Plant Photobiology 2001

Light has long been known to be an important signal for plant development, influencing nearly all aspects of the life cycle from germination to flowering. From May 30 to June 2, 2001, the mechanisms by which plants perceive and respond to their light environments were discussed with great excitement at Plant Photobiology 2001, hosted by the University of Missouri Interdisciplinary Plant Group at its 19th Annual Missouri Symposium. Nearly 150 researchers from 15 different countries converged on the small midwestern city of Columbia, Missouri to share their passion for plant photobiology. From our observations and from the comments of participants before, during, and after the meeting, it appears that Plant Photobiology 2001 was successful far beyond our hopes. So before we say anything about the science, we thank those who contributed to the success of this meeting. First, thanks to the University of Missouri Conference Office and members of the Interdisciplinary Plant Group who helped—students, postdocs, and Professor Doug Randall in particular—without whom this meeting would never have happened. Second, a humble thanks to all of the participants, whose presence, enthusiasm, and collegial natures justified all of the hard work of those previously mentioned.

“SPECKLES, SPOTS, AND DOTS”: INSIGHTS INTO PHOTORECEPTOR STRUCTURES, FUNCTIONS, AND LOCALIZATIONS

The first step in any plant photomorphogenic response is the perception of an incoming light signal. These signals can come in many forms, varying in both quality and quantity. To date, three classes of plant photoreceptors have been identified at the molecular level: the red/far-red-light–absorbing phytochromes (Fankhauser, 2001) and two distinct classes of blue light receptors, the cryptochromes and the phototropins (Briggs et al., 2001; Christie and Briggs, 2001). In this section, we will summarize the most significant new insights into the structures, functions, and localizations of these important plant proteins.

Phytochromes

Higher plant phytochromes are dimeric chromopeptides (monomer sizes of 120 to 130 kD) that can be divided into two major functional regions: an N-terminal “sensor domain” in which chromophore binding occurs, and a C-terminal “output domain” that contains two Per/Arnt/Sim (PAS)-related domains and a histidine kinase–related domain (Fankhauser, 2001). Signaling from the output domain is regulated through the interconversion of the molecule between its red light–absorbing “inactive” Pr form and its “active” far-red-light–absorbing Pfr form, a process dependent on the sensor domain (Fankhauser, 2001).

Phytochromes and phytochrome-like proteins are found in a broad assemblage of organisms ranging from nonphotosynthetic eubacteria to higher plants (Davis et al., 1999). One interesting difference between the phytochrome and phytochrome-like photoreceptors of different organisms that was discussed at this conference is the utilization of different linear tetrapyrrole bilins as chromophores. For example, Rick Vierstra (University of Wisconsin, Madison) presented evidence that the bacteriophytochromes present in Deinococcus radiodurans and Pseudomonas aeruginosa may use biliverdin as their native chromophores, whereas Tilman Lamparter (Freie Universität Berlin, Germany) and J. Clark Largarias (University of California, Davis) showed that the Synechocystis phytochrome Cph1 is likely to use phycocyanobilin (PCB) as its chromophore. Preliminary results from David Kehoe's laboratory (Indiana University, Bloomington) suggest that PCB also is likely to represent the chromophore for the phytochrome-like protein RcaE involved in chromatic adaptation of Fremyella diplosiphon. Although higher plant phytochromes generally are thought to use phytochromobilin (PφB) as their native chromophore (Fankhauser, 2001), Masaki Furuya's group (Hitachi Advanced Research Laboratory, Saitama, Japan) recently obtained data suggesting that phytochrome B (phyB) may use PCB in addition to PφB.

Why such a diversity of bilin chromophores? Certainly, there is a need for different photosensory capacities among the various organisms containing phytochrome and phytochrome-like molecules. Rick Vierstra speculated that the shift in chromophore use is related to the need of higher plants to efficiently sense neighbors by monitoring red/far-red ratios. Hence, whereas biliverdin-containing bacteriophytochromes are quite good far-red sensors, they are poor red sensors and would not be expected to exhibit good ratio sensing. The PCB-utilizing phytochromes of cyanobacteria can function as red/far-red sensors; however, given their blue-shifted absorption spectra, they are not as good at discriminating between red and far-red light as PφB-containing phytochromes of higher plants.

