Tests of the Link between Functional Innovation and Positive Selection at Phytochrome A:The Phylogenetic Distribution of Far-Red High-Irradiance Responses in Seedling DevelopmentAuthor(s): Sarah Mathews and Donna TremonteReviewed work(s):Source: International Journal of Plant Sciences, Vol. 173, No. 6 (July/August 2012), pp. 662-672Published by: The University of Chicago PressStable URL: http://www.jstor.org/stable/10.1086/665975 .Accessed: 17/07/2012 11:50

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access toInternational Journal of Plant Sciences.

http://www.jstor.org

TESTS OF THE LINK BETWEEN FUNCTIONAL INNOVATION AND POSITIVE SELECTIONAT PHYTOCHROME A: THE PHYLOGENETIC DISTRIBUTION OF FAR-RED

HIGH-IRRADIANCE RESPONSES IN SEEDLING DEVELOPMENT

Sarah Mathews1,* and Donna Tremonte*

*Arnold Arboretum of Harvard University, 22 Divinity Avenue, Cambridge, Massachusetts 02138, U.S.A.

We surveyed the phylogenetic distribution of far-red high-irradiance responses (FR-HIRs), which areimportant for early development of seedlings in shaded habitats, in exemplar species across seed plants to testwhether the responses are unique to angiosperms and may have helped early angiosperms colonize theunderstory of Mesozoic forests dominated by ferns and gymnosperms. We show that dark-grown seedlingsin most seed plants adopt an etiolated morphology and that being green in the dark is a derived condition insome conifers and gnetophytes. We also show that FR-HIR-induced seedling de-etiolation, mediated byphytochrome, is widely distributed in seed plants. Together, these data show that despite the ability ofgymnosperms to green in the dark, most of them do not do so but link seedling development to light signals, asdo angiosperms. Our data refute an obvious link between the origin of FR-HIRs and the signal of positiveselection at the PHYA locus early in the history of angiosperms.

Keywords: phytochrome, seed plants, angiosperms, phyA, far-red high-irradiance response (FR-HIR),phylogeny.

Online enhancement: appendix PDF.

Introduction

Seedling establishment is a critical phase of plant develop-ment during which the young embyronic plant must makethe transition from reliance on maternal reserves present inthe seed to autotrophic growth relying on a fully developedphotosynthetic apparatus. The probability of success is usu-ally small, and selective pressures are expected to be highduring this phase of a plant’s life cycle (Darwin 1859; Grubb1977; Harper 1977). Thus, conditions for seedling establish-ment play a significant role in determining species ranges andhabitat preferences. Phylogenetic reconstructions that trace theevolutionary history of ecophysiological traits on phylogenetictrees of early-diverging angiosperm lineages (Amborellaceae,Nymphaeales, Austrobaileyales, Chloranthaceae) suggest thattheir common ancestors exploited disturbed microsites in theunderstory of wet primary forests (Feild et al. 2003; Feild andArens 2007). Living representatives of these lineages are elimi-nated when disturbances create canopy gaps that result inlarge increases in light availability and vapor pressure deficits(Feild 2008). Investigation of Early Cretaceous fossil leavesindicated that comparative studies of these taxa reveal genu-ine signals of early-angiosperm ecophysiology (Feild et al.2011). Together, these studies suggest that traits that enhancesurvival in the shade were important to the early success ofthe angiosperms. Such traits include slow growth, relatively

large seed size, possession of lower compensation and satura-tion points than sun-adapted seedlings, optimal allocation ofresources across plant organs balanced with stressors such aslow nitrogen availability, other abiotic stresses, herbivory,and competition (Salisbury 1942; Grime 1979; Westobyet al. 1992; Facelli 2008). Signals about the presence and/ordegree of shade also are critical because they allow plants tofine-tune physiological, developmental, and growth responsesto the light environment. Photoreceptor systems, includingthose that have optima in the blue, green, red, and far-redregions of the visible spectrum, provide this important in-formation (Franklin 2008; Sellaro et al. 2010). Prominentamong these are the phytochromes, due to (1) their role inearly detection of neighboring but not yet shade-producingplants, (2) their ability to detect sun flecks (Sellaro et al.2011) and mediate rapid responses to transient light signals(Botto et al. 1996; Shinomura et al. 1996), and (3) their rolein early development of seedlings in the shade (Yanovskyet al. 1995). The question emphasized in this article iswhether responses mediated by one of the phytochromes,phyA, provided an adaptive advantage to early angiospermsfacing the challenge of colonizing the shaded understory ofMesozoic forests (Mathews et al. 2003).

The Phytochrome System

Light is a major regulator of plant growth, morphogenesis,and development, and it is particularly important in seedlingestablishment. In the absence of light signals, seedling de-velopment in eudicots and grasses is skotomorphogenetic, or

1 Author for correspondence; e-mail: smathews@oeb.harvard.edu.

Manuscript received November 2011; revised manuscript received March

2012.

662

Int. J. Plant Sci. 173(6):662–672. 2012.

� 2012 by The University of Chicago. All rights reserved.

