INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Light has three main effects on plant development (for review, see Mustilli and Bowler, 1997; Batschauer, 1998). First, it is the source of energy that fuels growth through photosynthesis. Second, light is a developmental signal that modulates morphogenesis, such as de-etiolation and the transition to reproductive development. Third, light is also deleterious for plants because excess light, absorbed by the photosynthetic apparatus, promotes the formation of dangerous compounds such as active oxygen species. Because plants must quickly respond to changing and often extreme light conditions, sophisticated photosensory networks have evolved that enable plants to maximize photosynthesis while minimizing damage. One of the main mechanisms of this overall control is accomplished through regulation of gene expression.

Light is perceived in plants by a sophisticated system of photoreceptors that detect different light wavelengths. Five phytochromes mediate red and far-red light responses in Arabidopsis. Among these, phytochrome A (PhyA) is primarily responsible for the perception of constant far-red light, whereas PhyB is primarily responsible for the perception of constant red light. Three photoreceptors for blue light have been identified in Arabidopsis (for review, see Lin, 2000). Crytpochrome 1 (Cry1) is the principal blue/UV-A light receptor, modulating growth at medium and high-light intensities. Cry2 has major functions in responding to low intensities of blue light (Lin et al., 1996). Phototropin, encoded by the non-phototropic hypocotyl 1 (NPH1) gene, is the photoreceptor for phototropism (Liscum and Briggs, 1995). Various reports have discussed the possibility of other blue-light receptors, though their identities were enigmatic (Zeiger and Zhu, 1998; Briggs and Huala, 1999; Frechilla et al., 1999). NPL1 (NPH-like 1), a fourth blue-light receptor, recently was identified that is partly functionally redundant with NPH1, and has a major role in the chloroplast high-blue-light avoidance response (Jarillo et al., 2001; Kagawa et al., 2001; Sakai et al., 2001).

Downstream from the photoreceptors are a plethora of positive and negative regulators of light signaling (for review, see Nagy et al., 2000; Neff et al., 2000). Among these, the COP9 signalosome (CSN) is a multisubunit regulatory complex that functions through unknown mechanisms as a master repressor of photomorphogenesis in the dark (for review, see Karniol and Chamovitz, 2000). One of the targets of the CSN-mediated repression is HY5. HY5 is a basic Leu zipper transcription factor directly involved in the expression of light-inducible genes (Oyama et al., 1997; Chattopadhyay et al., 1998). No role for these receptors or signaling molecules has been reported for responses to light stress.

The effect of light on plant development is particularly evident in seedling development and the transition from growth under soil (dark) to growth above the ground (light). As photomorphogenesis is initiated, cellular and subcellular processes are initiated to allow the development of photosynthetic capable tissues. This includes chloroplast development, pigment synthesis, and assembly of the photosystems in the thylakoids. All of these processes are accomplished by and depend on the differential expression of a large number of genes. However, before a chloroplast is competent for performing photochemistry, it is saturated with photons that have no outlet, and thus form toxic compounds that can kill the developing cell. In response to this light-induced stress, plants produce photoprotective pigments such as carotenoids and xanthophylls, and protective proteins.

An example of protective proteins are the early light-induced proteins (ELIPs), nuclear-encoded thylakoid membrane proteins that are transiently expressed immediately after light stress. Elip transcript and protein appear considerably faster than those of other light-induced genes during the early stage of de-etiolation, and disappear before chloroplast development is completed (Grimm and Kloppstech, 1987). In mature plants, ELIP accumulation under light stress conditions correlates with the photoinactivation of photosystem II (PSII), degradation of the D1 protein, and changes in the level of pigments (Adamska et al., 1992a, 1993). ELIPs bind chlorophyll a and lutein and have been proposed to function as transient pigment carriers or chlorophyll exchange proteins (Adamska et al., 1999).

The regulation of ELIP expression is modulated by light and other stress signals. Blue and red light induce Elip transcription in etiolated plumulas of pea (Pisum sativum) seedlings (Adamska, 1995), whereas blue and UV-A light induce ELIP in adult tissues (Adamska et al., 1992a, 1992b). ELIP homologs from various systems have been implicated in various stress responses. For example, one of the responses to extreme dehydration of the "resurrection" plant Craterostigma plantagineum is the expression of the ELIP homolog dsp-22 (Bartels et al., 1992).

