A zinc finger protein RHL41 mediates the light acclimatization response in Arabidopsis

Authors


*For correspondence (fax +81 797 74 2133; e-mail oeda@sc.sumitomo-chem.co.jp).

Summary

Arabidopsis thaliana plants showed an increased tolerance to high-intensity light when pre-exposed to medium-intensity light. This response, known as light acclimatization, depended on the quantity of light, the period of irradiation, and the quality of light. Among characterized acclimatization-induced cDNA clones, we identified a zinc finger protein rhl41 (responsive to high light) gene, that was rapidly up-regulated in proportion to the time of irradiation and the light intensity. Transgenic Arabidopsis plants over-expressing the rhl41 gene showed an increased tolerance to high-intensity light, and also morphological changes of thicker and dark green leaves. Interestingly, the palisade parenchyma was highly developed in the leaves of the transgenic plants, which is one of the long-term acclimatization responses in Arabidopsis plants. The anthocyanin content (a light protectant) as well as the chlorophyll content also increased. Antisense transgenic plants exhibited decreased tolerance to high irradiation. We propose that the RHL41 zinc finger protein has a key role in the acclimatization response to changes in light intensity.

Introduction

Acclimatization to light regimes is one of the most important and complex responses of plants to various environmental conditions. A number of reports have provided information on the physiological and biochemical characterization of light acclimatization and photoprotective responses to high-light stress ( Anderson and Osmond, 1987; Anderson et al., 1995 ; Aro et al., 1993 ; Demmig-Adams and Adams, 1992 ; Durnford and Falkowski, 1997). Marked modulations of the composition, function and structure of the photosynthetic apparatus occur in response to light intensity ( Walters and Horton, 1994; Walters and Horton, 1995a; Walters and Horton, 1995b). When plants are exposed to excess photons, adaptation mechanisms of chloroplasts become operative ( Asada et al., 1998 ), including down-regulation of photosystem II (PSII) (degradation of D1 proteins), supply of electron acceptors, and scavenging of active oxygen species. The xanthophyll cycle involving the de-epoxidation of violaxanthin to zeaxanthin is also a photoprotective response which dissipates excess energy in the antenna chlorophyll ( Demmig-Adams and Adams, 1992 ; Ruban and Horton, 1999). It has been proposed that the water–water cycle in chloroplasts dissipates excess excitation energy under environmental stress, accompanied by the scavenging of active oxygens in chloroplasts ( Asada, 1999). Karpinski et al. (1999) reported a mechanism of systemic acquired acclimatization, in which systemic redox changes in the proximity of PSII, hydrogen peroxide and the induction of anti-oxidant defences are key determinants.

Several transgenic studies defining the photoprotective mechanism have been reported. Kozaki and Takeba (1996) reported that photorespiration is one of the mechanisms that can protect plants from photo-oxidation, and that transgenic plants over-producing the glutamine synthetase gene exhibit increased tolerance to high irradiation. To analyse the potential of the active oxygen-scavenging system of chloroplasts, expression of E. coli catalase in tobacco chloroplasts has been attempted, and photosynthesis of transgenic plants proved tolerant to high irradiance under conditions of drought ( Shikanai et al., 1998 ).

Studies on the molecular basis of the cold acclimatization response or drought stress response have progressed, and a number of genes that respond to these stresses at the transcriptional level have been described ( Shinozaki and Yamaguchi-Shinozaki, 1996; Thomashow, 1994). To better understand the complex network of osmotic and cold signal transduction pathways, a genetic screen for Arabidopsis mutants with aberrant gene expression in response to low temperature was developed ( Ishitani et al., 1997 ). This new genetic approach will aid in analysing stress signal transduction pathways.

On the other hand, little is known of molecular mechanisms involved in modulating the expression of genes involved in light acclimatization in higher plants. In this report, factors affecting the acclimatization response of Arabidopsis, such as light quantity and quality, are described. We isolated cDNAs corresponding to mRNAs that are rapidly induced during light acclimatization, and identified a cDNA coding for the zinc finger protein RHL41 that is up-regulated in proportion to irradiation time and light intensity. To verify the function of RHL41 as a transcription factor involved in the acclimatization response, we analysed transgenic plants that over-express the rhl41 gene. The transgenic plants exhibited increased tolerance to irradiance, with morphological changes such as development of palisade parenchyma and increased anthocyanin and chlorophyll contents. Our findings indicate that RHL41 may regulate several acclimatization-responsive genes either directly or indirectly, which in turn confers tolerance to high irradiation.

Results

Acquisition of tolerance to high irradiation in Arabidopsis acclimatized to medium-light stress

When Arabidopsis plants that had been grown under normal light conditions (75 µE m−2 sec−1) were exposed to high-intensity light of 1800 µE m−2 sec−1 for 8 h, leaves were severely damaged and chlorophyll was bleached, especially in old leaves ( Figure 1a, W75). However, the plants that had been acclimatized to medium white light of 200 µE m−2 sec−1 for 24 h showed a decreased bleaching of chlorophyll after exposure to high-intensity light ( Figure 1a, W200), indicating that the tolerance to high-light stress was attained by the acclimatization treatment.

Figure 1.

Acquisition of tolerance to high irradiation in Arabidopsis acclimatized to medium-intensity light stress.