1058-5893/2012/17306-0009$15.00 DOI: 10.1086/665975

etiolated (fig. 1). Maternal resources are directed towardelongation, while the development of leaves, pigments, andthe photosynthetic apparatus is delayed. Cotyledons remainclosed and folded back toward the axis of elongation, pro-tecting the cotyledons as they are pushed through the soil orlitter. Light signals induce photomorphogenetic development,or de-etiolation. Extension growth is inhibited, cotyledons un-fold, protective anthocyanin pigments are expressed, and leavesand the photosynthetic machinery develop (fig. 1). The abil-ity of plants to modulate extension growth in response to thelight environment is an aspect of etiolation and de-etiolationthat underlies their ability to forage for light (Casal et al.1997). This plasticity is critical for emerging seedlings, wherethe ability to elongate during etiolated growth may allowthem to reach the light before maternal resources are depleted,while the ability to inhibit extension growth once they havereached the light facilitates their establishment.

Plants use several photoreceptor systems to detect and re-spond to light cues (Kami et al. 2010). Among these systems,the phytochromes comprise the sole system for mediatingdevelopmental responses to red (R) and far-red (FR) signals,which plants use to distinguish between open and shaded envi-ronments by measuring the proportions of R and FR wave-lengths in ambient light. Above a canopy (lightest blue shadedregion of fig. 2), the R : FR ratio varies from 1.05 to 1.25,whereas below a canopy (green shaded region in fig. 2), R isdepleted and the ratio varies from 0.05 to 1.15 (Smith 1982).Phytochromes measure the ratio R : FR in a particularly ele-gant way. They are soluble chromoproteins that exist in twoforms interconvertible by light via isomerization of a covalentlyattached bilin chromophore; the red-absorbing (Pr) form,

synthesized in the dark (absorption maximum, ;660 nm),is converted to the far-red-absorbing (Pfr) form (absorptionmaximum, ;730 nm; fig. 2) and is converted back to Pr

(Rockwell et al. 2006). Thus, there exists at any given mo-ment a dynamic cellular equilibrium of Pr and Pfr. Their pro-portions, and changes in the proportions, are capable ofinducing physiological and developmental responses in a man-ner that is more a function of light quality than quantity(Smith and Morgan 1983).

It is important to describe briefly the three physiological re-sponse modes of phytochrome action. Low-fluence responses(LFRs) are the classic R/FR-reversible responses described inthe germination of lettuce seed (Borthwick et al. 1952). Theyare induced when irradiation produces levels of Pfr/Ptot from10�3 to 0.87, the theoretical maximum due to the overlap-ping spectra of Pr and Pfr (Mancinelli 1994). These levels arereadily achieved in R-rich environments such as in large can-opy gaps and open environments, suggesting that an impor-tant ecological function of phytochrome acting in LFRs isto promote development when access to photosyntheticallyactive radiation is not limiting (Smith 1995). Phytochromesacting in this mode may inhibit development in the shade(Smith and Morgan 1983). Very low-fluence responses(VLFRs) are induced by levels of Pfr/Ptot from 10�6 to 10�3.VLFRs enable rapid responses in seedlings that are justemerging from the soil or in buried seeds that are very brieflyexposed to light via disturbance (Casal et al. 1997). Far-redhigh-irradiance responses (FR-HIRs) are induced by the main-tenance of relatively low levels of Pfr/Ptot for relatively long pe-riods of time. Although FR-HIRs are fluence and wavelengthdependent, they are induced by fluence rates that are lowerthan those typically found under deep-canopy shade, and lon-ger exposures (e.g., 8–16 h) induce FR-HIRs at rather lowlevels of irradiance (Downs and Siegelman 1963; Hartmann1967; Smith 1983). FR-HIRs predominate under canopies, un-der leaf litter, and in the first few millimeters below the soil

Fig. 1 Cartoon of dark-grown etiolated (left) and light-grown de-

etiolated (right) seedlings.

Fig. 2 Above- and below-canopy spectra and absorption maxima

for phytochrome in the red-absorbing (Pr) and far-red-absorbing (Pfr)

forms. Image by Ethan Levesque.

663MATHEWS & TREMONTE—PHYTOCHROME A IN EARLY ANGIOSPERMS

surface (Smith 1982), suggesting that an important ecologicalfunction of FR-HIRs is to moderate development in FR-enriched, shaded habitats.

These observations suggest that division of labor amongphytochromes creates a dual-sensing system to modulate de-velopment in the contrasting light environments found inopen versus shaded habitats. Studies of mutants in Arabidop-sis thaliana, rice, and maize (Quail et al. 1995; Takano et al.2005; Sheehan et al. 2007) have confirmed that such a dual-sensing system exists. Phytochrome A (phyA) is the principalmediator of early development in shaded habitats via bothVLFRs and FR-HIRs (Yanovsky et al. 1995; Botto et al.1996; Shinomura et al. 1996), whereas phyB is the principalmediator of early development in open habitats via LFRs(Reed et al. 1994). Consistent with the role of phyB to pro-mote development via LFRs in open habitats, phyB is alsothe principal mediator of shade avoidance in shade-intolerantspecies, induced by reduction in the R : FR ratio of ambientlight (Franklin 2008), which promotes a sharp decrease inPfr/Ptot (Smith 1982). An important feature of phyA may beits ability to counteract strong shade avoidance in early seed-ling development by promoting de-etiolation in the shade(McCormac et al. 1992; Reed et al. 1994). Mutants of A.thaliana mutants lacking a functional phyA fail to survive indense shade (Yanovsky et al. 1995). Presumably, photoauto-trophism cannot be achieved before maternal resources storedin the seed are expended.

