characterization of 6-kilodalton domain of biologi&d activity’ fileplant physiol. (1 997) 11...

Plant Physiol. (1 997) 11 5: 693-704

Characterization of Regions within the N-Terminal 6-Kilodalton Domain of Phytochrome A That Modulate

l t s Biologi&d Activity’

Emily T. Jordan, Jane M. Mar i ta , Richard C. Clough, and Richard D. Vierstra*

Department of Horticulture and the Cell and Molecular Biology Program, University of Wisconsin, Madison, Wisconsin 53706

Phytochrome A (phyA) is a red/far-red (FR) light photoreceptor responsible for initiating numerous light-mediated plant growth and developmental responses, especially in FR light-enriched environments. We previously showed that the first 70 amino acids of the polypeptide contain at least two regions with potentially opposite functions (E.T. Jordan, J.R. Cherry, J.M. Walker, R.D. Vierstra [I9961 Plant 1 9: 243-257). One region is required for activity and correct apoprotein/chromophore interactions, whereas the second appears to regulate phytochrome activity. We have further resolved these functional regions by analysis of N-terminal deletion and alanine-scanning mutants of oat (Avena sativa) phyA in transgenic tobacco (Nicotiana tabacum). l h e results indicate that the region involved in chromophore/apoprotein interactions contains two separate segments (residues 25-33 and 50-62) also required for biological activity. The region that regulates phyA activity requires only five adjacent serines (Sers) (residues 8-12). Removal or alteration of these Sers generates a photoreceptor that increases the sensitivity of transgenic seedlings to red and FR light more than intact phyA. Taken together, these data identify three distinct regions in the N-terminal domain necessary for photoreceptor activity, and further define the Ser-rich region as an important site for phyA regulation.

Plants have evolved a complex network of photoreceptors that utilize light signals to optimize growth and de- velopment. One important class of photoreceptors is the phytochromes, a family of homodimeric, approximately 120-kD chromoproteins that control a wide variety of mor- phogenic responses to R and FR light (Kendrick and Kro- nenberg, 1994). Phytochromes absorb light through a linear tetrapyrrole chromophore covalently attached to an inter- na1 Cys in each monomer (Vierstra, 1993; Quail et al., 1995). The chromoproteins exist in two conformational forms: Pr (A,,, = 666 nm) and Pfr (A,,, = 730 nm) that are inter- convertible by absorption of R and FR light, respectively.

* This work was made possible by financia1 support from a National Institutes of Health predoctoral training grant (no. GM- 07215 to E.T.J.), a U.S. Department of Energy grant (no. DE-FG02- 88ER13968 to R.D.V.), and a grant from the Department of Ener- gy/National Science Foundation/U.S. Department of Agriculture Collaboration Program on Research in Plant Biology (no. BIR 92-20331 to the University of Wisconsin).

* Corresponding author; e-mail [email protected]; fax 1-608-262-4743.

693

Phytochromes are initially synthesized as the biologically inactive Pr form and are converted to the active Pfr form only by irradiation with light (Smith, 1995). Although well studied both biochemically and genetically, the mechanism(s) of action of the phytochromes is still unknown.

phyA, the most abundant type of phytochrome in etiolated seedlings, is also the best characterized. Analysis of genetic phyA mutants and transgenic plants overexpressing phyA showed that this photoreceptor is most active in FR light-enriched environments, such as under a leaf canopy or the soil surface (Parks and Quail, 1993; Cherry and Vierstra, 1994; Smith, 1995). Under these conditions, phyA is predominantly in the stable Pr form (t1,* >100 h). Upon photoconversion of Pr to Pfr by exposure to R light- enriched environments, the levels of phyA drop dramati- cally because Pfr represes PHYA transcription and is rapidly degraded (t1,2 = 0.5-1 h; Vierstra, 1994).

One domain of particular importance for both the structure and function of phyA includes the N-terminal 70 amino acids (the 6-kD domain). Previous work with the purified photoreceptor showed that the 6-kD domain is required for correct chromophore! apoprotein interactions and undergoes a substantial conformational change upon photoconversion of Pr to Pfr (Hahn et al., 1984; Vierstra et al., 1987). Its role in phyA activity was studied by characterizing the effects of ectopically expressing N-terminal- deletion and site-directed mutants in transgenic plants. These investigations demonstrated that the 6-kD domain is necessary for phyA activity, but potentially in different ways.

