Arabidopsis fhl/fhy1 double mutant reveals a distinct cytoplasmic action of phytochrome A

Abstract

Phytochrome A (phyA) plays an important role during germination and early seedling development. Because phyA is the primary photoreceptor for the high-irradiance response and the very-low-fluence response, it can trigger development not only in red and far-red (FR) light but also in a wider range of light qualities. Although phyA action is generally associated with translocation to the nucleus and regulation of transcription, there is evidence for additional cytoplasmic functions. Because nuclear accumulation of phyA has been shown to depend on far-red-elongated hypocotyl 1 (FHY1) and FHL (FHY1-like), investigation of phyA function in a double fhl/fhy1 mutant might be valuable in revealing the mechanism of phyA translocation and possible cytoplasmic functions. In fhl/fhy1, the FR-triggered nuclear translocation of phyA could no longer be detected but could be restored by transgenic expression of CFP:FHY1. Whereas the fhl/fhy1 mutant showed a phyA phenotype in respect to hypocotyl elongation and cotyledon opening under high-irradiance response conditions as well as a typical phyA germination phenotype under very-low-fluence response conditions, fhl/fhy1 showed no phenotype with respect to the phyA-dependent abrogation of negative gravitropism in blue light and in red-enhanced phototropism, demonstrating clear cytoplasmic functions of phyA. Disturbance of phyA nuclear import in fhl/fhy1 led to formation of FR-induced phyA:GFP cytoplasmic foci resembling the sequestered areas of phytochrome. FHY1 and FHL play crucial roles in phyA nuclear translocation and signaling. Thus the double-mutant fhl/fhy1 allows nuclear and cytoplasmic phyA functions to be separated, leading to the novel identification of cytoplasmic phyA responses.

Because plants use light as a source of energy, their development strongly depends on the spectral quality, quantity, direction, and periodicity of light. To perceive these important environmental stimuli, plants possess several classes of photoreceptors: phototropins (PHOTs), cryptochromes (CRYs), the ZTL/FKF/LPK2 receptors, phytochromes, and yet unidentified UV-B photoreceptor(s). The PHOTs, CRYs, and ZTL/FKF/LPK2 receptors sense the blue region of the spectrum, mediating different responses, most of which are intertwined with those mediated by the red-light-sensing phytochromes. The phytochrome family in Arabidopsis thaliana comprises five members (phyA to phyE) (1). By definition, phytochromes are red light (R)/far-red light (FR) photochromic photoreceptors, but they also absorb blue light (B) effectively. Of all known plant photoreceptors, phyA is the only one activated by high fluences of FR. It is the predominant phytochrome in dark-adapted tissue but becomes rapidly degraded upon illumination with R (2). The physiology can be categorized as very-low-fluence responses (VLFRs) (3) to various wavelengths or high-irradiance responses (HIRs) (4) to FR, both of which are known to be mediated by phyA (5, 6). Because the FR–HIR can be used to address phyA responses specifically, it is widely used to study phyA signaling, although such conditions are nevertheless not encountered by plants in natural habitats. At least three general mechanisms seem to be involved in phyA signaling: compartmentation, differential degradation, and transcriptional regulation. Many phyA signal transduction components are transcription factors [HY5, LAF1, HFR1, PIF3, FHY3, FAR1 (7)], whereas many others are early targets of phyA responses (8). Several of the signal transduction components (HY5 and HFR1) and phyA itself are further regulated via ubiquitination and differential degradation. In this process, the cop/det/fus mutants (9), as well as other components of the protein degradation machinery, play crucial roles. Another central mechanism seems to be the compartmentation of phyA between cytoplasm and nucleus. It was observed that phytochrome becomes sequestered to small cytoplasmic foci soon after FR-absorbing phytochrome (Pfr) formation (10), although the physiological significance of this is not understood. Lately, research has focused more on the import of phyA Pfr into the nucleus, where it has generally been observed to accumulate and form nuclear foci in a light-quality- and quantity-dependent manner (11, 12).