Phytochromes and Angiosperm Success

Both FR-HIR-induced seedling development and shade avoid-ance have been put forth as factors promoting the success ofangiosperms. Mathews et al. (2003) posited that positive selec-tion detected at the PHYA locus, on the branch leading to allangiosperm PHYA sequences, might be linked to the evolu-tion of FR-HIRs and an enhanced capacity of angiosperms tocolonize shady habitats. Others posited that shade avoidanceis most highly expressed in angiosperms, providing them withmore plastic responses to the presence of neighbors than arefound in other clades of land plants (Smith 2000). These hy-potheses are testable using phylogenetic approaches. If FR-HIRs and/or shade avoidance are adaptive innovations thatpromote the success of angiosperms over other seed plants, wemight expect the innovations, and possibly the genes that en-code them, to be unique to angiosperms.

A systematic survey of shade avoidance responses acrossland plants has not been made, but based on reports in the lit-erature, elements of shade avoidance occur as deep in landplants as liverworts, which display classic shade avoidanceresponses in end-of-day FR experiments (Fredericq 1964;reviewed in Mathews 2006). Conversely, reports in the litera-ture of FR-HIRs in embryophytes are restricted to angio-sperms and ferns, where, based on phylogenetic evidence, theyare likely to have evolved independently (Mathews 2006).

Recent phylogenetic evidence suggests that all seed plantshave orthologs of PHYA and PHYB (Mathews 2010; fig. 3).Thus, the genes in A. thaliana that underlie FR-HIRs (PHYA)and shade avoidance (PHYB) originated early in the historyof seed plants and are not synapomorphies for angiosperms.

However, further tests of variable selection in the phytochromegene phylogeny (Mathews 2010) suggest that PHYA was sub-ject to major episodes of selection and bursts of radical aminoacid replacement during the divergence of the two genes (fig.3), which is consistent with an adaptive role for phyA duringseed plant evolution. In contrast, limited molecular innovationwas detected in the PHYB lineage (fig. 3). To test whether, inseed plants, FR-HIRs evolved earlier in angiosperms than theeudicot-monocot split or even deeper in the seed plant tree, wetested the ability of continuous FR to inhibit extension growthin seedlings from exemplar clades across angiosperms and gym-nosperms. To do so we used a standard measure of FR-HIRand phyA activity, the measurement of hypocotyl elongationin continuous FR (Fankhauser and Casal 2004).

Material and Methods

Detailed germination requirements and the duration of lighttreatments for each species can be found in the appendix, avail-able as a PDF in the online edition of the International Journalof Plant Sciences. Briefly, after being imbibed in the dark, seedswere placed in one of four light treatments: (1) continuouswhite (Wc), (2) continuous dark (Dc), (3) continuous FR(FRc), and (4) continuous R (Rc). The expected phytochrome-mediated morphologies in the Arabidopsis thaliana modelare shown in figure 4: dark-grown seedlings are etiolated, butlight signals induce the inhibition of extension growth throughthe activities of blue-light receptors and phytochromes (in Wc)and through phytochromes (in Rc and FRc). Note that the effi-cacy of FRc to inhibit extension growth is greater than that ofRc (Fankhauser and Casal 2004).

Growth Conditions

Seeds were sterilized (3–7 min 10% bleach, 3–7 min 70%ethanol) and rinsed repeatedly with double-distilled water.Seeds were then imbibed and exposed to 3–24 h of Rc or Wc

to promote synchronous germination. Imbibed, and in somecases stratified, seeds were planted in soilless seedling mix(Fafard’s), plated on filter paper, or plated on 1/2 MS mediaplates (Murashige and Skoog basal medium with sucrose andagar) and then incubated in darkness until germination. Seed-lings were immediately moved into light treatments when ei-ther the hypocotyl or the radicle emerged.

Light Sources

Seedlings were transferred to Wc (2 mmol m�2 s�1 fromfluorescent tubes, F40T12/Utility, 40 W, Philips), Rc (0.1mmol m�2 s�1 from 660-nm diffuse-line LED arrays, AppliedOptotech), FRc (1.0 mmol m�2 s�1 from 720-nm diffuse-lineLED arrays, Applied Optotech, or 2.5 mmol m�2 s�1 from735-nm arrays in a Percival LED chamber), or Dc for periodsof 7, 14, 21 or 35 d, depending on the species.

Growth Measurements

Seedlings were harvested and measured to the nearest mil-limeter. The first shoot organ that emerged from the seed and

664 INTERNATIONAL JOURNAL OF PLANT SCIENCES

that was potentially responsive to light differed among spe-cies, depending on their particular morphologies or modes ofdevelopment. For all conifer species, as well as Ephedra neva-densis and Welwitschia mirabilis, hypocotyl lengths were mea-sured. In Ginkgo biloba, the cotyledons are retained within theseed, and hypocotyls do not elongate; thus, epicotyl lengthswere measured. In cycads, the leaf is the first organ toemerge; thus, petiole length was measured in seedlings ofZamia furfuracea. In all angiosperms, except for monocotsand Nymphaeales, hypocotyl length was measured. In mono-cots, coleoptile or epicotyl length was measured, dependingon which organ was responsive to light. In Nuphar lutea andNymphaea odorata, first leaf length was measured; in thesespecies, both the mesocotyl and the first leaf were responsiveto light (cotyledons remained in the seed, and hypocotyls didnot elongate), but of first leaves, the most accurate measure-ments could be made.