In pea (Scharnhorst et al., 1985; Kolanus et al., 1987) and tobacco (Nicotiana tabacum; Blecken et al., 1994), ELIP is encoded by a single gene, whereas two ELIPs was reported to exist in barley (Hordeum vulgare; Grimm and Kloppstech, 1987) and in Arabidopsis (Moscovici-Kadouri and Chamovitz, 1997; Heddad and Adamska, 2000). The functions of the two genes are unclear, as is the genetic regulation of Elip transcription.

To understand genetic mechanisms regulating light and stress control of ELIP induction, we have initiated a study of ELIP in Arabidopsis. Previous studies on Elip in other plants were limited to characterizing the light quality and intensity that regulate Elip expression, and often in dismembered leaves. Arabidopsis provides a convenient system for studying Elip transcription due to the large collection of characterized light-signaling mutants available. We show here that multiple photoreceptors, including a cryptic blue-light receptor, regulate Elip transcription, and that the two ELIP genes have differing regulation patterns, which hint at different functions.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Development of Semiquantitative Reverse Transcriptase (RT)-PCR Experiment Conditions

Previous studies showed that light regulation of Elip steady-state transcript levels is manifested at the level of transcription, allowing for a correlation between steady-state transcript levels and transcriptional control (Adamska, 1995). To study the regulation of Elips in multiple genetic backgrounds under different conditions, we developed a sensitive and fast RT-PCR method for analyzing Elip transcript levels. We used two PCR techniques: conventional RT-PCR followed by hybridization, and the LightCycler system. Ubiquitin10 (Ubq10) was used as an internal RT-PCR control because it was previously shown by RNA gel-blot analysis to be constitutively expressed in light and dark conditions (Sun and Callis, 1997). To confirm this result by RT-PCR, and to determine if the steady-state levels of Ubq10 are affected by the developmental stage of the plant, we checked Ubq10 levels in dark- and light-grown seedlings by both RT-PCR techniques. As seen in Figure 1, A and B, the steady-state levels of Ubq10 in dark-grown seedlings kept in the dark or exposed to 1 h light are equal as determined by conventional RT-PCR and LightCycler RT-PCR. This endogenous mRNA standard has the advantage of serving as a control for RNA recovery and integrity, as well as for sample-to-sample variations in RT and PCR.

View larger version (47K): [in this window] [in a new window]   Figure 1.   Development of RT-PCR experimental conditions. A and B, Ubq10 is a constitutive control. Four-day-old Arabidopsis dark-grown wild-type seedlings were kept in the dark (Dark) or exposed to 1 h light (Light), RNA was extracted, and amounts of Ubq10 were determined by conventional RT-PCR followed by Southern blot (A) or by the LightCycler (B). The amount of Ubq10 PCR products as a function of polymerization cycles is shown in A. The calculated initial amount of Ubq10 transcript is shown in B. C and D, RT-PCR controls for Elip1 and Elip2. C, Elip1 and Elip2 primer pairs were used for PCR on genomic DNA and cDNA (from "Light" above) templates, and in the absence of RT (RT). D, "Light" sample from above was used as template to determine the kinetics of PC reaction as a function of number of PCR cycles, as detected by dot blot.

To avoid artifacts due to genomic DNA contamination in an RNA preparation, PCR primer pairs of Elip1 and Elip2 were designed around introns (see "Materials and Methods"). In addition, to avoid artifacts due to the high identity between the two Elip transcripts, the forward primers of Elip1 and Elip2 anneal to the divergent 5' end of the genes. As shown in Figure 1C, PCR on genomic DNA and cDNA yield product sizes for Elip1 of 414 and 330 bp, respectively, and for Elip2, 705 and 400 bp, respectively. In addition, RNA was treated with DNase (during total RNA preparation) prior to cDNA synthesis and the control reaction was performed in which RT is omitted (RT; Fig. 1C). Conventional PCR was carried out for 15 cycles where the kinetics of the PCR reactions allowed quantitative analysis for all three genes, Elip1, Elip2, and Ubq10 (Fig. 1, A and D).

