Photoreceptors in Signal Transduction

Sunlight is the origin of nearly all the metabolic energy that drives life processes, and plants and algae are by and large responsible for the conversion of solar energy into chemical forms that are used by all organisms. These simple truths are perhaps obvious to plant biologists, but I think they merit enunciation every once in a while, if only to remind zoologists that, ultimately, plants rule. Even the fossil fuels that support much of modern human culture are owed to the photosynthetic activity of plants. Of course, plants have not evolved to photosynthesize on behalf of the less self-reliant inhabitants of the planet, but rather to provide for their own nourishment, growth, and development. Thus, virtually every aspect of plant life, from germination through reproduction and senescence, is affected by light.

The biology of many nonphotosynthetic organisms is also linked to sunlight. The reflective and refractive properties of light are visually perceived by many animals so that they may gauge within their varied and changeable environments the availability of food, water, shelter, and even reproductive partners, as well as evade the threat of predators and other dangers. Vertebrate sleep cycles, and sleep-like states even in invertebrates, are also linked to the sun's daily periodicity. Circadian rhythms, in fact, appear to be basic to organisms as diverse as fungi, plants, and mammals.

Not all aspects of sunlight, however, are conducive to life, and it is certain that the evolution of photoperception arose in large part from the selective advantage of being able to avoid the deleterious effects of solar radiation, as opposed to exploiting solar energy and periodicity. UV light is indeed one of the most common mutagens within the biosphere. Accordingly, a range of biomolecular mechanisms that mitigate the harmful effects of UV light occur throughout the biological kingdom.

One protective strategy relies on the photolyases, chromophore-bearing enzymes that repair UV-induced pyrimidine adducts within the genomes of bacteria, plants, and animals. An intriguing aspect of photolyases is that they themselves undergo photoactivation. By virtue of a light-harvesting chromophore—either a pterin or a deazaflavin derivative—that absorbs light from the blue portion of the visible spectrum, photolyases utilize light energy to reverse DNA damage such as thymine dimers. Specifically, the excitation energy imparted to the light-harvesting chromophore is transferred to an active site–bound molecule of (reduced) flavin adenine dinucleotide (FAD). In a free-radical mechanism, a single excited electron from FAD can be transiently transferred to a thymine dimer lesion, thereby cleaving the adduct, before returning back to regenerate the fully reduced FAD moiety (Park et al., 1995; see also Cashmore et al., 1999). In this way, organisms exploit for purposes of DNA repair the very same source (i.e., the sun) that induces mutagenic lesions.

The existence in photosynthetic organisms of molecular events that are light-regulated by virtue of the ATP and reducing equivalents that accumulate during photosynthesis has been recognized for some time. Enzyme activation, transcription, and translation are all subject to the photosynthetic energy charge of the plant cell. But photolyase is distinctively instructive as a specific example of how light cues can directly trigger a series of biomolecular events that are not directly related to photosynthetic metabolism. Admittedly, the series of events occur in the case of photolyase within a single enzymatic context: light energy is collected by a chromophore (i.e., a photoreceptor); the light energy is transferred transiently to a distinct molecular species (e.g., FAD); and a biomolecular reaction is effected (in this case, DNA repair). In other cases, however, activation of photoreceptors can result in a cascade of intermolecular events that qualify as bona fide mechanisms of signal transduction.

The phytochromes, photoreceptors that respond to red and far-red light, are the most familiar nonphotosynthetic plant photoreceptors, but the importance of blue and UV light–absorbing photoreceptors—the cryptochromes, in particular—is currently gaining wide recognition (see Cashmore et al., 1999). As in the case of the phytochrome family of proteins, cryptochromes appear throughout the plant kingdom and arise from multigene families: two cryptochrome-encoding genes have been identified in Arabidopsis, and three have heretofore been isolated in the fern, Adantium capillus-veneris. Sequence analyses, moreover, show the cryptochromes to be remarkably similar to the blue light–dependent photolyases discussed above. Cryptochromes also resemble photolyases in that both categories of blue light–absorbing protein extend beyond the plant kingdom; indeed, cryptochromes have been implicated in the entrainment of circadian behavior in Drosophila, humans, and mice. Although no cryptochrome to date has been found to possess DNA repair activity, the cryptochromes are presumed to have evolved from the photolyases in such a way that enzyme activity was discarded and photoreceptor activity was elaborated.

