SUB1, an Arabidopsis Ca2+-Binding Protein Involved in Cryptochrome and Phytochrome Coaction

See allHide authors and affiliations

Science  19 Jan 2001:
Vol. 291, Issue 5503, pp. 487-490
DOI: 10.1126/science.291.5503.487


Cryptochromes and phytochromes are the major photosensory receptors in plants and often regulate similar photomorphogenic responses. The molecular mechanisms underlying functional interactions of cryptochromes and phytochromes remain largely unclear. We have identified an Arabidopsis photomorphogenic mutant,sub1, which exhibits hypersensitive responses to blue light and far-red light. Genetic analyses indicate that SUB1functions as a component of a cryptochrome signaling pathway and as a modulator of a phytochrome signaling pathway. The SUB1 gene encodes a Ca2+-binding protein that suppresses light-dependent accumulation of the transcription factor HY5.

Plants rely on multiple photosensory receptors to perceive changes of light quality and quantity and to regulate growth and development. The blue/ultraviolet-A light receptors (cryptochromes) and red/far-red light receptors (phytochromes) are major photoreceptors mediating light responses such as inhibition of hypocotyl elongation and stimulation of anthocyanin accumulation (1, 2). The molecular mechanism of photoreceptor signal transduction, especially that of cryptochromes, remains largely unclear. Recent studies have demonstrated that protein phosphorylation and transcriptional regulation are important mechanisms of phytochrome signal transduction (3–5). The involvement of calcium homeostasis has also been implicated in the signaling processes of both phytochromes and cryptochromes (6, 7). Although genes encoding phytochromes and cryptochromes appear to be evolutionarily unrelated, these two types of photoreceptors often elicit the same light responses. Moreover, for various light responses in different plant species, phytochromes and cryptochromes often affect each other's function, resulting in synergistic or antagonistic light responses (9). Such phenomena, collectively referred to as the coaction of phytochrome and cryptochrome (9), have also been found for the photomorphogenic responses in Arabidopsis(10–15). It has been reported that phytochromes and cryptochromes may physically interact to affect each other's activity and that the signaling molecules of one photoreceptor may modulate the function of another photoreceptor (13–18).

To investigate cryptochrome signal transduction, we sought to identify mutations affecting hypocotyl growth in blue light (19). One of the resulting mutants was referred to assub1 (short under blue light). However, sub1 was later found to have a short hypocotyl phenotype not only in blue light but also in far-red light (Fig. 1A). Thesub1 mutant shows no sign of photomorphogenic development in the dark, and it appears to grow normally in red light at fluence rates tested (Fig. 1) (20). In addition to hypocotyl inhibition, the sub1 mutant also exhibits hypersensitive light responses in cotyledon expansion and gene expression (20). For example, the blue and far-red light–induced expression of theCHS and CHI genes, encoding flavonoid biosynthetic enzymes, was elevated to a relatively higher level in thesub1 mutant than in the wild type (Fig. 1, B and C). The function of SUB1 is dependent on the light fluence rate. Thesub1 mutant demonstrates a more pronounced short-hypocotyl phenotype in blue light or far-red light of relatively low fluence rates (<10 μmol m−2 s−1) (Fig. 1, D to G). When grown in light of higher fluence rates (>10 μmol m−2 s−1), the relative difference in hypocotyl length between sub1 and wild-type seedlings diminished (Fig. 1, D to G), suggesting that SUB1 functions primarily in low light.

Figure 1

The sub1 mutant. (A) Five-day-old wild-type (WT) or sub1 mutant (sub1) seedlings were grown in continuous blue light (2 μmol m−2 s−1), far-red light (2 μmol m−2 s−1), or red light (4 μmol m−2 s−1). (B) An RNA blot showingCHS (chalcone synthase) and CHI (chalcone isomerase) gene expression in etiolated wild-type (WT) or sub1 plants transferred from dark to blue light (D to B; ∼40 μmol m−2 s−1) or to far-red light (D to FR; ∼40 μmol m−2 s−1), for 1, 3, or 6 hours. (C) The relative mRNA levels of CHS andCHI shown in (B) were normalized by the rRNA signal. (D to G) Response to fluence rate by hypocotyl growth of 5-day-old sub1, sub1cry2 (D),sub1cry1 (E), and sub1phyA (F and G). Twenty seedlings were measured for each sample, and the standard deviations are shown.