(a) Comparison of photo-oxidation damage of non-acclimatized and acclimatized leaves. The plants grown under continuous white light of 75 µE m−2 sec−1 (W75) were acclimatized to white light of 200 µE m−2 sec−1 (W200), blue light of 200 µE m−2 sec−1 (B200) or red light of 200 µE m−2 sec−1 (R200) for 24 h. Then the acclimatized plants were exposed to white light of 1800 µE m−2 sec−1 for 24 h ‘Not exposed’ represents the plants that were not acclimatized nor exposed to high irradiation. (b) Effects of light intensity and light quality on acclimatization. The plants grown under continuous white light of 75 µE m−2 sec−1 (W75) were acclimatized to white light of 200, 400, 600 and 800 µE m−2 sec−1 (W200, W400, W600 and W800), blue light of 200 µE m−2 sec−1 (B200) and red light of 200 µE m−2 sec−1 (R200) for 24 h. The acclimatized plants were exposed to white light of 1800 µE m−2 sec−1 for 4, 8 and 24 h, then chlorophyll fluorescence was measured. The younger five leaves for each plant were measured. (c) Time course of the acclimatization response. Plants grown under continuous white light of 75 µE m−2 sec−1 were acclimatized to white light of 630 µE m−2 sec−1 for the indicated periods, and then exposed to white light of 1800 µE m−2 sec−1 for 8 h. The chlorophyll fluorescence was measured before and after high-intensity irradiation.

Various factors that affect the acclimatization response were examined ( Figure 1b,c). In these experiments, PSII activity was determined by chlorophyll fluorescence measurement ( Kraus and Weis, 1991; Schreiber et al., 1995 ), before and after exposure to white light of 2000 µE m−2 sec−1. To examine the effect of light intensity on acclimatization, the plants were exposed to white light of 200–800 µE m−2 sec−1 for 24 h prior to exposure to the high irradiation ( Figure 1b). As expected, PSII activity after the high irradiation was increased in proportion to light intensity up to 600 µE m−2 sec−1 during the acclimatization process. The acclimatization treatment itself caused significant photo-inhibition ( Figure 1b, W200 to W800, 0 h), and the acclimatization by light of 800 µE m−2 sec−1 lead to photo-inhibition, the result being decreased tolerance after high irradiation.

The time course of the response to acclimatization at 630 µE m−2 sec−1 was then examined ( Figure 1c). The remaining PSII activity after high light exposure gradually increased with an increase in the period of acclimatization, and the response to acclimatization reached a plateau within 15 h. Thus, exposure to white light of 600 µE m−2 sec−1 for 15–24 h is sufficient for the acclimatization response.

To examine the effects of the quality of light, the plants were acclimatized to blue or red light ( Figure 1b, B200, R200). The PSII activity on exposure to high-intensity light after acclimatization to blue light of 200 µE m−2 sec−1 was comparable to that after acclimatization to white light of 200 µE m−2 sec−1. On the other hand, the plants acclimatized to red light of 200 µE m−2 sec−1 showed decreased PSII activity compared with that of blue light-acclimatized plants after high irradiation. Consistent with these results, the blue light-acclimatized plants did not show apparent photo-bleaching ( Figure 1a, B200), whereas the red light-acclimatized plants showed photo-bleaching ( Figure 1a, R200).

Molecular cloning of the light-stress inducible cDNA for zinc finger protein RHL41 using a differential display technique

We considered that the expression of several genes involved in photoprotection is probably induced during the acclimatization process, which in turn confers tolerance to high irradiation. Therefore, genes induced during the acclimatization period were cloned and characterized using a differential display technique. Total RNA isolated from leaves acclimatized to white light of 600 µE m−2 sec−1 for 10 min to 15 h was compared to non-acclimatized controls, and 49 differentially expressed cDNA clones were isolated and sequenced. The identified clones include photosynthesis-related genes, such as those encoding ferredoxin-NADP+ reductase ( Newman and Gray, 1988), D2 protein ( Jansen et al., 1999 ) and chlorophyll a/b binding protein ( Castresana et al., 1988 ). They also included stress-inducible genes such as trypsin inhibitor and catalase genes ( Frugoli et al., 1996 ). A factor for potential transcriptional regulation such as zinc finger protein RHL41, a factor for translational regulation such as acidic ribosomal protein P1 ( Newman et al., 1994 ; Wool et al., 1991 ) and the nitrate reductase gene ( Crawford et al., 1988 ) were also identified. Northern blot analysis revealed that mRNAs for the D2 protein, trypsin inhibitor, catalase, zinc finger protein RHL41 and ribosomal protein P1 were specifically induced by exposure to white light of 600 µE m−2 sec−1 (data not shown).

We focused on the cDNA for the zinc finger protein RHL41, because it could function as a potential regulatory gene in signal transduction pathways for photoprotection. The full-length cDNA for the rhl41 gene was cloned and analysed. Figure 2 shows the amino acid sequence of RHL41 deduced from the nucleotide sequence of cDNA. RHL41 consists of 162 amino acid residues with a calculated molecular mass of 17.3 kDa. RHL41 was found to be identical to the ZAT12 protein, one of a variety of zinc finger proteins of Arabidopsis that had been cloned using PCR techniques ( Meissner and Michael, 1997), but its function has not been assigned. RHL41 is a TFIIIA-type zinc finger protein that has two Cys2/His2 zinc finger motifs ( Takatsuji, 1998). The amino acid sequence of RHL41 was compared with those of other plant TFIIIA-type zinc finger proteins, i.e. petunia zinc finger DNA-binding proteins EPF1 and EPF2-4 ( Takatsuji et al., 1992 ), a wheat zinc finger DNA-binding protein WZF1 ( Sakamoto et al., 1993 ) and an yeast salt-tolerance zinc finger protein STZ ( Lippuner et al., 1996 ) ( Figure 2). The alignment shows that the sequence homology of RHL41 with the other zinc finger proteins is very low outside of the motifs. The length of the spacers between motifs, which is important for target sequence recognition in RHL41, is the shortest among these TFIIIA-type zinc finger proteins. RHL41 contains the QALGGH sequence in the two motifs, which was found to be essential for DNA binding ( Kubo et al., 1998 ). It does not contain the basic B-box which has been found in the other proteins ( Lippuner et al., 1996 ).