Although the photosensory action of phytochromes in different taxa may be related to differences in chromophore use, the way in which functional differences arise between duplicated phytochromes within a single organism is still an open question. Bob Sharrock (Montana State University, Bozeman) discussed the use of transgenic Arabidopsis overexpressing chimeric phytochromes as a tool to understand the functional differences between phyB and phyD, proteins that exhibit ∼80% amino acid identity (Clack et al., 1994). At least one determinant in the differential activity of these receptors may be related to regions of the proteins spanning the C-terminal region of the “photosensory domain” and the PAS-related domains. In particular, Sharrock's group has observed that overexpression of a chimeric phytochrome containing this region from PHYB confers apparent phyB function to an otherwise phyD molecule, whereas the reverse situation does not result in a phyB overexpressor phenotype.

Upon the discovery that phytochromes can migrate to the nucleus in response to light (Sakamoto and Nagatani, 1996), a dramatic conceptual and experimental shift in how we study phytochromes and their action was made. One area of interest has been in understanding how the movement of phytochromes between the cytoplasm and the nucleus is regulated and how this process relates to particular photomorphogenic responses. Eberhard Schäfer (Albert-Ludwigs Universität, Freiburg, Germany) provided a progress report from his laboratory's studies of the intracellular localization and partitioning of phyA in response to different light conditions. First, nuclear translocation of a phyA–green fluorescent protein (GFP) fusion protein can occur in response to either a single pulse of far-red light or continuous far-red light. However, a large portion of the phyA-GFP pool apparently is retained within the cytoplasm under both irradiation conditions. Schäfer also reported that these two pools of phyA are different in both their visible appearance—with nuclear phyA forming clear “speckles” and cytosolic phyA being diffuse—and their stability—with nuclear phyA exhibiting very high stability compared with cytosolic phyA. Although a diffuse signal of phyA was predominant in the cytoplasm after far-red treatment, pulsed red light given concomitant with continuous far-red light resulted in the formation of numerous cytosolic speckles.

The formation of nuclear phyA speckles appears to be under circadian control to some extent according to Schäfer. Ferenc Nagy (Biological Research Center, Szeged, Hungary) reported that the nuclear localization of phyB also is regulated, at least in part, by the activity of the circadian oscillator, because the numbers and size of the nuclear speckles varies diurnally. The physiological role of such nuclear and cytoplasmic speckles remains unknown. However, given phytochrome's ability to interact directly with transcription factors (Martinez-Garcia et al., 2000), it is possible that the speckles represent foci for transcriptional activity, as has been observed for nuclear speckles in metazoan systems (Wei et al., 1999). Interestingly, Nagy also reported that physiologically inactive mutant versions of phyB do not form nuclear speckles.

Cryptochromes

The cryptochrome (cry) blue light receptors were identified first in plants but have been found since in a broad assemblage of organisms, including humans (Christie and Briggs, 2001). Although the N-terminal regions of both cry1 and cry2 are similar to DNA photolyases, the C-terminal regions are relatively novel and diverged between the two cryptochromes (Christie and Briggs, 2001). It has been found that overexpression of the C-terminal region of cry1 and cry2 confers a constitutive photomorphogenic (COP) phenotype in dark-grown Arabidopsis seedlings, leading to the hypothesis that the photolyase-like domain normally acts as a repressor of the C-terminal domain in darkness and that upon irradiation, this repressive action is relieved (Yang et al., 2001). Tony Cashmore (University of Pennsylvania, Philadelphia) reported that in yeast and in vitro, the C-terminal region of cry1 interacts with COP1, a nuclear protein that is key to the switch from skotomorphogenesis to photomorphogenesis (Ang et al., 1998). Because cry1 translocates to the nucleus in response to light (Christie and Briggs, 2001), these results suggest potential direct connections between cry1 and the nuclear events mediating the deetiolation program. Relative to cry2 signaling events, Alfred Batschauer (Phillips-Universität, Marburg, Germany) reported the identification, by yeast two-hybrid screening, of two cry2-interacting proteins, an F-box protein and a novel protein that has been designated ROC1. Analysis of ROC1 function by overexpression suggests that it functions as a negative regulator of cry2. However, the way in which ROC1 potentially represses cry2 function is unclear at present, because ROC1 appears to be a cytosolic protein, in contrast to the apparent constitutive nuclear localization of cry2 (Christie and Briggs, 2001).