Cherry et al. (1992) showed that an N-terminal deletion mutant of oat (Avena sativa) phyA lacking amino acids 7 through 69 (deletion NA, A7-69) is inactive in transgenic tobacco (Nicotiana tabacum). Whereas ectopic expression of full-length oat phyA induced a “light-exaggerated” phenotype characterized in part by decreased internode/ hypocotyl length (Keller et al., 1989; Cherry et al., 1991; Nagatani et al., 1991), deletion NA did not elicit this phenotype, despite its accumulation of similar levels of phyA. Subse- quently, Boylan et al. (1994) found that Arabidopsis seedlings expressing an oat phyA deletion mutant lacking amino acids 1 to 52 (AN52) were taller than wild-type

Abbreviations: FR, far-red; FRc, continuous F R phyA, phytochrome A; R, red; Rc, continuous R.

www.plantphysiol.orgon May 25, 2019 - Published by Downloaded from Copyright © 1997 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

694 Jordan et al. Plant Physiol. Vol. 115, 1997

tobacco seedlings under continuous FR light, suggesting that the AN52 mutant had a dominant negative effect. Stockhaus et al. (1992) showed that tobacco seedlings expressing rice phyA mutant S/A, which contained Ala in place of a11 10 Ser near the N terminus, were more sensitive to R light than seedlings expressing equivalent levels of intact phyA. This latter study suggested that alteration of the N-terminal Ser increases the activity of phyA.

To better define N-terminal sequences important for phyA activity, we previously created and analyzed a variety of N-terminal deletion mutants of oat phyA (Jordan et al., 1996). We found that the 6-kD domain of oat phyA contains two regions involved in photoreceptor function. One region, located between amino acids 13 and 62, is necessary for correct apoprotein / chromophore interactions and biological activity; deletion of even a small portion of this region substantially reduced or eliminated the activity of the photoreceptor. The second region consists of seven residues near the N terminus (amino acids 6-12), five of which are Ser. Deletion of these seven residues resulted in a photoreceptor that made plants more sensitive to R and FR light than full-length oat phyA.

We further characterize N-terminal sequences of oat phyA involved in photoreceptor structure and function using a series of more refined deletion and Ala-scanning mutants. The data show that the 6-kD domain of phyA contains three regions important for function. The region between amino acids 19 and 62 that is responsible for the spectral integrity and conformational stability of Pfr can be resolved into two smaller regions (amino acids 25-33 and 50-62) that are both required for biological activity. A third region (amino acids 8-12), consisting of five adjacent Ser, is not required for chromophore f apoprotein interactions of Pfr, but may function in regulating phyA activity.

MATERIALS A N D METHODS

Construction of Deletion and Site-Directed Mutants

The construction of the phyA-deletion mutant NF (A6- 12) was described previously in Jordan et al. (1996). Dele- tions NG (A2-22), NH (A22-47), NI (A22-30), and NJ (A31- 47) (see Fig. 1A) were created using a reconstruction of the full-length oat (Avena sativa) PHYA gene designated pWTRE. It contains an altered 5’-coding sequence re- engineered to include severa1 unique, translationally silent restriction sites to facilitate further modifications of the N-terminal amino acid sequence (Jordan et al., 1996). To construct deletion NG (A2-22), the 54-bp double-stranded linker GATCCATTACTTGAGGCAGGCGATGTTAGCA- CAGACAACCCTTGATGCCGAGCT was cloned into pWTRE digested with BamHI and SacI. This linker contained a BamHI site at its 5’ end, included the initiator Met codon of PHYA followed by codons 22 through 31, and ended with a 3’ SacI site. Deletion NH (A2247) was produced by replacing the sequence between the XkoI and SpeI sites of pWTRE with the double-stranded oligonucleotide

GCCAGAGTTCCCAAGCAAGAGATC, containing codons TCGAGCAGGCCTGCTTCCAGTTCTTCCTCGCGAAACC-

3 through 21 and codon 48 of PHYA that were flanked by a 5’ XkoI site and a 3‘ SpeI site.

Deletion NI (A22-30) was similarly constructed by replacing the sequence between the XkoI and SacI sites of pWTRE with the double-stranded oligonucleotide TCGA-

GAGTTCCCAAGCAAGGCGAGCT containing codons 3 through 21 followed by codons 31 and 32, and began and ended with a XkoI and a SacI site, respectively. To construct deletion NJ (A3147), two overlapping oligonucleotides

GCGAAACCGCCAGAGTTCC and GCTTCAACTAGTGC-

CTCTGGCGGTTTCGCGAGG) were annealed and ex- tended to produce a 90:bp double-stranded bridge containing the XkoI and SpeI sites of pWTRE-flanking codons 3 through 30 followed by codon 48. This XhoI1 SpeI-digested bridge was ligated into pWTRE digested with XkoI and SpeI.

To construct the Ala-scanning mutants NTSA (2-18), ASMl (25-33), ASM2 (3547), and ASM3 (50-62) (see Fig. 1 A), an oligonucleotide-based mutagenesis kit (Altered Sites, Promega) was used to change selected codons to those designating Ala. The mutagenic oligonucleotide sequences used were as follows: for NTSA (2-18), GAGG-

CTGCTCGAAACCGCCAGGCTGCTCAAGCAAGGG; for

GCTGCCGCTCTCGCTGCTGAATATG; for ASM2 (3547),

GCCGCTGCCGCCGCCGCTCTAGTTGAAG; and for ASM3

CAGCTGTGGCGGCAGGGGCATCTGAGAAGG. NTSA (2-18) encoded Ala in place of Ser-2, Ser-3, Ser-4, Ser-8, Ser-9, Ser-10, Ser-11, Ser-12, Ser-17, and Ser-18; ASMl (25-33) contained Ala in place of Gln-25, Thr-26, Thr-27, Asp-29, Glu-31, and Asn-33; ASM2 (35-47) encoded Ala in place of Glu-35, Tyr-36, Glu-37, Glu-38, Ser-39, Asp-41, Ser-42, Phe-43, Asp- 44, Tyr-45, Ser-46, and Lys-47; and ASM3 (50-62) contained Ala in place of Glu-50, Gln-52, Arg-53, Asp-54, Pro-56, Pro- 57, Gln-59, Gln-60, and Arg-62 (see Fig. l). The sequences of a11 deletion and Ala-scanning mutants were verified by dideoxy chain-termination sequencing.