Figure 2.

Amino acid sequence of zinc finger protein RHL41.

The deduced amino acid sequence of RHL41 was aligned with the sequences of petunia zinc finger DNA-binding proteins EPF1 and EPF2-4, yeast salt-tolerance zinc finger protein STZ, and wheat zinc finger DNA-binding protein WZF1. The basic B-box is indicated by double underlining, and zinc finger motifs are indicated by bold underlining. Identical amino acids are indicated by open boxes, and gaps introduced to maximize alignment are indicated by dashes.

The light-stress response of rhl41 gene expression was investigated by Northern blot analysis. rhl41 gene expression was induced within 1 h after exposure to medium-intensity light of 600 µE m−2 sec−1, and slightly increased up to 16 h. It was strongly induced by exposure to high-intensity light (1900 µE m−2 sec−1) for 5 h. The accumulation of rhl41 mRNA in response to the high-intensity light was more than 10 times that in response to the medium-intensity light. rhl41 gene expression was also induced by either blue or red light, but the expression level was about 1.5-fold higher by blue light than that by red light ( Figure 3a). These results clearly show that the rhl4l gene is highly responsive to light stress, indicating that it may play an regulatory role in the acclimatization process.

Figure 3.

RNA gel blot analysis of rhl41 transcripts.

(a) Expression of the rhl41 gene in response to medium-intensity, high-intensity, blue and red light. Arabidopsis plants grown under continuous white light of 75 µE m−2 sec−1 were exposed to medium-intensity light of 600 µE m−2 sec−1 for 1–16 h, high-intensity light of 1900 µE m−2 sec−1 for 5 h, blue light of 300 µE m−2 sec−1 for 6 h, or red light of 450 µE m−2 sec−1 for 6 h. (b) Time course of rhl41 expression under high-intensity light stress. Arabidopsis plants grown under continuous white light of 75 µE m−2 sec−1 were exposed to high-intensity light of 1900 µE m−2 sec−1 for 0.5, 1, 2, 3, 5 and 7 h (+ High light), transferred to normal light conditions (75 µE m−2 sec−1) and then incubated for 0.5, 1, 2 and 16 h (– High light). (c) Expression of the rhl41 gene in response to various intensities of light. Arabidopsis plants grown under continuous white light of 75 µE m−2 sec−1 were exposed to white light of 300, 600, 1000 1500 or 2000 µE m−2 sec−1 for 3 h. Total RNA was extracted from leaves after light exposure, and 20 µg of each treatment was fractionated on gels, transferred to a nylon membrane, and hybridized to rhl41 probe.

rhl41 gene expression was induced within 30 min after exposure to high-intensity light of 1900 µE m−2 sec−1, and it was strongly expressed after 5–7 h of irradiation ( Figure 3b). When plants under high-light stress were subsequently transferred to a weak light condition, a reduction of > 50% in the rhl41 mRNA accumulated under high-light conditions occurred within 0.5 h ( Figure 3b). However, low level of rhl41 mRNA remained for up to 16 h after termination of the high-light stress.

The level of rhl41 mRNA gradually increased when the plants were exposed to light intensity of 300–1500 µE m−2 sec−1. Interestingly, a markedly increase in the level of rhl41 mRNA was observed when plants were exposed to 2000 µE m−2 sec−1 light. The amount of rhl41 transcript at 2000 µE m−2 sec−1 was five times greater than that at 1500 µE m−2 sec−1 ( Figure 3c).

Tolerance to high irradiation in transgenic plants that over-express the rhl41 gene

We generated transgenic plants to verify the functions of RHL41 in photoprotective responses, in which rhl41 cDNA was introduced to over-express RHL41 protein. Arabidopsis plants were transformed with a binary vector carrying fusions of the CaMV 35S promoter and the rhl41 cDNA in the sense orientation (pIG-RHL41-S). Fifty-two hygromycin-resistant independent transformants were obtained by Agrobacterium-mediated root transformation. PCR analysis of the transgenic plants showed that 79% of the analysed clones had the introduced rhl41 gene (data not shown). The leaves of transgenic plants harbouring pIG-RHL41-S grown under normal light conditions (75 µE m−2 sec−1) were darker green in colour compared with those of the wild-type plants ( Figure 4a). Leaf morphology was also modified in transgenic plants, which exhibited a rounder shape with a shorter petiole compared with leaves of the wild-type plants. All of the transgenic plants examined had grown to bolting, flowered and set seeds. Transgenic plants of the T2 generation were used for further analysis. After exposure to high-intensity light of 2000 µE m−2 sec−1 for 8 h, the leaves of the wild-type plants were severely damaged ( Figure 4b). On the other hand, the leaves of sense transgenic plants were unaffected. These results clearly show that constitutive expression of the rhl41 gene conferred enhanced tolerance to high irradiation.

