Molecular Plant Advance Access originally published online on October 31, 2007
Molecular Plant 2008 1(1):84-102; doi:10.1093/mp/ssm010
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2007. Published by Oxford University Press on behalf of CSPP and IPPE, SIBS, CAS.
A Rice Phytochrome A in Arabidopsis: The Role of the N-terminus under red and far-red light
a Ludwig-Maximilians-Universität München, Bereich Botanik, Menzinger Str. 67, 80638 München, Germany
b Hitachi Advanced Research Laboratory, Hatoyama, Saitama 350-0395, Japan
1 To whom correspondence should be addressed. E-mail c.bolle{at}lrz.uni-muenchen.de, tel ++49 (0)89-17861288, fax ++49 (0)89-1782274
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONRESULTSDISCUSSIONMATERIAL AND METHODS |
|---|
The phytochrome (phy)A and phyB photoreceptors mediate three photobiological response modes in plants; whereas phyA can mediate the very-low-fluence response (VLFR), the high-irradiance response (HIR) and, to some extent, the low fluence response (LFR), phyB and other type II phytochromes only mediate the LFR. To investigate to what level a rice phyA can complement for Arabidopsis phyA or phyB function and to evaluate the role of the serine residues in the first 20 amino acids of the N-terminus of phyA, we examined VLFR, LFR, and HIR responses in phyB and phyAphyB mutant plants transformed with rice PHYA cDNA or a mutant rice PHYA cDNA in which the first 10 serine residues were mutated to alanines (phyA SA). Utilizing mutants without endogenous phyB allowed the evaluation of red-light-derived responses sensed by the rice phyA. In summary, the WT rice phyA could complement VLFR and LFR responses such as inhibition of hypocotyl elongation under pulses of FR or continuous R light, induction of flowering and leaf expansion, whereas the phyA SA was more specific for HIR responses (e.g. inhibition of hypocotyl elongation and anthocyanin accumulation under continuous far-red light). As the N-terminal serines can no longer be phosphorylated in the phyA SA mutant, this suggests a role for phosphorylation discriminating between the different phyA-dependent responses. The efficacy of the rice phyA expressed in Arabidopsis was dependent upon the developmental age of the plants analyzed and on the physiological response, suggesting a stage-dependent downstream modulation of phytochrome signaling.
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMATERIAL AND METHODS |
|---|
Monitoring environmental light is the most important clue for plants to evaluate their living conditions. During evolution, three major classes of photoreceptors have been developed in higher plants: the red/far-red-absorbing phytochromes, the blue/UV-A-absorbing cryptochromes, and the phototropins (Schäfer and Nagy, 2006; Wada et al., 2006). The phytochromes are a family of red/far-red-responsive photoreceptors, ubiquitous in plants, but also identified in many prokaryotic species and fungi (Lamparter, 2004; Mathews, 2006). These approximately 125-kDA proteins act as dimers and each protein contains a covalently attached tetrapyrole (bilin) chromophore. With the help of the chromophore, the protein can photoconvert between two different states—the red-absorbing (Pr) and far-red-absorbing (Pfr, active) forms (Quail, 1997). It is generally assumed that the Pfr form is the active state, which initiates the signal transduction.
Phytochromes have a conserved N-terminal photosensory domain (input) and a C-terminal regulatory domain (output), which typically includes two PAS-domains and a histidine-kinase-related region (Montgomery and Lagarias, 2002; Rockwell et al., 2006). Homodimerization and light-dependent nuclear targeting are mediated by the C-terminus. The photosensory domain contains an N-terminal Ser/Thr-rich extension (NTE) and a bilin-lyase domain can be discriminated. The bilin-lyase domain includes a GAF domain, which contains the chromophore attachment site.
The crystal structure of a bacteriophytochrome photosensory core gives the first indication on how the photosensory domain might function (Wagner et al., 2005, 2007). But, in contrast to bacteria, plant phytochromes have an N-terminal Ser/Thr-rich extension, whose function is not yet understood. Recently, it has been shown that the N-terminal photosensory domain can act as an output domain. Fused to a dimerization domain and a nuclear localization signal, the N-terminal domains of both phyA and phyB are still able to induce light-dependent signaling (Matsushita et al., 2003; Mateos et al., 2006). The phytochrome Pr:Pfr phototransformation has been proposed to involve phototransformation-dependent conformational changes. The NTE region undergoes a conformational change from random coil to amphiphilic 
-helix, which then interacts with the chromophore in the Pfr form (Deforce et al., 1994). Also, two Trp residues near the core regulatory region of oat phyA become preferentially exposed in the Pfr form (Wells et al., 1994).
The photoreceptor pigments for these responses have long been obscure, but physiological and photochemical studies have established that light-mediated responses (or photomorphogenesis) in plants can be classified into three different response modes according to their energy requirements (Briggs et al., 1984):
- (1) The low fluence response (LFR) that is characterized by its red (R)/far-red (FR) reversibility and requires 10–1000 µmol m–2 of R light for induction (Borthwick et al., 1952).
- (2) The very low fluence response (VLFR), which is induced by fluences below 1 µmol m–2 of light (Blaauw et al., 1968). In the laboratory, this can be simulated by short hourly pulses of R or FR light.
- (3) The FR-dependent high irradiance response (FR-HIR), which requires prolonged exposure to FR light of relatively high photon flux (Mohr and Wehrung, 1960).
- (2) The very low fluence response (VLFR), which is induced by fluences below 1 µmol m–2 of light (Blaauw et al., 1968). In the laboratory, this can be simulated by short hourly pulses of R or FR light.
In Arabidopsis, five discrete phytochrome-encoding genes—PHYA to PHYE—are present. They can be clustered by amino acid sequence similarity of the encoded proteins into two subfamilies—phyA and phyC, and phyB, phyD, and phyE (Clack et al., 1994). Phytochrome mutants and transgenic lines that overexpress individual phytochromes have been instrumental in determining the function of these photoreceptors. Physiological studies and action spectra for different responses revealed distinct and overlapping roles of different members of the Arabidopsis phytochrome family throughout photomorphogenesis (Quail, 1998; Whitelam et al., 1998; Shinomura et al., 1996, 2000). In Arabidopsis, phyA mutants were isolated by screening for a long hypocotyl phenotype under far-red light (Nagatani et al., 1993; Parks and Quail, 1993; Whitelam et al., 1993). Physiological analyses indicated that phyA mediates responses mainly to continuous FR light (HIR) and very low fluences (VLFR), but can also be responsible for red/far-red reversible low fluence responses (R/FR LFR) (Shinomura et al., 1996, 2000). As phyA is the only photoreceptor which can perceive FR light, it plays a major role in mediating hypocotyl inhibition under FR and FR-enriched environments, but it is also involved in blue light-sensing (Neff and Chory, 1998) and floral promotion (Johnson et al., 1994; Reed et al., 1994). Recent analyses have indicated that phyA also plays a role in perceiving high fluences (>160 µmol m–2 s–1) of R light (Franklin et al., 2007).
R-light perception is usually attributed only to the light-stable phytochromes (phyB–E), with phyB playing the predominant role. Mutants deficient in phyB exhibit reduced sensitivity to R light, especially noticeable by the inhibition of hypocotyl elongation under R light, and show a constitutive shade-avoidance phenotype (Somers et al., 1991; Reed et al., 1993, 1994; Halliday et al., 1994; Devlin et al., 1996). A pronounced petiole elongation, retarded leaf development, and early flowering are the phenotypes characteristic of the shade-avoidance syndrome (Franklin and Whitelam, 2005). The phytochromes phyC, D, and E have been identified as weak R-light sensors, with slightly diverse function in rosette leaf morphology, shade-avoidance responses and modulation of blue-light sensing (Franklin et al., 2003; Monte et al., 2003; Aukerman et al., 1997; Devlin et al., 1998, 1999; Hennig et al., 1999, 2002).