GCAGGCCTGCTTCCAGTTCTTCCTCGCGAAACCGCCA-

(CGATGTCCTCGAGCAGGCCTGCTTCCAGTTCTTCCTC-

ATCAAGGGTTGTCTGTGCTAACACCCTTGCTTGGGAA-

CAGGCGATGGCTGCTGCTAGGCCTGCTGCTGCTGCTG-

ASMl(25-33), GCAAGGGTGTTAGCAGCTGCTGCCCTT-

GCCGAGCTCAATGCTGCTGCTGCTGCAGCTGGTGCC-

(50-62), CTCCAAACTAGTTGCAGCAGCTGCTGCTGGTG-

Plant Transformation and Growth Conditions

The mutated PHYA genes were inserted into the binary vector pBinl9 (Bevan, 1984) upstream of the PHYA 3’- untranslated region and expressed under the control of the cauliflower mosaic virus 35s promoter. The genes were then introduced into tobacco (Nicotiana tabacum, cv Xanthi) using Agrobacterium tumefaciens-mediated transformation with the A. tumefaciens strain LBA4404. Kanamycin- resistant regenerated plants were screened via immunoblot analysis for expression of the various oat phyA mutants (see below). For the phenotypic analyses of mature tobacco plants, five independently transformed T1 lines each of NG (A2-22), NH (A2247), NI (A22-30), NJ (A3147), and ASM3 (50-62); four lines of NTSA (2-18) and ASMl (25-33); two lines of ASM2 (35-47); six lines of full-length oat phyA-



N-Terminal Functional Regions of Phytochrome A 695

expressing plants; and wild-type tobacco plants were grown simultaneously under natural diurna1 light cycles in a greenhouse. The height of each plant was measured from the base of the stem to the apex when flower primordia were first visible (60 d after sowing).

Progeny of plants used in the mature plant assay were screened for segregation of kanamycin resistance by ger- mination on media containing 100 mg/L kanamycin. Re- sistant T2 seedlings were screened for oat phyA expression by immunoblot analysis using polyclonal antibodies raised against oat phyA (Shanklin et al., 1987). For the hypocotyl assays, two to three independent, homozygous T2 lines expressing each mutant were sown on 100-mm plates containing 0.5X Murashige and Skoog media and 0.8% agar. Plates were stored in the dark at 4°C for 2 d and then placed at 25°C either in darkness or under various fluence rates of FRc or Rc light for 6 d. Fluence rate was controlled by covering the plates with varying numbers of sheets of white chromatography paper (3MM, Whatman). FR and R light sources and calibration of fluence rates were as described in Jordan et al. (1996). Hypocotyl lengths (in milli- meters) were measured after 6 d growth in FRc or Rc light. The hypocotyl growth experiments were performed at least twice for each line.

Phytochrome Extractions and Quantitation

For the phytochrome dose/ phenotypic response analysis of mature plants, phytochrome content was determined either spectrophotometrically or by ELIsA (Jordan et al., 1996). Extracts were prepared from 10 g of green leaf tissue and concentrated by (NH,),SO, precipitation as described in Jordan et al. (1996). Phytochrome was quantitated spectrophotometrically by measuring the A(AA730-AA665) of etiolated tissue extracts or the A(AA730-AA800) of green tissue extracts, the latter being multiplied by the conversion factor 2.18 to obtain A(AA730-AA665). Alternatively, phytochrome content was determined by sandwich ELISA as described in Cherry et al. (1993). A polyclonal antibody raised against oat phyA was used as the capture antibody (20 pg mLpl; Shanklin et al., 1987) and a monoclonal antibody specific for oat phyA was used as the detection antibody (Oat-22, 10 pg mL-'; Cordonnier et al., 1984). A dilution series of purified oat phytochrome was used as the standard (Vierstra and Quail, 1983).

lmmunoblot Analyses

Immunoblots were performed on protein extracts sepa- rated by SDS-PAGE and electrophoretically transferred to PVDF membranes as described in Cherry et al. (1993). Primary antibodies were either affinity-purified polyclonal antibodies raised against oat phytochrome (1 pg mL-l; Shanklin et al., 1987) or the monoclonal antibody Oat-22 (2 pg mL-'). Goat anti-rabbit or anti-mouse immunoglobu- lins conjugated to alkaline phosphatase (1 pg mLpl) were used as secondary antibodies.

