An apparent reversal in floral symmetry in the legume Cadia is a homeotic transformation

Abstract

Within papilionoid legumes, characterized by flowers with strong bilateral symmetry, a derived condition within angiosperms, Cadia (Cadia purpurea) has reverted to radially symmetrical flowers. Here, we investigate the genetic basis of this morphological reversal. Two orthologues of the floral symmetry gene CYCLOIDEA (CYC) demarcate the adaxial (dorsal) region of the flower in typical papilionoid legumes. In the model legume Lotus japonicus, one of these LegCYC genes has been shown, like CYC, to be required for the establishment of floral bilateral symmetry. This study shows that these genes are expressed in the adaxial region of the typical papilionoid flower of Lupinus, which belongs to the same papilionoid subclade as Cadia. In Cadia, these genes also are expressed, but the expression pattern of one of these has expanded from the adaxial to the lateral and abaxial regions of the corolla. This result suggests that the radial flowers of Cadia are dorsalized and, therefore, are not a true evolutionary reversal but an innovative homeotic transformation, where, in this case, all petals have acquired dorsal identity. This study raises a question over other putative reversals in animals and plants, which also may be cryptic innovations.

Most members of the plant family Leguminosae (Fabaceae) have flowers that are characterized by a single axis of symmetry (bilateral symmetry or zygomorphy). The majority of legume species, from the subfamily Papilionoideae and accounting for ≈5% of all flowering plant species, are characterized by having typical “pea” (papilionoid) flowers with pronounced bilateral symmetry particularly in the corolla (illustrated here by Lupinus nanus; Fig. 1a). The typical papilionoid flower has three distinctive petal types: a single upper (adaxial or dorsal) petal (the “standard”), two lateral petals (“wings”), and two lower (abaxial or ventral) petals (“keel” petals). The androecium of typical papilionoids also is bilaterally symmetrical, with stamens of different lengths along the dorsoventral axis, the adaxial stamens being shorter (Fig. 1a). However, within the Papilionoideae, a number of unrelated genera have radially or nearly radially symmetrical flowers. The flowers of these atypical species frequently have been considered primitive (reviewed in ref. 1), but more recent phylogenetic studies suggest that they are derived from a zygomorphic ancestral state (2, 3). An example is Cadia Forsk., a genus of seven species of small shrubs from Arabia, Madagascar, and eastern Africa. Cadia has actinomorphic pendent flowers in solitary or few-flowered axillary racemes (ref. 4; Fig. 1b). These flowers produce abundant nectar but no scent, suggesting these may be pollinated by birds (1). Cadia occupies a relatively isolated phylogenetic position (J. S. Boatwright and D. Edwards, personal communications) within the genistoid clade of papilionoid legumes (83 genera and 2,350 species; ref. 5). Its closest relatives are from the tribe Podalyrieae (J. S. Boatwright and D. Edwards, personal communications), which have typical zygomorphic papilionoid flowers.

Fig. 1.

Mature whole and dissected flower of L. nanus and C. purpurea. (a) Detail of an L. nanus flower, showing strong bilateral symmetry typical of papilionoid flowers. Individual standard (D, dorsal), wing (L, lateral), and keel (V, ventral) petals are shown, as well as the monadelphous (with united filaments) androecium with stamens of different lengths (lateral view on the left, and dissected dorsal and lateral/ventral stamens on the right; the three dorsal stamens are shorter than the lateral and ventral ones). (b) C. purpurea flowers are radially symmetrical, with equal dorsal (D), lateral (L), and ventral (V) petals and stamens.

The early development of Cadia (Cadia purpurea) flowers is similar to that of most papilionoid species with zygomorphic flowers where sepals, petals, and stamens are initiated unidirectionally, starting on the abaxial side (6). In typical papilionoid legumes, although organogenesis is asymmetric, a phase of uniform organ growth precedes the unequal development along the dorsoventral axis, and differential organ development occurs at an advanced stage of floral ontogeny (7). By contrast, in C. purpurea, floral organs continue to develop equally after organogenesis until maturity (6). Tucker (6) therefore interpreted the phenotype of C. purpurea as “neotenous,” that is, retaining the characteristics of early flower development (uniform growth) and not entering the asymmetric phase. In genetic terms, this interpretation suggests that the radial phenotype of C. purpurea may be caused by a loss of expression of genes that affect floral organ differentiation during the latter stage of development of typical zygomorphic papilionoid flowers.