Expression of Elip Is Mediated by Red, Far-Red, and Blue Light and by Heat Shock

To examine the regulation of Elip expression during photomorphgenesis in wild-type Arabidopsis seedlings, the mRNA levels of Elip1 and Elip2 in seedlings grown under different light qualities were determined by RT-PCR. As seen in Figure 2A, both Elips are apparently not expressed, or expressed at levels below detection in 4-d-old dark-grown seedlings. Exposure of these seedlings to 1 h red, far-red, and blue light resulted in a detectable increase in transcript levels of both Elips. Elip expression is also induced in dark-grown seedlings following heat shock at 37°C for 1 h. This indicates that Elip expression is induced by high irradiance response (HIR) light in etiolated seedlings exposed to 1 h light during the etioplast/chloroplast conversion, and raises the question of which photoreceptors are involved in the positive regulation of Elip steady-state mRNA levels.

View larger version (53K): [in this window] [in a new window]   Figure 2.   Effect of light on expression of Elips. A, RT-PCR analysis of 4-d-old Arabidopsis dark-grown wild-type seedlings that were kept in the dark or exposed to 1 h of white, blue, red, or far-red lights, or to heat shock in the dark at 37°C. One-fifth of the PCR products were resolved in 1.7% (w/v) agarose gel and blotted onto Hybond N+. The membrane was hybridized with DIG labeled Elip1 and Elip2 probes. B, RT-PCR LightCycler analysis reveals the differences between Elip1 and Elip2 transcript levels. Four-day-old Arabidopsis dark-grown wild-type seedlings were kept in the dark (D) or exposed to 1 h of white light (L) before total RNA extraction. One-tenth of the cDNA was amplified by the LightCycler. Top, Amplification curves of Elip1 and Elip2. The diagram documents the fluorescence intensity (approximate PCR product concentration) plotted against the number of PCR cycles. Bottom, Melting curve analysis of PCR products. The diagram documents the negative derivative fluorescence plotted against temperature, where the peak therefore highlights the melting of DNA.

Elip1 and Elip2 Have Different Dark Expression Patterns

The Elip2 cDNA was isolated from a cDNA library made from etiolated seedlings (see "Materials and Methods"). This led us to question the significance of the lack of Elip expression found in dark-grown seedlings (Fig. 2A; Heddad and Adamska, 2000). To further study Elip1 and Elip2 regulation, the experiments were repeated using the highly sensitive LightCycler method. LightCycler RT-PCR was performed with cDNA samples from 4-d-old dark-grown seedlings treated with 1 h of white light or kept in darkness. The amplification curves of this experiment are shown in Figure 2B (top). The mRNA levels of Elip1 and Elip2 in light-treated seedlings are similar, with fluorescence levels rising after cycle 16. However, the mRNA levels of Elip1 and Elip2 in dark-grown seedlings are different. The fluorescence signal of the Elip2 transcript starts to rise after 20 cycles, whereas no Elip1 product is detected even after 45 cycles. The melting curves analysis (Fig. 2B, bottom) of this experiment confirms that there is no Elip1 PCR product in the dark-grown seedlings, as opposed to Elip2 product at the same condition, or to Elip1 and Elip2 in the light-exposed seedlings. These data indicate that Elip1 and Elip2 are differentially regulated in darkness. The result further shows the advantage of using the LightCycler system in that it is highly sensitive and allows the detection of very small amounts of transcript, as compared with the conventional PCR experiments or RNA gel-blot experiments.

Phytochromes Regulate Elip Expression via Red and Far-Red Light

To examine the role of different phytochromes in the regulation of Elip expression, we used phyA, phyB, and phyA/phyB double mutants. Four-day-old dark-grown wild-type and mutant seedlings were kept in the dark or exposed to 1 h of red or far-red light. Total RNA was isolated and conventional PCR experiments were performed. Red and far-red lights have similar positive effects on Elips in wild-type seedlings (Table I). This effect of far-red light on Elip mRNA was lost in the phyA mutant, indicating that PhyA positively regulates Elips. The phyA mutant also showed reduced red-light induction of both Elips. It is surprising that in phyB, the red and far-red induction of Elips was not impaired. However, the absence of both phytochromes in the phyA/phyB mutant resulted in a loss of red and far-red induction of both Elips. This suggests an essential requirement of PhyA for red-light induction of Elips. Analysis with the LightCycler identified very low levels of Elip transcripts in the phyA/phyB mutant (not shown).