In rice, phyA seedlings are partially insensitive to continuous far-red light (HIR) and to pulses of R light (VLFR). phyAphyC double mutants showed no significant residual phytochrome responses under continuous far-red light (FRc), indicating that both are involved in the photoperception of FRc in rice (Takano et al., 2001, 2005; Xie et al., 2007). Responses to continuous R light were completely abolished in phyAphyB double mutants, but only partially in phyB and phyBphyC mutants, indicating that phyA and phyB act in a highly redundant manner to control de-etiolation under R light. The phyBphyC double mutant displayed a clear R/FR reversibility, suggesting that both phyA and phyB can mediate the low-fluence response (LFR). Rice is a short-day plant, and mutation in either phyB or phyC caused moderate early flowering under the long-day photoperiod, while monogenic phyA mutation had little effect on the flowering time. Both the phyAphyB and the phyAphyC double mutant, however, flowered very early. Therefore, rice phyA induces photomorphogenesis in a similar fashion as in Arabidopsis using two distinct modes of photoperception—the FR-HIR and the VLFR. Additionally, evidence accumulates for both dicots and monocots that phyA also mediates R/FR LFR (Long and Iino, 2001; Stowe-Evans et al., 2001; Takano et al., 2005).
phyA accumulates to relatively high levels in dark-grown seedlings, but most of it is rapidly lost upon transfer to light. The abundance of phyA is regulated at numerous levels, beginning at the expression level of the PHYA gene, which is under negative feedback control (Bruce et al., 1991). Combined with the high turnover rate of the PHYA mRNA, the transition from darkness to light causes a rapid drop in the levels of the PHYA mRNA and the synthesis of the PHYA apoprotein. Furthermore, phyA is subject to rapid light-induced proteolytic degradation, probably via the ubiquitin/26S proteasome pathway, as it could be shown that COP1 (an E3 ligase) interacts with phyA (Clough et al., 1999; Seo et al., 2004). Upon photoconversion to Pfr, phyA moves quickly into the nucleus, where it aggregates in small foci (Kircher et al., 1999; Hisada et al., 2000; Kim et al., 2000; Nagatani, 2004). The functions of these nuclear foci are still unknown, but seem to be essential for phytochrome function.
Phytochromes have been shown to be reversibly phosphorylated in vivo and in vitro on specific serine residues and possibly to fine-tune the phyA-dependent light-signaling pathway. Phosphorylation of phyA could reduce the affinity to bind for downstream targets, thereby leading to an attenuation of plant photoresponses (Kim et al., 2004).
In oat phyA, three main serine phosphorylation sites have been identified—Ser-7, Ser-17, and Ser-598. Phosphorylation at Ser-7 is similar in both Pr and Pfr forms; Ser-17 is phosphorylated preferentially in the Pr form, whereas Ser-598 in the Pfr form (McMichael and Lagarias, 1990; Lapko et al., 1996, 1997, 1999). Ser-598 is positioned in the hinge region, and transgenic Arabidopsis plants expressing oat phyA with an alanine substitution in this position are hypersensitive to light compared with the transgenic plants expressing wild-type oat phyA. This suggests that phytochrome phosphorylation at Ser-598 plays an inhibitory role (Kim et al., 2004).
Reversible protein phosphorylation would imply the action of both kinases and phosphatases on phytochromes. Until now, only phosphatases have been identified-, the cytoplasmic Ser/Thr-specific protein phosphatase 2A (FyFF) and the type 5 Ser/Thr protein phosphatase, PAPP5, which preferentially bind and dephosphorylate active phytochromes (Kim et al., 2002; Ryu et al., 2005). Dephosphorylation can amplify phytochrome signaling. The phosphorylation blocks the interaction with its signal transducers, whereas the dephosphorylation enforces the interaction. By mutating the Ser-598, a higher level of unphosphorylated, active phytochrome can be maintained, which leads to the hypersensitive phenotype observed.
The two other phosphorylation sites (Ser-7 and Ser-17) are in the N-terminal extension. Several studies have focused on this region. Deletions in the N-terminus did not affect dimer formation and chromophore attachment. An oat phyA 
7-69 in tobacco was no longer active (Cherry et al., 1992) and 1-52 of oat phyA and 1-80 of rice phyA cause a dominant-negative suppression phenotype in transgenic Arabidopsis or tobacco seedlings grown under FR (Boylan et al., 1994; Emmler et al., 1995). With respect to hypocotyl elongation, oat phyA 6-12 behaved normally under VLFR in Arabidopsis, but was hyperactive in tobacco. In contrast, oat phyA 6-12 showed a dominant-negative suppression of HIR in both species (Casal et al., 2002). Alanine-scanning and deletions in oat phyA defined the regions between residues 25 and 33 and between residues 50 and 62 as necessary for biological activity and for a correct Pfr apoprotein/chromophore interaction (Jordan et al., 1995, 1997). Substitutions of serine to alanine in the first 10 serine residues of rice phyA resulted in an increased biological activity of phyA in tobacco, suggesting that phytochrome responses might be desensitized (Stockhaus et al., 1992).
With a better understanding of both the phyA in Arabidopsis and rice, we revisited the function of rice phyA expressed in Arabidopsis. An ectopic expression of the full-length rice phyA or a rice phyA in which the first 10 serine residues have been substituted by alanines to prevent phosphorylation were utilized. In contrast to all previous studies, the lines expressing the rice phyA were produced in the phyAphyB or phyB mutant background, which allows a better assessment under R-light conditions. The lines were physiologically and biochemically analyzed respectively their VLFR, LFR and HIR responses.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMATERIAL AND METHODS |
|---|
Arabidopsis Lines Expressing Rice PHYA and Rice PHYA S/A
Rice and Arabidopsis phyA have similar functions during de-etiolation but also some functional differences have been documented. To investigate to what level the rice phyA can complement for Arabidopsis phyA or phyB, we examined VLFR, LFR, and HIR responses in phyB and phyAphyB mutant plants transformed with a rice PHYA cDNA under the control of the CaMV 35S promoter. The amino terminus of phyA from all flowering plant species analyzed to date is very rich in serine residues (Figure 1A). This is especially true for the first 20 amino acids, which contain from seven to 11 serine residues. In contrast, fewer serine residues can be found in other phytochromes, with the exception of phyC. To study to what level the N-terminal serine residues are required for phytochrome function in Arabidopsis, we transformed phyB and phyAphyB double mutants with a mutant rice PHYA cDNA (PHYA SA) in which the first 10 serine codons encoding amino acid residues 2–4, 10–14, 19, and 20 were changed to alanine codons (Figure 1B).
|
Immunoblot analysis was employed to evaluate phyA protein levels in light-grown plants (Figure 2). The protein levels were high under white-light conditions, reflecting the constitutive expression by the CaMV 35S promoter. Only for the PHYA S/A construct in the phyAphyB background could no high-level expression lines be identified. We selected two representative lines from each construct to analyze them for their very low fluence (VLFR), low fluence (LFR), and high irradiance (HIR) responses during the lifecycle of a plant.
|
Germination Assays under R, FR, and Blue Light
Germination in Arabidopsis can be triggered by a phyA-dependent VLFR and by an LFR mediated via phyB, C, and E. The VLFR and the phyB-dependent LFR can be differentiated by analyzing germination rates after different periods of imbibition. The phyB-dependent LFR response is induced after 3 h imbibition with a R-light pulse, whereas the VLFR requires 48 h imbibition and a light pulse of any light quality (Botto et al., 1996; Shinomura et al., 1996; Poppe and Schäfer, 1997). phyA, in contrast to phyB, is not present in dry seeds and requires de-novo synthesis to accumulate. Germination rates were analyzed to ascertain the response of WT, mutant, and transgenic plant lines to red (R) and far-red (FR) light pulses after imbibition. Sterilized seeds received a FR-light pulse to revert all phytochromes to the Pr form and then imbibed for 3 h, 24 h (1 d), 47 h (2 d) or 72 h (3 d) in the dark, after which they were treated with varying light intensities of R, B, or FR monochromatic light pulses. Germination was evaluated after an additional 7 d in darkness.