Figure 4.

Transgenic plants that over-express the rhl41 gene.

(a) Transgenic Arabidopsis plants that over-expressed the rhl41 gene (right, line S18–2) and wild-type plants (left) were grown under continuous white light of 75 µE m−2 sec−1 for 26 days. The stem was cut off to obtain a clear photograph of rosette leaves. (b) Comparison of photo-oxidation damage of leaves from wild-type (left) and transgenic (line S10-3) (right) Arabidopsis plants. Plants grown under continuous white light of 75 µE m−2 sec−1 were exposed to high-intensity light of 2000 µE m−2 sec−1 for 8 h.

To test whether introduction of the rhl41 gene enhances tolerance to high irradiation, wild-type plants and transgenic plants harbouring either a vector control or pIG-RHL41-S were exposed to high-intensity light of 2000 µE m−2 sec−1 for 6 h, and the PSII activity was examined by determination of chlorophyll fluorescence ( Figure 5a). The PSII activity in leaves of transgenic plants harbouring pIG-RHL41-S after high-light exposure was higher than that of wild-type plants or transgenic plants harbouring a vector control. We measured the PSII activity in 183 T2 plants from 52 independent lines and confirmed that almost all sense transgenic plants exhibited a higher tolerance to high irradiation compared to wild-type plants. We then selected five relatively highly tolerant clones, S3-2, S12-1, S16-1, S18-2 and S18-3, for further analysis.

Figure 5.

Tolerance to high irradiation in transgenic plants that over-express the rhl41 gene.

(a) High-intensity light tolerance of transgenic Arabidopsis plant leaves estimated by the change in PSII activity. Wild-type plants (WT), transgenic plants harbouring the vector control (HM), and transgenic plants harbouring pIG-RHL41-S (S16-1, S18-2, S18-3), which had been grown under continuous white light of 75 µE m−2 sec−1, were exposed to high-intensity light of 2000 µE m−2 sec−1 for 6 h. The chlorophyll fluorescence was measured before (–HL) and after (+HL) high irradiation. (b) Expression of the rhl41 gene in transgenic plants. Total RNA was extracted from high-intensity light-exposed (+) and non-exposed (–) leaves under the same conditions as in (a), and 20 µg of each was fractionated on gels, transferred to a nylon membrane, and hybridized to rhl41 probe. (c) Decreased tolerance to high irradiation in antisense transgenic plants. Wild-type plants (WT), transgenic plants harbouring pIG-RHL41-S (S3-2) and transgenic plants harbouring pIG-RHL41-A (A2-1, A4-1), which had been grown under continuous white light of 75 µE m−2 sec−1, were exposed to high-intensity light of 2000 µE m−2 sec−1 for 6 h, and the chlorophyll fluorescence was measured.

The light-stress response of the rhl41 mRNA in transgenic plants of S16-1, S18-2 and S18-3 was examined using RNA blot analysis ( Figure 5b). The expression pattern of the rhl41 gene in transgenic plants harbouring a vector control was comparable to that in wild-type plants; little rhl41 gene expression was detected under unstressed conditions, and was strongly induced only by exposure to high-intensity light of 2000 µE m−2 sec−1 for 6 h. On the other hand, a high level of rhl41 mRNA was detected in transgenic plants harbouring pIG-RHL41-S even under unstressed conditions. The level of rhl41 mRNA after high irradiation varied among the clones, probably due to the expression of the endogenous rhl41 gene induced by high-intensity light.

We also generated transgenic plants harbouring the rhl41 cDNA in the antisense orientation (pIG-RHL41-A), in which accumulation of rhl41 transcripts was reduced by 35–63% compared with that in wild-type plants. As expected, the PSII activity in antisense transgenic plants was lower than that in wild-type plants during exposure to high irradiation ( Figure 5c).

Acclimatization response in transgenic plants

We then determined whether the acclimatized transgenic plants would exhibit increased tolerance to high irradiation as compared with non-acclimatized transgenic plants, and whether non-acclimatized transgenic plants would show a comparable tolerance to acclimatized wild-type plants. Non-acclimatized ( Figure 6a, NA) and acclimatized ( Figure 6b, A) transgenic or wild-type plants were exposed to high-intensity light of 2000 µE m−2 sec−1 for 6 h, and the time course of PSII activity was examined. The non-acclimatized transgenic plants always showed higher PSII activity than non-acclimatized wild-type plants during 1–6 h of exposure to high irradiation ( Figure 6a). The PSII activity in acclimatized transgenic plants was slightly higher than that in acclimatized wild-type plants, and in particular the line S18-2 showed significantly high PSII activity compared with wild-type plants ( Figure 6b). An enhancement of high-light tolerance in acclimatized transgenic plants was apparent compared with non-acclimatized transgenic plants, and non-acclimatized transgenic plants showed almost the same or a slightly lower level of tolerance than did the wild-type plants after acclimatization.

Figure 6.

Increased tolerance to high irradiance in transgenic plants by light acclimatization.

(a) Time course of PSII activity in non-acclimatized wild-type (WT) and transgenic plants during exposure to high irradiation. The wild-type plants and transgenic plants (S3-2, S18-2 and S18-3) were exposed to 2000 µE m−2 sec−1 for 6 h, and the chlorophyll fluorescence was measured every hour. (b) Time course of PSII activity in acclimatized wild-type and transgenic plants during exposure to high irradiation. The wild-type plants (WT) and transgenic plants (S3-2, S18-2 and S18-3) were acclimatized to medium light of 600 µE m−2 sec−1 for 17 h, and then exposed to high-intensity light of 2000 µE m−2 sec−1 for 6 h. The chlorophyll fluorescence was measured every hour.

