DE1, a 12-base pair cis-regulatory element sufficient to confer dark-inducible and light down-regulated expression to a minimal promoter in pea.

We found a cis-regulatory element of 12 base pairs (bp) (GGATTTTACAGT) capable of conferring light responsiveness to a minimal promoter, CaMV 35S46, in pea. The 12-bp sequence is located in the 5' upstream region of the light down-regulated gene pra2, which encodes a small GTPase belonging to the YPT/rab family. Here we examined gain-of-function analyses using synthetic promoter-luciferase constructs in a transient assay and found that the 12-bp element alone was sufficient to confer dark induction, as well as light down-regulation on the minimal promoter. We named this dark inducible element DE1. Effects of various light conditions on the reporter gene activity showed that DE1 received signals from phytochrome A, phytochrome B, and blue light photoreceptors. Using phytochrome-deficient mutants, we showed that the pra2 protein level in seedlings was also regulated by these photoreceptors. The changes in the immunoblotting pattern of the pra2 protein in these mutants were correlated with the changes in epicotyl elongation. Results from transient assays using these mutants showed that the DE1 received signals from phytochromes A and B, demonstrating that this element is indeed a light-responsive element. To our knowledge, this is the first cis-element that by itself confers light responsiveness to a minimal promoter.

Plants use light not only as an energy source for photosynthesis but also as an environmental signal. Light signals received by photoreceptors control many aspects of plant development. There are several photoreceptors including the blue/ UV-A receptor called phototropin, the blue/UV-A light receiving cryptochromes that control processes such as cotyledon expansion and hypocotyl inhibition (1,2), and the red/farred absorbing phytochromes that control processes such as seed germination, hypocotyl growth, leaf expansion, and floral induction (1). Many of these developmental changes are accompanied by changes in gene expression. The mechanism of lightregulated transcription has been widely investigated, but the accumulated evidence has shown that the mechanism is very complex. For example, whereas cis-elements for up-regulation have been reported, there is no report of a cis-element that is sufficient to confer light responsiveness by itself to a minimal promoter (3). Compared with up-regulation, the mechanism for light down-regulation of transcription is even less clear. Some genes down-regulated by light have been identified such as AS1 (4), ATHB-2 (5), NPRs (6), TUB1 (7), and PHYA (8), but only a few cis-elements for down-regulation have been reported (8 -10), and the details remain unclear.
Previously we have reported that a small GTPase gene in pea, pra2, which belongs to YPT/rab family (11), is one of the genes whose expression is down-regulated by phytochrome (12). The pra2 gene is mainly expressed in the growing zone of etiolated epicotyls, and its expression is repressed when the plant is illuminated (10,13). Because small GTPases are molecular switches that are turned on by GTP and off by the hydrolysis of GTP to GDP, and members of YPT/rab family play important roles during intracellular transport, we propose that pra2 protein may participate in vesicle transport during stem elongation of etiolated seedlings (13). We were interested in the down-regulation of the pra2 gene, and we surveyed cis-elements located in the 5Ј upstream region using a reporter gene in a transient assay (10). We found that a 12-bp 1 sequence in the 5Ј upstream region of this gene was important for the light down-regulation of the reporter gene and that this response was mediated by phytochrome B. However, we do not know whether the 12-bp element alone confers light down-regulation to a minimal promoter and whether the 12-bp element receives signals from photoreceptors other than phytochrome B. To answer these questions, we have examined the role of the 12-bp element by gain-of-function analysis in a transient assay to determine the effect of light conditions on the reporter gene. We found that the 12-bp element was sufficient to confer light down-regulation to a minimal promoter and capable of receiving signals from various photoreceptors. We named this dark inducible element DE1.
Because the DE1 has a capacity to respond to various light signals, there is a possibility that expression of the pra2 gene is indeed regulated by various light signals. Previously we had only shown the involvement of phytochrome B (10). Here we used mutants to examine the involvement of other photoreceptors in the regulation of pra2 expression. In Arabidopsis, 5 phytochrome genes (PHYA-PHYE) have been isolated, and it has been revealed that each phytochrome has distinct but at times overlapping functions (14). Many photomorphogenic mutants have been isolated (15,16) and shown to be powerful tools for investigating light-regulated expression (17,18). In pea, several phytochrome mutants has been isolated. The pcd1 and pcd2 mutations block steps leading to the formation of the phytochrome chromophore, phytochromobilin (19,20), whereas the fun1 and lv mutants are deficient in phytochrome A and phytochrome B apoproteins, respectively (21,22). We have used these mutants to investigate whether photoreceptors, in addition to phytochrome B, are involved in the expression of the pra2 protein and whether DE1 confers light down-regulation as expected.