Phototropins

Like the phytochromes and cryptochromes, the phototropin (phot) blue light receptors, phot1 and phot2, can be split into two major polypeptide domains: an N-terminal sensory domain containing two PAS-like domains, LOV1 (light, oxygen, and voltage) and LOV2, that serve as flavin mononucleotide binding sites, and a C-terminal serine/threonine protein kinase domain (Christie and Briggs, 2001). Although phot1 was identified originally as a sensor for low-fluence-rate phototropic stimuli, it has since been found to play a role in the movement of chloroplasts under low- light conditions (Sakai et al., 2001) as well as in the rapid blue light–induced inhibition of hypocotyl growth (Folta and Spalding, 2001). phot2 has been shown to play a role in phototropism under high-fluence-rate conditions (Sakai et al., 2001) and to mediate the high light–induced chloroplast avoidance response (Jarillo et al., 2001; Kagawa et al., 2001; Sakai et al., 2001).

Although it is assumed that light activation of the phototropin kinase domain is important for signaling (Liscum and Stowe-Evans, 2000; Christie and Briggs, 2001), little is known about the intramolecular reactions occurring within the phototropins upon light absorption. However, given the recent crystallization of an LOV2 domain (Crosson and Moffat, 2001), this is not likely to be the case for long. In a talk befitting a top faculty candidate, third year graduate student Sean Crosson (University of Chicago) reported on his latest studies with the LOV2 crystals. Surprisingly, these crystals remain photoactive, thus allowing Crosson to demonstrate the formation of a flavin C(4a)–cysteinyl adduct within the LOV2 domain in response to blue light, something previously proposed from absorption and fluorescence spectroscopy (Salomon et al., 2000). Although previous circular dichroism measurements suggested that LOV2 undergoes a conformational change in response to the formation of this adduct (Salomon et al., 2000), no obvious change in conformation has been observed in the LOV2 crystals used by Crosson.

Preliminary results presented by Winslow Briggs (Carnegie Institution of Washington, Stanford, CA) suggest that phot1 will join the list of mobile photoreceptors. Using transgenic Arabidopsis expressing phot1-GFP, the Briggs laboratory has confirmed that phot1 is plasmalemma associated with dark-grown seedlings. Although expression of phot1 appeared fairly ubiquitous throughout the plant, higher GFP signals were observed in vascular parenchyma cells and in the end walls of hypocotyl and root cells. These patterns of apparently concentrated phot1 are curiously similar to the patterns of expression for the putative auxin efflux carriers PIN1 and PIN2 (Morris, 2000). Examination of phot1-GFP signals in blue light–irradiated seedlings also has revealed that at least some portion of the phot1 protein becomes cytosolic in response to light. Mannie Liscum (University of Missouri, Columbia) proposed that the phot1-interacting protein NPH3 (Motchoulski and Liscum, 1999) might act as a molecular scaffold to bring phot1 together with the colocalized auxin transporter proteins. In this way, activated phot1 might phosphorylate the auxin carrier, thus changing its activity and promoting the generation of a lateral auxin gradient across the curving organ, an apparent prerequisite for tropic responses (Liscum and Stowe-Evans, 2000). The mobility of phot1 in response to light could be representative of other intracellular protein movements (e.g., shifting of PIN localization from basal to lateral walls) or of some sensor sensitization/regeneration event. Like any exciting new discovery, the phot1 expression/localization studies have raised more questions than they have answered.

