embryos and flowers and is expressed at meristem and primordia boundaries
.
Cell
85
,
159
170
.
Souer
,
E.
(
1997
).
Genetic analysis of meristem and organ-primordium formation in Petunia hybrida
. Ph. D. thesis,
Free University
,
Amsterdam
.
Tucker
,
S. C.
(
1989
).
Overlapping organ initiation and common primordia in flowers of Pisum sativum (Leguminosae: Papilionoideae
).
Amer. J. Bot.
76
,
714
729
.
van der Krol
,
A. R.
,
Brunelle
,
A.
,
Tsuchimoto
,
s.
and
Chua
,
N.-H.
(
1993
).
Functional analysis of Petunia floral homeotic MADS-box gene pMADS1
.
Genes Dev.
7
,
1214
1228
.
van Hoof
,
A.
and
Green
,
P. J.
(
1996
).
Premature nonsense codons decrease the stability of phytohemagglutinin mRNA in a position-dependent manner
.
Plant J.
10
,
415
424
.
Weberling
,
F.
(
1989
).
Morphology of Flowers and Inflorescences
.
Cambridge University Press
,
Cambridge
.
Weigel
,
D.
,
Alvarez
,
J.
,
Smyth
,
D. R.
,
Yanofsky
,
M. F.
and
Meyerowitz
,
E. M.
(
1992
).
LEAFY controls floral meristem identity in Arabidopsis
.
Cell
69
,
843
860
.
Weigel
,
D.
(
1995
).
The genetics of flower development: From floral induction to ovule morphogenesis
.
Annu. Rev. Genet.
29
,
19
39
.
Weigel
,
D.
and
Nilsson
,
O.
(
1995
).
A developmental switch sufficient for flower initiation in diverse plants
.
Nature
377
,
495
500
.