In this paper, we show that the DE1 is indeed a single light-responsive element and actually received signals from various photoreceptors.

EXPERIMENTAL PROCEDURES
Plant Materials and Growth Conditions-For the protein level measurements, pea seeds (Pisum sativum L.) were soaked in water and sown on irrigated rock wool. Seedlings were grown in the dark for 5 days at 25°C. Before light treatment, the epicotyl between 0 and 1 cm from the apical hook was marked by pen under a dim green safe light. Plants were then irradiated with brief red (2 min) or far-red (5 min) light, returned to darkness, and kept for 12 h. In the experiments to examine the high irradiance response (HIR), plants were continuously irradiated with the appropriate light for 12 h. After irradiation, the stem length was measured. Proteins were extracted from the upper 1-cm portion of the epicotyl.
For the transient expression assay, pea seeds were imbibed and sown individually in plastic pots. Seedlings were grown in the dark for 5 or 6 days at 25°C. Seedlings were then placed horizontally in the bombardment device (Model GIE-III, Tanaka Co. Ltd.). The bombardment procedure was as described by Inaba et al. (10). The gold particles were coated with a mixture of two kinds of plasmids, a plasmid containing one of the synthetic promoter-luciferase (LUC) constructs and a plasmid containing a 35 S-GUS construct as the internal standard. All steps during bombardments were performed in a darkroom under a dim green safe light. After bombardment, the seedlings were illuminated by the appropriate light conditions.
Construction of Synthetic Promoter-The 35S46-LUC vector was created by inserting the 35S46 promoter into the promoterless pBI221-LUCϩ vector. The 35S46 promoter was created by PCR. The oligonucleotide, 5Ј-TGAGGATTTTACAGTAATTGAGGATTTTACAGTAATTG-AGGATTTTACAGTAAT-3Ј (WT1), containing three copies of the 18-bp sequence, 5Ј-TGAGGATTTTACAGTAAT-3Ј, was synthesized, phosphorylated, and ligated. The 18-bp oligonucleotide contained the 12-bp cis-regulatory element with neighboring 3 bp at both ends. Then, the complementary oligonucleotide, 5Ј-ATTACTGTAAAATCCTCAAT-TACTGTAAAATCCTCAATTACTGTAAAATCCTCA (WT2), was annealed and subcloned into the EcoRV site of the pZErO-2 plasmid (Invitrogen, San Diego, CA). A plasmid containing nine tandem repeats of the 18-bp sequence was obtained. Its sequence was confirmed. To remove the sequence derived from the pZErO-2 vector, the nine tandem repeats were amplified by PCR using two primers that contained BamHI and EcoRV restriction sites. PCR products were purified by a QIAquick PCR purification kit (QIAGEN) and then digested by EcoRV and BamHI. The purified DNA fragment was subcloned into BamHI-EcoRV-digested 35S46-LUC vector. To make the 3 single-base pair mutated construct, the two oligonucleotides used were as follows: 5Ј-T-GAGGCTTTTCCCGTAATTGAGGCTTTTCCCGTAATTGAGGCTTTT-CCCGTAAT-3Ј (MT1) and 5Ј-ATTACGGGAAAAGCCTCAATTACGGG-AAAAGCCTCAATTACGGGAAAAGCCTCA-3Ј (MT2). The mutated construct was prepared by the same method.
Light Sources-We used light-emitting diodes (LEDs) as the light sources for continuous irradiation. LEDs used in the experiments were as follows: Red, STICK-mR LED ( max ϭ 660 nm at 30 mol m Ϫ2 s Ϫ1 ; TOKYO RIKA Co. Ltd., Tokyo, Japan); Far-red, STICK-mFR ( max ϭ 735 nm at 25 mol m Ϫ2 s Ϫ1 ); Blue, STICK-mB LED ( max ϭ 470 nm at 25 mol m Ϫ2 s Ϫ1 ). Equipment for the red/far-red reversible experiments was the same as described previously (10,12). The length of exposure to red was 2 min at 30.5 mol m Ϫ2 sec Ϫ1 , and far-red was 5 min at 36.5 mol m Ϫ2 sec Ϫ1 .