Because phyA does not contain any known nuclear localization sequence (NLS), its translocation appears to depend on other components. FR-elongated hypocotyl 1 (FHY1) and FHL (FHY1-like) have been shown to be specifically impaired in phyA signaling (13–16). Because both components have been demonstrated to contain functional nuclear export sequence and NLS motifs from which at least the NLS is essential for phyA responses (17), it has been suggested that FHY1 and FHL mediate phyA nuclear translocation. Indeed both components must be present for normal nuclear accumulation of phyA (18, 19). According to the strong phenotype of fhy1, this component is thought to act early in signaling by a direct interaction with phyA. Under R and FR, an interaction between FHY1 and holo-phyA was detected in yeast two-hybrid screens (18). This interaction could also be demonstrated for FHL (19), although the fhl-null phenotype in Arabidopsis is rather subtle (16).

In several plant species, including Arabidopsis, phytochromes switch off negative gravitropism in R and FR (20, 21). In Arabidopsis, exclusively phyA and phyB contribute to this response (21) but act in different modes. Because the inhibition of gravitropism through phyA is a VLFR (20, 21) it can be stimulated by either FR or B. In FR, the inhibition of gravitropism leads to a randomization of hypocotyl orientation, whereas the situation in B is more complex, because hypocotyl orientation is not only determined through inhibition of gravitropism but also through B-induced positive phototropism (22). Furthermore, phyA has been suggested to enhance positive phototropism (23). To date, little is known about the signal transduction pathway of PHOT-mediated phototropism. Recently, PKS1 was found to interact not only with phyA but also with PHOT1 and to modulate both phyA VLFR- and PHOT1-mediated phototropism (24). The molecular mechanisms underlying the gravitropic response are poorly understood, although several loci have been identified with specific deficiency in phyA-mediated gravitropism, among them HFR1 (25), which is involved in CRY-mediated deetiolation responses under B conditions (26). Furthermore, PKS1 and NDPK2, which were shown to interact with phyA, are located in the cytoplasm (27) or at the plasma membrane (24). Because residual phyA Pfr remains in the cytoplasm even under the most favorable light conditions (12), it was impossible to discriminate between the phyA pools in the cytoplasm and in the nucleus; thus, it has been difficult to disentangle their individual actions. A double-knockout mutant for both FHY1 and FHL loci can simulate a condition with only cytoplasmic detectable phyA and therefore is suitable for investigating early, cytoplasmic phyA responses. By investigating the phyA:GFP migration behavior and the phenotype in FR–HIR and VLFR conditions, we were able to identify phyA-mediated inhibition of gravitropism, inhibition of hypocotyl elongation in B and R-enhanced phototropism as cytoplasmic phyA-mediated events.

Results

fhl/fhy1 Homozygous Mutants: Disruption of both FHY1 and FHL Genomic Loci.

FHY1 is a central player in phyA signaling, probably acting close to the photoreceptor itself. FHL, the only FHY1 homolog in Arabidopsis, also seems to be involved in signaling (16). FHL RNAi expression in fhy1–3 leads to a phyA-like phenotype. To analyze the function of FHY1 and FHL in a clean fhy1- and fhl-null background, a double mutant was obtained by crossing fhl1 with fhy1–3. Likely homozygous lines were confirmed by genomic PCR [supporting information (SI) Fig. 7B]. The physiology of the double mutant was then studied in detail.

fhl/fhy1 Is Insensitive to FR but Not to B in Its Inhibition of Hypocotyl Elongation.