Cryptochrome1, Cryptochrome2, and Phototropin Are Not Vital for the Blue-Light Induction of Elip

To identify the blue-light photoreceptor(s) involved in Elip expression, we used mutants defective in photoreceptors for blue light. Four-day-old Arabidopsis dark-grown wild type, cry1, cry2, cry1/cry2 double mutant, nph1, and phyA/phyB double mutant seedlings were kept in the dark or exposed to 1 h of blue light before total RNA extraction. Blue light clearly results in an increase in Elip1 levels in wild type and in the mutants (Table I), indicating that none of these photoreceptors or pairs of photoreceptors have essential roles in the transcriptional regulation of Elip1.

To examine the expression level of Elip2 in these mutants, the experiments were continued using the LightCycler because conventional RT-PCR experiments yielded contradictory results (not shown). Figure 3 shows normalized initial amount of Elip2 mRNA. Like Elip1, Elip2 was induced by blue light in all mutants. However, Elip2 transcript levels were reduced in the cry1, cry2, cry1/cry2, and nph1 mutants relative to wild type, which may be indicative of a redundant function for these photoreceptors.

View larger version (9K): [in this window] [in a new window]   Figure 3.   RT-PCR analysis of Elip2 expression in blue-light receptor mutants. Four-day-old Arabidopsis dark-grown wild-type (w.t.), cry1, cry2, cry1/cry2, nph1, and phyA/phyB seedlings strains were kept in the dark or exposed to 1 h of blue light before total RNA extraction. One-tenth of the RT-reaction was amplified by the LightCycler using Elip2 and Ubq10 primers. The normalized initial amount of Elip2 mRNA is shown.

Downstream Regulators of Elip Expression

To determine the role of CSN in Elip regulation, RT-PCR was performed on dark-grown 4-d-old wild-type and CSN mutant cop9 seedlings. The normalized results in Figure 4A show that although in dark-grown wild-type seedlings Elip1 is not expressed and Elip2 is found in very low levels, transcripts of both genes accumulate at high levels in dark-grown cop9 seedlings. This indicates that the CSN fuctions in the repression of Elip expression in darkness.

View larger version (44K): [in this window] [in a new window]   Figure 4.   Analysis of Elip in downstream light-signaling mutants. A, RT-PCR analysis of Elip expression in cop9. RNA was isolated from 4-d-old dark-grown Arabidopsis wild-type (w.t.) and cop9 seedlings. One-tenth of the cDNA was amplified by the LightCycler using Elip1, Elip2, and ubq10 primers. The normalized initial amounts of Elip1 and Elip2 are shown. B, RT-PCR analysis of Elip expression in hy5. Four-day-old hy5 dark-grown seedlings were kept in the dark (D) or exposed to 1 h of white light (L) or to heat shock (H) before total RNA extraction. One-tenth of the cDNA was amplified by the LightCycler. Amplification curves of Elip1, Elip2, and Ubq10 are shown in the graph, with normalized initial amounts of Elip1 and Elip2 mRNA presented in the bar graph. Error bars represent SD based on three replicates of the same sample.

To identify a potential positive regulator of Elip expression, we next studied the hy5 mutant. HY5 is a light-regulated transcription factor for light-inducible genes (Oyama et al., 1997; Chattopadhyay et al., 1998). RNA was isolated from 4-d-old dark-grown hy5 mutant seedlings that were kept in the dark or exposed to 1 h of white light. As in the wild type (Fig. 2), Elip2 levels increase in the light in hy5 (Fig. 4B). In contrast, Elip1 levels do not increase in light in hy5. This suggests that HY5 is involved in the transcription of Elip1 gene directly, or promotes transcription of genes that are upstream of Elip1. To determine if Elip1 expression is totally silenced in hy5, or only in a light-dependant manner, we examined the effect of heat shock on Elips in this mutant. As shown in Figure 4B, both Elip1 and Elip2 are induced in a light-independent fashion by heat shock in hy5, indicating that HY5 acts downstream in a light signal transduction pathway that positively regulates Elip1, but that heat shock acts through independent signaling pathways.