Figure 3 (B and D) shows that WT and phyAphyB seeds that received a FR-light pulse were not able to germinate after 3 h imbibition. An increase in germination efficiency can be observed in WT after 1 d imbibition, with a further increase after 3 d. phyAphyB double mutant seeds, however, did not germinate, even after 3 d imbibition, indicating that phyA is required for FR-induced VLFR-inducible germination. Seeds transgenic for the rice WT PHYA in the phyAphyB double mutant require 2 d imbibition for elevated germination rates, suggesting that phyA has first to been synthesized and can at least partially substitute for the missing Arabidopsis phyA. But only very few seeds (less than 15%) of the phyAphyB double mutant expressing the rice WT PHYA can germinate under these conditions (Figure 3F).
|
WT seeds that received LFR treatment germinated efficiently after 3 h imbibition and photon fluence above 100 µmol m–2 of R light (Figure 3A). The phyAphyB double mutant and lines expressing rice PHYA in phyAphyB did not exhibit any significant increase in the germination rate under these conditions (less than 10% germination rate; Figure 3C and 3E), as this response requires phyB. Twenty-four hours imbibition enhanced the germination rate of WT seeds that received between either 100 and 10 000 µmol m–2 (LFR) or 0.03–10 µmol m–2 (VLFR) R light. For the phyAphyB seeds, only higher fluences of the R pulses (100–10 000 µmol m–2) led to an improvement in germination. This response is likely due to the function of phyC and phyE (Figure 3C). A longer imbibition period (up to 3 d) did not significantly further increase the germination efficiency of the double mutant. In phyAphyB lines expressing rice PHYA, an enhanced germination rate is only observed for fluences higher than 0.1 µmol m–2. These results suggest that the rice phyA can detect R light above 0.1 µmol m–2 and mediate a R-light-mediated LFR and induce signaling, but only after imbibition for at least 24 h.
An improvement of germination efficiency of phyAphyB PHYA seeds for R-light treatment requires 1 d imbibition for the LFR and 2 d imbibition for the VLFR, whereas WT requires 3 h and 2 d, respectively. That active transgenic phyA is not observed after 3 h imbibition can be explained by the fact that the rice PHYA cDNA is under the transcriptional control of the CaMV 35S promoter, which is generally not active in immature cotyledons (Benfey et al., 1989). Therefore, a time lag of at least 24 h is needed before active rice phyA protein levels are high enough to activate germination.
For a fuller evaluation of the functionality of the rice phyA in Arabidopsis, germination efficiency was examined in WT and phyB and phyAphyB mutant seeds after 47 h imbibition and various intensities of monochromatic light treatments. Germination rates after a FR-light pulse are slightly lower than after R-light pulses for WT and phyB and no significant germination can be detected in phyAphyB seeds in FR, even at higher fluences (less than 10% germination) (Figure 4B). Expression of the rice PHYA in the phyAphyB double mutants increased the germination rate after 47 h of imbibition and exposure to not only R light, but also FR and B light pulses, albeit at a much lower percentage. The germination efficiency for phyB lines overexpressing PHYA does not improve compared with the parental line; actually, a reduction can be observed under most light intensities. Germination under B light is, in all cases, very similar to their germination under FR light (Figure 4B and 4C). These results indicate that FR and B can also be detected by the rice phyA in Arabidopsis and signaling initiated.
|
Interestingly, phyB seeds are able to germinate more efficiently than WT upon exposure to R, B, or FR light (Figure 4A, 4C and 4E). Because, in R light, a difference was only observed for fluences between 0.01 and 1 µmol m–2, this effect seems specific for the VLFR (Figure 4A). Reduced phyB levels could enhance phyA function for the VLFR. phyB seeds transgenic for the rice PHYA cDNA germinate less efficiently than both the WT and phyB under the low fluences of R light (Figure 4A). Increased phyA content affected by transgenic rice PHYA could suppress phyA-dependent VLFR-activated germination, suggesting that rice phyA is perceived like a phyB under these conditions.
Figure 4D compares the photon fluence of R, B, and FR light necessary to obtain 40% germination. WT and phyB exhibit very similar curves, although phyB requires higher fluences in R and B light to germinate, whereas phyAphyB can only germinate efficiently after R-light treatment. Germination of phyAphyB seeds can, however, be rescued by the expression of rice PHYA. Though this complementation was observed for R, B, and FR-light pulses, it was strongest for R light. These data suggest that rice phyA under these circumstances is more efficient in substituting for the lack of phyB than for phyA in these lines. In contrast, expression of PHYA in the phyB mutant background reduced germination efficiency for all three light qualities tested.
R (0.3 µmol m–2) or FR (75 µmol m–2) light pulses after 48 h of imbibition were used to determine the effectiveness of the mutated phyA SA for germination in comparison to the WT rice phyA (Figure 4E). Both in the phyB and the phyAphyB mutant background germination of the PHYA SA-expressing lines was more efficient than the WT PHYA-expressing lines and this effect was stronger under R light. This indicates that phyA SA does not interfere with the endogenous phyA under R light to reduce germination efficiency, as does the WT rice phyA, but it can sense the R and FR-light pulses. Nevertheless, we cannot rule out that phyA SA still interferes with the endogenous phyA under R light, but that this effect is masked by more efficient signaling via the phyB-signaling pathway. With the methods used, we are not able to discriminate whether phyA SA is sensing these light pulses as very low or low fluences.
Hypocotyl Elongation and Cotyledon Opening under R and FR Light
Because of its fluence dependency, hypocotyl elongation is considered the standard test for light-responsiveness. Cell expansion of the hypocotyl is maximal in dark-grown seedlings in an effort to reach favorable light conditions. To protect the apical meristem in darkness, cotyledons are folded and an apical hook is formed. Upon illumination, the hook opens and cotyledons unfold and expand. Cotyledon opening is considered a more light-sensitive process than hypocotyl elongation. The hypocotyl elongation of the phyAphyB mutant under FR-light conditions is comparable to dark-grown seedlings. Under R light, a slight reduction in hypocotyl elongation can be observed, as phyC, D, and E are still functional. Under R and FR-light, apical hook formation and inhibition of cotyledon expansion can be noted in the phyAphyB double mutant. This indicates that both responses are mainly dependent on phyA and phyB.
Under continuous FR light (HIR), the expression of WT rice PHYA in the phyAphyB mutant reduces the hypocotyl length slightly but not to WT levels, which suggests that rice phyA cannot completely substitute for phyA under FR light (Figure 5). Nevertheless, a complete opening of the cotyledons can be observed. In the phyB mutant, expression leads to no statistically significant hypersensitive phenotype (p > 0.05). On the other hand, overexpression in Ler causes a hypersensitive phenotype, indicating that phyB might be necessary for a hypersensitive phenotype (data not shown).
|
The rice phyA SA seems to be more active than the WT rice phyA under continuous FR, as the hypocotyl of the transgenic phyAphyB double mutant lines is strongly reduced and even shorter than WT. However, no hypersensitive phenotype can be observed in the phyB lines and WT (data not shown) expressing PHYA SA.