Morphological changes in leaves of transgenic plants

Transgenic plants grown under normal light conditions had rounder shaped dark green leaves ( Figure 4), an observation which suggests changes in modulation of the structure of leaf tissue and/or the accumulation of pigments in the leaves. In cross-sections of leaves of 20-day-old plants, there was a significant increase in leaf thickness in transgenic plants ( Figure 7a). In wild-type plants, leaf thickness was 298 ± 52 µm on average, while the leaf thickness in transgenic lines S3-2 and S18-2 was 515 ± 92 and 386 ± 69 µm, corresponding to 178% and 130% of the wild-type value, respectively. In wild-type plants, leaves consisted of layers of two or three cells, whereas in transgenic plants, leaves showed a well-developed structure, consisting of a typical spongy and palisade parenchyma. Significant changes in individual cell size were not observed. To examine whether long-term acclimatization induces the formation of palisade parenchyma in wild-type plants, leaf sections of wild-type plants that had been acclimatized for 2 weeks under the 200 µE m−2 sec−1 light were analysed ( Figure 7a). As expected, the acclimatized wild-type leaves developed palisade parenchyma and the thickness of the leaves was 512 ± 93 µm. In accordance with a high population of chloroplasts in leaf tissues of the transgenic plants, the amount of chlorophyll on a fresh weight basis in transgenic plants was 1.5–1.8 times greater than that in wild-type plants ( Figure 7c), but no difference in the chlorophyll a/b ratio was observed between transgenic and wild-type plants (data not shown).

Figure 7.

Phenotypic changes in transgenic plants.

(a) Cross-sections through wild-type (WT), long-term acclimatized wild-type (WT + A) and transgenic (S3-2 and S18-2) Arabidopsis leaves. The four photographs shown are of the same magnification. Bar = 100 µm. The average thickness in 15 leaves from three plants for each line was determined and is indicated in parentheses. (b) Accumulation of anthocyanin in leaves of transgenic plants in response to light stress. Wild-type plants (WT) and transgenic plants (S3-2, S18-2 and S18-3) grown under continuous white light of 75 µE m−2 sec−1 (NA) were acclimatized to medium light of 600 µE m−2 sec−1 for 16.5 h (A), and then exposed to high-intensity light of 2000 µE m−2 sec−1 for 6 h (A + HL). Anthocyanin was extracted from leaves that received each treatment, as described in Experimental procedures. The data are the averaged value of three independent experiments. (c) Increased chlorophyll content in the leaves of transgenic plants. Chlorophyll was extracted from leaves of the wild-type plants (WT) and transgenic plants (S3-2, S12-1 and S18-3) grown under continuous white light of 75 µE m−2 sec−1 as described in Experimental procedures. The data are the averaged values of three to five independent experiments.

We then investigated the induction and accumulation of anthocyanin in the leaves in response to light stress ( Figure 7b). Transgenic plants grown under normal light conditions (NA) contained elevated amounts of anthocyanin compared with wild-type plants, and the amount in transgenic plants was 9–21 times greater that that in wild-type plants. Induction and accumulation of anthocyanin by acclimatization treatment (A) was observed in both wild-type and transgenic plants, and further accumulation of anthocyanin was observed after high irradiation (A + HL). A remarkable accumulation was observed in line S3-2, in which thickness of the leaves was much increased.

Discussion

We found that acclimatization to medium-intensity light stress markedly enhanced tolerance to high irradiation in Arabidopsis plants, and that the acclimatization response depended on the quantity of light ( Figure 1). Plants became acclimatized to high light stress even by 1 h exposure to the medium-intensity light ( Figure 1c). Karpinski et al. (1999) reported that systemic acclimatization occurs rapidly after high irradiation . They showed that partial exposure to excess light for 30 min in low light-adapted Arabidopsis plants results in a systemic acclimatization to excess excitation energy in unexposed leaves. The acclimatization response proceeded by further exposure to medium-intensity light stress, and the tolerance level was gradually increased by the biological and physiological changes that occur with the acclimatization response.

Response to light acclimatization also depended on its spectrum ( Figure 1b). Blue light was most effective for light acclimatization, suggesting that blue light is the major signal in the rapid acclimatization response. It has been reported that wild-type Arabidopsis plants grown in red light show abolished acclimatization to light intensity, with marked changes in the maximum rates of photosynthesis and chlorophyll a/b ratio, which also suggests a key role for blue light in the regulation of acclimatization to irradiance ( Walters and Horton, 1995b) . Mutants defective in phytochrome photoreceptors also showed photosynthetic changes comparable to those observed in the wild-type plants, indicating that phytochrome is not directly involved in light acclimatization ( Anderson et al., 1995 ; Walters and Horton, 1994; Walters and Horton, 1995a; Walters and Horton, 1995b).