Spectrophotometric Analyses and Dark Reversion

Phytochrome used for difference spectroscopy was partially purified from 6-d-old etiolated seedlings by (NH,),SO, precipitation and hydroxyapatite chromatography under a green safelight (Jordan et al., 1996). Phytochrome samples were incubated with an affinity matrix consisting of Oat-22 antibody covalently attached to Affigel-10 beads (Bio-Rad). The matrix was prepared by mixing Oat-22, purified by Pro- tein A chromatography, with the Affigel-10 beads at a con- centration of 4 mg Oat-22 mL-' beads (Jordan et al., 1996). Phytochrome extracts (approximately 0.1 A[AA] units of phytochrome mL-l) were gently mixed with 500 pL of Oat-22/ Affigel-10 beads for 1 h in the dark. The beads were washed three times with 1 mL of 100 mM Tris-HC1 (pH 7.8,4"C). The amount of bound phytochrome was determined indirectly by subtracting the amount of phytochrome remaining in the supernatant and washes from the amount of phytochrome added to the beads.

FR minus R light difference spectroscopy was performed on phytochrome immobilized to the Oat-22/ Affigel-10 matrix as described in Jordan et al. (1996). Thermal reversion rates of Pfr to Pr were similarly determined using the R light minus FR light difference spectroscopy on immobilized phytochromes treated with 5 mM sodium dithionite.

Helical Hydrophobic Moment Analysis

The helical hydrophobic moments were derived from the equation:

H,,Sin(Sn)]' + [ ~~-lH,Cos(Sn) ' "'/N I 1 where n is a particular residue in the amino acid sequence, N is the window over which the mean values of hydrophobicity are averaged, 6 is the distance between residues as they would be aligned in a helix (6 = 100" for an a-helix), and H is a hydrophobic value assigned to a residue n, based on the average free energy associated with transfer from a nonpolar to a polar solvent (Eisenberg et al., 1984; Finer-Moore and Stroud, 1984). The average (pH) value for each residue was calculated from the above equation using the program PepPlot from the Genetics Computer Group (Madison, WI). Because the (pH) value used a window of 8 residues and a starting sequence 73 amino acids in length, the first 5 amino acids (1-5) and the last 5 amino acids (69-73) were not included in the evaluation.

RESULTS

Production of Tobacco Expressing Deletion and Ala-Scanning Mutants of Oat phyA

By a number of criteria, the N-terminal 6-kD domain is important not only for structural integrity but also for biological activity of phyA. Previously, we characterized transgenic tobacco expressing six deletion mutants of this domain (Jordan et al., 1996). Analysis of the activity and spectrophotometric properties of these deletions indicated that the 6-kD domain contains at least two functional re-




gions. The first region, located between amino acids 13 and 62, is necessary for Pfr conformational stability and spectral integrity and is essential for activity. Deletions lacking a11 or part of this region were less conformationally stable as Pfr in vitro and had their Pfr absorbance maxima shifted to shorter wavelengths, indicating that their Pfr structures were altered specifically. Moreover, these mutants were much less effective in inducing the light-exaggerated phenotype in tobacco plants than was full-length oat phyA. The second functional region, between amino acids 6 and 12, is not required for Pfr apoprotein/chromophore interactions, but appears to be involved in attenuating phyA activity. Seedlings expressing deletion NF (A6-12), which lacked this region, were 10 to 100 times more sensitive to R and FR light than those expressing equivalent amounts of full-length oat phyA.

To better define critica1 regions within the N-terminal 6-kD domain, we examined a series of more refined deletion and Ala-scanning mutants of oat phyA (Fig. 1A). These mutations were designed to further map the previously identified functional regions, and to determine whether changes in activity were due to either substitutions of specific amino acids or to alterations in the spatial posi- tioning of existing amino acids. Deletions NH (A2247), NI (A22-30), and NJ (A3147) were missing a11 or part of the previously identified N-terminal region required for eliciting the light-exaggerated response in mature tobacco (Jor- dan et al., 1996). Deletion NH (A2247) lacked a11 26 amino acids in this region, whereas deletions NI (A22-30) and NJ (113147) were missing the first 9 and last 17 amino acids, respectively. Deletion NG (A2-22) lacked the entire Ser-rich region plus four additional residues (Gln-19, Ala-20, Arg-

Figure 1 . A, Diagram of the various deletion and Ala-scanning mutants of oat phyA. All mutants were derived from an oat PHYA gene that was modified to contain severa1 silent restriction sites within the coding region for the first 70 amino acids (Jordan et al., 1996). The amino acid sequences shown for the deletion mutants represent residues that were removed. Residues underscored by the black symbols in the de- scription of the Ala-scanning mutants indicate amino acids that were changed to alanines. Deletion NF (A6-12) was described in Jordan et al. (1996). €3, lmmunoblot detection of the various deletion and Ala-scanning mutants of oat phyA expressed in tobacco seedlings. Equal aliquots of crude extracts from 6-d-old etiolated seedlings were subjected to SDS-PAGE and im- munoblotted using Oat-22 antibody. WT, Wild-type seedlings; FL, seedlings expressing full-length oat phyA.