To study how SUB1 is involved in the cryptochrome function, we examined the genetic interactions of sub1 with cryptochrome mutants. When grown in blue light of relatively low fluence rates,cry2 and sub1 exhibited a long- or short-hypocotyl phenotype, respectively, whereas thesub1cry2 double mutant showed hypocotyl growth comparable to the sub1 parent (Fig. 1D). This result indicates thatsub1 is epistatic to cry2 and that SUB1 is likely to function downstream from the cry2 photoreceptor. Because cry2 itself functions primarily in low light, presumably because of the degradation of cry2 protein in high light (21, 22), it is not surprising that all three genotypes showed a less pronounced phenotype in high light (Fig. 1D). The sub1 and cry1mutations exhibited a more complex, epistatic relation dependent on fluence rate. When grown in blue light with relatively low fluence rates, the sub1cry1 double mutant exhibited a short-hypocotyl phenotype just like that of the sub1monogenic parent and the sub1cry2 double mutant (Fig. 1, D and E), suggesting that SUB1 also acts downstream from the cry1 photoreceptor. When plants were grown under blue light of higher fluence rates, the phenotype of the sub1cry1 double mutant became increasingly like its cry1 monogenic parent (Fig. 1E). These results are consistent with the hypothesis that, with respect to the hypocotyl inhibition, SUB1 normally functions in low light. The lack of SUB1 activity in high light may result from a light-dependent suppression of the expression or activity of SUB1. The observation that cry1 activity is dependent on SUB1 only in low light suggests that cry1 mediates multiple signaling pathways, resulting in an inhibition of cell elongation, and that the function of SUB1 is associated with a pathway that operates primarily in low light.

Because sub1 also showed an enhanced response to far-red light and phytochrome phyA is the major photoreceptor mediating far-red light responses (1), we next examined the genetic interaction between sub1 and phyA (Fig. 1, F and G). Compared with wild-type plants, the sub1 andphyA mutant seedlings grown in far-red light developed short and long hypocotyls, respectively, but the sub1phyA double mutant resembled the phyA parent at all the fluence rates of far-red light tested (Fig. 1F). Because phyA is also known to mediate hypocotyl inhibition in blue light (13, 23), especially in low light (Fig. 1G), we further analyzed howsub1 and phyA mutations interacted in blue light (Fig. 1G). In contrast to the sub1cry2 andsub1cry1 double mutants, the sub1phyA double mutant again showed a hypocotyl length very similar to that of thephyA parent in all the fluence rates of blue light tested (Fig. 1G). We conclude that phyA is epistatic tosub1 in both far-red light and blue light. These results suggest that the activity of cry2 and cry1 is dependent, at least partially, on SUB1, whereas the activity of phyA is not dependent on SUB1. Therefore, SUB1 is likely to act as a signal transducer of cry1 and cry2 but as a modulator of phyA signal transduction.

The sub1 loss-of-function mutation results from a transferred DNA (T-DNA) insertion in the 3′-end untranslated region of the SUB1 gene, which causes significantly decreased SUB1 mRNA expression, and consequently, a markedly lower SUB1 protein level in the sub1 mutant (19). Increasing the SUB1 level in sub1 mutant plants by transgenic expression of the SUB1 cDNA rescued the defects caused by the sub1 mutation (19). SUB1encodes a novel 552-residue polypeptide containing EF-hand-like Ca2+-binding motifs at the COOH-terminal region (Fig. 2A). SUB1 also has two regions enriched in basic residues that resemble nuclear localization signals (Fig. 2A). However, SUB1 does not seem to accumulate in the nucleoplasm. The SUB1-GUS fusion protein expressed in plant cells can be found throughout the cytosol, and it is apparently enriched in the nuclear periphery region surrounding the nucleus (Fig. 2B) (20). It is likely that SUB1 may be associated with nuclear envelope or endoplasmic reticular membranes. In addition to SUB1,Arabidopsis has at least two SUB1-like (SUL) genes, which we refer to as SUL1 andSUL2 (Fig. 2A). The conceptual translation products of theSUL1 and SUL2 genes are approximately 50% identical to that of SUB1. Genes showing high (>50%) amino acid identity toSUB1 are also found in other plants, including monocotyledons and conifers; but genes similar to SUB1 were not found in cyanobacteria, yeast, Caenorhabditis elegans, or Drosophila, for which the genomes have been completely sequenced (20). These results suggest that theSUB/SUL genes may be unique to terrestrial plants. LikeSUB1, SUL1 and SUL2 also contain EF-hand–like motifs in the COOH-terminal region. The EF hand is a Ca2+-binding motif composed of two α helices connected by a loop that coordinates Ca2+ binding (24,25). To investigate whether SUB1 may be a calcium-binding protein, we expressed and purified a COOH-terminal fragment of SUB1 (SUB1c) that contains EF-hand–like motifs (Fig. 2C). SUB1c indeed showed a calcium-binding activity in an in vitro45Ca2+-binding assay, in which proteins bound to a nitrocellulose are allowed to bind to the radioactive45Ca2+ (Fig. 2C). Compared with calmodulin, SUB1c has a lower affinity to Ca2+ (Fig. 2C). This may not be surprising because, like some other EF-hand proteins that have relatively lower affinity to Ca2+, the primary structure of the EF-hand of SUB1 deviates from that of the canonical EF-hand motifs found in calmodulin (24, 25).