NEWLY IDENTIFIED LOCI INVOLVED IN PHOTOMORPHOGENIC SIGNALING

Mutational and two-hybrid screening approaches continue to represent the most frequent sources of new players in photomorphogenic signaling. In this section, we will provide an overview of new loci that have been identified by such methods, restricting our discussion to genes/proteins whose characterizations remain unpublished or that appeared in print only within a couple of months of the meeting. Of 16 genes/proteins presented at this meeting that appear to represent new photomorphogenic signaling molecules, 12 appear to function primarily in phytochrome signaling pathways, 5 (EID1, EVE1, MAP7, MAP8, and PKS2) in phyA pathways, 4 (DSR1, ELF3, SHL5, and SOB1) in phyB pathways, and 3 (DDWF1, PAB1, and PRP1) being used by multiple phytochromes. The remaining four genes/proteins discussed at this meeting (SHL1, SHL2, SHL3, and SHL4) appear to function downstream of both phytochromes and cryptochromes (Pepper et al., 2001).

Apparent Phytochrome A Signaling Molecules

Eberhard Schäfer discussed the recent cloning and characterization of EID1 (mutant designation, empfindlicher im dunkelroten Licht 1) (Dieterle et al., 2001). EID1 is an F-box protein that appears to be involved in targeting positively acting components of phyA signaling pathways for proteolysis. The EVE1 locus (mutant designation, enhanced very low fluence response 1) was reported by Jorge Casal (University of Buenos Aires, Argentina) to be allelic to DIMINUTO/DWARF1, which encodes a brassinosteroid biosynthetic enzyme (Klahre et al., 1998). The MAP7 and MAP8 loci (mutant designation, modifier of arf7 phenotypes) were proposed by Emily Stowe-Evans (University of Missouri, Columbia/Indiana University, Bloomington) to function in phyA-dependent enhancement of phot1-mediated phototropism. Neither gene has been cloned yet. As Christian Fankhauser (Université de Geneva, Switzerland) discussed, PKS2, identified by its homology with PKS1 (phytochrome kinase substrate 1) (Fankhauser, 2001), can interact with both phyA and phyB in yeast and appears to act as a negative regulator of very-low-fluence responses.

Apparent Phytochrome B Signaling Molecules

Mutations in the DSR1 locus confer decreased sensitivity to red light and resemble weak phyB mutants. Christian Fankhauser reported that DSR1 encodes a novel protein with no apparent homolog in Arabidopsis. As discussed by David Alabadi (Scripps Research Institute, La Jolla, CA), ELF3 was cloned recently and found to encode a novel nucleus-localized protein that functions in an input pathway to the circadian oscillator (Covington et al., 2001; Hicks et al., 2001) as well as in a phyB pathway regulating stem elongation (Liu et al., 2001). In a large screen for seedlings hyper-responsive to light (SHL), Alan Pepper's group (Texas A&M University, College Station) identified the SHL5 locus that appears to function as a negative regulator of phyB signaling (Pepper et al., 2001). The molecular identities of the SHL proteins are unknown at present. Finally, Michael Neff (Washington University, St. Louis, MO) presented results showing that gain-of-function mutations in the SOB1 locus (mutant designation, suppressor of phyB) are capable of suppressing phenotypes conferred by both weak and null phyB mutations. Neff's group has cloned SOB1 and found that it encodes a Dof-type transcription factor.