A Deletion Mutants

NF (A 6-12)

NG (A 2-22).

NH (A 22-47)

NI (A 22-30)

NJ (A 31-47)

Alanine Scanning Mutants

B

21, and Val-22) directly downstream of the last Ser (Fig. 1A).

The Ala-scanning mutants were produced by changing groups of conserved and/ or charged residues to Ala (Zoller, 1991). For the mutants ASMl (25-33), ASM2 (35- 47), and ASM3 (50-62), we altered residues (mostly hydrophilic) between amino acids 25 and 62 that are conserved among known phyAs (Fig. 1A). The NTSA (2-18) mutant resembled the S/A mutant of rice phyA (Stockhaus et al., 1992), in that a11 10 Ser within the first 20 amino acids of oat phyA were changed to Ala. Because five of the seven residues removed in hyperactive deletion NF (A6-12) were Ser, we designed NTSA (2-18) to test if the Ser in particular were responsible for the hyperactive phenotype induced by the NF (A6-12) deletion (Jordan et al., 1996).

The coding regions of each mutant were fused to the cauliflower mosaic virus 35s promoter and introduced into tobacco. Regenerated plants expressing the mutant phyAs were identified by immunoblot analysis of green leaf extracts using monoclonal antibody Oat-22, which is specific for oat phyA (Cordonnier et al., 1984). The immunoblot in Figure 1B shows that the mutant phyAs migrated during SDS-PAGE with apparent molecular masses similar to those predicted from polypeptide composition. Full-length oat phyA migrated with an apparent mass of 124 kD. Deletions NG (112-22) and NH (A2247) migrated at 121 kD, deletion NI (A22-30) at 123 kD, deletion NJ (A3147) at 122 kD, and Ala-scanning mutants NTSA (2-18), ASMl (25-33), ASM2 (3547), and ASM3 (50-62) at 124 kD. In transgenic lines selected for further analyses, the mutant phyAs were a11 ex'pressed to similar levels, which allowed for direct comparison of their in vivo activities.

10 20 30 40 50 60 70 I I I I I I I I MSSSRPASSSSSRNRQSSQARVLAQTTLDAELNAEYEESGDSFDYSKLVEAQRDGPPVQQGRSEKVIAYLQ

LAQTTL3AELNAEYEESSDSFDYSK

NTSA ASMl ASMZ ASM3 (2-18) (25-33) (35-47) (50-62)

MSSSRPASSSSSRNRQSSQAVLAQTTLDAELNAEYEESGDSFDYSKL~AQR~PPVQ~RSE~IAYLQ -0 .. 0.0 O O O A A A A A AAAAAAA T TTT TT TT T




Phenotypic Analysis of Mature Tobacco Expressing phyA Mutants

leve1 of phytochrome as wild-type tobacco plants (determined spectrophotometrically) were approximately four

Because the deletion and Ala-scanning mutants affected the two putative functional regions of the 6-kD domain, either by remova1 or alteration of existing amino acids, we predicted that three types of mutants would emerge: inactive or partially active phyAs, phyAs with normal activity, or hyperactive phyAs. As a first step to determine the activity of the various mutants, we examined the relationship between the amount of phytochrome expressed in mature plants and the degree of light responsiveness as assayed by plant height. Representative lines expressing each of the mutant phyAs were grown simultaneously under natural, diurna1 light conditions in a greenhouse alongside wild-type tobacco plants and Vthose expressing full-length oat phyA. When flower primordia first appeared (60 d after sowing), we measured plant height and harvested green leaf tissue from each plant. Phytochrome was partially purified from this tissue and assayed in two ways: spectrophotometrically by measuring A(AA730- AAso0) following R and FR irradiation, and immunologi- cally by sandwich ELISA using Oat-22. The first assay detected a11 photoreversible phytochromes, whereas the second assay detected only oat phyA. The height of each plant was then plotted relative to its phytochrome content determined by either assay.

As observed previously, tobacco plants were highly sensitive to phytochrome content, as reflected by a nonlinear dose/ response relationship that was saturated at low levels of active phyA (Fig. 2; Cherry et al., 1991). For example, full-length oat phyA plants expressing at least twice the

times shorter than wild-type plants (23.1 ? 7.3 and 81.0 2 6.2 cm, for full-length oat phyA and wild-type tobacco plants, respectively) (Fig. 2, A and B, left panels). Full- length oat phyA plants expressing levels of phyA lower than this threshold were about the same height as wild- type tobacco plants, whereas plants expressing higher levels of chromoprotein were saturated for the dwarf phenotype. A similar dose/response curve was observed if phytochrome content was determined by ELISA (Fig. 2, A and B, right panels). This indicated that the dwarf response was elicited by increased levels of photoreversible oat phyA, and not by elevated levels of tobacco phytochromes.