In Arabidopsis, both CRY1 and CRY2 have been implicated in developmental functions such as inhibition of hypocotyl elongation and cotyledon expansion, and CRY1 has been linked to additional processes of development, circadian cycle, and light-dependent gene regulation (see Ahmad et al., 1998). Current research into the photobiology of plants places emphasis on further specifying the roles of individual cryptochromes, with the ultimate goal being to elucidate varying cascades of blue light signal transduction. In this regard, the fascinating relationship between the photolyases and cryptochromes is sure to be informative. Sequence comparisons can speak not only to matters of phylogeny per se, but they also can potentially offer mechanistic insights into the structure–function bases of cryptochromes as signal transducers.

In this issue of THE PLANT CELL, on pages 81–95, Imaizumi et al. describe their ongoing efforts to understand cryptochrome photobiology. Specifically, they have identified two additional cryptochromes in A. capillus-veneris, bringing the total number of discovered cryptochrome genes in this fern to five and thereby establishing the largest family of cryptochromes to be yet defined. Besides possessing a set of variant cryptochromes that can be compared within the biological context of a single plant species, A. capillus-veneris represents a promising experimental organism in that several aspects of its development are regulated by blue light. Previous studies in the fern, moreover, have suggested a correlation between blue light responses and intracellular localization of the pertinent photoreceptor(s) (Kadota et al., 1986). By analogy to the emerging picture in which distinct phytochromes are differentially translocated into the nucleus during red light photoperception and signal transduction (e.g., Kircher et al., 1999), the intracellular localization of cryptochromes and the factors that may control localization are of particular interest.

Imaizumi et al. have examined the intracellular distribution of the five fern cryptochromes by creating functional cryptochrome–β-glucuronidase fusion proteins. In this way, a tendency for two of the cryptochromes, CRY3 and CRY4, to accumulate in the nucleus has been observed; CRY3 does so specifically in response to red light. The nuclear localization signal for both proteins appears to reside within the proteins' C-terminal domains; interestingly, the C termini of cryptochromes are not related to photolyases, the presumed evolutionary forerunners that function as enzymes but not as signal transducers proper.

The translocation of cryptochrome species into the nucleus echoes data confirming that phytochromes undergo translocation into the nucleus in response to red light (Kircher et al., 1999). Whether other aspects of cryptochrome signaling will prove to mirror phytochrome-mediated signal transduction remains to be seen. It is in fact only recently that mechanisms and components of phytochrome signal transduction, beyond the prosthetic tetrapyrrole chromophore, have been identified. In addition to the involvement of classic signal transducing agents such as G proteins, calcium, and calmodulin (Bowler et al., 1994), for example, it now appears that at least one phytochrome species has kinase activity (Fankhauser et al., 1999; see Reed, 1999). In Arabidopsis, moreover, a kinase that is related to a tumor suppressor in animals has been implicated in phytochrome signaling (Choi et al., 1999), as has been a nuclear protein that binds specifically to the C terminus of phytochromes (Ni et al., 1999).

Whether transduction cascades that initiate with cryptochromes involve signal transducers that are similar to those recruited by phytochrome remains to be seen. Nevertheless, Imaizumi et al. have made an important step toward refining our understanding of cryptochrome-regulated developmental events. Specifically, they have examined the age-dependent expression profiles of the five A. capillus-veneris CRY genes and their responses to light. Whereas CRY1, CRY2, and CRY3 are transcribed in a regulated manner throughout the A. capillus-veneris life cycle, expression of the newly described CRY4 and CRY5 genes differs strikingly over the course of development. CRY4, as opposed to CRY5, for example, is expressed preferentially in spores and is strongly downregulated in response to light. CRY5 transcripts, on the other hand, are virtually undetectable in spores but appear during other stages of the fern life cycle and are induced by light during spore germination. The analysis of these light effects in terms of blue and red/far-red wavelengths shows that both regions of the spectrum are linked to expression of the cryptochrome genes. Intriguingly, crosstalk between cryptochrome and phytochrome signal transduction pathways has been suggested previously in Arabidopsis (e.g., Neff and Chory, 1998), and a case can indeed be made for the involvement of phytochromes in the regulation of both CRY4 and CRY5 gene expression in A. capillus-veneris. This fern system can thus be regarded as an attractive model for further studying the dynamics of photoreceptor signaling in multiple aspects of plant development.