Microarray Analysis of Elip1 Expression

Elip expression studies have concentrated primarily on studying Elip responses to singular phenomena, and until the present study, only by northern analysis. However, Elip1 was serendipitously included in the microarray distributed through the Arabidopsis Functional Genomics Consortium. Data from the publicly available experiments can help to clarify the factors involved in Elip regulation and its functions in Arabidopsis. Table II presents a summary of the relevant experiments having significant results for Elip1. The anatomical experiments show that Elip1 levels are higher in leaves relative to flowers, consistent with the chloroplast localization of ELIP (Kruse and Kloppstech, 1992). In addition to light, two other environmental stress situations, elevation of CO₂ and aluminum, positively effect Elip expression.

The phototropic stimulation experiments are compatible with our results provided above, and display a significant increase in Elip levels in the dark-grown wild type after exposure to 1 h blue-light illumination. In addition, the 1.03 R/G normalization ratio present in experiment number 6,619 confirms our result in Figure 2 that the Phototropin pathway has no effect on blue-light induction of Elip1 transcription. We demonstrate this by using the phototropin photoreceptor mutant, nph1, whereas in the microarray experiment it was shown by analysis of nph4. NPH4 functions downstream of phototropin (NPH1; Harper et al., 2000).

Experiment 7,230 shows that far-red light induces Elip1 in adult Arabidopsis plants. This information complements our result that far-red induces Elip1 via PhyA in de-etiolation, but does not support earlier findings in green pea in which no Elip transcript or protein could be detected under light of 480 to 780 nm (Adamska et al., 1992a).

The Development of PSII in hy5 Is Retarded

Because Elip1 could not be detected during greening of etiolated hy5 seedlings, further characterization of the hy5 phenotype may reveal potential functions of ELIP1. ELIP was shown to play a role during the greening process (for review, see Adamska, 1997); therefore, the functional state of PSII as measured by chlorophyll fluorescence induction was studied during greening of hy5 as compared with wild-type seedlings. Figure 5 shows the photosynthetic efficiency of 4-d-old wild-type and hy5 dark-grown seedlings that were exposed to light for increasing periods and under different light intensities. At both 25 and 100 µmol m² s¹ white light, the efficiency of the photosynthetic apparatus is lower in hy5 than in the wild type. However, the PSII efficiency values of 14-d-old light-grown wild-type and hy5 seedlings were essentially identical, indicating that hy5 does reach the same efficiency level as that of the wild type. These results suggest that the development of PSII activity in the mutant is temporally retarded relative to that of the wild type. The fact that this phenotype is expressed only for a limited time in the plastid development could be due to the expression of other ELIP family genes that are not affected by the mutation in hy5. Therefore, these results suggest that hy5 mutant seedlings display a slower formation of PSII activity.

View larger version (16K): [in this window] [in a new window]   Figure 5.   Development of variable chlorophyll fluorescence during greening of hy5 and wild type. Four-day-old wild-type (WT) and hy5 dark-grown seedlings were exposed to 25 µmol m2 s1 or 100 µmol m2 s1 white light for different periods of time. A pulse amplitude-modulated fluorimeter (PAM) was used to calculate the maximum photochemical efficiency of PSII in the dark-adapted state (Fv/Fm) parameter. Error bars represent SD based on six different measurements.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The work presented here addresses the question how the two Elip genes in Arabidopsis are regulated at the genetic level during photomorphogenesis. For this purpose, we have investigated the transcript levels of these genes under different physiological conditions and genetic backgrounds by two quantitative RT-PCR techniques. It has been shown previously that steady-state levels of Elip mRNA correlate to changes in transcriptional activity (Adamska, 1995). This point is further emphasized in the microarray experiment where the transcriptional inhibitor cordycepin inhibited induction of Elip1 (Table II).

Our results that both Elips can be induced in dark-grown Arabidopsis seedlings by illumination with high irradiant red, far-red, and blue lights are consistent with earlier results showing Elip induction in etiolated pea seedlings (Adamska et al., 1992b). The low irradiance response was not addressed here. The microarray experiments present in Table II also confirm our results for the effects of blue light, far-red light, and the phototropin pathway on Elip1 expression. These far-red light results differ from those obtained in adult pea plants where only blue light was found to result in ELIP induction in adult plants (Adamska et al., 1992a).