Statistics

We used one-way ANOVAs to assess the differences amongthe means of axis length across light treatments. Where sam-ple size for any one of the light treatments was small, weused the nonparametric Krusgal-Wallis test to assess differ-ences among the means. Test statistics for each species are inthe appendix, and they were the basis for coding of speciesstates for reconstruction of the elongation response on thephylogenies of seed plants. In just one case did we code anaxis as short in response to FRc in the absence of a significanttest statistic: responses of seedlings of N. lutea were very sim-

ilar to those of N. odorata (fig. 6A; appendix), and we codedthem both as short, even though the test statistic from thesmall-sampled Nuphar experiment was not significant.

Reconstruction of Ancestral States for Greening in theDark and FRc-Induced Inhibition of Elongation

The use of a dark control for the measurement of elonga-tion provided the opportunity to make a qualitative evalua-

Fig. 3 Summary phylogeny of seed plant phytochromes. Hatch marks on a branch represent amino acid sites inferred to have been subject topositive selection along that branch. Numbers on branches indicate how many radical amino acid changes map to a particular branch in the tree,

using a parsimony-based approach (data from Mathews 2010).

Fig. 4 Seedlings of Arabidopsis thaliana grown for 3 d in differentlight treatments. W, continuous white; D, continuous dark; FR,

continuous far-red; R, continuous red.

665MATHEWS & TREMONTE—PHYTOCHROME A IN EARLY ANGIOSPERMS

tion of the distribution of seedling greening in the dark. Ourobservations contradicted an assumption held by many thatall gymnosperms green in the dark during seedling develop-ment. Thus, we summarized our observations of both green-ing and elongation on phylogenetic trees of seed plants.

We used the ‘‘Trace Character History’’ module in Mesquite(Maddison and Maddison 2010) to reconstruct the history ofboth traits on a simplified (to match the taxon sampling ofthe light response experiments) PHYA maximum likelihoodtree, from which an ultrametric tree was derived using r8s(Sanderson 2002). Alternate topologies were generated by in-ferring trees with implementation of topological constraints inRAxML (Stamatakis 2006). The presence or absence of green-ing in the dark is variable in conifers and gnetophytes but notin other seed plant clades; thus, we used trees that differedwith respect to placement of gnetophytes as sister either to allconifers (gnetifer topology) or to Pinaceae (gnepine topology).The position of gnetophytes remains controversial, but analy-ses of most nucleotide data sets support the gnepine topology(Mathews 2009), while unpublished DNA data sets show sup-port for conifer monophyly. The presence or absence of FRc-induced inhibition of elongation varies in gymnosperms and, toa lesser degree, angiosperms. Thus, we used trees that differedwith respect to the root, which was placed on either the branchthat separated angiosperms from all gymnosperms (acrogym-nosperm topology; Cantino 2007) or the branch between cy-cads þ angiosperms and Ginkgo, gnetophytes, and conifers.While it is likely that gymnosperms as a whole (living andextinct) are paraphyletic with respect to angiosperms, thesetwo trees reflect the uncertainty surrounding the rooting ofthe tree of living seed plants in trees inferred from moleculardata (Mathews et al. 2010). For further discussion of thesetopologies, see Mathews (2009) and Mathews et al. (2010).

In the ‘‘Trace Character History’’ module, we used a likeli-hood reconstruction method (Schluter et al. 1997; Pagel 1999),which for each node finds the state assignment that maximizesthe probability of arriving at the observed states in the terminaltaxa, given the model of evolution and allowing the states at allother nodes to vary. We used the Markov k–state 1–parametermodel, which corresponds to Lewis’s (2001) Mk model, andthe Markov k–state 2–parameter model (AsymmMk) model,which has two parameters, one each for the change from 0 > 1and 1 > 0 (Maddison and Maddison 2010). The AsymmMkmodel has two options for handling the root: one in which theroot state frequencies are equal and one in which the root statefrequencies are equal to the equilibrium frequencies.

Results

The Distribution of Etiolation in Seed Plants

Etiolation, the absence of apparent chlorophyll and an elon-gated phenotype (fig. 1), was nearly universal in dark-grownseedlings from species across seed plants (figs. 5–7). The excep-tions were the gnetophyte Ephedra nevadensis (fig. 5E) and theconifers Araucaria cunninghamia (fig. 5F) and Pinus elliottii(not shown). Conversely, other gnetophytes and conifers re-quired a light signal to green visibly, including Welwitschia mir-abilis (fig. 5B), Larix laricina (fig. 5C), Thuja occidentalis(appendix, fig. A27), and Gnetum montanum (not shown). As

expected, all angiosperm seedlings were etiolated in the dark(figs. 6, 7). Results from reconstructing the history of greeningin the dark suggest that this is a derived condition in the conifer-gnetophyte clade (fig. 7). The probability that the ancestor ofextant seed plants greened in the dark is only 0.22, and thatof the ancestor of Ginkgo conifers and gnetophytes is only0.23. The probability of greening in the dark increases to0.78 at the base of the conifer-gnetophyte clade, regardless oftree topology (fig. 7A, 7B). The reconstruction of greening inthe dark on both the gnetifer and gnepine trees suggests thatgreening in the dark may be a synapomorphy of conifers andin gnetophytes (fig. 7A, 7B). Among the models implementedto infer ancestral states, the AsymmMk model with root statefrequencies equal to the equilibrium frequencies gave the bestlikelihoods (2DL ¼ 6:48, df ¼ 1, p < 0:25).