The phenotype of the double mutant resembled that of FHL RNAi in fhy1–3 lines (16). In FR, fhl/fhy1 fails to inhibit hypocotyl elongation and exhibits a hypocotyl as long as phyA with closed cotyledons, whereas WT and fhl show a short hypocotyl and open cotyledons, undergoing a partial deetiolation (Fig. 1A and SI Fig. 7C). The observed phenotype of fhl/fhy1 is even stronger than the already severe phenotype of fhy1 in FR (Fig. 1A), and moreover, both fhl/fhy1 and fhy1 are resistant to FR-induced block of greening, which is typical of phyA. The hypocotyl of phyB is as short as that of WT in FR but is not inhibited in R, whereas hypocotyl length of phyA, fhy1, fhl, and fhl/fhy1 in R did not significantly differ from that of the WT (SI Fig. 7D). WT, fhy1, fhl, and fhl/fhy1 all showed a short, deetiolated phenotype in B, whereas phyA and phyB both had a longer hypocotyl with open cotyledons over a range of fluence rates (Fig. 1B). The difference of hypocotyl length between phyA and fhl/fhy1 was significant (Student's t test p < 0.01). To investigate this apparent FHY1/FHL-independent phyA signaling pathway in B, we analyzed other B-mediated phyA responses.

Fig. 1.

Fluence-rate–response curves for the inhibition of hypocotyl elongation in FR (A) and B (B). Data points represent mean hypocotyl lengths from three independent experiments (20–30 seedlings each experiment). SEs are indicated.

Germination of fhl/fhy1 Is Impaired in R and FR Light.

Germination can be controlled either R/FR photoreversibly or induced under VLFR conditions in which both R and FR stimulate. Seeds were irradiated with a monochromatic R or FR pulse either 3 or 48 h after imbibition started (Fig. 2A). In WT seeds, germination could be promoted by a single R or FR pulse given after 48 h but not after 3 h, matching the time of highest phyA abundance (6), whereas germination was no longer promoted in phyA in either treatment (Fig. 2B). fhl showed only a slight reduction of germination compared with the WT in each treatment, whereas fhy1 showed a strong phenotype under FR conditions. The fhl/fhy1 double mutant exhibited almost no promotion of germination, following the phyA phenotype (Fig. 2B).

Fig. 2.

Induction of germination. (A) Scheme of light treatments. Dots indicate an FR pulse of 3 mmol·m⁻² after 1 h of imbibition and 5 μmol·m⁻² after 3 or 48 h; bars indicate an R pulse of 0.05 μmol·m⁻². Germination was scored after 5 d. (B) Average germination (three experiments with at least 100 seeds each). SEs are indicated.

Abrogation of Gravitropism in fhl/fhy1 Differs from That of phyA in B but Not in FR.

A randomization of hypocotyl growth orientation is induced through FR as a result of repression of gravitropism through phyA (20, 21): this is observed for WT seedlings (Fig. 3). This randomization was even enhanced in the phyB mutant (Fig. 3B and SI Fig. 8), suggesting a negative influence of phyB on phyA and/or its signal transduction. Indeed, a negative interference of overexpressed phyB on phyA responses has been implicated in hypocotyl elongation (28). fhy1, fhl, and fhl/fhy1 all exhibited an insensitivity toward FR abrogation of gravitropism, showing a phenotype intermediate between phyA and phyB (Fig. 3). Eighty-eight percent of fhl/fhy1 hypocotyls grew within ±25° of the vertical, whereas 98% of phyA hypocotyls were found in that angular range. This suggests that a FHY1/FHL-independent route exists for complete inhibition of gravitropism. Thus FHY1 and FHL can be added to the short list of components known to mediate phyA-induced agravitropism (25, 29).

Fig. 3.

Hypocotyl growth direction in FR. (A) Mean hypocotyl angle from vertical. Perfect negative gravitropism is represented by 0°. (B) Bar charts showing the relative distribution of hypocotyl growth direction in 10° classes of 4-d-old seedlings grown in 5 μmol·m⁻²·s⁻¹ FR from the front. Angles from three independent experiments were grouped in 10° classes and plotted as the percentage of the total seedlings (30–40 in each experiment).