Extraction of Protein and Immunoblotting-Total protein for immunoblotting was extracted by grinding tissue in a mortar and pestle on ice with sand together with 0.3 ml of buffer and 3 stem pieces (between 0 and 1 cm below the apical hook). The buffer contained 125 mM Tris-HCl (pH 6.8), 6% SDS, and 20% glycerol. The mixture was heated at 100°C for 3 min and centrifuged. The supernatant protein was separated by SDS-polyacrylamide gel electrophoresis, blotted onto a nitrocellulose membrane, probed with monoclonal IgG against pra2 protein (13) and goat anti-mouse IgG conjugated to peroxidase (Bio-Rad), and developed with an ECL kit (Amersham Pharmacia Biotech).
Measurement of Enzyme Activity-The stem of the bombarded pea seedling (between 0 and 1 cm below the apical hook) was ground in liquid nitrogen using a chilled mortar and pestle. The powder was dispensed into a microcentrifuge tube and mixed with 300 l of the buffer, which consists of 100 mM potassium phosphate (pH 7.8), 1 mM dithiothreitol, 1% Triton X-100, and 1 mM EDTA and then centrifuged at 15,000 g at 4°C for 5 min. The supernatant was frozen at Ϫ80°C until the enzyme assay was conducted. Luciferase assays were performed as described by Miller et al. (23) using the Pica Gene luciferase assay kit (Wako, Osaka, Japan). Photon emission derived from LUC activity was counted by AUTO LUMAT LB953 (Berthold, Bad Wildbad, Germany). GUS assays were conducted using the method of Jefferson et al. (24), with some modification (25). 4-Methylumbelliferone solutions dissolved in 0.2 M Na 2 CO 3 were used as standards. All LUC values were normalized to the corresponding GUS values. Samples of at least 6 bombardments were independently assayed for each construct.

The 12-bp Element, DE1, Was Sufficient to Confer Light
Down-regulation-Previously, we identified a cis-element, a 12-bp sequence, GGATTTTACAGT, involved in phytochrome down-regulated expression of the pra2 gene of pea (10). Gainof-function analysis showed that the 93-bp sequence containing the 12-bp element could confer light down-regulation when fused to a heterologous, CaMV 35S90 promoter (cauliflower mosaic virus 35 S promoter of 90 bp from the transcriptional start site). The CaMV 35S90 promoter has the as-1 element, which interacts with the fused element and complicates interpretation of the results (3). It is necessary to confirm the role of the 12-bp element in light responsiveness using a minimal promoter, CaMV 35S46, that has 46 bp from the transcriptional start site and is a smaller version of CaMV 35S90. We constructed the reporter plasmid, pGF9, containing a ninetandem repeat of the cis-element, CaMV 35S46, and the luciferase gene (Fig. 1). Using the plasmid pGF9 in a transient assay, we tested whether the 12-bp element was sufficient to confer light down-regulation of the minimal promoter in a red/far-red reversible manner. After bombardment, we placed the plants under a range of light conditions, before incubating them for 12 h in darkness, prior to measurement of their luciferase activity ( Fig. 2A, pGF9). The expression of the reporter gene was highest in darkness (lane D). The brief red light treatment repressed the expression (lane R), and subsequent far-red light immediately after red light treatment reversed the effect (lane R/F). The 3 single-base pair mutated construct, pGF9M, in which the adenines in the 12-bp element were changed to cytosines, decreased to 20% of the level observed in the case of pGF9 in darkness and abolished the red/far-red reversible response. These results indicate that the 12-bp element is dark-inducible and sufficient to confer red/farred reversible light down-regulation to the minimal promoter. When the nine tandem copies of the cis-element were fused to CaMV 35S46 promoter in the reverse direction, we could also observe red/far-red reversible response, indicating that the function of this element is independent of its direction (data not shown). We named this dark inducible element DE1.
The DE1 Received Signals from Various Photoreceptors-The red/far-red reversible reaction in Fig. 2A is a low fluence response (LFR) usually mediated by phytochrome B (14). There is a possibility that DE1 also responds to light signals received by photoreceptors other than phytochrome B. To examine this possibility, we determined the effects of various continuous light irradiations on the expression of the reporter gene as shown in Fig. 2B. A response to continuous irradiation is termed an HIR, and red-HIR and far-red-HIR are thought to be mediated by phytochrome B and phytochrome A, respectively (14). Responses to blue light may be mediated by phytochrome A and blue light photoreceptors (1). We exposed plants to continuous red (lane Rc), far-red (lane Fc), or blue light (lane Bc) irradiation for 12 h and measured the reporter gene activity (Fig. 2B). Each treatment repressed the expression of the reporter gene (pGF9), compared with that of darkness (lane D), and the 3 single-base pair mutations abolished dark induction, as well as the HIR (pGF9M). These results suggest that the DE1 receives signals not only from phytochrome B, but also from phytochrome A and blue-light photoreceptors. It seems probable that the various photoreceptors transduce the light signal via the DE1.