Northern blot analysis showed that the rhl41 gene was highly responsive to light stress and the expression was greatly induced by exposure to high- rather than medium-intensity irradiation ( Figure 3a,c). The light quality experiment showed that rhl41 gene expression was induced by blue rather than red light ( Figure 3a), and the acclimatization response was mainly induced by blue light ( Figure 1a,b). The exact relationship between RHL41 expression level and acclimatization response to light quality may be revealed by examining the RHL41 expression pattern at each spectrum. A possible explanation for this result is that there might be post-transcriptional regulation of the rhl41 gene in the response to red light. Continuous light may not be physiologically appropriate for Arabidopsis plants as such conditions affect the phytochrome responses ( Anderson et al., 1997 ; Yamaguchi et al., 1999 ). A different light response of the rhl41 gene might be obtained if light/dark conditions were used instead of continuous light. The rhl41 mRNA showed a rapid turnover in response to light stress; its expression was induced within 30 min after light stress and decreased within 30 min after removal of light stress ( Figure 3b). The level of rhl41 mRNA was maximal after 5 h exposure to high-intensity light, at which photo-oxidation in the leaves had already begun ( Figure 3b). It could be considered that the photoprotective response was rapidly triggered by rhl41 expression, but the prepared defensive state was insufficient to overcome excess photo-oxidation damage. It is interesting that there was a threshold of light intensity in the induction of rhl41 expression at between 1500 and 2000 µE m−2 sec−1 ( Figure 3c). Up-regulation of the rhl41 gene may be controlled by at least two different mechanisms.

The function of RHL41 as a regulatory factor in the photoprotective process is supported by our analysis of transgenic plants that over-expressed the rhl41 cDNA. Transgenic plants exhibited enhanced tolerance to irradiation ( Figure 5), with phenotypic changes such as development of palisade parenchyma and increased anthocyanin and chlorophyll contents ( Figure 7). Northern blot analysis of the transgenic plants confirmed that the rhl41 gene was constitutively expressed even under unstressed conditions ( Figure 5). Additional increases in rhl41 expression derived from the endogenous rhl41 gene by exposure to high-intensity light varied in the tested clones. The high level of induction under stressed conditions observed in clone S16-1 could be explained as follows ( Figure 5b). The expression pattern of the rhl41 gene is thought to consist of two phases, as shown in the time course experiment ( Figure 3b). During exposure to high-intensity light for 3 h, rhl41 expression is relatively low, and photo-oxidative damage of the plants is not apparent (first stage). After exposure to high light for 5–7 h, the transcript level is strikingly increased, and photo-oxidation of the leaves is proceeding (second stage). In the second stage, no correlation between the expression level of rhl41 and tolerance to high irradiation is observed. It could be considered that the photoprotective response was rapidly triggered by rhl41 expression, but the prepared defensive state was no longer enough to overcome excess photo-oxidation damage in the second stage, as discussed above. The high level of transcript in line S16-1 after high-intensity light treatment is due to endogenous rhl41 gene expression in the second stage. From the data on PSII activity, photo-oxidation could be proceeding in the line S16-1 in this second stage. The level of tolerance to high irradiation could correlate with the level of RHL41 expression under unstressed conditions. In our experiment, all transgenic plants, including line S16-1, which expressed the rhl41 gene under unstressed conditions, exhibited statistically increased PSII activity, i.e. increased tolerance to high light.

To investigate the PSII status in detail, we analysed the relative variable fluorescence (Fm′ − Ft)/Fm′ ( Genty et al., 1989 ) in wild-type and transgenic plants under unstressed and stressed conditions. The results show that transgenic plants have a relatively high (Fm′ − Ft)/Fm′ value under unstressed conditions, suggesting that transgenic plants essentially have a large electron flux from PSII to PSI. Detailed analysis will be published elsewhere.

Antisense transgenic plants exhibited decreased tolerance to high irradiation, which may also support the function of RHL41 as a regulatory factor in the photoprotective process ( Figure 5c). More definite results may be obtained using an rhl41 gene-disrupted line, in which the effect of rhl41 expression is completely removed. Thus, screening from T-DNA tag lines is under way.

Non-acclimatized transgenic plants showed almost the same or only a slightly lower level of tolerance compared with the acclimatized wild-type plants ( Figure 6), which suggests that the RHL41 protein plays a key role in the adaptive response to increased light intensity. In contrast, tolerance to high-intensity irradiation in transgenic plants was slightly increased by acclimatization treatment, suggesting that enforced expression of the rhl41 gene alone is not sufficient to orchestrate the full array of biological and physiological changes that occur during the acclimatization response, and that an alternative or additional pathway, probably complementary to the RHL41 protein-mediated process, might be involved in the acclimatization response.

Transgenic plants had thick leaves as the palisade parenchyma was developed ( Figure 7a). Wild-type plants that had been acclimatized for 2 weeks under medium-intensity light stress also developed palisade parenchyma. As the rhl41 gene is constitutively expressed under these light conditions in wild-type plants, this result strongly suggests that RHL41 could control palisade parenchyma formation. In general, high-intensity light-grown plants have a developed palisade parenchyma allowing greater utilization of high light levels compared with low-intensity light-grown plants ( Kimura et al., 1998 ). Thus, the thickness of the leaves could be an essential factor for enhancement of tolerance to irradiation in transgenic plants. It is clear that elevated anthocyanin production is caused by expression of the rhl41 gene ( Figure 7b). The RHL41 may regulate some genes involved in anthocyanin biosynthesis, either directly or indirectly, leading to increased anthocyanin production in transgenic plants under both unstressed and stressed conditions. Anthocyanin production is triggered by numerous stress conditions such as wounding, low temperature, pathogen attack and exposure to ozone ( Mol et al., 1996 ). Expression of the rhl41 gene may induce this stress response, thereby increasing anthocyanin accumulation. Accumulation of anthocyanin may be one explanation for the development of tolerance to irradiation in transgenic plants, since it is known to be produced by light and UV irradiation and function as a photoprotective pigment ( Caldwell et al., 1983 ). Chlorophyll content was also increased in transgenic plants ( Figure 7c), suggesting that chlorophyll biosynthesis or the organization of the chloroplast may be affected by rhl41 expression. Photosynthetic efficiency may be increased by increased amounts of chlorophyll, resulting in enhanced tolerance to excess light.