Molecules Apparently Involved in Signaling from Multiple Phytochromes

Pill-Soon Song (Kumho Life and Environmental Science Laboratory, Kwangju, Korea/University of Nebraska, Lincoln) discussed the recent identification of DDWF1 (dark-induced DWF-like protein 1) by two-hybrid screening with a small G protein, Pra2, which has been shown to be involved in phytochrome-regulated hypocotyl growth (Kang et al., 2001). Because DDWF1 is a cytochrome P450 involved in brassinolide synthesis, it may represent a potential direct connection between phytochrome and brassinosteroid signaling. Meng Chen (Salk Institute, La Jolla, CA) described the isolation of PAB1 (phytochrome and actin-binding protein 1) and PRP1 (phytochrome-related phosphatase 1) by two-hybrid screening using the C-terminal output domain of phyB as bait. PRP1 has been shown to exhibit serine/threonine protein phosphatase activity, and the phenotypes of loss-of-function prp1 mutants suggest that PRP1 acts as a positive regulator of phyA signaling and a negative regulator of phyB signaling. Phenotypic analysis of pab1 loss-of-function mutants and overexpressing lines suggest that PAB1 functions as a positive regulator of both phyA- and phyB- dependent responses. Chen also reported that the PAB1-CFP fusion protein is found both in the nucleus and associated with the endoplasmic reticulum (ER). Interestingly, he reported a similar ER association for cytoplasmically localized phyB-GFP. The importance of such an ER association for phyB and PAB1 remains to be determined.

THE PHOTOSIGNALING SUPERHIGHWAY: INTEGRATIVE WEBS AND NETWORKS PREVAIL

One of the major themes emerging in plant photobiology and in plant biology as a whole is that of interacting networks of signal-response pathways. Much of Jorge Casal's presentation was devoted to describing the complex interactions his group has observed among and between the different photoreceptor signaling pathways. Such interactions can be positive, negative, additive, or synergistic, and they can be different for one pair of receptor pathways for different output responses (Casal, 2000). Casal concluded his talk by proposing that the most important photomorphogenic response of a plant in the natural environment is the shift from a heterotrophic to an autotrophic growth form and that the many observed photoresponse pathway interactions may serve to reduce the noise of the light environment such that plants can respond most effectively to the dark-to-light transition that induces this developmental change.

As should be obvious from many of the new photomorphogenic loci reported at the meeting, the networks include not only photoreceptor pathways but also hormone response pathways, most notably brassinosteroid and auxin pathways. At least six of the speakers at Plant Photobiology 2001 could have been transplanted to a Plant Growth Regulator Conference and retained a captive audience. Jason Reed (University of North Carolina, Chapel Hill) presented an update on his laboratory's studies to understand the connection between phyB and auxin responses. Several of the genes Reed's group has identified in mutant screens for suppressors of weak phyB mutations have turned out to encode members of the Aux/IAA protein family, which appear to function as repressors of auxin-induced gene expression (Liscum and Reed, 2001). Reed reported that his group recently found that Escherichia coli produced a SHY2/IAA3 (mutant designation, short hypocotyl 2)-CBD fusion protein that can be used to “pull down” phyB from plant extracts, providing, as in the case of Pra2-DDWF1, a potential direct connection between a photoresponse and a hormone response pathway. Emily Stowe-Evans presented her work showing interactions between two receptor pathways, phot1 and phyA, and auxin response pathways (Stowe-Evans et al., 2001). Studies in Karen Halliday's laboratory (Bristol University, UK) have revealed yet another road of the “superhighway,” namely, the influence of temperature responses on auxin and phytochrome pathway interactions. It is abundantly clear from the studies presented at this meeting that photomorphogenesis represents not just a collection of receptors, their functions, and a few downstream elements arranged in neatly packaged linear response pathways but rather a broad integration of multiple environmental inputs, of which light is just one.

THE BIG PICTURE: MAKING SENSE OF THE PHOTOSIGNALING SUPERHIGHWAY

One of the most difficult tasks currently facing the plant photobiology community is how to experimentally approach the vastness and complexity of the interactive network of response pathways regulating photomorphogenesis. For example, what kinds of experiments do we initiate that go beyond one gene–one response questions? And once high-throughput photobiology is being done, how do we put the results obtained back into the context of the plant?