Genes that are important for the control of floral symmetry have been identified in Antirrhinum majus L. (Veronicaceae, Lamiales) (8–10). Two closely related genes CYCLOIDEA (CYC) and DICHOTOMA (DICH) demarcate the dorsal region of the flower and affect the development of floral organs along the dorsoventral axis. CYC and DICH expression is maintained throughout floral development in the adaxial region (8, 9). Double knockout mutants for CYC and DICH have a fully radially symmetric phenotype characterized by ventralization of floral organs (i.e., resembling the ventral phenotype of wild-type flowers) (8).

CYC/DICH homologues have been identified in legumes, referred to as LegCYC (11). There is strong functional evidence that at least one LegCYC gene, LjCYC2 (Lotus japonicus 1 in ref. 11), is involved in the control of floral symmetry in the model papilionoid legume Lo. japonicus (12), a member of the robinioid clade. Like CYC, LjCYC2 is expressed in the adaxial region of developing flowers and has been found to be important for establishing dorsal identity (12). Lotus plants expressing LjCYC2 constitutively have lateral and ventral petals that are similar to dorsal petals in shape and epidermal cell type, whereas those with reduced LjCYC2 activity have lateral petals with a ventralized phenotype (12). We show here that two LegCYC genes, belonging to the same gene clade (LegCYC1) as LjCYC2 (11) (sequences given in Fig. 4, which is published as supporting information on the PNAS web site) are expressed on the adaxial side throughout floral development in another papilionoid legume, Lupinus (L. nanus), from the genistoid clade. L. nanus has typical zygomorphic pea flowers, suggesting that these genes are candidates for the control of floral symmetry in genistoid legumes. The expression of these two LegCYC genes is examined in the radially symmetrical flowers of Cadia (C. purpurea), which, like Lupinus, is a genistoid legume (2). Comparison of LegCYC expression pattern between an unusual legume such as C. purpurea and a related zygomorphic species can illuminate the genetic basis of morphological reversal in relation to floral symmetry in legumes. The reversibility of evolution is an important problem, but few examples of apparent phenotypic reversals have been investigated genetically (reviewed in ref. 13). Here, we establish whether the apparent reversal of floral symmetry in Cadia is the result of loss of expression of candidate genes, which could be described as “true reversal,” or is due to modified gene expression causing the replication of ancestral states.

Results

LegCYC1A and LegCYC1B RNA was detected in floral tissue of L. nanus in a pattern similar to Antirrhinum CYC (ref. 8; Fig. 2). Both genes were detected in floral meristems before organogenesis, on the adaxial side of the meristem (Fig. 2 d and g). At more advanced developmental stages, both genes were detected in the corolla (Fig. 2 e, f, h, and i). Similar to CYC, expression of LegCYC1B in the dorsal petal was found in the inner cell layers at the site where cell division was repressed early in organogenesis. Although the expression domains of LegCYC1A and LegCYC1B are largely overlapping, suggesting functional redundancy, LegCYC1A appears to have a reduced expression domain relative to LegCYC1B.

Fig. 2.

Patterns of RNA in situ hybridization of LegCYC1A and LegCYC1B in L. nanus inflorescenses (a and b, respectively), with details of floral meristems at different development stages (d–f, LegCYC1A; g–i, LegCYC1B). Longitudinal sections of L. nanus inflorescences show floral meristems (fm) in the axil of bracts (B) on the abaxial side. More mature flowers are at the base of the inflorescence. RNA from LegCYC1A and LegCYC1B is detected in the adaxial part of floral meristems before organogenesis (d and g) and during floral organ development (e, f, h, and i). Although both copies have a similar expression pattern, LegCYC1B appears to have a wider expression domain than LegCYC1A. Negative control (sense probe) is shown (c). Ad, adaxial; Ab, abaxial; St, stamen; AbS, abaxial sepal. (Scale bars: a–c, 500 μm; d–h, 200 μm; f and i, 250 μm.)

RT-PCR results in L. nanus suggest that, in agreement with the findings in situ, both LegCYC1A and LegCYC1B are florally expressed (Fig. 3) and only in the adaxial part of the developing flower (Fig. 3). RT-PCR also shows that this expression is found in floral buds at more advanced developmental stages. Both copies are transcribed at this stage not only in the standard (dorsal) petal but also in the dorsal anthers.