Compared with full-length oat phyA, none of the deletions examined in this study displayed full biological activity in mature plants (Fig. 2A). Deletions NG (A2-22) and NJ (A3147) were partially active; NG (A2-22) and NJ (A31- 47) plants, which contained levels of phytochrome greater than twice those of wild-type tobacco plants, were shorter than the nontransformed plants but never as short as full- length oat phyA plants. Deletions NH (A22-47) and NI (A22-30) were completely inactive. Despite accumulating levels of spectrophotometrically active photoreceptor up to 4-fold higher than that sufficient for intact phyA to induce the light-exaggerated phenotype, NH (A2247) and NI (A22-30) plants grew to similar heights as wild- type tobacco.

The activities of the Ala-scanning mutants NTSA (2-18), ASMl (25-33), ASM2 (3547), and ASM3 (50-62) were assayed in a similar fashion (Fig. 2B). Mutants ASMl (25-33)

A B

0- o- - o o k % T i z % T ! s o + k - 3 ò 7 d w

A(AAYg Protein X 10 pg phytoehromel g protein 00-!s o-!

A(AA)/g Protein X 10 pg phytochromd g protein

Figure 2. Relationship between phytochrome dose and phenotypic response of tobacco plants expressing deletion mutants (A) or Ala- scanning mutants (B) of oat phyA. Wild-type plants (A), plants expressing full-length oat phyA (O), or plants expressing the various mutants (O) were simultaneously grown to matu- rity in a greenhouse under natural lighvdark cycles. Phytochrome was extracted from api- cal leaf tissue and quantitated either by a spectrophotometric assay (left panels of A and B), which detects both oat and tobacco chromoproteins, or by a sandwich ELISA (right panels of A and B) using the Oat-22 monoclonal antibody, which i s specific for oat phyA. For each plant, phytochrome content was plotted against the plant height at flowering. The plant populations for each phyA construction shown were derived from two to five independently transformed lines. The phytochrome dose/ phenotypic responses of full-length oat phyA- expressing plants are included in each pane1 for comparison.




and ASMS (50-62) were biologically inactive; matureplants containing elevated levels of these phytochromeswere similar in height to wild-type tobacco plants. Al-though the levels of spectrophotometrically detectablephyA in the ASMS (50-62) plants appeared low (Fig. 2B,left panels), subsequent biochemical analysis showed thatmutant ASMS (50-62) efficiently bound chromophore andwas fully photoreversible (Fig. 6B and data not shown).Unlike mutants ASM1 (25-33) and ASMS (50-62), mutantsNTSA (2-18) and ASM2 (35^7) were active in matureplants, in that plants expressing these mutants were asshort as full-length oat phyA plants.

Phenotypic Analysis of Seedlings Expressing phyA Mutants

In seedlings phyA mediates the light-dependent inhibi-tion of hypocotyl elongation, especially in FR light-enriched environments (Smith, 1995). To further define thebiological activity of the deletion and Ala-scanning mu-tants, we measured the hypocotyl lengths of seedlingsexpressing these mutant phyAs grown under either FRc orRe light. For the assay, we generated T2 lines homozygousfor each deletion, as identified by their failure to segregatefor both kanamycin resistance and oat phyA expression(data not shown) and used those lines that expressedequivalent levels of immunodetectable phyA protein (Fig.3, A and B, upper panels).

Consistent with previous results (Stockhaus et al., 1992;Boylan et al., 1994; Jordan et al., 1996), expression of full-length oat phyA substantially reduced hypocotyl elonga-tion in seedlings grown under continuous light. Six-day-old full-length oat phyA seedlings were approximately 2.5-and 2-fold shorter than wild-type tobacco seedlings whengrown under FRc light (0.15 jumol m~2 s"1; Fig. 3A) and Relight (0.04 jumol m~2 s'1; Fig. 3B), respectively. When thephyA mutants were assayed under identical fluence ratesof FRc and Re light, they induced hypocotyl growth re-sponses characteristic of either inactive, active, or hyperac-tive photoreceptors (Fig. 3). The mutant phyAs that wereinactive in mature plants, NH (A22-47), NI (A22-30), ASM1(25-33), and ASMS (50-62), were also inactive in seedlingsirradiated with either FRc or Re light; these seedlings wereas tall as wild-type tobacco seedlings. The Ala-scanningmutant ASM2 (35^7), which was as active as the full-length oat phyA in mature plants, was also active in seed-lings; ASM2 (35-47) seedlings were as short as full-lengthoat phyA seedlings.

In contrast to being partially active in mature plants,deletions NG (A2-22) and NJ (A31^47) were not active inseedlings grown under low fluence rates of FRc or Re light.Both NG (A2-22) and NJ (A31-47) seedlings were as tall aswild-type tobacco seedlings under these light conditions(Fig. 3). It was interesting that the Ala-scanning mutantNTSA (2-18), which had activity in mature plants compa-rable to that of full-length oat phyA, appeared to be moreeffective than full-length oat phyA in seedlings. NTSA(2-18) seedlings were approximately 2-fold shorter thanfull-length oat phyA seedlings under FRc or Re light, sug-gesting that replacement of the 10 N-terminal Ser with Alacreated a hyperactive photoreceptor.