Functional Genomics for High-Throughput Photobiology?

With the recent completion of the Arabidopsis genome sequencing project, a variety of whole-genome functional genomics approaches have opened up (Theologis, 2001) that reductionist plant photobiologists should look to as a potential means of asking larger, more encompassing questions about photomorphogenesis. Talks by Peter Quail (Plant Gene Expression Center/University of California, Berkeley) and Xing-Wang Deng (Yale University, New Haven, CT) illustrated this point forcefully. Both laboratories have been using DNA microarray technology to determine what global changes in gene expression occur in response to a variety of light conditions. Quail's group has used Affymetrix chips (http://www.affymetrix.com/products/arabidopsis.html) to establish expression profiles under different far-red light irradiation conditions. By comparing wild-type and phyA profiles, they have been able to establish that 812 genes (of the 8600 represented on the array) exhibit reproducible phyA-dependent changes in expression in response to far-red light (Tepperman et al., 2001).

Deng's group has used customized DNA chips (http://info.med.yale.edu/wmkeck/dna_arrays.htm) to examine profiles under a number of conditions (dark versus white, red, blue, and far-red light; dark/light transitions; light/dark transitions) in wild type, loss-of-function photoreceptor mutants, and photoreceptor-overexpressing lines. One of the more interesting findings of the Deng studies is that large numbers of biologically coordinated genes exhibit coordinate changes in expression in response to photoreceptor signaling, quite consistent with what has been reported already for the circadian regulation of gene expression (Harmer et al., 2000). The Quail and Deng laboratories now face the monumental task of wading through all of the data to determine what is biologically meaningful and what is not.

Can Physiological Ecology and Ecological Genetics Bring Together Reductionist Biology and Real World Biology?

We think that those in attendance at Plant Photobiology 2001 would agree that on the basis of the last session of the meeting, the answer to the above question is a resounding yes. In what was a highlight of an already impressive meeting, several of our colleagues keyed the ecological lockbox of photobiology, approaching photomorphogenesis in a way that was both informative and eye-opening (spoken from the perspective of reductionist biologists).

Carlos Ballare (University of Buenos Aires, Argentina) presented recent work from his group aimed at understanding the photomorphogenic effects of UV-B light on plants in the natural environment. By coupling bench and field studies, Ballare's group has been able to determine that UV-B has positive influences on plant growth through modification of phyB-dependent processes (Boccalandro et al., 2001) and that plant defenses against herbivory also are enhanced by UV-B. Joanna Schmitt (Brown University, Providence, RI) discussed the work of her group in assessing the fitness benefits and costs of the phytochrome-dependent shade avoidance response in field grown Arabidopsis and Impatiens capensis (touch- me-not). In short, Schmitt and colleagues have found that the shade avoidance response does provide an adaptive advantage in both species, although it can be offset by costs incurred by other responses occurring in the field, such as water stress. Cynthia Weinig (Brown University/University of Minnesota, Minneapolis) presented her quantitative trait locus analysis of multiple environmentally influenced traits (e.g., timing of flowering) in the Landsberg erecta and Columbia accessions of Arabidopsis under both controlled growth (phytotron) and natural field conditions. Weinig showed, very convincingly, how her studies and others like them have the power to uncover traits important for plant fitness and to identify the genes responsible for environmentally dependent variations in those traits.

PLANT PHOTOBIOLOGY 2003: MARBURG, GERMANY

All in all, Plant Photobiology 2001 was a great success. We had a terrific mix of science, from structural biology and molecular genetics to ecology and evolution, and from microorganisms to higher plants. Although the speakers represented a wide variety of scientific thinking, we believe that by the end of the meeting there was a great appreciation for all of the science, which we hope will be cultivated into productive collaborations. With that, we extend the invitation from the plant photobiology community for readers to attend the next meeting, tentatively planned for 2003 in Marburg, Germany.