The introduction of one transcription factor gene could lead to the expression of several protective genes against stresses, resulting in a greater increase in tolerance to environmental stresses. It was reported that increased expression of Arabidopsis CBF1, a transcriptional activator that binds to the DNA regulatory element of cor (cold-regulated) genes, induced cor gene expression and increased freezing tolerance ( Jaglo-Ottosen et al., 1998 ). Liu et al. (1998) reported that transgenic Arabidopsis plants that over-express transcription factor DREB1, the DRE (dehydration-responsive element) binding protein, showed tolerance to freezing and dehydration. Transgenic plants over-expressing RHL41 protein showed enhanced tolerance to high irradiation with drastic morphological changes, such as palisade parenchyma formation and increased amounts of the light protectant anthocyanin and of chlorophyll. These changes could have been induced during the long-term acclimatization process in the Arabidopsis plants, and probably function in concert to overcome excess light stress. Thus, our data suggest that RHL41 as a transcription factor could regulate several acclimatization-responsive genes, especially genes relating to palisade parenchyma formation. Transgenic plants are considered to exhibit the orchestrated response caused by long-term acclimatization because the rhl41 gene and genes under its control are constitutively expressed. The use of a chemical-induced promoter will provide information about the early light acclimatization events after induction of rhl41 gene expression, and analysis of an rhl41 gene-disrupted line will make it possible to identify target genes of zinc finger protein RHL41.

Experimental procedures

Plant material and light conditions

Arabidopsis (Arabidopsis thaliana ecotypes Landsberg and Wassilewskija (WS)) plants were grown at 23°C in growth chambers at 90 µE m−2 sec−1 with continuous light. Two-week-old Arabidopsis plants (ecotype landsberg) were used for light acclimatization experiments. In the case of transgenic plants, surface-sterilized seeds of Arabidopsis (ecotype WS) were placed on MS medium containing 20 µg ml−1 hygromycin and 0.2% gelrite (Wako, Osaka, Japan) in a 150 mm diameter plastic Petri dish, and cultured at 90 µE m−2 sec−1 with continuous light at 22°C for 2 weeks. The hygromycin-resistant seedlings were transferred to soil, grown at 23°C in growth chambers at 90 µE m−2 sec−1 with continuous light for a week, and used for experiments.

Plants were exposed to various intensities of light (200–2000 µE m−2 sec−1) at 23°C at 70% humidity in a specially constructed photo-environmental simulator (Tabai Espec, Osaka, Japan) , equipped with xenon lamps (Ushio Inc., Osaka, Japan) as a light source. To obtain a light spectrum similar to sunlight, a special filter, which reduces the xenon lamp-specific spectrum present in the infra-red region, was attached to the xenon lamps. An infra-red cut filter was used to prevent radiation heat. Different light qualities were obtained using blue (350–560 nm) and red (560–800 nm) band filters (Tabai Espec, Osaka, Japan). The photon fluence rates of the red and blue light were measured using a photometer (Opto Research, Tokyo, Japan) . The plants were exposed to red light of 450 µE m−2 sec−1 or blue light of 300 µE m−2 sec−1 for 24 h in the light acclimatization experiment and for 6 h in Northern blot analysis.

In the long-term acclimatization experiment, surface-sterilized seeds were placed on MS medium containing 20 µg ml−1 hygromycin and 0.2% gelrite in a 150 mm diameter plastic Petri dish, and cultured at 90 µE m−2 sec−1 for a week, and then transferred to 200 µE m−2 sec−1 and cultured for a week. The hygromycin-resistant seedlings were transferred to soil, and cultured at 200 µE m−2 sec−1 for another week.

Measurement of photosynthetic activity of leaves

The photochemical efficiency of photosystem II in intact leaves was determined as the ratio of variable to maximum chlorophyll fluorescence (Fv/Fm) ( Schreiber et al., 1995 ), using a portable fluorometer PAM-2000 (Walz, Effeltrich, Germany). The values were calculated from data for 15 leaves from three plants.

Differential display

Plants were acclimatized to medium-intensity light (600 µE m−2 sec−1) for 10 min to 15 h, then total RNA from leaves of acclimatized and non-acclimatized plants was extracted using ISOGEN kits (Nippon Gene, Tokyo, Japan) .

Differential display of mRNA was performed according to Liang and Pardee (1992) using the RNAmapTM mRNA differential display system (GenHunter Corporation) . T12MN or HT11N (GenHunter Corporation) were used for anchored primers, and AP-1–AP-5, HAP1 and HAP10–HAP15 were used for arbitrary primers (GenHunter Corporation, MA, USA) were used. Differentially displayed bands were cut out and eluted in 100 µl of distilled water at 100°C for 15 min. DNA was ethanol-precipitated, re-amplified by PCR, and then analysed on agarose gels, after which bands of the expected size were purified and cloned into pCRTM2.1 vector (Invitrogen). When HT11N and HAP1,10 or 11 were used as primers, the purified bands were digested with HindIII and cloned into the HindIII site of the pUC119 vector. Cloned fragments were sequenced and homology was determined using the DDBJ Homology Search System .