Fig. 3.

RT-PCR analysis of LegCYC1A (1A) and LegCYC1B (1B) from whole-flower buds at early developmental stages (≈1 mm diameter) and dissected floral tissue from L. nanus and C. purpurea flowers are later stages. The corolla and androecium in both species were dissected as described in Materials and Methods. Intron splicing in both genes allowed to distinguish between cDNA and genomic DNA. cDNA products amplified by actin-specific primers were used as a positive control. gDNA, genomic DNA; −ve, negative control (no template); D, dorsal; L, lateral; V, ventral.

In C. purpurea, RT-PCR from young flower buds revealed that LegCYC1A and LegCYC1B also are expressed in young floral buds (Fig. 3). However, the pattern of expression of these genes in dissected tissue from the corolla and androecium at an advanced developmental stage are different from each other and from their L. nanus orthologues. In the corolla of C. purpurea, LegCYC1A is expressed only in the dorsal petal, and its level of expression appears moderate to weak (Fig. 3). LegCYC1B, however, is expressed in all petals (Fig. 3), showing an expansion of the expression domain of this gene with respect to L. nanus, which is consistent with the radial phenotype of the corolla. In addition, unlike in L. nanus, neither LegCYC1A nor LegCYC1B appear to be expressed in the androecium of C. purpurea (Fig. 3).

Discussion

This study shows that two CYC-like genes, based on their expression pattern in the zygomorphic flowers of the typical papilionoid legume Lupinus, are likely to control floral symmetry in the genistoid clade to which Lupinus belongs. These genes, which have duplicated independently from CYC and DICH (11), may be functionally redundant. Therefore, not only do CYC-like genes appear to have been recruited independently for the control of zygomorphy in Lamiales and legumes (12), but maintenance of partially redundant copies also has occurred independently in these lineages. The role of these genes in the development of zygomorphic flowers in this legume clade is supported by functional evidence from the orthologous LjCYC2 in Lo. japonicus (12). This result provides a framework for examining the genetic control of floral symmetry in a clade that contains atypical species with actinomorphic flowers.

In C. purpurea, like in Lupinus, both LegCYC1A and LegCYC1B are transcribed in early developing flowers. Taking ontogeny into account, this result is not surprising because early development of C. purpurea flowers is similar to that of most papilionoid species with zygomorphic flowers, with unidirectional organ initiation starting on the abaxial side (6). It remains to be demonstrated whether LegCYC genes have the same expression pattern in early floral development in C. purpurea compared with typical papilionioids. During the latter stages of floral development, whereas floral organs undergo unequal growth along the dorsoventral axis in typical papilionoid legumes such as Lupinus, they develop equally in C. purpurea (6). This equal development could suggest that genes that control dorsoventral asymmetry are not expressed after organogenesis in C. purpurea. The molecular data presented here, however, suggest that rather than failing to maintain CYC expression during the later stages of flower development, one CYC-like gene, LegCYC1B, is expressed in all five petals of C. purpurea. Another copy, LegCYC1A, is expressed adaxially but may be down-regulated. The expansion of LgCYC1B expression appears to be whorl-specific. In the androecium, neither LegCYC1B or LegCYC1A mRNA was detected in C. purpurea, although expression was found in the dorsal region in L. nanus. The apparent loss of CYC expression in the stamens is intriguing but difficult to interpret. Mature stamens in C. purpurea are identical morphologically, and the Cadia androecium is actinomorphic, but it is hard to establish whether the stamens have a ventralized phenotype. Without more information on the role of CYC in stamen development of other legume species, it is difficult to place our observations in context.

The expression of LegCYC1B in the corolla is reminiscent of the backpetals mutation in Antirrhinum (9). This mutant has ectopic expression of CYC in the lateral and ventral petals. A transposon insertion in an AT-rich site ≈4.2 Kb upstream of start codon is believed to affect a cis-acting region that normally suppresses CYC transcription during the latter stages of development in wild-type Antirrhinum flowers (9). It may be that a change in cis regulation also has led to the expansion of the expression domain of LegCYC1B in C. purpurea.