Absence of the FR Block of Greening

The expectation of seedlings grown in FRc is that the FRblock of greening would result in seedlings with the inabilityto accumulate chlorophyll, possibly through phyA-dependentrepression of the genes for the light-dependent protochlo-rophyllide oxioreductase genes that angiosperms dependon for reduction of protochlorophyllide to chlorophyllide(Barnes et al. 1996). While our Arabidopsis control accu-mulated minimal visible chlorophyll in the FRc treatment(fig. 4), seedlings of many of the other species we surveyedappeared to accumulate high levels of chlorophyll in FRc

(figs. 5, 6). We do not know whether this is because the FRblock of greening has a limited distribution in seed plants,perhaps occurring only in eudicots, where it has beenreported in Arabidopsis thaliana and tomato (Barnes et al.1996); whether the duration-dependent nature of the re-sponse differs among species, with longer exposure to FRc

required to induce the response in other species; or whetherthere was enough light leakage, of either W or R, into ourFRc treatment, or during transfer of germinants to differentlight treatments, to allow seedlings to escape this block.Since the tomato seedlings from our FRc treatment didgreen (not shown), we assume that light leakage at somestage is the explanation in at least some cases. However,with only two data points, both from within eudicots(Barnes et al. 1996), there are too few data to infer whetherthe FR block of greening occurs broadly in flowering andother seed plants.

The Distribution of FR-HIRs

FRc-induced inhibition of elongation via FR-HIRs wasnearly universal in seedlings of angiosperms. The two excep-tions were in seedlings of Sarcandra glabra and Ceratophyllumdemersum (fig. 6; appendix). In gymnosperms, FRc inhibitedextension growth in Zamia furfuracea, in both gnetophytes,and in the conifer T. occidentalis (Cupressaceae; fig. 5; appen-dix), whereas FRc failed to inhibit extension growth in Ginkgobiloba and in the conifers A. cunninghamia (Araucariaceae),L. laricina, and P. elliottii (both Pinaceae; fig. 5; appendix). Inthe case of C. demersum, FRc significantly increased elonga-tion, leaving open the possibility that FR-HIRs function differ-ently in this aquatic angiosperm (fig. 6E; appendix).

666 INTERNATIONAL JOURNAL OF PLANT SCIENCES

Results from reconstructing FR-induced inhibition of elon-gation on different seed plant phylogenetic trees suggest thatFR-HIRs are ancestral in living seed plants. As with the recon-struction of greening in the dark, inference with the AsymmMkmodel with root state frequencies equal to the equilibriumfrequencies gave the best likelihoods (2DL ¼ 5:14, df ¼ 1,

p < 0:05), but tree topology made no difference to node prob-abilities (fig. 8A, 8B). The probabilities that FR-HIRs werepresent in the ancestors of living seed plants; of cycads þangiosperms; of angiosperms alone; of Ginkgo, conifers,and gnetophytes; and of conifers þ gnetophytes were all0.84 (fig. 8).

Fig. 5 Seedlings of gymnosperms grown in continuous white, dark, far-red, and red light (left–right). A, Zamia furfuracea. B, Welwitschiamirabilis. C, Larix laricina. D, Ginkgo biloba. E, Ephedra nevadensis. F, Araucaria cunninghamii.

667MATHEWS & TREMONTE—PHYTOCHROME A IN EARLY ANGIOSPERMS

Discussion

We used a common assay, the ability of FRc to inhibit elon-gation, to test the link between the origin of FR-HIR-inducedde-etiolation and episodic selection at the PHYA locus, which

encodes the mediator of FR-HIRs in Arabidopsis thalianaand rice (Quail 1995; Takano et al. 2005). We noted firstthat etiolated growth (fig. 1) was a feature of nearly all dark-grown seedlings (fig. 6; appendix). We found that seedlingsof just a handful of those species that etiolated in the dark did

Fig. 6 Seedlings of angiosperms grown in continuous white, dark, far-red, and red light (left–right). A, Nymphaea odorata. B, Schisandra sphenanthera.

C, Calycanthus floridus. D, Saururus chinensis. E, Ceratophyllum demersum. F, Sarcandra glabra. G, Aquilegia vulgaris. H, Acorus calamus.

668 INTERNATIONAL JOURNAL OF PLANT SCIENCES

not de-etiolate in FRc. In one of these, Ceratophyllum demer-sum, FRc significantly enhanced elongation. The historical recon-structions of greening in the dark and of FR-HIRs in seedling

development (figs. 7, 8) indicate that, despite the ability of gym-nosperm seedlings to green in the dark (due to the presenceof a light-independent system for reduction of protochloro-

Fig. 7 History of greening in the dark on two seed plant phylogenetic trees. Likelihood character state reconstructions indicating the

probability of being green in the dark (green shading) or of requiring a light signal to green (no shading). A, Tree with conifers monophyletic andsister to gnetophytes. B, Tree with gnetophytes nested within conifers as sister to Pinaceae. Reconstructions were inferred using Mesquite

(Maddison and Maddison 2010) and the AsymmMk model, with root state frequencies equal to the equilibrium frequencies.