Hypocotyl orientation in B is determined through PHOT-mediated phototropism but also through inhibition of gravitropism via phyA (22). WT seedlings irradiated laterally with low fluence B grew straight toward the light source irrespective of gravity, whereas phyA grew at an intermediate angle, responding to both stimuli (Fig. 4 A and B). This phenotype is apparently not caused by a disturbance in sensing phototropism (30) but rather results from an incomplete inhibition of negative gravitropism. The phenotype is seen in neither the fhy1 nor the fhl single mutant nor the fhl/fhy1 double mutant (Fig. 4 A and B). Comparing the means of WT with all mutants by single-factor ANOVA, highly significant differences were revealed (P₀ = 10⁻⁸) below 1.8 μmol·m⁻²·s⁻¹, whereas if phyA was excluded, no significant difference was seen (P₀ = 0.09), suggesting that neither FHY1 nor FHL plays a role in mediating this response. CRYs also contribute partially to the repression of gravitropism (22). To exclude the influence of CRY on our observations, we repeated the analysis at 500 nm and 520 nm, thereby lowering absorption by CRY but still providing sufficient Pfr for that response (20, 21). WT and all of the tested mutants grew directly toward the light (SI Fig. 9), but phyA, unable to switch off gravitropism, did not grow more than 20° from the vertical at the fluence rates tested. Therefore, suppression of gravitropism via phyA is differentially transduced in FR and B.

Fig. 4.

Hypocotyl growth orientation of 4-d-old seedlings in unilateral B. (A) Hypocotyl growth angle of 3-d-old seedlings grown in different fluence rates of 470 nm B. (B) Hypocotyl orientations of representative seedlings grown for 3 d in unilateral B (0.6 μmol·m⁻²·s⁻¹). All pictures were scaled to show differences in hypocotyl length. (C) R enhancement of phototropism. Seedlings were grown for 4 h in unilateral B (470 nm; 0.6 μmol·m⁻²·s⁻¹) and simultaneously irradiated with R (2 μmol·m⁻²·s⁻¹) from above. R exposure time varied depending on the indicated fluences. Data points represent mean angles from three independent experiments (20–30 seedlings in each experiment). Error bars indicate the SEs.

R-Enhanced Phototropism Is Normal in fhl/fhy1 but Not in phyA.

Phototropism is promoted by R preirradiation, an effect mediated by phyA (23). Gravitropism and phototropism have been shown to be genetically separated. Nevertheless, it is not known whether both responses, the promotion of phototropism and the abrogation of gravitropism, use the same mechanism. The repression of gravitropism itself might enhance phototropism and both responses are seen at very low fluences. Therefore we tested R-enhanced phototropism by illuminating 2-d-old, dark-grown, negatively gravitropic seedlings with different fluences of R and simultaneously with unilateral B. In accordance with the previous findings, phyA was not able to respond with enhanced phototropism, whereas fhl/fhy1 reacted similarly to the WT and the other mutants tested (Fig. 4C).

Nuclear Import of phyA Is Impaired in fhl/fhy1.

The intracellular distribution of phyA changes in a light-quality-dependent manner. WT protoplasts transformed with 35S::phyA:GFP showed an exclusively cytoplasmic fluorescence if incubated in darkness (D) immediately after transformation. A nuclear translocation of phyA:GFP was clearly detectable after irradiation with white light (W) or FR. To investigate the influence of light conditions during cultivation and transformation procedure, the experiment was repeated with protoplasts from etiolated seedlings under a green safelight. Even under these conditions, phyA:GFP is distributed throughout the cytoplasm in D and accumulates in the nucleus upon irradiation with W or FR (Fig. 5B). The observed change in localization is not caused by selective degradation (see below). In contrast, cyan fluorescent protein (CFP) fused to FHY1 was always seen in the nucleus, regardless of whether protoplasts were cultivated and transformed in light or D (Fig. 5 A and B), indicating that these conditions have no influence on phyA:GFP and CFP:FHY1 localization. However, whereas residual cytoplasmic fluorescence remained after incubation in D or in W, nuclear accumulation was enhanced by FR irradiation. Thus, FHY1 undergoes only a partial redistribution in FR (Fig. 5 A and B). Both phyA:GFP and CFP:FHY1 colocalize in distinct nuclear subcompartments after 6 h of FR-illumination, leading to areas of strong fluorescence (Fig. 5C). To investigate further whether a functional relationship between phyA and FHY1 translocation exists, phyA localization was analyzed in fhl/fhy1 protoplasts. phyA:GFP nuclear import was not detectable in protoplasts incubated in W and FR but could be restored completely if protoplasts were cotransformed with CFP:FHY1. fhl/fhy1 protoplasts cotransformed with phyA:GFP and CFP:FHY1 also showed accumulation of both proteins in distinct nuclear foci after irradiation with either W or FR (Fig. 6B). Furthermore, in the absence of both FHY1 and FHL, phyA:GFP formed numerous cytoplasmic foci in an FR-induced manner (Fig. 6A).