Under VLF-FR light, the expression of rice PHYA in the phyAphyB mutant inhibits hypocotyl elongation, which results in a hypocotyl shorter than WT. This indicates that rice phyA can rescue the loss of Arabidopsis phyA under very low fluence for hypocotyl elongation and cotyledon opening. This hypersensitive phenotype can also be observed in the phyB background. On the other hand, the PHYA SA expression in the phyB background has little effect and only a slight, but statistically significant (p < 0.05), reduction in hypocotyl elongation can be observed in the phyAphyB double mutant. These lines did not display completely unfolded cotyledons and retained a partial hook under these conditions.
Under R light, the rice PHYA is active in the phyB and the phyAphyB double mutant lines. Especially in the phyAphyB double-mutant background, a strong hypersensitive phenotype can be observed. These results are similar to the data by Halliday et al. (1999), which demonstrate that rice phyA in phyB can substitute for phyB in EOD-FR-mediated hypocotyl elongation. Therefore, under R light, the expression of PHYA in the phyB and phyAphyB mutants restores WT hypocotyl length and cotyledon opening, indicating that the rice phyA can substitute for phyB under R light. In contrast, under R light in phyB and in phyAphyB mutants, only a marginal reduction in hypocotyl elongation and only partially unfolded cotyledons can be observed when the PHYA SA construct is expressed. In conclusion, for hypocotyl elongation, WT phyA can substitute in the phyAphyB double mutant under VLFR and R light, whereas the rice phyA SA variant complements better under continuous FR light (HIR conditions) and not under R light.
Pigment Analysis: FR Light Killing Effect and Anthocyanin Accumulation
Synthesis of chlorophyll is not possible under FR light because the committing step in its biosynthesis, which is catalyzed by the POR enzyme, needs light of higher energy for induction (van Tuinen et al., 1995; Barnes et al., 1996; Sperling et al., 1997). Furthermore, the level of protochlorophyllide is also reduced under FR light and the expression of the PORA gene is down-regulated by a phyA-dependent HIR. This, and perhaps some not so well understood physiological processes, lead to the phenotype that WT seedlings, after incubation under FR light (continuous and pulses), are no longer able to green efficiently in the following white light. phyA mutants have been shown to be resistant to this so-called ‘far-red light killing effect’. We analyzed our lines for their efficiency in the greening response after 3 d of continuous FR light or after 3 d of pulsed FR light (0.5 µmol m–2 s–1; 5 min h–1). Under the HIR conditions, only the phyAphyB line can green efficiently, while none of the complemented lines is able to accumulate chlorophyll, suggesting complementation of the phenotype (data not shown). The picture is more differentiated under very low fluence conditions. Under these conditions, WT and the phyB mutant can green to some extent, but the phyAphyB mutant accumulates chlorophyll to a much higher level (Figure 6). Expression of rice PHYA in phyAphyB reduces the greening efficiency drastically, whereas the expression of PHYA SA does not inhibit chlorophyll accumulation. A similar trend can be observed in the expression lines in the phyB background, as the lines expressing PHYA SA are greener than phyB.
|
Interestingly, expression of PHYA in phyB and phyAphyB mutant lines leads to less accumulation of chlorophyll under R and white-light conditions and the cotyledons appear palish green (Figure 5A).
Anthocyanin accumulation under FR light is a phyA-dependent process; therefore, the phyAphyB double mutant is unable to accumulate anthocyanin under continuous FR light. Nevertheless, the rice PHYA expression in phyAphyB does not increase the accumulation significantly, whereas the expression of rice PHYA SA leads to a level similar to WT (data not shown).
Agravitopic Responses of the Hypocotyl
The growth pattern of the hypocotyl is usually not equal and seedlings constantly perform lateral movements during their growth. This can also be interpreted as a possibility to determine the optimal orientation towards light. In darkness, the gravitropic response is dominant, leading to an upright growth of the seedlings. Under R and FR-light, enhanced movements can be observed, resulting in a random orientation (Poppe et al., 1996). Under FR light, this effect is phyA-dependent; therefore, the phyA and the phyAphyB mutant are not able to perform lateral movements and grow with a low variation from the perpendicular orientation. Wild-type seedlings, on the other hand, when grown on vertical plates, which allow them movement in three directions, are evenly distributed in all directions after 3 d in continuous FR light (Figure 7A). As the agravitropic growth response under light can best be observed under conditions in which the hypocotyl is slightly elongated, we quantified the response under lower continuous FR light (0.5 µmol m–2 s–1). Expression of PHYA SA in the phyAphyB mutant did not reduce the agravitropic growth, but the expression of rice WT PHYA in Arabidopsis is able to complement the endogenous phyA (Figure 7B and 7C). Under hourly pulses of FR light, similar effects can be observed (data not shown).
|
Loss of the agravitropic response can also be observed under R light. In this case, it is phyA- and phyB-dependent. phyB mutants demonstrate a reduced random orientation and the phyAphyB mutant grows completely upright. The expression of WT PHYA in phyB and phyAphyB leads to a random growth similar to WT. PHYA SA expression either in phyB or phyAphyB mutants is not as efficient in compensating, as the randomization is comparable to phyB (data not shown).
Petiole Length and Leaf Size: Rosette Formation
The adult growth habit of WT Arabidopsis plants consists of leaves in a compact rosette. Upon change to reproductive growth, the stem develops nodes with elongated internodes, generating the inflorescence stem. The loss of the rosette habit has been described in phyAphyB double mutants following R/FR treatments (Devlin et al., 1996). We tested the growth of soil grown plants under white light (80 µmol m–2 s–1, 16 h light/8 h dark). Under these conditions, phyB and phyAphyB mutants develop longer petioles and shorter leaf blades compared with WT. Expression of PHYA in the phyB and the phyAphyB lines did not reduce the length of petiole elongation significantly (Figure 8A and 8B). The rosette of a phyAphyB double mutant contains only a low number of leaves. Most leaves are very small and roundish, but the oldest leaf can be enlarged, which leads to a high diversity in leaf form. To quantify these effects, the length of the leaf blade was measured (Figure 8C). The expression of WT PHYA leads to increased leaf size, whereas the expression of PHYA SA only led to a marginal effect. In the phyB background, the effect of phyA SA is stronger in line 411 compared with line 221, which reflects the higher protein level in line 411. No significant variation between the leaf width could be observed. In summary, WT phyA seems more efficient than phyA SA to complement some of the shade-avoidance responses inflicted by the loss of the phyB in the mutant lines.
|
Flowering Time
The transition of a vegetative apex to a reproductive apex is controlled by several signals, including light (Chory et al., 1996; Cerdán and Chory, 2003). Light can be perceived as day length, allowing the plants to distinguish between short and long day, a seasonal signal, or the R:FR ratio, which is dependent on canopy density. A low R:FR ratio tends to accelerate flowering as part of the shade-avoiding response. Arabidopsis (a long-day plant) flowers earlier under long-day conditions. Because phyA mutants flower later under these conditions, this suggests that phyA also plays a role in day-length perception in Arabidopsis. The flowering time for the phyAphyB mutant is even earlier than phyB, suggesting a cross-talk between both phytochromes. Flowering time was analyzed by counting days until bolting (data not shown) and the number of leaves upon bolting. Expression of PHYA WT and PHYA SA in phyB mutants led to a slightly delayed flowering time compared with phyB. In the phyAphyB double mutant, only the WT phyA was able to delay flowering similarly to WT, whereas PHYA SA-expressing lines still flowered earlier, albeit not as early as the phyAphyB mutant (Figure 8D). These results are very similar to those of the other adult phenotypes observed, with phyA WT being more active than phyA SA under these developmental stages.
