Current Research
 Evolution of novel floral organ identity in Aquilegia
Researchers: Bharti Sharma, Elena Kramer, Levi Yant, Emily Gleason, Lynn Holappa, Joshua Puzey (past researcher), Faye Rosin (past researcher)
    One of the main attractions to working with Aquilegia is its novel floral morphology (Fig. 1 above). The perianth is composed of two whorls of morphologically distinct petaloid organs: the first whorl petaloid sepals are ovate and flat and the second whorl petals are characterized by large nectar spurs (Fig. 1B). The 4-7 whorls of stamens have typical dicot morphology but internal to the stamens is one whorl of sterile organs termed staminodia, which are composed of a central filament flanked by lateral wings (Fig. 1C). These organs are clearly dissimilar from the outer stamens or inner carpels, and intermediates between the organ types are not commonly observed. While the ecological importance of the nectar spur is well established (Hodges 1997; Whittall and Hodges, 2007), the function of the staminodia is not well understood. It has been hypothesized, however, that they play a role in protecting the young ovaries from herbivory damage (Voelckel et al., 2010). The close phylogenetic relatives of Aquilegia put the evolution of these novel traits in context (Fig. 1D). The sister genus Semiaquilegia has a nectary pocket but no spur (Fig. 1E) and its inner stamens are reduced to irregular, sterile filaments that are not laterally expanded (Fig. 1F; Tucker and Hodges 2005). In contrast, closely related genera, such as Isopyrum, exhibit only a slight nectar cup and no sterilized stamens.
    Understanding the genetic basis of this distinct morphology began with a full characterization of B gene homologs in Aquilegia (Fig. 2 below; Kramer et al., 2007). There are three paralogs of AP3, termed AqAP3-1, -2 and -3, and one PI homolog, AqPI. The AqAP3-1 expression domain is broad at early developmental stages but quickly becomes stronger in the staminodia, suggesting that it may play an important role in promoting staminodial identity but could also contribute to petal or stamen identity. AqAP3-2 is primarily expressed in stamens but also in later petal development. In contrast, AqAP3-3 expression is petal-specific and we suggest that it is essential to the development of this organ type. AqPI is likely to contribute to all of these functions, both based on its broad expression and the fact that AqPI heterodimerizes with all three AP3s. Lastly, at very late stages, AqPI and AqAP3-1 and -2 are detected by RT-PCR in the sepals, raising the possibility that they contribute to petaloidy of the sepals. We have now used RNAi-based VIGS (Gould and Kramer 2007) with the tobacco rattle virus platform to obtain specific silencing, confirmed by qRT-PCR, of AqPI, AqAP3-1, -2 and -3 as well as dual silencing of AqAP3-1/2.
    Silencing of AqPI in A. vulgaris produced a strong B class phenotype of petal to sepal and stamen to carpel transformation (Kramer et al. 2007). In addition, the novel staminodia were transformed into carpels, confirming that these organs are controlled by the Aquilegia B gene homologs. The strong transformations observed in these plants were expected since all three AP3 homologs appear to function as heterodimers with AqPI. In addition, we observed decreases in AqAP3-2 and AqAP3-3 expression. This could be due to off-target silencing but, since it did not affect AqAP3-1, we hypothesize that it reflects auto-regulatory feedback similar to what is observed for AP3 and PI in Arabidopsis (Jack et al., 1992; Jack et al., 1994). We are now targeting each of the AP3 paralogs for knockdown, both individually and in combination. Silencing of AqAP3-3 results in transformation of petals into sepals with no effect on the development of the other floral organs (Sharma et al. 2011) and a manuscript is in preparation describing the AqAP3-1, -2 and double -1/2 phenotypes. In addition, we are studying the roles of the Aquilegia AGAMOUS paralogs AqAG1 and AqAG2 in establishing staminodium identity.
Beyond this candidate gene approach, we plan to examine 1) the evolution of the novel staminodium organ identity program (see below), and 2) the developmental and genetic basis of nectar spur evolution (see the Petal Spurs page).