669MATHEWS & TREMONTE—PHYTOCHROME A IN EARLY ANGIOSPERMS

phyllide to chlorophyllide; Adamson et al. 1997; Armstrong1998; Reinbothe et al. 2010), seedling development is stronglylinked with light signals in most seed plants.

Our results refute the suggestion that positive selection atPHYA early in the history of angiosperms was linked to theorigin of FR-HIR-induced seedling de-etiolation, which might

Fig. 8 History of far-red high-irradiance responses on two seed plant phylogenetic trees. Likelihood character state reconstructions indicating

the probability of having short (green shading) or long (no shading) hypocotyls, epicotyls, or petioles in continuous far-red light. A, Tree rooted

between Ginkgo and cycads. B, Tree rooted between angiosperms and extant gymnosperms. Reconstructions were inferred using Mesquite

(Maddison and Maddison 2010) and the AsymmMk model, with root state frequencies equal to the equilibrium frequencies.

670 INTERNATIONAL JOURNAL OF PLANT SCIENCES

have enhanced the ability of angiosperms to colonize the un-derstory of ferns and gymnosperms (Mathews et al. 2003).However, it does not rule out the possibility that this episodeof selection was linked with the innovation in FR-HIRs earlyin the history of angiosperms. Two episodes of strong selec-tion, along with bursts of radical amino acid replacements, oc-curred in the PHYA lineage during its divergence from thePHYB lineage, whereas in contrast, the PHYB lineage was lit-tle impacted by positive selection or radical amino replace-ments (fig. 3; Mathews et al. 2003; Mathews 2010).

Based on our findings of the distribution of FR-HIRs, to-gether with the data on variable selection, we outline twoscenarios for further testing. (1) Innovation in FR-HIRs oc-curred early in the history of angiosperms, and it providedan adaptive advantage, even if the function did not originateat that time. (2) The two major activities of phyA that havebeen characterized in model systems, mediation of FR-HIRsand VLFRs, originated at different times, and their originswere associated with earlier and later episodes of selection,respectively. These two activities are genetically independent(Yanovsky et al. 1997, 2002), and it is possible that FR-HIRswere more important early in the history of seed plants, whereasVLFRs were more important early in the history of angio-sperms. PhyA acting in the VLFR mode detects light levelsthat other phytochromes cannot distinguish from darkness,relying on the accumulation of high levels of Pr in the dark(Casal et al. 1997; Sharrock and Clack 2002). The exquisiteresponsiveness to transient light signals provided by VLFRsmay have been advantageous for angiosperms, with their re-latively small seeds (Feild et al. 2003; Moles et al. 2005),allowing dark-imbibed seeds to utilize more efficiently the briefpulses of light delivered by sun flecks and during small-scaledisturbances.

These hypotheses can be tested by determining the phylo-genetic distribution of VLFRs in seed plants and by mutagen-esis experiments. Positively selected sites and sites at which

radical replacements occurred and were fixed in the PHYAgene lineage are candidate determinants of phyA-mediatedFR-HIRs and VLFRs. We are creating synthetic mutants, inwhich amino acids at positively selected sites are switchedback to their ancestral states for testing in an A. thalianamutant lacking a functional phyA. If the derived PHYA char-acter state is a determinant of FR-HIRs or VLFRs, then thesynthetic mutant, with the ancestral amino acid, will not res-cue the mutant phenotype. In this way, the link betweenfunctional divergence and molecular adaptation can be tested,providing reciprocal illumination to studies of protein func-tion and adaptive evolution (Dean and Thornton 2007).Studies of molecular evolution can highlight functional traitsthat may be less obvious candidates for key innovations thancarpels, net-veined leaves, and xylem vessels but that are can-didates for innovations that contributed to the early successof angiosperms faced with the challenge of colonizing the un-derstory of Mesozoic forests dominated by gymnosperms andferns.

Acknowledgments

We thank Patrick Griffith and the Montgomery BotanicalCenter, the Holden Arboretum, and Charles Tubesing, TimDevine (California State University), Harold W. Blazier(Ohio University), Ernesto Sandoval (Universitiy of Califor-nia–Davis), Teddi Bloniarz (University of Massachusetts–Amherst), Donald J. Padgett (Bridgewater State University),Peter Weston (Royal Botanic Garden, Sydney), Taylor S.Feild (University of Tennessee), Tim Brodribb, and Bill Cull-ina (NEWFS) for seeds, plant material, and/or germinationinformation. Special thanks to Gustavo Romero, Kanchi N.Gandhi, Jack Alexander, Tom Ward, Joel McNeal, DougGoldman, and Kurt Schellenberg. This work was supportedby NSF IBN-0214449 to S. Mathews and the Arnold Arbo-retum.

Literature Cited

Adamson HY, RG Hiller, J Walmsley 1997 Protochlorophyllide

reduction and greening in angiosperms: an evolutionary perspective.

J Photochem Photobiol B 41:201–221.