The use of photoreceptor mutants provides direct evidence for the involvement of both PhyA and PhyB in Elip regulation. PhyA acts to induce Elip expression by HIR far-red light, but also has a role in perceiving HIR red light. In contrast, absence of PhyB by itself does not effect red and far-red induction of Elips. In this case, PhyA and maybe the other phytochromes (PhyC-E) compensate for the lack of PhyB. However, PhyB appears to work synergistically with PhyA in controlling the red-HIR induction of Elips as shown by the complete loss of red-light induction of both Elips in the phyA/phyB double mutant. Other phenomena are also known to be under HIR control of both PhyA and PhyB in Arabidopsis, including the control of hypocotyl elongation (Quail et al., 1995; Smith 1995) and expression of chlorophyll a/b binding protein (CAB) genes (Reed et al., 1994).

The identity of the photoreceptor involved in blue-light induction of Elips remains cryptic. The blue-light-induced up-regulation of Elips was not silenced in cry1/cry2 double mutants, or in the nph1 mutants. Microarray experiment 6,619 also indicates that Elip1 expression is not effected by at least one of the pathways regulated by the phototropin receptor. While the present manuscript was in review, a fourth blue-light receptor, NPL1, was identified (Jarillo et al., 2001; Kagawa et al., 2001). To determine if NPL1 has an essential role in mediating the blue-light induction of Elips, we examined the npl1 mutant by RT-PCR. As seen in Figure 6, both Elip1 and Elip2 are induced by blue-light in npl1. These results lead to two hypotheses: (a) The known blue-light photoreceptors have redundant functions in regulation of Elips by blue light. To further test this hypothesis, Elip induction in an Arabidopsis quadruple mutant cry1/cry2/nph1/npl1 would need to be analyzed. (b) A novel blue-light photoreceptor is involved in the regulation of Elips. This photoreceptor would work either independently or in co-action with the four known blue-light photoreceptors. This last possibility needs further consideration because there is accumulating evidence for such a receptor. For example, the stomatal responses of light-grown cry1, cry2, cry1/cry2, nph1, and nph1/cry1 plants did not differ from those of wild type (Lasceve et al., 1999; Eckert and Kaldenhoff, 2000). The potential role of a carotenoid derivative as a blue-light chromophore has been controversial (Palmer et al., 1996; Frechilla et al., 1999; Lasceve et al., 1999; Tlalka et al., 2001; Eckert and Kaldenhoff, 2000; Jin et al., 2001). We attempted to dissect the role of carotenoids or xanthophylls in Elip regulation through the use of the carotnoid inhibiting herbicide noflurazon (Chamovitz et al., 1991). It is unfortunate that this norflurazon treatment itself induces Elips (not shown).

View larger version (102K): [in this window] [in a new window]   Figure 6.   NPL1 is not necessary for blue-light induction of Elip s. Four-day-old Arabidopsis dark-grown wild-type (w.t.) and cav1-1 (npl-1) seedlings strains exposed to 1 h of blue light before total RNA extraction. One-fifth of the PCR products with Elip1, Elip2, and Ubq10 primers were resolved in 1.7% (w/v) agarose gel.

Both Elip1 and Elip2 were induced by blue light in the phyA/phyB double mutant. This result appears to exclude the possibility that the blue-light induction of Elips is mediated by phytochrome. Other blue-light-regulated processes have also been shown to be independent of phytochrome in various systems. For example, a pea phyA/phyB double mutant was also recently reported to maintain normal blue-light responses, and CRY1 was shown to act independently of both PhyA and PhyB in tomato (Lycopersicon esculentum; Weller et al., 2001a, 2001b). However, the slight reduction in Elip2 levels in phyA/phyB could indicate a possible phytochrome involvement in blue-light regulation of Elips, similar to that reported for the dependence of Cry1 on phytochrome for regulating hypocotyl growth and anthocyanin accumulation (Ahmad and Cashmore, 1997).