Characterization of Phytochrome Mutants on pra2 Expression-To examine whether the expression of the pra2 gene is indeed regulated by various photoreceptors, we investigated pra2 expression using photoreceptor mutants. In pea there are several mutants deficient in various phytochromes. We used the fun1-1 and lv-5 mutants and the pcd1pcd2 double mutant. The pcd1pcd2 double mutant is blocked at two steps in chro-mophore biosynthesis (19,20). The fun1-1 and lv-5 mutants are deficient in phytochromes A and B, respectively (21, 22). We exposed these mutants to various light conditions and examined the pra2 protein level by immunoblotting (Fig. 3). When grown in complete darkness, all plants expressed the pra2 protein (lane D). Red/far-red light reversible down-regulation was observed in wild type and fun1-1 plants, but not in the lv-5 and pcd1pcd2 plants (Fig. 3, upper row). These results indicate that this LFR is not mediated by phytochrome A but is mediated by phytochrome B. The results from pcd1pcd2 plants show that chromophore biosynthesis is required for functional phytochrome B activity. This result is in agreement with expectations from our previous results (12,13).
To examine whether photoreceptors other than phytochrome B are involved in pra2 expression, the effects of continuous irradiation were also examined in these mutants (Fig. 3, lower  row). The effect of continuous red irradiation on pcd1pcd2 plants was different from that of wild type (lane Rc), indicating that phytochromes mediate this red-HIR as expected. Red-HIR is usually mediated by phytochrome B but not by phytochrome A (14), and the result from the phytochrome A-deficient fun1-1 plants was in agreement with this suggestion. However, there was some down-regulation in the phytochrome B-deficient lv-5 mutant, suggesting some other phytochrome(s) may also play a role in the red-HIR. The effect of continuous far-red on mutant lv5 was similar to that on wild type. The expression by fun 1-1 or pcd1pcd2 plants was reduced in comparison to wild type plants implicating phytochrome A in the far-red-HIR. The effect of continuous blue light on all three mutants was similar to that of wild type (lane Bc), indicating that phytochromes are not involved in this response. Presumably blue light photoreceptors are involved in this response. Thus, the mutant studies indicate that light received by phytochrome A, phytochrome B, and blue-light receptors affected the expression of the pra2 protein.
Previously we observed that the pra2 protein level was correlated with stem elongation in etiolated pea seedlings (13). To test this observation for the mutants, we measured the epicotyl elongation (Fig. 4) and compared it with the pra2 protein level represented in Fig. 3 (lower row). Changes in stem elongation were correlated with the pra2 expression pattern, consistent with the proposition that the pra2 protein may be involved in stem elongation.