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMATERIAL AND METHODS |
|---|
Phytochrome (phy) A mediates three distinct photobiological responses in plants: the very-low-fluence response (VLFR), which can be saturated by short pulses of very-low-fluence light; the high-irradiance response (HIR), which requires prolonged irradiation with higher fluences of far-red light (FR); and, to some extent, the low-fluence response (LFR). For a better understanding of the role of a monocot phyA in a dicot plant, we expressed rice phyA in the phyB and phyAphyB double mutants. The lines were analyzed under different light intensities and qualities and various well defined responses were quantified.
So far, monocot phyA from oat and rice have been expressed in tobacco and different Arabidopsis ecotypes (Stockhaus et al., 1992; Boylan et al., 1994; Emmler et al., 1995; Jordan et al., 1995, 1997; Casal et al., 2002). The response was mainly evaluated according to the inhibition of hypocotyl elongation under different light conditions. In most cases reported, expression of PHYA resulted in a hypersensitive phenotype under high irradiance of FR light. Indeed, we could show also that expression of rice phyA in WT (Ler) leads to a hypersensitive phenotype (data not shown, Halliday et al., 1999). Expression of rice PHYA in the phyAphyB double-mutant background rescued the FR HIR phenotype only partially, as the transgenic phyAphyB lines were still significantly longer than WT and phyB. Furthermore, no hypersensitivity to HIR conditions could be observed in the phyB background. This suggests that phyB is necessary for the hypersensitive phenotype observed in WT. On the other hand, rice PHYA expressed in tobacco shows no significant hypersensitive phenotype (Emmler et al., 1995). Therefore, besides the structure of the phyA, a plant-specific component might be important.
A hypersensitive phenotype was described for tobacco or Arabidopsis lines expressing oat phyA under hourly pulses of FR light (VLFR; Casal et al., 2002). In our experiments, expression of rice PHYA in both the phyAphyB double-mutant or phyB lines led likewise to a hypersensitive phenotype. This indicates that in contrast to high fluences of FR light, very low fluences are sufficient to activate rice phyA in Arabidopsis, thereby leading to a complementation of the phenotype of the phyAphyB double mutant. This implies that either the rice phyA is more active under VLFR conditions in Arabidopsis, that the response is saturated earlier, or that the synergistic effect of phyB is not needed for the VLFR, but only for the HIR.
Expression of monocot phyA was also described to lead to a hypersensitive phenotype in tobacco, and Arabidopsis Ler under R-light conditions (Halliday et al., 1999; Casal et al., 2002). This is confirmed by our results because expression of rice PHYA in phyB and phyAphyB mutants led to a shortened hypocotyl. Expression in plants without the endogenous phyB allows us to distinguish between R-light-mediated responses of the endogenous phyB and the rice phyA. Our results clearly suggest that rice phyA can act as a R-light sensor for hypocotyl elongation.
Besides the inhibition of hypocotyl elongation, we used several other physiological parameters to analyze the phenotype of rice PHYA-expressing lines under HIR, LFR, and VLFR conditions (Figure 9). Under most LFR conditions, PHYA expressed in phyAphyB could either complement the mutant phenotype (agravitropic growth and cotyledon opening under R light, leaf size of the adult plant) or is acting in a hypersensitive way (hypocotyl elongation under R light, flowering time). In contrast, germination after a R-light pulse is only partially complemented in the transgenic lines, as only R-light fluences higher than 1 µmol m–2 were able to induce germination in these transgenic lines.
|
Analyzing expression of rice PHYA in the phyB background under R-light conditions resulted in a different outcome from the expression in phyAphyB lines. Whereas agravitropic growth and cotyledon opening under R light and leaf size of adult plants are complemented in the phyB mutant, flowering time is only partially complemented. These findings suggest an effect of the endogenous phyA. Interestingly, germination after a R-light pulse is decreased compared with the phyB mutant.
HIR was tested by hypocotyl elongation, cotyledon opening and anthocyanin accumulation under continuous FR light and by analyzing the greening efficiency in W light after exposure to continuous FR light (far-red light-killing effect). As mentioned before, hypocotyl length was only partially complemented and anthocyanin accumulation was only slightly increased compared with the phyAphyB mutant. Both of the other physiological parameters, however, were similar to WT in the phyAphyB double mutant. This could suggest that both the cotyledon opening and the FR killing effect under continuous FR conditions are not exclusively HIR responses. Indeed, cotyledon opening has been characterized as being very sensitive to very low fluences (Yanovsky et al., 1997). Recently, it has been shown that lines expressing oat PHYA in the Arabidopsis ecotype RLD could accumulate less anthocyanin (Casal et al., 2002). Furthermore, only a partial complementation of the far-red light-killing effect could be observed in these lines, but they were also hypersensitive with respect to hypocotyl elongation. Our rice PHYA expression lines in Arabidopsis ecotype Ler showed a very similar phenotype to the lines described by Casal et al. (2002). The differences between the responses in WT and the phyB or phyAphyB double mutant can only be explained that phyB is needed for the HIR response of rice phyA, although effects due to differences in protein levels cannot be excluded.
Under VLFR conditions, a substantial difference between the rice phyA expressing lines in phyB or phyAphyB double mutant can be observed. In transgenic phyAphyB lines, the far-red light-killing effect after pulsed FR light and the hypocotyl length are hypersensitive, cotyledon opening under FR-light pulses is complemented and germination after a FR-light pulse is partially complemented. In the phyB background, hypocotyl elongation of the rice phyA-expressing lines is also hypersensitive. However, neither the agravitropic growth under FR-light pulses nor the far-red light-killing effect after FR-light pulses is hypersensitive compared to the phyB mutant, indicating that the residual phyA in the phyB mutant has already saturated the response.
Rice phyA in the Arabidopsis phyAphyB mutant seems to be not very efficient in substituting for the endogenous phyA and FR light, as several responses, especially under high irradiances, are only partially complemented or not at all. Part of the reason could be the ectopic expression. We favor the explanation that the differences between the responses are due to the monocot phyA in a dicot background. This is supported by the fact that some responses can be complemented by the rice phyA whereas others cannot, suggesting additional plant-specific signaling factors.
Our results clearly suggest that rice phyA can act as a R-light sensor, especially for hypocotyl elongation and cotyledon unfolding, agravitropic growth, flowering time and leaf size. For germination, R-light photon fluences above 0.3 µmol m–2 were sufficient to activate the response, albeit not as efficiently as the endogenous phyB. It is intriguing to reason that this R-light responsiveness might be mediated by the VLFR. This notion is supported by the fact that expressed rice PHYA in Arabidopsis shows a hypersensitive phenotype under FR-very-low fluences and R light, but not under HIR conditions.