Fig. 5.

Localization of GFP, phyA:GFP, and CFP:FHY1 in Arabidopsis WT protoplasts after 16 h of incubation in D or irradiation with 30 μmol·m⁻²·s⁻¹ W or 5 μmol·m⁻²·s⁻¹ FR. Arrows point to the nuclei as apparent from the differential interference contrast (DIC) image. (A) Localization in protoplasts from light grown plants. (B) Localization in protoplasts from etiolated seedlings. The upper row shows fluorescent images, and the lower row shows DIC images. (C) Colocalization of phyA:GFP and CFP:FHY1 after 14 h of incubation in D and 6 h of irradiation with 5 μmol·m⁻²·s⁻¹ FR. (Scale bars, 25 μm.)

Fig. 6.

Localization of phyA:GFP in Arabidopsis fhl/fhy1 protoplasts. (A) Localization after incubation in D or irradiation with 30 μmol m⁻² s⁻¹ W or 5 μmol·m⁻²·s⁻¹ FR. (B) Localization of phyA:GFP and CFP:FHY1 in cotransformed Arabidopsis fhl/fhy1 protoplasts after 16 h of incubation in D or irradiation with 30 μmol·m⁻²·s⁻¹ W or 5 μmol·m⁻²·s⁻¹ FR. (Scale bar, 25 μm.)

phyA Degradation Is Not Altered in fhl/fhy1.

A dramatically reduced level of phyA might result from the fhl/fhy1 mutation, providing a trivial explanation of the data. Thus, the phyA abundance in R- and FR-irradiated seedlings was analyzed by Western blotting with an anti-phyA antibody (SI Fig. 10). A decrease in phyA levels in WT was detectable after 3 h of continuous irradiation with R. The same decrease was noticeable in fhl/fhy1 seedlings under the same conditions. Six hours of irradiation with R were sufficient to deplete phyA to an undetectable level in both WT and fhl/fhy1 (SI Fig. 10). As expected, depletion of phyA in FR proceeded much more slowly than in R. A significant reduction of phyA levels did not occur within 48 h of continuous irradiation with FR in both WT and fhl/fhy1, showing no differences between their phyA degradation kinetics (SI Fig. 10). We could thereby exclude any significant effect of fhl/fhy1 on phyA levels.