The occurrence of a putative ancestral state such as radial symmetry within a clade that has a derived character (zygomorphy) is frequently referred to as an evolutionary reversal. Our results indicate that from a genetic point of view, the radial symmetry of Cadia is an evolutionary innovation correlated, in part, by the expansion of the expression domain of a CYC-like gene in the corolla. The functional significance of this expression pattern is suggested by transgenic experiments in Lo. japonicus (12). Overexpression mutants showing an expansion of expression of LjCYC2 displayed a dorsalized phenotype (12), supporting the hypothesis that changes in LegCYC1B expression in Cadia have an effect on flower development. Because CYC is a marker for dorsal identity, these changes can be considered homeotic, with the lateral and ventral petals of Cadia assuming a dorsal phenotype. This interpretation also is supported by morphology. In Cadia, the five petals are large and bilaterally symmetric, features that are typical of the papilionoid standard petal. By contrast, wing and keel petals in typical papilionoid flowers are asymmetric and small relative to the standard. Partial dorsalization also has been noted in Mohavea confertiflora, where stamen number is reduced from four to two compared with its close relative A. majus and is correlated with expansion of CYC and DICH expression from the adaxial to the lateral region (14). Although loss of function is often perceived as the simplest scenario to explain possible reversals to radial symmetry, it has been suggested that, on close morphological inspection, many shifts zygomorphy to actinomorphy involve other genetic or developmental changes (15). Our findings support the hypothesis that such homeotic-like transformations may play an important role in establishing morphological diversity.

Materials and Methods

RNA in Situ Hybridization.

RNA in situ hybridization was carried out as described by K. Barton (available upon request; The Carnegie Institute of Washington, Stanford, CA) on 7-μm longitudinal sections of L. nanus inflorescences fixed in 4% paraformaldehyde. Digoxygenin-labeled probes specific to LegCYC1A and LegCYC1B were prepared from linearized PCR fragments, amplified from the 5′ region of the ORF by using primers LEGCYC1-F2 (5′-CTT TCY TTA ACC CTG AAA ATG CTT C-3′) at the start of the ORF and LEGCYC1A-R1 (5′-CTA CYA CTA CCC CTT CTG G-3′)/LEGCYC1B-R1 (5′-CAA GCS GGT TCC TTY TGT G-3′), respectively (primers and PCR conditions described in ref. 16), and cloned into pCR4 plasmid (Invitrogen, Carlsbad, CA).

Extensive and exhaustive attempts to provide in situs of Cadia flowers were unsuccessful. Fixation and sectioning were very problematic, probably because of the adaptation of the flowers to their xeric habitat (tough sepals covered in filamentous trichomes).

RT-PCR.

Tissue for RT-PCR was collected from young floral buds (≈1 mm diameter). Tissue also was collected from the corolla and androecium of flowers at more advanced and comparable developmental stages in L. nanus and C. purpurea, during the latter stages of organ enlargement where these had undergone a high degree of dorsoventral differentiation in L. nanus and where the flowers, still in bud, could be easily dissected. Floral orientation in C. purpurea was determined with respect to the gynoecium, which is the only part of the flower showing dorsoventral asymmetry. Total RNA was extracted by using Qiagen RNeasy RNA extraction kit (Valencia, CA) from the following tissues: the dorsal, lateral, and ventral petals in L. nanus, and each individual petal from a single flower in C. purpurea, the three adaxial stamens separated by a groove from the seven lateral and ventral stamens (combined) in L. nanus, and the dorsal (three uppermost) lateral (two on either side) and ventral (three lowermost) stamens in C. purpurea. cDNA was synthesized by using Qiagen Omniscript RT kit. RT-PCR was carried out by using the locus-specific primers LEGCYC1A-F1 (5′- CCA GAA GGG GTA GTR GTA G-3′)/LEGCYC1B-F1 (5′-CAC ARA AGG AAC CWG CTT G-3′) and primer LEGCYC1-R1 (5′-CAC TCY TCC CAR GAY TTT CC-3′), which is downstream of the short intron at the 3′ end of the LEGCYC ORFs (primers described in ref. 16). Locus specificity was confirmed by sequencing of RT-PCR products. RT-PCRs were repeated with cDNA from separate RNA extractions of tissue from individual flowers. Actin was used as a positive control (using primers described in ref. 17). To ensure that the quantity of PCR products was comparable between samples, actin products were assayed between 20 and 35 cycles. PCR products are shown in the linear phase after 30 cycles.