onB

£>oI>,s

t

IaIS -

10 -

5 -

n -

T1

TX

1

T1 T

1

TJL

n

T1

TnT1

Figure 3. Hypocotyl growth response of seedlings expressing dele-tion and Ala-scanning mutants of oat phyA compared with wild-type(WT) seedlings and seedlings expressing full-length (FL) oat phyA.The transgenic lines used were homozygous for the transgene andexpressed equivalent amounts of oat phyA. Top, Equal aliquots ofcrude extracts from 6-d-old seedlings irradiated with FRc (0.15 j^molm~2s^1) light (A) or Re (0.04 /j,mol m~2 s^ ' ) (B) were immunoblottedwith Oat-22 antibody. Bottom, The average hypocotyl lengths (inmm) of 20 to 40 seedlings for each line were determined after 6 dunder either light condition. Error bars reflect the so. The mutants aredescribed in Figure 1.

To more accurately compare the biological activity ofthe phyA mutants, we measured their hypocotyl growthresponses over a range of FRc and Re light fluence rates(Fig. 4). NH (A22-47), NI (A22-30), NJ (A31-47), ASM1(25-33), and ASMS (50-62) seedlings resembled wild-typetobacco seedlings over a 1000-fold range of FRc or Re lightfluence rates, indicating that these mutant phyAs had nomeasurable activity in seedlings. NG (A2-22) seedlingsresembled wild-type tobacco seedlings under low fluencerates of FRc light, but resembled full-length oat phyAseedlings under the highest fluence rate of FRc light tested(14.3 jumol m~2 s^1). Under all Re light fluence rates, NG(A2-22) seedlings were slightly shorter than wild-typetobacco seedlings (Fig. 4). These data suggest that deletion www.plantphysiol.orgon May 25, 2019 - Published by Downloaded from

Copyright © 1997 American Society of Plant Biologists. All rights reserved.



-

100 s g 75 M

c;i

E 50 2 R x 25 t3

O D -3 -2 -1 O 1

log [Fluence Rate] (pmol m-2 s-1)

I 1 100

B g 75

F

M

4 m

50 8 E 25 &?

O

B


100

75

50

25

O

log Fluence Rate] (pmol m-2 s-1) D -3 -2 -1 O 1 2

100

50

25

D 3 -2 -1 O 1 2 log [Fluence Rate] (pmol m-2 s-1)

NG (h2-22) retained some activity under Rc light and high fluence rates of FRc light. ASM2 (35-47) seedlings were the same height as full-length oat phyA seedlings under a11 fluence rates tested, indicating that the ASM2 (35-47) chromoprotein was active. In contrast, NTSA (2- 18) seedlings were shorter than full-length oat phyA seedlings under a11 fluence rates of FRc light and a11 but the highest fluence rates of Rc light, consistent with a hyperactive behavior. Like NF (A6-12) seedlings (Jordan et al., 1996), NTSA seedlings were shorter as the Rc light fluence rates decreased.

Functional Comparison of Ala-Scanning Mutant NTSA (2-18) and Deletion NF (A6-12)

The hypersensitive phenotype can be elicited by expression of deletion (NF [A6-12]) or Ala-scanning (NTSA [2- 181) mutants of the Ser-rich region (Jordan et al., 1996; this report). To determine if these mutations were equally effective in increasing light sensitivity beyond the activity of full-length oat phyA, we simultaneously compared the hypocotyl growth responses of the wild-type tobacco, full- length oat phyA, NF (A6-12), and NTSA (2-18) seedlings over a range of FRc or Rc light fluence rates (Fig. 5). We examined three homozygous lines of NTSA (2-18), NTSA21, NTSA25, and NTSA36, which expressed levels of phyA similar to those of the chosen full-length oat phyA and NF (116-12) lines. Line NTSA36 accumulated phyA levels equivalent to those of NF (A6-12) and full-length oat phyA seedlings (Fig. 5, A and B, upper panels). Lines NTSA21 and NTSA25 expressed approximately 1.3- and 2-fold less phyA than NTSA36, respectively.

Figure 4. Effects of varying fluence rates of FRc or Rc light on hypocotyl lengths of 6-d-old seedlings expressing deletion mutants (A and C) and Ala-scanning mutants (B and D) of oat phyA. The responses of wild-type (WT) seedlings and seedlings expressing full-length (FL) oat phyA are included for comparison. The hypocotyl lengths are expressed as the percentage of the length of dark-grown seedlings. Each point represents the average length of 20 to 40 seedlings. Error bars are omitted for clarity. Representative relative standard deviations for each line are as follows. Under 0.1 5 pmol m-2 s-' FRc light (A and 6): WT, 26%; FL, 26%; N C (A2-22), 26%; NH (A22-47), ?8%; NI (A22-30), 28%; NJ (A31-47), 27%; NTSA (2-18), 22%; ASMl (25-33), 27%; ASM2 (35-47), 24%; and ASM3 (50-62), 27%. Under 0.12 pmol m-'s-' Rc light (C and D): WT, 58%; FL, 25%; N C (A2-22), 56%; NH (A22-47), 29%; NI (A22-30), ?5%; NJ (h31-47), 26%; NTSA (2-1 8), 22%; ASMl (25-33), 25%; ASM2 (35-47), 26%; and ASM3 (50-62), %7'/0. A, WT; 0, FL; O, N C (A2-22); V, NH (A22-47); O, NI (A22- 30); A, NJ (A31-47); b, NTSA (2-18); O, ASMl (25-33); t, ASM2 (35-47); and O, ASM3 (50-62).