Discussion

The translocation of phyA into the nucleus upon irradiation seems to be a central process in transduction of light signals into physiological responses. In this process, the components FHY1 and FHL are crucial (18, 19). A functional dependence of phyA on FHY1 in relation to intracellular trafficking is further supported by the observations that nuclear accumulation of phyA:GFP could be restored in fhl/fhy1 if CFP:FHY1 were supplied transgenically (Fig. 6B) and that phyA:GFP and CFP:FHY1 colocalize in nuclear subcompartments (Fig. 5C). Furthermore, enhanced formation of cytoplasmic foci was induced through FR in the absence of FHY1 and FHL (Fig. 6A). Formation of cytoplasmic spots of higher phyA abundance could also be observed in WT plants (SI Fig. 11) (12). Cytoplasmic foci detected in WT and fhl/fhy1 protoplasts may be the same, and they both possibly connect to the phenomenon described earlier as sequestered areas of phytochrome (SAP) (10), although different methods lead to their observation. Neither composition nor function of these phyA aggregates is known, but because they occurred exaggeratingly in the absence of FHY1 and FHL, they might reflect a disturbance in the nuclear import mechanism of phyA. Without FHY1/FHL, phyA cannot be translocated and consequently aggregates in the cytoplasm in a pathological manner. Another possible explanation is the necessity of FHY1/FHL as a scaffold for correct phyA signalosome complex formation, missing partners would thus lead to cytoplasmic aggregation of phyA and other components. In either case, the cytoplasmic foci observed in fhl/fhy1 would be directly related to the molecular function of FHY1 and FHL with respect to phyA transport and signaling. In the fhl/fhy1 double mutant, nuclear accumulation of phyA:GFP is undetectable (Fig. 6A), possibly deleting nuclear signaling while leaving the cytoplasmic pathway intact.

Indeed, the phenotype of the fhl/fhy1 mutant strongly resembles phyA, apparently as a result of disturbed phyA nuclear import. However, because R-enhanced phototropism, abrogation of gravitropism, and inhibition of hypocotyl elongation in B seem to be independent of FHY1 and FHL (Figs. 1B and 4), translocation of the photoreceptor might not be involved in this signaling pathway, although we cannot exclude the possibility that a very small number of phyA molecules could nevertheless be present in the nucleus. Because abrogation of gravitropism is a VLFR requiring <3% of maximal phyA Pfr (20, 21), it is possible that a few nuclear-localized phyA Pfr molecules might be sufficient to induce this response. There is, on the other hand, little doubt that a cytoplasmic phytochrome signaling pathway exists (31). Sensing of light direction and polarization in mosses (32) is incompatible with the concept of freely migrating phytochrome regulating transcription in the nucleus, implying a separate cytoplasmic phytochrome action mechanism in lower plants. Cytoplasmic actions of phytochrome have long been suggested in higher plants, but lately, research has focused more on nuclear-localized phytochrome (12) and its influence on transcriptional regulation (8). Early signaling events, such as ion influx and calcium level changes, however, occur too fast to be activated by translocated phytochrome (33). The most rapid effects, the pelletability and sequestering of phytochrome in the cytoplasm, possibly enhanced in the fhl/fhy1 mutant as discussed above, have a half-time of 2 s after activation (34, 35). Within 2.5 s of irradiation, cytoplasmic streaming in the higher plant Vallisneria is stimulated (36). The photoreversible attachment and detachment of barley root tips to a glass slide, also known as the Tanada effect, occurs within 15 s, extending the list of cytoplasmic phytochrome functions in higher plants (37). Furthermore, it is well established that Arabidopsis phyA does interact directly with both PKS1 and NDPK2, which are cytoplasmic or plasma membrane-associated (24, 27), also implying cytoplasmic phyA functions.

Although most specific phyA responses, such as inhibition of hypocotyl elongation in FR, can be attributed to nuclear phyA, the fhl/fhy1 double mutant allowed us to identify responses in the cytoplasm. Because these responses appear in B, the situation becomes more complex. The two CRYs, CRY1 and CRY2, are crucial for deetiolation in B (38, 39), whereas phyA seems to be involved only partially in B detection (6). phyA mutants show a reduced inhibition of hypocotyl elongation in B (26). Furthermore phyA has been shown to interact with CRY1, the soluble photoreceptor of high-fluence B (40). Signal transduction of CRY1 involves components distributed within the nucleus and/or the cytoplasm, allowing for some components to act in both phyA and CRY signaling (26) in both compartments. CRYs are also involved in suppression of gravitropism but play a minor role in low-irradiance B in this respect (22). Therefore the abrogation of gravitropism might be mediated through integration of both signal transduction chains in the cytoplasm. This might occur through direct interaction of phyA and CRY1 or a cytoplasmic intermediate involved in both CRY and phyA signal transduction, such as SUB1 (41).