Armstrong G 1998 Greening in the dark: light-independent chloro-

phyll biosynthesis from anoxygenic photosynthetic bacteria to gymno-

sperms. J Photochem Photobiol B 43:87–100.

Barnes SA, NK Nishizawa, RB Quaggio, GC Whitelam, NH

Chua 1996 Far-red light blocks greening of Arabidopsis seedlings

via a phytochrome A–mediated change in plastid development.

Plant Cell 8:601–615.

Borthwick HA, SB Hendricks, MW Parker, EM Toole, VK

Toole 1952 A reversible photoreaction controlling seed germina-

tion. Proc Natl Acad Sci USA 38:662–666.

Botto JF, RA Sanchez, GC Whitelam, JJ Casal 1996 Phytochrome A

mediates the promotion of seed germination by very low fluences

of light and canopy shade light in Arabidopsis. Plant Physiol 110:

439–444.

Cantino PD, JA Doyle, SW Graham, WS Judd, RG Olmstead, DE Soltis,

PS Soltis, MJ Donoghue 2007 Towards a phylogenetic nomenclature

of Tracheophyta. Taxon 56:822–846.

Casal JJ, RA Sanchez, MJ Yanovsky 1997 The function of phyto-

chrome A. Plant Cell Environ 20:813–819.

Darwin C 1859 On the origin of species by means of natural se-

lection, or the preservation of favored races in the struggle for life.

Appleton, New York.Dean AM, JW Thornton 2007 Mechanistic approaches to the

study of evolution: the functional synthesis. Nat Rev Genet 8:

675–688.

Downs RJ, HW Siegelman 1963 Photocontrol of anthocyanin

synthesis in milo seedlings. Plant Physiol 38:25–30.

Facelli JM 2008 Specialized strategies. I. Seedlings in stressful

environments. Pages 56–78 in MA Leck, VT Parker, RL Simpson,

eds. Seedling ecology and evolution. Cambridge University Press,

Cambridge.Fankhauser C, JJ Casal 2004 Phenotypic characterization of a pho-

tomorphogenic mutant. Plant J 39:747–760.

Feild TS 2008 Regeneration ecology of early angiosperm seeds

and seedlings: integrating inferences from extant basal lineages

and fossils. Pages 130–149 in MA Leck, VT Parker, RL Simpson,

eds. Seedling ecology and evolution. Cambridge University Press,

Cambridge.

Feild TS, NC Arens 2007 The ecophysiology of early angiosperms.

Plant Cell Environ 30:291–309.Feild TS, NC Arens, TE Dawson 2003 The ancestral ecology of

671MATHEWS & TREMONTE—PHYTOCHROME A IN EARLY ANGIOSPERMS

angiosperms: emerging perspectives from extant basal lineages. Int JPlant Sci 164(suppl):S129–S142.

Feild TS, TJ Brodribb, A Iglesias, DS Chatelet, A Baresch, GR

Upchurch, B Gomez, et al 2011 Fossil evidence for cretaceousescalation in angiosperm leaf vein evolution. Proc Natl Acad Sci

USA 108:8363–8366.

Franklin KA 2008 Shade avoidance. New Phytol 179:930–944.

Fredericq H 1964 Influence formatrice de la lumiere rouge-fonce surle developpement des thalles de Marchantia polymorpha L. Bull

Soc R Bot Belg 98:67–76.

Grime JP 1979 Plant strategies and vegetation processes. Wiley,

New York.Grubb PJ 1977 The maintenance of species richness in plant

communities: the importance of the regeneration niche. Biol Rev

52:107–145.Harper JL 1977 Population biology of plants. Academic Press,

London.

Hartmann KM 1967 Ein Wirkungsspectrum der Photomorphogenese

unter Hochenergiebedingungen und seine Interpretation auf derBasis des Phytochrome. Z Naturforsch 22B:1172–1175.

Kami C, S Lorrain, P Hornitschek, C Fankhauser 2010 Light-

regulated plant growth and development. Curr Top Dev Biol 91:

29–66.Lewis PO 2001 A likelihood approach to estimating phylogeny from

discrete morphological character data. Syst Biol 50:913–925.

Maddison WP, DR Maddison 2010 Mesquite: a modular system forevolutionary analysis. Version 2.74. http://mesquiteproject.org.

Mancinelli AL 1994 The physiology of phytochrome action. Pages

211–269 in RE Kendrick, GHM Kronenberg, eds. Photomorpho-

genesis in plants. Kluwer Academic, Dordrecht.Mathews S 2006 Phytochrome-mediated development in land plants:

red light sensing evolves to meet the challenges of changing light

environments. Mol Ecol 15:3483–3503.

——— 2009 Phylogenetic relationships among seed plants: persistentquestions and the limits of molecular data. Am J Bot 96:228–236.

——— 2010 Evolutionary studies illuminate the structural-functional

model of plant phytochromes. Plant Cell 22:4–16.

Mathews S, JG Burleigh, MJ Donoghue 2003 Adaptive evolution inthe photosensory domain of phytochrome A in early angiosperms.

Mol Biol Evol 20:1087–1097.

Mathews S, MD Clements, MA Beilstein 2010 A duplicate generooting of seed plants and the phylogenetic position of flowering

plants. Philos Trans R Soc B 365:383–395.