Investigating the genetic basis of staminodium identity
    In order to move beyond our characterization of floral organ identity homologs, we have used the PALM Laser Microbeam to isolate multiple developmental stages of stamen and staminodia primordia from early stage Aquilegia floral meristems. A total of 1 million m2 of tissue was microdissected for each of three replicates of five sample classes (Fig. 3 below): stamen primordia at inception before staminodium initiation (class st0); separate stamen and staminodium primordia immediately after carpel initiation (classes st1 and sd1), at which stage staminodia can be unequivocally identified; and separate stamen filaments and staminodia (classes st2 and sd2). We chose to use only stamen filaments for class st2 because the anthers express a high number of genes related to microsporogenesis, which we already know will be differentially expressed (Voelckel et al, 2010). Also, rare stamen/staminodium chimeras consist of a complete staminodium with a small terminal filament and aborted anther, suggesting that the entire staminodium corresponds to the stamen filament rather than any part of the anther. All samples were sequenced on an Illumina Genome Analyzer II. These runs obtained approximately 12 million ~35 bp reads for each class of RNA, which were mapped using TopHat (v.1.0.12) to a set of gene models we developed based on the two currently available EST datasets and the initial assembly of the Aquilegia coerulea genome. This RNA-seq dataset is now being analyzed to identify candidate genes for both the upstream regulation of AqAP3-1 and -2, and the potential downstream targets of these proteins. In particular, we are interested in the pathways that may be responsible for delimiting the staminodium whorl from the rest of the stamens, that promote lateral outgrowth of the staminodia but limit expansion in the stamen filament, and that prevent post-anthesis abscission of the staminodium but promote it in the stamen.
Gould, B. and Kramer, E. M. (2007) Virus-induced gene silencing as a tool for functional analyses in the emerging model plant Aquilegia (columbine, Ranunculaceae). Plant Methods, 3:6.
Hodges, S.A. (1997) Rapid radiation due to a key innovation in columbines., in Molecular evolution and adaptive radiation. T.J. Givnish and K.J. Sytsma, Eds. Cambridge University Press: Cambridge. p. 391-405.
Jack, T., Brockman, L. L. and Meyerowitz, E. M. (1992) The homeotic gene APETALA3 of Arabidopsis thaliana encodes a MADS box and is expressed in petals and stamens. Cell, 68:683-697.
Jack, T., Fox, G. L. and Meyerowitz, E. M. (1994) Arabidopsis homeotic gene APETALA3 ectopic expression: transcriptional and posttranscriptional regulation determine floral organ identity. Cell, 76:703-716.
Kramer, E. M., L. Holappa, B. Gould, M. A. Jaramillo, D. Setnikov, and Santiago, P. (2007) Elaboration of B gene function to include the identity of novel floral organs in the lower eudicot Aquilegia (Ranunculaceae). Plant Cell, 19:750-766.
B. Sharma, C. Guo, H. Kong, E. M. Kramer. (2011) Petal-specific subfunctionalization of an APETALA3 paralog in the Ranunculales and its implications for petal evolution. New Phytologist, 190:870-883.
Tucker, S.C. and Hodges, S. A. (2005). Floral ontogeny of Aquilegia, Semiaquilegia, and Isopyrum (Ranunculaceae). Int’l J Plant Sci, 166:557-574.
Voelckel, C., Borevitz, J., Kramer, E. M., and Hodges, S. A. (2010) Within and between whorls: comparative transcriptional profiling of Aquilegia and Arabidopsis. PLoS ONE, 5:e9735.
Whittall, J.B. and Hodges, S.A. (2007) Pollinator shifts drive increasingly long nectar spurs in columbine flowers. Nature, 447:706-710.
Fig. 1. A. Aquilegia floral diagram. B. Side view of flower. C. Whorl of staminodia (std) surrounding the carpels (car). D. Evolutionary relationships of Aquilegia and its sister genera. A nectar pocket and inner sterile filaments (fil) appear to have evolved in the common ancestor of Aquilegia and Semiaquilegia. E. Mature Semiaquilegia petal with nectar pocket (arrowhead). F. Semiaquilegia flower with sterile filaments (arrowhead) between fertile stamens (A) and carpels (C). Size bars in E, F = 200 μm. Panels E. and F. are from Tucker and Hodges 2005.
Fig. 2. Summarized expression paterns of three AP3 paralogs and the single PI homolog.
Fig. 3. Left three panels: AqAP3-2 in situs are used to show the five classes (st0-st2 and sd1-2) of tissue collected using LMD. Right two panels: Example LMD sections.