Another aspect supporting a cytoplasmic role of phyA is the enhancement of phototropism by R. Several mutually compatible theories about possible mechanisms were proposed earlier, one of them connecting attenuation of gravitropism with promotion of phototropism and/or the requirement of phyA to modulate the activity of phototropism signaling components (24). PKS1, a modulator of phyA VLFR, recently has been found to be necessary for phototropism. It is known to bind to PHOT1, NPH3, and phyA directly and might therefore represent the link between PHOT1 and phyA signaling (24). Both PHOT1 and PKS1 are located cytoplasmically within the plasma membrane (24), supporting our findings that this response is also independent of fhl/fhy1 and therefore independent of nuclear phyA.

With the help of fhl/fhy1, we identified phyA-specific cytoplasmic responses, namely the R-enhanced phototropism, abrogation of gravitropism, and the inhibition of hypocotyl elongation in B. Simultaneously, inhibition of hypocotyl elongation and abrogation of gravitropism in FR (Figs. 1A and 3) now can be considered nuclear-phyA-dependent responses. Similar phyB responses in R were also shown to be mediated exclusively by nuclear phyB (42). An autonomous cytoplasmic signaling system would provide an explanation for why residual phyA Pfr remains in the cytoplasm. It is still unknown, however, what restricts a specific part of it from migrating to the nucleus. Low-temperature-fluorescence measurements hint at two different isoforms of phyA, possibly related to specific phosphorylation (43). This might itself regulate localization or, alternatively, the composition and occurrence of multiprotein complexes might regulate nucleocytoplasmic phyA ratio. The existence of an FHY1-independent signaling route implies other molecules to be part of a putative cytoplasmic phyA signalosome. Indeed, as phyA is cytoplasmic in D, the first events in both cytoplasmic and nuclear signaling routes occur in the cytoplasm. Possibly, some of these events and components are shared by both routes and perhaps with CRY and PHOT signaling, too. The central question of whether or not all photoresponses mediated by phyA derive from one and the same primary reaction partner still persists. An adequate answer to this question will probably first appear when the still unknown mechanism of phytochrome nuclear transport has been described. fhl/fhy1, a mutant in which normal phyA levels are present but nuclear import is blocked, might prove an important tool in revealing this import mechanism. With this tool, we hope to open the door to a thoroughgoing analysis of phytochrome cytoplasmic signaling, migration to the nucleus, transcriptional regulation, and their integration in the living cell.

Materials and Methods

Light Sources.

W sources were L58 fluorescent tubes (Osram, Berlin, Germany) at 100 μmol·m⁻²·s⁻¹ PAR in 20°C growth chambers. The same light source was used for induction of germination and irradiation of protoplasts. The R source was an light-emitting diode (LED) array (LED660) at 660 nm [20 nm full width at half-heights (FWHH)]. FR was provided by LED arrays (LED740, 30 nm FWHH) with a 2-mm Plexiglas filter (90053; Röhm, Darmstadt, Germany) producing an FR field with neglectable R (>700 nm) contamination. The B source was an LED array at 470 nm (25 nm FWHH, Luxeon Blue LXHL; Philips Lumileds Lighting, San Jose, CA). All other LEDs were supplied by Roithner Laser Technik (Vienna, Austria). A projector (150 W) fitted with 500 nm and 520 nm interference filters (15 nm FWHH; Schott, Mainz, Germany) was used as cyan light source.

Plant Material and Growth Conditions.