McCormac A, G Whitelam, H Smith 1992 Light-grown plants of

transgenic tobacco expressing an introduced oat phytochrome Agene under the control of a constitutive viral promoter exhibit

persistent growth inhibition by far-red light. Planta 188:173–181.

Moles A, D Ackerly, C Webb, J Tweddle, J Dickie, M West-

oby 2005 A brief history of seed size. Science 307:576–580.Pagel M 1999 The maximum likelihood approach to reconstructing

ancestral character states of discrete characters on phylogenies. Syst

Biol 48:612–622.Quail PH, MT Boylan, BM Parks, TW Short, Y Xu, D Wag-

ner 1995 Phytochromes: photosensory perception and signal trans-

duction. Science 268:675–680.

Reed JW, A Nagatani, TD Elich, M Fagan, J Chory 1994 PhytochromeA and phytochrome B have overlapping but distinct functions in

Arabidopsis development. Plant Physiol 104:1139–1149.

Reinbothe C, ME Bakkouri, F Buhr, N Muraki, J Nomata, G Kurisu,

Y Fujita, S Reinbothe 2010 Chlorophyll biosynthesis: spotlight onprotochlorophyllide reduction. Trends Plant Sci 15:614–624.

Rockwell NC, Y-S Su, JC Lagarias 2006 Phytochrome structure andsignaling mechanisms. Annu Rev Plant Biol 57:837–858.

Salisbury EJ 1942 The reproductive capacity of plants: studies in

quantitative biology. Bell, London.Sanderson MJ 2002 Estimating absolute rates of molecular evolution

and divergence times: a penalized likelihood approach. Mol Biol

Evol 19:101–109.

Schluter D, T Price, AO Mooers, D Ludwig 1997 Likelihood ofancestor states in adaptive radiation. Evolution 51:1699–1711.

Sellaro R, M Crepy, SA Trupkin, E Karayekov, AS Buchovsky, C Rossi,

JJ Casal 2010 Cryptochrome as a sensor of the blue/green ratio of

natural radiation in Arabidopsis. Plant Physiol 154:401–409.Sellaro R, MJ Yanovsky, JJ Casal 2011 Repression of shade-

avoidance reactions by sunfleck induction of HY5 expression in

Arabidopsis. Plant J 68:919–928.Sharrock RA, T Clack 2002 Patterns of expression and normalized

levels of the five Arabidopsis phytochromes. Plant Physiol 130:

442–456.

Sheehan MJ, LM Kennedy, DE Costich, TP Brutnell 2007 Subfunc-tionalization of PhyB1 and PhyB2 in the control of seedling and

mature plant traits in maize. Plant J 49:338–353.

Shinomura T, A Nagatani, H Hanzawa, M Kubota, M Watanabe,

M Furuya 1996 Action spectra for phytochrome A– and B–specificphotoinduction of seed germination in Arabidopsis thaliana. Proc

Natl Acad Sci USA 93:8129–8133.

Smith H 1982 Light quality, photoperception, and plant strategy.Annu Rev Plant Physiol 33:481–518.

——— 1983 The natural radiation environment: limitations on the

biology of photoreceptors: phytochrome as a case study. Symp Soc

Exp Biol 36:1–18.——— 1995 Physiological and ecological function within the

phytochrome family. Annu Rev Plant Physiol Plant Mol Biol 46:

289–315.

——— 2000 Phytochromes and light signal perception by plant: anemerging synthesis. Nature 407:585–591.

Smith H, DC Morgan 1983 The function of phytochrome in nature.

Pages 491–517 in W Shropshire Jr, H Mohr, eds. Encyclopedia of

plant physiology: photomorphogenesis. Vol 16. Springer, Berlin.Stamatakis A 2006 Raxml-vi-hpc: maximum likelihood-based phy-

logenetic analyses with thousands of taxa and mixed models.

Bioinformatics 22:2688–2690.Takano M, N Inagaki, X Xie, N Yuzurihara, F Hihara, T Ishizuka,

M Yano, et al 2005 Distinct and cooperative functions of phy-

tochromes A, B, and C in the control of de-etiolation and flowering

in rice. Plant Cell 17:3311–3325.Westoby M, E Jurado, M Leishman 1992 Comparative evolutionary

ecology of seed size. Trends Ecol Evol 7:368–372.

Yanovsky MJ, JJ Casal, JP Luppi 1997 The VLF loci, polymorphic

between ecotypes Landsberg erecta and Columbia, dissect twobranches of phytochrome A signal transduction that correspond

to very-low-fluence and high-irradiance responses. Plant J 12:

659–667.Yanovsky MJ, JJ Casal, GC Whitelam 1995 Phytochrome A,

phytochrome B and HY4 are involved in hypocotyl growth

responses to natural radiation in Arabidopsis: weak de-etiolation

of the phyA mutant under dense canopies. Plant Cell Environ 18:788–794.

Yanovsky MJ, JP Luppi, D Kirchbauer, OB Ogorodnikova, VA

Sineshchekov, E Adam, S Kircher, et al 2002 Missense mutation in

the PAS2 domain of phytochrome A impairs subnuclear localizationand a subset of responses. Plant Cell 14:1591–1603.

672 INTERNATIONAL JOURNAL OF PLANT SCIENCES