In addition to light signals, Elips are also controlled by heat shock. It was shown previously that the accumulation of Elip transcript in etiolated barley and pea seedlings was induced by cyclic heat shock applied for several days (Beator et al., 1992; Otto et al., 1992). However, in other experiments performed with heat-treated etiolated pea, heat shock could not induce Elip without involvement of short illumination (Kloppstech et al., 1991). A recent study in adult light-grown Arabidopsis plants tested the possibility that Elips are induced by other environmental stresses other than light, but RNA gel-blot analysis could not detect induction of either Elip following heat shock (Heddad and Adamska, 2000).

Despite the similar light and heat shock regulation patterns of Elip1 and Elip2, the two genes have different accumulation patterns in the dark. Elip2 is expressed at low levels in dark-grown seedlings, whereas no Elip1 mRNA could be detected. This first evidence for presence of Elip transcript under dark conditions results from the sensitivity of the LightCycler system because no Elip transcript could be detected in the earlier studies performed by RNA gel blot, or in this study as shown in Figure 2, by conventional RT-PCR.

Our results demonstrate the positive role of HY5 on Elip1 transcription during photomorphgenesis. The absence of Elip1 mRNA in the hy5 light-treated seedlings is not surprising because this transcription factor has been already shown to bind directly to G-box DNA sequences, well-characterized light-responsive elements in light-responsive promoters (Chattopadhyay et al., 1998). A study in transgenic pea plants had demonstrated that two light-responsive elements are involved in light-regulated expression of Elip. One element is similar to the GT1 binding site and the other resembles a G-box-like ACGT element (Blecken et al., 1994).

The CSN mutant, cop9, which mimics light growth while grown in darkness, shows high expression levels of the classic light-inducible genes such as CAB, Chs, and PsbA when grown in a total darkness (Wei and Deng, 1992). This study shows that the light induction of both Elips is repressed in the dark by the CSN. From this result, together with the results of hy5 experiment, it can be concluded that light signals abolish the CSN-mediated degradation of the transcription factor HY5, and thus allow it to activate directly or indirectly the transcription of Elip1 gene. However, the reduction of Elip2 expression in the dark must be mediated by a different mechanism because HY5 does not regulate Elip2 expression. Based on earlier data, a working model for the signal transduction pathways that regulate Elip expression in Arabidopsis seedlings is suggested in Figure 7.

View larger version (19K): [in this window] [in a new window]   Figure 7.   Working model for the signal transduction pathways that regulate the expression of the Elip genes in Arabidopsis seedlings. Different light qualities are sensed by at least three different photoreceptors, PhyA, PhyB, and a novel blue-light receptor, to initiate a signal transduction cascade that abolishes the repressory action of the CSN, leading to the expression of both Elip1 and Elip2. The expression of Elip1 is positively effected by HY5, whereas other transcription factors regulate Elip2. Heat shock stimulates the expression of Elip through an independent signaling pathway.

Our results that the dark levels of Elip2 are higher than the dark level of Elip1, together with the result that HY5 is not involved in the regulation of Elip2, hint that the light regulation of Elip1 occurs at the level of transcription, whereas other mechanisms may be involved in the regulation for Elip2. These two differences in regulation imply different function. Because Elip1 is strictly light induced and apparently responds as a light stress protein, Elip2 may be a "housekeeping gene" that is constantly expressed at low levels, ready to be translated under stress conditions.

The microarray data further indicate that the ELIPs are stress-related proteins. This is consistent with previous evidence that Elip steady-state levels are regulated by other environmental stresses. Low temperature, for example, positively regulated Elip transcription and stabilization (Adamska and Kloppstech, 1994). Because low temperature noticeably increases excitation accumulation of PSII, it was proposed that this cold-induction accumulation of Elip mRNA could be interpreted in terms of redox control of gene expression (Montane et al., 1998). In Crateostigma plantagineum, an ELIP-related Dsp-22 protein is induced during desiccation; in this case, abscisic acid (ABA) and light were simultaneously required for Elip induction (Bartels et al., 1992). In a similar experiment when 6-d-old green barley plants were treated with a combination of high light and ABA, Elip level increased in comparison with light-treated control. However, no effect was observed with just ABA (Potter and Kloppstech, 1993). This environmental induction of Elip1 transcription implicates additional functions that are controlled by signal transduction pathways other than those described here.