The phyA-211 (44), phyB-9 (45), fhy1–3 (15), and fhl1 (16) mutants are in the Columbia ecotype background and therefore compared with the Columbia WT in all analyses presented. The fhl1/fhy1–3 double mutant was obtained by crossing fhl1 with fhy1–3. Homozygous lines were selected for long hypocotyls and resistance to FR-induced killing after 4 d of irradiation with FR (5 μmol·m⁻²·s⁻¹), followed by 3 d of W (30 μmol·m⁻²·s⁻¹). Homozygous lines were confirmed via PCR. Surface sterilized seeds were germinated on Murashige and Skoog agar without additional carbohydrates. After 4 d at 4°C in D, the seeds were exposed for 1 h to W (30 μmol·m⁻²·s⁻¹) to stimulate germination, and they were transferred immediately to appropriate light conditions. Hypocotyl lengths of seedlings were measured after 4 d. Fluence rates used to irradiate seedlings for protein extraction and for protoplast light treatments were 5 μmol·m⁻²·s⁻¹ FR, 5 μmol·m⁻²·s⁻¹ R, and 20 μmol·m⁻²·s⁻¹ W. To determine gravitropism in FR, seedlings were sown in horizontal rows on upright, square Petri dishes while illuminated perpendicularly. Gravitropism in B was analyzed as described (22). Seeds for germination assays were simultaneously harvested from WT and mutants grown under long-day conditions and stored in 4°C for 3 months and incubated 16 h at 35°C before experiments and treated as described (6). VLFR pulses had a total fluence of 0.05 μmol·m⁻² R, and FR pulses had a total fluence of 5 μmol·m⁻² FR. For R-enhanced phototropism assays, seedlings were treated as described (46).

Protoplast Transient Transformation.

The protoplast transient transformation procedure used was based on Sheen (47). In the case of protoplasts from etiolated tissue, seedlings were harvested and incubated intact in protoplast solution (47). A total of 5 μg of DNA of each plasmid was used for transfection. For construction of 35S::CFP:FHY1, the GFP coding region of plasmid 35S::GFP:FHY1 (15) was exchanged with CFP. The plasmid 35S::phyA:GFP (12) was a gift from F. Nagy (Biological Research Center, Szeged, Hungary). Immediately after PEG transformation, light-grown protoplasts were transferred for 16 h to the appropriate light conditions. For protoplast harvest and transfection from etiolated seedlings, the entire procedure was carried out in dim green light from 527-nm LEDs.

Microscopy.

For epifluorescence and light microscopy, protoplasts were analyzed in a fluorescence microscope (DM 6000; Leica, Deerfield, IL) with standard bandpass filter cubes for yellow fluorescent protein and CFP and documented with a digital DFX 500 camera (Leica). D-incubated or FR-treated protoplasts were prepared under dim green safelight. Photographs were taken during the first 10 min of analysis and assembled with CorelDraw X3 software (Corel, Ottawa, Canada).

Protein Extraction and Western Blots.

Seedlings were harvested into liquid nitrogen. A total of 100 μl of 2× SDS buffer was added to every 100 mg of seedlings before they were homogenized (Mixer Mill; Qiagen, Valencia, CA). Samples were immediately boiled for 5 min and cleared by centrifugation. Proteins in the supernatant were separated on 10% SDS gels and transferred to PVDF membranes for Western blot analysis as described previously (48), by using an anti-phyA primary antibody (a gift from A. Nagatani, Kyoto University, Kyoto, Japan) and an anti-actin primary antibody (Abcam, Cambridge, UK).

Measurements and Statistics.

Hypocotyl length and angle measurements were done with Image Tool Software (University of Texas Health Science Center, San Antonio, TX), and scripts written by J. Hughes (Justus Liebig University Giessen). All physiological experiments were repeated three times under equal conditions. Circular bar charts were created by using Excel (Microsoft, Redmond, WA) and Origin 7.5 (OriginLab, Northampton, MA) software.

Abbreviations

B

blue light

FR

far-red light

R

red light

CFP

cyan fluorescent protein

CRY

cryptochrome

FHY1

FR-elongated hypocotyl 1

HIR

high irradiance response

LED

light-emitting diode

Pfr

far-red-absorbing phytochrome

phyA

phytochrome A

phyB

phytochrome B

VLFR

very-low-fluence response

W

white light

D

darkness

PHOT

phototropin.