Specific Response Mediated by the DE1 Was Diminished in Phytochrome Mutants-It is important to demonstrate that the DE1 is capable of receiving signals from photoreceptors using the above-characterized mutants. We examined the effect of light on the reporter gene expression by the transient assay using the plasmid pGF9. If the DE1 actually receives signals from a photoreceptor, deficiency of the photoreceptor should reduce the specific response mediated by it. In Fig. 3, we showed that a phytochrome A-deficient mutant, fun1-1, is deficient in the far-red-HIR, and a phytochrome B-deficient mutant, lv-5, is deficient in the LFR for pra2 expression. Here, we tested the effect of continuous far-red light or brief red light on the reporter gene expression of fun1-1 and lv-5 plants. We used the activities in darkness as the induction control and activities in continuous blue light as the repression control, because both activities in the two mutants were similar to those of wild type plants (Fig. 3). In both mutants, the activities of the reporter gene were highest in darkness and repressed by blue light (Fig.  5, lanes D and Bc). In the phytochrome A-deficient mutant fun1-1, response to continuous far-red light (lane Fc) was somewhat attenuated, whereas a significant response to brief red light (lane R) was retained. In the phytochrome B-deficient mutant lv-5, the response to brief red light (lane R) was atten-

DISCUSSION
In this report, we have shown that a 12-bp element, DE1, could confer light responsiveness to a minimal promoter, CaMV 35S46, using a reporter gene in a transient assay and that the element received the signal from phytochrome A, phytochrome B, and blue light receptors (Fig. 2). These findings were confirmed by the use of phytochrome-deficient mutants, because the DE1 lost the ability to confer light responsiveness only for the specific light conditions relevant to that photoreceptor (Fig.  5). It seems probable that various photoreceptors transduce the light signals to the DE1. To date there are some reports for gain-of-function analysis of various cis-elements. Most experiments have been done using the CaMV 35S90 promoter that has the as-1 element. The GT-1 tetramer can confer light responsiveness to the CaMV 35S90 promoter, but not to the CaMV 35S46 promoter, probably because the as-1 element is needed to build a light-responsive module with the GT-1-binding site (3). In case of light down-regulation, the RE3 could confer light responsiveness to the modified CaMV 35S46 promoter, probably interacting with the B domain of the 35 S promoter (9). Thus, combinatorial interaction of light-respon- The pGF9 construct was introduced into the growing region of etiolated pea shoots by particle bombardment with the 35 S-GUS construct as the internal standard. Plants were irradiated with various lights, and the reporter enzyme activity was measured after a 12-h incubation. D, dark; R, brief red; Fc, continuous far-red; Bc, continuous blue. Relative activity was defined in the legend to Fig. 2. Values are the means of at least six independently bombarded samples, with error bars representing S.E. sive elements determines the response characteristics of lightregulated promoters (18,26). The smallest sequence unit sufficient to confer light responsiveness to the CaMV 35S46 promoter has been a 52-bp element of the CHS unit I (3,27). This 52-bp element is long compared with the DE1. To our knowledge, DE1 is the first small cis-element reported that, by itself, confers light responsiveness to a minimal promoter, 35S46.
We propose that the cis-elements for light down-regulation should be divided into two groups. One is the repressor-binding site under light, and the other is the activator-binding site in darkness. Previously, we proposed that the 12-bp element regulates induction in darkness rather than repression by light (10). The present data show that introduction of the pGF9 construct resulted in dark induction of the reporter gene and that the 3 single-base pair mutations abolished dark induction rather than light-repressed expression (Fig. 2), supporting the hypothesis that the DE1 is an activator-binding site in darkness. Previous gel mobility shift assays also suggested the possibility that DE1 is an activator-binding site in darkness (10). Moreover, pra2 expression itself shows a dark-inducible pattern. When etiolated plants were illuminated for one day and then returned to darkness, pra2 expression was restored (13). Thus, these data support the view that DE1 is an activator-binding site in darkness. The RE1 element on the PHYA promoter and the RE3 element on the AS1 promoter have been identified as cis-elements involved in phytochrome downregulation, but these two elements are thought to be binding sites for transcriptional repressors (8,9). By contrast, DE1 may be responsible for dark induction rather than light down-regulation.
The pra2 protein level was regulated by phytochrome A, phytochrome B, and blue-light photoreceptors (Fig. 3). DE1 received signals from these photoreceptors, suggesting that DE1 is likely to regulate pra2 expression mediated by various photoreceptors. In lv-5 plants, the pra2 protein level did show a red-HIR, whereas the expression did not show an LFR. In general, both red-HIR and LFR are mediated by phytochrome B. However, our data suggest that the red-HIR of the pra2 gene is not mediated solely by phytochrome B. Because the various phytochromes have overlapping functions, it is not surprising that the red-HIR was regulated by other phytochromes. For example, the Arabidopsis phyB mutant retained an LFR for CAB gene expression, which is generally mediated by phytochrome B (17). In lv-5 plants, other phytochromes are likely to regulate the red-HIR for pra2 expression, as well as phytochrome B.
The observed correlation between pra2 gene expression and epicotyl elongation observed in the mutants is consistent with our previous hypothesis that the pra2 protein may be involved with epicotyl elongation in etiolated pea seedlings. One possibility is that the pra2 protein is induced in darkness and leads to stem elongation. Upon illumination, various photoreceptors send a signal to the DE1 and repress pra2 expression, which then results in reduced elongation. This hypothesis requires detailed testing, and the interaction of pra2 expression with the hormones involved with elongation deserves examination.
In conclusion we have shown that a 12-bp cis-regulatory element, DE1, confers dark induction to a minimal promoter. Introduction of the DE1 into other species should reveal whether this element is functionally conserved between species