Although accumulating correlative evidence indicates that ELIPs are involved in protection of the photosynthetic apparatus, the elucidation of the physiological role of the ELIPs has been hampered by lack of a genetic system. The discovery that hy5 lacks Elip1 expression provides a preliminary model system to study the role of ELIP1. A defect in PSII activity displayed by hy5 seedlings may result from the loss of Elip1 expression in this mutant. The development of photosynthetic activity in hy5 is retarded relative to that of the wild type. The low F_v/F_m ratio in the hy5 seedlings seems to be due mostly to a relatively high initial (minimum) PSII fluorescence in the dark-adapted state (F₀) level. This could be due to PSII centers in which electron flow to the plastoquinone pool is partially inhibited (that is, closed PSII centers), a situation that can be induced by light stress. The fact that the F_v/F_m ratio in the mutant seedlings exposed to 100 µmol m² s¹, as compared with those exposed to 20 µmol m² s¹, reached lower levels supports this suggestion. However, as the time of illumination and thus of the development of the thylakoids continues, hy5 seedlings recover from this initial light stress and the F_v/F_m ratio reaches the same values in mutant seedlings exposed to both low and high light. Following prolonged illumination, hy5 completely recovers from the light stress and the F_v/F_m ratio is similar in the mutant to the wild type. We hypothesize that the phenotype exhibited by hy5 during the early phase of the greening process results from a lack of Elip1 transcription. Further study is needed to validate this hypothesis.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Materials, Growth, and Illumination Conditions

Arabidopsis seedlings were grown for 4 d on Murashhige and Skoog medium (Sigma, St. Louis) with 1% (w/v) agar, in darkness, at 22°C. Different light qualities were obtained by using a cool-white fluorescent light (100 µmol m² s¹; OSRAM, Munich) with the filters (Chris James, London): blue (380-500 nm, 40 µmol m² s¹), red (600-700 nm, 50 µmol m² s¹), or with far-red enriched lights with a far-red filter (700-780 nm, 2 µmol m² s¹). For heat shock treatment, the seedlings were incubated at 37°C in the dark. Seedlings were harvested and frozen in liquid nitrogen after 1 h of light or heat shock treatment. All subsequent manipulations were done under a green safe light.

The mutants used in this work were: cry1-304 (Ahmad and Cashmore, 1993), cry2-1 (Guo et al., 1998), cry1-304/cry2-1 (Guo et al., 1998), nph1-5 (Liscum and Briggs, 1995), phyA (Reed et al., 1993), phyB (Koornneef et al., 1980), phyA/phyB (Reed et al., 1993), hy5-1 Koornneef et al., 1980), cop9-1 (Wei and Deng, 1992), and cav1-1 (npl1; Kagawa et al., 2001).

Measurement of Transcript Levels by RT-PCR

RNA Preparation and RT Reaction

Conventional PCR Followed by Hybridization

Real-Time PCR by LightCycler

3

Plasmids Used in This Work

The cDNA clones for Elip1 (clone Id174P5T7) and Elip2 (clone IdVCVCD09) were obtained through AIMS. Elip1 was isolated from the mixed tissue cDNA library Lambda PRL2 (Newman et al., 1994). Elip2 was isolated from a cDNA library made from 5-d-old etiolated seedlings (T. Desprez, J. Amselem, H. Chiapello, P. Rouze, M. Caboche, and H. Hofte, unpublished data).

Chlorophyll Fluorescence Measurements

Wild-type and mutant seedlings were grown on agar plates and sowed in such a way as to form clusters of several seedlings so one could measure simultaneously the fluorescence emission of at least five to seven seedlings, thus obtaining an average result. Several clusters were measured on each plate. Variable fluorescence was measured using a Pulse Modulated Fluorimeter (PAM-101, Waltz, Germany). The modulated beam (650 nm) intensity at the seedlings level was about 1 mmol m² s¹ at 1.5 kHz and the intensity of the saturation light pulse was 3,000 mmol m² s¹ for 1-s duration. The variable fluorescence, F_v, was calculated as (F_m  F₀). F₀ is the minimal fluorescence and was determined with the modulated beam. F_m is the maximal fluorescence and was determined with the saturation light pulse. The ratio F_v/F_m is normalized with the concentration of chlorophyll and interpreted as functional state of PSII. The plants were dark adapted for several minutes before the onset of measurements and maintained thereafter in dim-green light.