Time course experiments of the germination efficiency under monochromatic R, FR and B light at the Okazaki Large Spectrograph could be used to define the photon fluences necessary for induction of VLFR (photon fluence under R and FR light: 0.01 µmol m–2) and LFR (photon fluence under R light: above 30 µmol m–2) (see also Shinomura et al., 1996). The phyAphyB double mutant was still able to germinate after higher fluences of R light (above 50 µmol m–2), but only after an imbibition for at least 24 h. This means that the activity of phyC, D, and E, similar to phyA, is dependent on de-novo synthesis of the photoreceptors (Shinomura et al., 1996). Under higher fluences of FR light, no increase in germination could be observed, confirming that phyA is the only phytochrome important for germination under FR light (VLFR). Expression lines were only responsive to light induction after 24–48 h of imbibition, which can be explained by the fact that the CaMV 35S promoter is not efficient in expressing genes in seeds (Benfey et al., 1989). Therefore, at least 2 d of imbibition are necessary to sensitize the seeds for a light pulse. Expression of rice WT PHYA led to an increase in germination under R light in Arabidopsis, whereas the effect under B and FR light was only marginal. Expression of PHYA in the phyB background led to the opposite phenotype, as it decreased germination efficiency under all light conditions. A phyB mutant on the other hand had an increased germination rate after a R-light pulse of lower photon fluences (below 1 µmol m–2, VLFR). These findings suggest that phyB negatively regulates the VLFR mediated by phyA. The expressed rice phyA could simulate the action of phyB, thereby reducing the germination efficiency even more than in WT. Also, under FR and B light conditions, typically attributed to the VLFR, phyB germinates more efficiently than WT, whereas the expression lines germinate less efficiently.
Adult phenotypes are usually attributed to the light-stable phytochromes (phyB–E) and, only recently, the importance of phyA in these processes has been established (Franklin et al., 2007). Nevertheless, the delay in flowering time of phyA mutants had already suggested a role for phyA beyond de-etiolation. Interestingly, the adult phenotype of the expression lines suggests that rice phyA can compensate for the loss of phyB in the phyB mutants and the response is even hypersensitive in the phyAphyB mutant. But this is only true for leaf length and flowering time. Petiole length seems to be regulated by a different pathway, as no reduction in length can be observed. These observations are congruent with the results from Halliday et al. (1999). It was shown that the expressed PHYA in phyB mutants was not effective in restoring the petiole elongation response to EOD FR treatments, whereas this was possible for hypocotyl elongation. This means that the rice phyA cannot supplement for phyB in the developmental process, leading to the inhibition of petiole elongation.
Monocot phyA contains three serine-rich blocks within the first 20 amino acids (Figure 1A). The second block contains four to six serines, including Ser-7, which is phosphorylated in vivo (Lapko et al., 1999). Ser-17, in the third block, is the second residue that has been shown to be phosphorylated in oat phyA, preferentially in the Pr form of the phytochrome (Lapko et al., 1999). In dicots, these blocks are maximal three serine residues long and are only partially in comparable positions to monocots. Nevertheless, this does not exclude that the serines in dicots might also be differentially phosphorylated. Indeed, the deletion of the amino acids 6–12 in the context of the Arabidopsis phyA suggests an important role for signaling of this domain (Trupkin et al., 2007). Changes or deletions within the first 20 AA of monocot phyA have been shown to be not affecting chromophore binding, Pfr to Pr dark reversion and localization to the nucleus (Stockhaus et al., 1992; Jordan et al., 1995, 1997). Only in the 6-12 oat phyA-GFP under pulsed FR light could more nuclei with an increased number of subnuclear foci be observed compared to WT phyA-GFP (Casal et al., 2002).
To investigate the role of the serine residues in the N-terminal part of the protein that have been suggested to be important for phosphorylation events, we expressed a construct in which the N-terminal serines were substituted to alanines, in order to prevent phosphorylation. In a previous study, it had been shown with assays measuring inhibition of hypocotyl elongation that transgenic tobacco lines expressing the rice PHYA SA construct were hypersensitive under continuous R and FR light, whereas the transgenic WT PHYA lines were less active, especially under FR-light conditions (Stockhaus et al., 1992; Emmler et al., 1995). It was demonstrated that the chromophore is attached to the mutant PhyA apoprotein, and the mutant photoreceptor is photoreversible, giving a difference spectrum indistinguishable from that of the rice phyA. Jordan et al. (1997) showed that oat phyA was hyperactive under R and FR light in tobacco, but not as hyperactive as the oat phyA SA (2-18) mutant. Expression of a deletion of oat phyA 6-12 was slightly more active than the oat phyA SA transgenic lines, but the oat phyA 2-22 had lost some of its activity.
The expression of the SA mutant in the phyAphyB background is more efficient compared to WT phyA in compensating for the loss of phyA and phyB in hypocotyl elongation and anthocyanin accumulation under continuous FR light, the agravitropic growth under pulses of FR light and germination under FR and R light. For other responses tested, the phyA SA mutant was less efficient than the WT phyA: inhibition of hypocotyl elongation and agravitropic responses under R light, inhibition of hypocotyl elongation and cotyledon opening in pulses of FR light, far-red killing effect after FR-light pulses, leaf expansion and induction of flowering. Greening after continuous FR light and cotyledon opening under FR and R light are similar in both lines. Like the WT phyA expression lines, the responses seem mixed, although the phyA SA lines are preferentially hyperactive under HIR and VLFR conditions. The most surprising fact for us was the reciprocal behavior of the inhibition of hypocotyl elongation, especially obvious in the phyAphyB double mutants (Figure 9). Expression of PHYA SA leads to a hypersensitive effect under continuous FR light and no strong inhibition of hypocotyl elongation under R light, whereas the WT PHYA-expressing lines are hypersensitive under R light, not under FR light. Inhibition of hypocotyl elongation under very low fluence conditions was stronger in transgenic lines carrying the WT PHYA than the PHYA SA. These effects were not ecotype-specific, as backcrossing of the expressing lines into the Col-0 phenotype did not change the responses (data not shown). Complementation of agravitropic growth under R light was more efficient in WT PHYA, whereas under continuous FR light, this was true for PHYA SA. These results reflect those from the hypocotyl elongation assays. For induction of germination, on the other hand, phyA SA was more effective after pulses of R and FR light. In summary, it seems that phyA SA is more effective in HIR than LFR and VLFR, whereas WT phyA is more efficient for VLFR and LFR than HIR, with the exception of the germination responses.
Arabidopsis plants expressing 6-12 oat phyA showed a hyperactive phenotype of hypocotyl growth and cotyledon unfolding under VLFR conditions. These results were similar with WT phyA. In tobacco, on the other hand, the 6-12 oat phyA showed an increased hyperactivity compared to the WT phyA. In contrast, both WT and mutated phyA showed a dominant-negative suppression of HIR in both species (Casal et al., 2002). When the same 6-12 deletion was introduced into Arabidopsis phyA and expressed in Arabidopsis under its own promoter, transgenic lines showed normal responses to pulses of FR light (VLFR) and impaired responses to continuous FR light (HIR). These data suggest that the deletion of the second serine block enhances responses to VLFR, but reduced HIR. This stands in contrast to the rice phyA SA lines in Arabidopsis, which are more active under HIR conditions. As in the 6-12 phyA construct, only the second block of serines is missing; other serines could still be phosphorylated, especially Ser-17. The phyA SA mutant abolishes phosphorylation in all of the first 20 AA from the N-terminus. These differences could explain the different phenotype and highlight the importance of the third block of serines. A deletion of the first serine-rich block in oat phyA (2-5) did not lead to any strong phenotype under HIR and VLFR conditions (Jordan et al., 1995, 1997; Casal et al., 2002).
Phytochromes can interact with the catalytic subunit of a Ser/Thr-specific protein phosphatase 2A-, FyPP (Kim et al., 2002). FyPP modulates mainly phytochrome-mediated light signals in the timing of flowering, as transgenic Arabidopsis plants with expressed FyPP levels exhibited delayed flowering. Therefore, the loss of phosphorylation leads to later flowering. As the PHYA SA lines are not able to complement for the phyB-dependent early flowering, in contrast to the full-length phyA, the reduced phosphorylation status of the mutant phyA is not delaying flowering, at least in a phyAphyB background, indicating that FyPP is probably not involved in dephosphorylation of the N-terminus of rice phyA.