RNA isolation and Northern blot analysis

The leaves were frozen in liquid nitrogen and stored at −80°C until RNA extraction. Total RNA was extracted from leaves, using ISOGEN kits (Nippon Gene, Tokyo, Japan). The extracted RNA (20 µg for each sample) was electrophoresed on formaldehyde–agarose gels ( Sambrook et al., 1989 ), transferred onto nylon membranes (Hybond N+, Amersham) by capillary blot, and fixed by incubation at 80°C for 2 h. Blots were hybridized using cloned fragments obtained from display gels or fragments of the full-length cDNA clones as probes. The probes were labelled with 32P (Boehringer Mannheim) following the manufacturer's instructions. Membranes were hybridized overnight with the probe at 55°C in 6.0 × SSC containing 1% SDS and 100 µg ml−1 calf thymus DNA, then were washed twice with 0.1 ×SSC containing 0.1% SDS for 15 min at room temperature and once with 0.1 × SSC containing 0.1% SDS for 15 min at 50°C.

Cloning of the rhl41 gene

The full length of the rhl41 gene was cloned by RACE according to the method of Frohman (1995). λDNA from the cDNA library of 5-day-old seedlings was used as template, and KS30 primer (5′-CCGGGCCCCCCCTCGAGGTCGACGGTATCG-3′) specific to the left arm of λ DNA and G142 primer (5′-GATAATCTCATAACAAATCTCCAAT GCTAC-3′) specific to the 3′ end of the rhl41 gene were used for PCR. The reaction mixture was incubated at 94°C for 5 min, and then 40 PCR cycles (94°C for 1 min, 65°C for 2 min and 72°C for 3 min) were run, followed by 5 min of elongation at 72°C. The 0.7 kb PCR fragments were gel-purified, cloned into pCRTM2.1 vector (Invitrogen), and sequenced.

Construction of vector plasmid and transformation of Arabidopsis thaliana

cDNA encoding the rhl41 gene was excised at NotI/SacI sites from the pCRTM2.1 vector and inserted into a binary vector plasmid pIG121-HM obtained from K. Nakamura of Nagoya University ( Akama et al., 1992 ; Hiei et al., 1994 ) in place of the gene for β-glucuronidase ( Jefferson et al., 1987 ) in the sense or antisense orientation between 35S cauliflower mosaic virus promoter ( Benfey et al., 1989 ) and nopaline synthase terminator to construct pIG-RHL41-S or pIG-RHL41-A, respectively. The plasmids pIG121-HM, pIG-RHL41-S and pIG-RHL41-A were introduced into Agrobacterium tumefaciens C58C1 rifr (pGV2260) (recA+). Arabidopsis thaliana ecotype WS was transformed according to the Agrobacterium-based procedure described by Valvekens et al. (1988) . The root explants were cultivated on callus-inducing medium for 3 days, infected with Agrobacterium tumefaciens harbouring pIG121-HM, pIG-RHL41-S or pIG-RHL41-A, and co-cultivated for 3 days. After removal of Agrobacterium by washing with 250 µg ml−1 Claforan (Hoechst Japan Ltd, Tokyo, Japan), the explants were transferred to shoot-inducing medium that contained 20 µg ml−1 hygromycin B and cultivated for 4 weeks. Shoots induced from green calli of the transformed cells were transferred to root-inducing medium, and seeds were harvested after 4 weeks. The antibiotic resistance of the transgenic seeds was examined by sowing them on germination medium containing 20 µg ml−1 hygromycin B.

Extraction and analysis of anthocyanin and chlorophyll pigments

The leaves were frozen in liquid nitrogen and stored at −80°C until extraction. Leaves (0.3–0.5 g) were homogenized with 1 ml of extraction buffer from the ISOGEN kit (Nippon Gene, Tokyo, Japan). The homogenates were stored for 5 min at room temperature, and 0.2 ml of chloroform was added to the homogenate. After vigorous shaking, the homogenate was centrifuged at 18 000 g for 15 min at 4°C. The quantity of anthocyanins was determined by spectrophotometric measurements of the aqueous phase (A530 − A657) and normalized to the total fresh weight of tissue ( Rabino and Mancinelli, 1986).

Chlorophyll was extracted with buffered 80% aqueous acetone according to the method of Porra et al. (1989) . The quantity of chlorophyll was normalized to the total fresh weight of tissue.

Preparation of leaf sections

Leaf tissues were sliced manually using a razor. After stretching the cuttings on water, they were placed on slides and analysed using an UFX-II microscope (Nikon). Leaf thickness was determined by observation of photographs taken under a microscope. The leaf thickness was determined on 28–60 cross-sections obtained from three plants for each sample. It was measured at one to three positions in each section. Data are given as mean ± SD.

Acknowledgements

We thank M. Ohara for providing helpful comments on the manuscript. We are grateful to Professor K. Nakamura, Nagoya University, for critical reading of the manuscript. We also thank Professor K. Asada, Fukuyama University, for critical reading of and commenting on the manuscript. We also thank Toshinori Shiraishi for his excellent technical assistance and commenting on the manuscript. This work was supported by the Petroleum Energy Center (PEC) and the Research Association for Biotechnology (RAB) of Japan.

Ancillary

Advertisement