Figure 2

SUB1 is a calcium-binding protein enriched in the nuclear periphery. (A) A comparison of the amino acid sequence of the Arabidopsis SUB1, SUL1, andSUL2 gene products. Boxed areas represent identical (black) or similar (gray) amino acids. Broken lines above the SUB1 sequence indicate basic regions resembling nuclear localization motifs. The hatched box connected by an underline indicates the EF-hand–like motif. Stars indicate residues potentially important for calcium binding. (B) Cellular localization of the GUS-SUB1 fusion protein in transiently transfected onion epidermal cells. Cells stained for GUS (left top) and DAPI (left bottom) are shown. An enlarged overlay (right) of the boxed areas is to highlight GUS stain in the nuclear periphery. Arrows indicate positions of the nucleus. (C) In the calcium-binding assay (31, 34), proteins (10 μg) were immobilized to a nitrocellulose membrane, incubated with radioactive 45Ca2+, washed, autoradiographed (bottom), and quantified for the45Ca2+ retained to the membrane by liquid scintillation (top). BSA, bovine serum albumin; SUB1c, purified SUB1 COOH-terminal fragment; CaM, bovine brain calmodulin. Inset shows the purified SUB1c (10 μg) fractionated in a 10% SDS-PAGE. Mr, molecular weight marker.

The sub1 mutation does not affect blue light–induced degradation of cry2 or phyA, nor the level of cry1 protein (Fig. 3A) (22,26). This is consistent with our hypothesis that SUB1 is a component of the cryptochrome signaling pathway that modulates phyA signal transduction. According to this hypothesis, SUB1 is a negative regulator of photomorphogenesis, whereas the cryptochromes suppress the activity of SUB1 to activate the light response (Fig. 3B). To account for the absence of a mutant phenotype in dark-grown sub1 plants, our model further predicts that SUB1 acts upstream of another component that is inactive in the dark. A bZIP transcription factor, HY5, appears to be a good candidate for such a component. The Arabidopsis hy5 mutant exhibits a long hypocotyl when grown in blue, red, or far-red light, but not in the dark (27, 28). It has been shown that HY5 undergoes COP1-dependent degradation in the dark and photoreceptor-dependent accumulation in light, and that the light-induced accumulation of HY5 protein correlates with light inhibition of hypocotyl growth (29, 30). These results indicate that HY5 acts downstream from both phytochromes and cryptochromes and that HY5 is inactive in the dark. To test whether SUB1 acts on HY5 by affecting the expression of HY5, we compared the expression of HY5 protein in wild-type and sub1 mutant plants. As previously reported, the HY5 protein starts to accumulate when etiolated seedlings are exposed to blue light. However, the light-induced accumulation of HY5 protein occurs much faster in thesub1 mutant than in the wild type (Fig. 3A). This result is consistent with SUB1's being a negative regulator of the light-induced accumulation of HY5 protein. Furthermore, a comparably low level of HY5 protein is detected in the dark-grown sub1 mutant and wild-type plants (Fig. 3A), which explains why sub1seedlings exhibit normal hypocotyl elongation in the dark. The hypothesis that SUB1 may act upstream of HY5 is further confirmed by an analysis of the sub1hy5 double mutant. In comparison with the wild type, the sub1hy5 double mutant showed long hypocotyls in both blue and far-red light, indicating thathy5 is epistatic to sub1 (Fig. 3C).

Figure 3

SUB1 is a negative regulator of HY5. (A) Immunoblots showing that the light-regulated expression of HY5, but not PHYA or CRY2, is affected by the sub1mutation. Five-day-old seedlings grown in the dark (0 hour) were transferred to blue light (∼40 μmol m−2s−1) for 1, 3, or 6 hours, and total protein extracts were analyzed by immunoblots probed with respective antibodies indicated at the right side. (B) A working model depicting the function of SUB1 in photoreceptor signal transduction. Arrows and T-bars represent positive or negative effect, respectively. Broken lines indicate that there may be more steps between the two indicated genes. (C) Light-inhibition of hypocotyl growth of the sub1hy5 double mutant. Seedlings were grown in continuous blue (2 μmol m−2 s−1) or far-red (4 μmol m−2 s−1) for 5 days. Hypocotyl lengths of 20 seedlings were measured for each sample, and standard deviations are shown.

SUB1 defines a point of crosstalk between cryptochrome and phyA signal transduction pathways. The position of SUB1 in a cryptochrome signaling pathway appears to be between photoreceptors and HY5. The finding that SUB1 is a calcium-binding protein suggests that SUB1 plays an important role in photomorphogenic responses resulting from the light-induced changes in ion homeostasis. Elucidation of the biochemical mechanisms of SUB1 would further our understanding of how photoreceptors function in the cell and how signaling from different photoreceptors interacts.

  • * To whom correspondence should be addressed. E-mail: clin{at}


Stay Connected to Science

Navigate This Article