The type 5 Ser/Thr protein phosphatase PAPP5 enhances phytochrome-mediated photoresponses (Ryu et al., 2005). Expression lines are characterized by a shorter hypocotyl under continuous R and FR light and a later flowering phenotype. The rice phyA SA expressing seedlings show a shorter hypocotyl under continuous FR light, fitting with the phenotype of the PAPP5 expression lines. This suggests that the phosphorylation state of the N-terminus of rice phyA might be regulated by PAPP5. On the other hand, under R-light conditions, no correlation can be observed. As the phyA SA mutant is only mimicking the dephosphorylated state, we cannot exclude that PAPP5 is involved in dephosphorylating Ser-7 or Ser-17. Its role might be more subtle and the influence of PAPP5 on the VLFR has not been shown in physiological experiments.
Rice phyA is active under VLFR, LFR, and HIR and, in all these conditions, it can substitute for several phyB and phyA responses. Interestingly, no clear clustering of the responses can be performed (Figure 9). This could be due to the ectopic expression via the CaMV 35S promoter. Recent results could show that the expression of Arabidopsis phyA under its own promoter shows significant differences from the expression of similar constructs of oat phyA under the control of the CaMV 35S promoter (Trupkin et al., 2007). Furthermore, an increased stability of the monocot phyA remains to be determined, especially as the rice phyA has been shown be important for R-light signaling. A rice mutant lacking phyB and phyC still displays a R/FR reversibility for induction of Lhcb gene expression, which indicates that rice phyA in rice can mediate the R/FR-reversible LFR (Takano et al., 2005). This is no direct evidence for an increased stability of rice phyA compared with Arabidopsis phyA, but a physiological indication.
For rice phyA, different absorption spectra have been shown between different developmental stages. As action spectra for Arabidopsis have only been performed for germination and hypocotyl elongation (Shinomura et al., 1996, 2000) but not for any other stages or responses, the differences observed here could be due to developmental or cell-specific modifications of the downstream signals or phosphorylation.
Genetic analyses have indicated that the VLFR and HIR are mediated by different transduction pathways. For example, in Arabidopsis, two loci have been identified which are affected only in the VLFR, but not HIR (Yanovsky et al., 1997). Other mutants have been identified that retain the VLFR, but are severely deficient in HIR, such as fhy3 or pat1-1 (Yanovsky et al., 2000; Bolle et al., 2000). The open question is whether the phyA molecule itself can switch signaling between the VLFR and the HIR. It has been proposed that the serine-rich domain is involved in a mechanism that down-regulates the VLFR, possibly through cellular context-dependent phosphorylation of the domain (Casal et al., 2002). Here, we can also demonstrate that subtle changes in the N-terminal region of a rice phyA can lead to changes between VLFR and HIR. The difference in how the different physiological responses respond to the expression of different variants of phyA clearly suggests that besides determinants on the phytochrome molecule itself, downstream factors that can vary, depending on the developmental status or cellular context, play an important role in light signal transduction.
Cross-talks between phyA and phyB can be either synergistic or antagonistic, depending on the action of phyA in the VLFR or HIR modes, respectively (Cerdán et al., 1999; Hennig et al., 1999, 2001; Casal et al. 2000; Cerdán and Chory, 2003; Torres-Galea et al., 2006). With the help of our transgenic lines, a larger role for interactions between phyA and phyB could be established. phyB seems to be important for the FR-HIR; on the other hand, it negatively affects VLFR during germination. A suppression of germination under very low fluence conditions by phyB, which is present in the dormant seed, seems a sensible regulatory mechanism to prevent seeds from germinating under adverse conditions. Studies on Arabidopsis mutants have shown that phyB can partially suppress phyA-dependent inhibition of hypocotyl elongation in FR light and induction of seed germination by FR light (Hennig et al., 2001). The data are in agreement for the VLFR response, but are divergent for the HIR response. As the rice phyA seems less capable of inducing HIR in Arabidopsis than the endogenous phyA, HIR-related distinctions could well be attributed to the structural differences between rice and Arabidopsis phyA.
|
MATERIAL AND METHODS |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMATERIAL AND METHODS |
|---|
Plant Material
The mutant alleles used for this study were phyA-201 (fre1-1) and phyB-1(hy3) in A. thaliana. The respective mutants (phyB = hy3-Bo64, phyAphyB = fre1-1xhy3) were transformed with the constructs described in Stockhaus et al. (1992). For the phenotypic characterization, T3 and T4 generations were used.
Plant Growth Conditions
For all phenotypical analysis except flowering, seeds were sterilized in 50% bleach (v/v), 0.1
Triton-X 100 for 2 min. After washing twice in sterile water, seeds were sown on half-strength MS-Medium (Duchefa), containing 0.8% agar (w/v).
Physiological Characterization
For all applications except germination, seeds were cold treated after sowing for 4 d at 4°C in the dark. After this stratification, seeds were exposed to R light (1 h, 20 µmol m–2s–1) to induce germination and incubated in the dark at room temperature over night. Thereafter, plates were transferred into appropriate light conditions.
Analysis of hypocotyl length was performed by irradiation for 3 d with hourly 5-min pulses of FR light (0.5 µmol m–2s–1) for VLFR conditions, continuous FR light (0.5 µmol m–2s–1) for HIR conditions and continuous R light (0.3 µmol m–2s–1) for LFR conditions.
For measurements of the gravitropic growth, seeds were grown on vertical plates for 3 d in continuous FR light (0.5 µmol m–2s–1).
For analysis of the FR-light-killing effect, seedlings were grown for 3 days in hourly 5-min pulses of FR light (0.5 µmol m–2s–1) for VLFR conditions and continuous FR light (0.5 µmol m–2s–1) for HIR conditions. Subsequently, seedlings were transferred for 3 d into white light before chlorophyll content was determined (see below).
To determine flowering time and the size of adult petioles, seeds were sown directly on soil and grown in a growth chamber under long daylight conditions (80 µmol m–2s–1 fluorescent white light, 16 h light, 8 h dark) at 22–24°C.
All experiments were repeated at least three times. Statistical significance was calculated by applying t-tests.
Germination Assays
All seeds were surface-sterilized and plated in lots of 50–100 individuals on plates containing aqueous agar medium and immediately exposed to FR light (3 µmol m–2) to inhibit phyB-dependent dark germination. Plates were kept in darkness for the indicated time (25°C) and exposed to monochromatic light generated by the Okazaki large spectrograph (Watanabe, 1982). Monochromatic light of 440, 726, or 660 nm was applied and total fluence was varied by changing the duration of the radiation and/or the fluence rate with the help of threshold boxes. This allowed performing several parallel assays. After the exposure to monochromatic light, seeds were kept in darkness for 7 d, and percent germination was measured. Germination was scored using a microscope to assess radicle emergence.
For germination experiments in the R or FR-light chambers, seeds were sterilized and plated on half-strength MS plates. Immediately after sowing, the seedlings were exposed to a 15-min FR-light pulse (2 µmol m–2s–1) and then transferred into complete darkness. After 48 h incubation, seedlings were exposed to a 30-s R-light pulse (0.3 µmol m–2) or a 3-min FR-light pulse (75 µmol m–2). Thereafter, seeds were incubated for 4 d in the dark before the germination rate was calculated.
As references, germination without the second light pulse (dark germination rate) and the germination under saturating pulses of R light (3 µmol m–2) were measured. To normalize the experiments percent, dark germination was subtracted from germination efficiencies and germination under saturating light conditions were taken as 100%.
Light Sources
Light intensities were determined with spectroradiometers (W, B and R light: model Li-1800, LiCor, Lincoln, NE; FR light: model SKP200 with a sensor for 730 nm, Skye Instruments, UK). The blue, R and FR-light sources were generated by LED using diodes with a maximum at 469, 660, or 740 nm, respectively (Quantum Devices, Barneveld and PVP GmbH).
In the growth chamber, fluorescent lamps (FH 21W/830, Osram) were used.
Measurements of Hypocotyl and Petiole Length and Gravitropic Growth
Hypocotyl and petiole lengths, cotyledon sizes and growth angles were documented using a digital camera (Coolpix 700, Nikon) and measured with the NIH Image software (ImageJ, National Institutes of Health). For the determination of the angle of growth, the deviation from the vertical was measured by plotting the vertical reference line and applying the angle tool of ImageJ.
Determination of Chlorophyll Content
Total chlorophyll was extracted by shaking ca. 50 seedlings in 80% acetone (v/v) over night in the dark. Chlorophyll absorption was determined photometrically by measuring the absorption at 660, 647, and 720 nm.
Chlorophyll content was calculated by the formula:
Chlorophyll a + b = 7,15 x (E660nm – E720nm) + 18,71 x (E647nm – E720nm) per mg fresh weight.
Protein Extraction and Immunoblotting
Three-week-old plants were extracted in ice-cold homogenization buffer (50 mM Tris/HCl, pH 8; 100 mM KAc; 1 mM EDTA; 1 mM DTT; 20% glycerol (v/v); 5 protease inhibitor; Sigma-Aldrich) using a micro pestle. The homogenized extract was clarified by centrifugation at 10.000 g for 5 min. Then, 2 x SDS loading buffer was applied and samples were incubated at 98°C for 5 min. SDS-PAGE was performed in Mini Protean 3 cells and immunoblotting in Mini Trans-Blot cells (Bio-Rad) according to manufacturer instructions. Immunoblot was blocked and incubated overnight with oat phyA-antibody (dilution 1:1000 in 1 x TBS) and washed twice for 20 min in 1 x TBS, 0.1% Tween 20. Thereafter, a 1-h incubation with horseradish-conjugated anti-rabbit antibody (dilution 1:10.000 in 1 x TBS; Sigma-Aldrich) was performed. After two additional washing steps, the secondary antibody was detected by chemiluminescence.
|
Acknowledgements |
|---|
We would like to thank Prof. Nam-Hai Chua for his strong support and his helpful discussions. Dr W.M. Leu was instrumental in generating the lines used in this work. We thank Prof. N. Murata for hosting us at the National Institute for Basic Biology, Dr R. Foster for critical reading of the manuscript and helpful discussions, and Prof. Dr D. Leister for his support.
Germination assays were carried out under the National Institute for Basic Biology Cooperative Research Programs for the Okazaki Large Spectrograph. CB was funded during the initial stages of the project by the 'Deutscher Akademischer Austauschdienst' (DAAD) and, later on, the work was supported by the 'Deutsche Forschungsgemeinschaft' (DFG) to CB (BO1146).
-
Aukerman MJ, Hirschfeld M, Wester L, Weaver M, Clack T, Amasino RM, Sharrock RA. A deletion in the PHYD gene of the Arabidopsis Wassilewskija ecotype defines a role for phytochrome D in red/far-red light sensing. Plant Cell (1997) 9:1317–1326.[Abstract]
Barnes SA, Nishizawa NK, Quaggio RB, Whitelam GC, Chua NH. Far-red light blocks greening of Arabidopsis seedlings via a phytochrome A-mediated change in plastid development. Plant Cell (1996) 8:601–615.[Abstract]
Benfey PN, Ren L, Chua NH. The CaMV 35S enhancer contains at least two domains which can confer different developmental and tissue-specific expression patterns. EMBO J (1989) 8:2195–2202.[ISI][Medline]
Blaauw OH, Blaauw-Jansen G, van Leeuwen WJ. An irreversible red-light-induced growth response. Avena. Planta (1968) 82:87–104.[CrossRef]
Bolle C, Koncz C, Chua NH. PAT1, a new member of the GRAS family, is involved in phytochrome A signal transduction. Genes Dev. (2000) 14:1269–1278.
Borthwick HA, Hendricks SB, Parker MW, Toole EH, Toole VK. A reversible photoreaction controlling seed germination. Proc. Natl Acad. Sci. U S A (1952) 38:662–666.
Botto JF, Sanchez RA, Whitelam GC, Casal JJ. Phytochrome A mediates the promotion of seed germination by very low fluences of light and canopy shade light in Arabidopsis. Plant Physiol (1996) 110:439–444.[Abstract]
Boylan M, Douglas N, Quail PH. Dominant negative suppression of arabidopsis photoresponses by mutant phytochrome A sequences identifies spatially discrete regulatory domains in the photoreceptor. Plant Cell (1994) 6:449–460.[Abstract]
Briggs WR, Mandoli DF, Shinkle JR, Kaufman LS, Watson JC, Thompson WF. Phytochrome regulation of plant development at the whole plant, physiological, and molecular levels. In: Sensory Perception and Transduction in Aneural Organisms—Colombetti G, Lenci F, Song P-S, eds. (1984) New York: Plenum Press. 265–280.
Bruce WB, Deng XW, Quail PH. A negatively acting DNA sequence element mediates phytochrome-directed repression of phyA gene transcription. EMBO J (1991) 10:3015–3024.[ISI][Medline]
Casal JJ, Cerdan PD, Staneloni RJ, Cattaneo L. Different phototransduction kinetics of phytochrome A and phytochrome B in Arabidopsis thaliana. Plant Physiol (1998) 116:1533–1538.
Casal JJ, Davis SJ, Kirchenbauer D, Viczian A, Yanovsky MJ, Clough RC, Kircher S, Jordan-Beebe ET, Schäfer E, Nagy F, Vierstra RD. The serine-rich N-terminal domain of oat phytochrome a helps regulate light responses and subnuclear localization of the photoreceptor. Plant Physiol (2002) 129:1127–1137.
Casal JJ, Yanovsky MJ, Luppi JP. Two photobiological pathways of phytochrome A activity, only one of which shows dominant negative suppression by phytochrome B. Photochem. Photobiol. (2000) 71:481–486.[CrossRef]
Cerdán PD, Chory J. Regulation of flowering time by light quality. Nature (2003) 423:881–885.[CrossRef][Medline]
Cerdán PD, Yanovsky MJ, Reymundo FC, Nagatani A, Staneloni RJ, Whitelam GC, Casal JJ. Regulation of phytochrome B signaling by phytochrome A and FHY1 in Arabidopsis thaliana. Plant J (1999) 18:499–507.[CrossRef][ISI][Medline]
Cherry JR, Hondred D, Walker JM, Vierstra RD. Phytochrome requires the 6-kDa N-terminal domain for full biological activity. Proc. Natl Acad. Sci. U S A (1992) 89:5039–5043.
Chory J, et al. From seed germination to flowering, light controls plant development via the pigment phytochrome. Proc. Natl Acad. Sci. U S A (1996) 93:12066–12071.
Clack T, Mathews S, Sharrock RA. The phytochrome apoprotein family in Arabidopsis is encoded by five genes: the sequences and expression of PHYD and PHYE. Plant Mol. Biol. (1994) 25:413–427.[CrossRef][ISI][Medline]
Clough RC, Jordan-Beebe ET, Lohman KN, Marita JM, Walker JM, Gatz C, Vierstra RD. Sequences within both the N- and C-terminal domains of phytochrome A are required for PFR ubiquitination and degradation. Plant J (1999) 17:155–167.[CrossRef][ISI][Medline]