As can be seen in Figure 5, the NTSA36 and NTSA21 seedlings grew to similar heights as NF (A6-12) seedlings under a11 fluence rates of FRc and Rc light tested. Seedlings from line NTSA25 were intermediate in height compared with those from the NF (A6-12) and full-length oat phyA lines. Although the NTSA25 seedlings were taller than NF (A6-12), NTSA36, and NTSA21 seedlings, they were still shorter than full-length oat phyA seedlings under a11 but the highest fluence rates of Rc light examined, suggesting that the intermediate leve1 of hyperactivity for line NTSA25 resulted from lower chromoprotein levels. A11 four mutants showed a greater repression of growth as the Rc light fluence rates decreased. Based on their similar hyperactive light responses, we concluded that the phyA mutants NTSA (2-18) and NF (A6-12) were equally active on a per molecule basis.

Spectrophotometric Characteristics of the phyA Mutants

As described in Jordan et al. (1996), the N-terminal region between amino acids 13 and 62 of oat phyA is important for both appropriate light absorption and stability of the Pfr form. By examining the spectral integrity and conformational stability of the deletion and Ala- scanning mutants, we further defined the domains important for these functions. One way to detect subtle changes in apoproteinf chromophore interactions is by analysis of phytochrome difference spectra following R and FR irradiation. The FR minus R light difference spectrum of full-length oat phyA has a R light maximum at 666 nm (86% Pr), and a FR light minimum at 730 nm (exclusively Pfr). Cherry et al. (1992) showed previously that a mutant



700 Jordan et al. Plant Physiol. Vol. 11 5, 1997

B

...o----

I I

! 2


-e--- - o-& I I I I I I D ’-3 -2 -1 O 1 2

log [Fluence Ratel (pmol m-2 s-1)

Figure 5. Comparison of hypocotyl lengths for seedlings expressing deletion N F (A6-12) or Ala-scanning mutant NTSA (2-18) under varying fluence rates of FRc light (A) or Rc light (B). Three independent, homozygous lines of NTSA (2-1 8) tobacco were examined in this study. Wild-type (WT) seedlings or seedlings expressing full- length (FL) oat phyA were grown alongside those expressing the mutant phyAs for comparison. Top, lmmunoblots of extracts from seedlings grown under 0.04 pmol m-’ s-l FRc or Rc light, using Oat-22 antibody. Bottom, The hypocotyl lengths of the various lines expressed as the percentage of the length of dark-grown seedlings. Each point represents the average length of 20 to 40 seedlings. Error bars are omitted for clarity. Representative relative standard deviations are as follows. Under 0.15 pmole m-2 s-l FRc light (A): WT,

and NTSA36, 22%. Under 0.12 pmole m-’s-’ Rc light (B): WT,

and NTSA36, 25%. A, WT; O, FL; O, N F (156-1 2); b, NTSA2l; W, NTSA25; and O , NTSA36.

27%; FL, 24%; N F (A6-12), 21%; NTSA21, 22%; NTSA25, 23%;

21 1%; FL, 25%; N F (A6-12), 2 2 % ; NTSA21, %2%, NTSA25, 54%;

phyA lacking most of the 6-kD domain (deletion NA [A7-69]) has both its maximum and minimum shifted to slightly shorter wavelengths (660 and 726 nm, respective-

ly), indicative of spectral perturbations in Pr and Pfr. We showed that two smaller deletions within this 70-amino acid region (NB [A49-621 and NC [A6-471) also have altered FR minus R light difference spectra, particularly in their FR light minima (Jordan et al., 1996). As a11 three deletions had impaired biological activity, their spectral integrity and function may be connected.

Prior to spectrophotometric analysis of the deletion and Ala-scanning mutants, we partially purified them using an immunoaffinity technique that separates oat phyA from tobacco phytochromes (Jordan et al., 1996). The technique involved immunoadsorption of oat phyA to Affigel-10 beads containing covalently bound Oat-22 antibody. After washing the beads to remove most tobacco proteins, the immobilized phyAs were used directly in the spectrophotometric assays. Figure 6 shows the difference spectra of the phyA mutants using the immunoaffinity procedure. Like full-length oat phyA, mutants NG (A2-22) (Fig. 6A) and ASM2 (3547) and NTSA (2-18) (Fig. 6B) had maxima and minima at 666 nm and 730 nm, respectively, indicating that their respective lesions did not affect regions important for the spectral integrity of Pr or Pfr. Mutants NH (A2247), NI (1122-30), and NJ (83147) (Fig. 6A), and ASMl (25-33) and ASM3 (50-62) (Fig. 6B) had R light maxima at 666 nm, but unlike full-length oat phyA, they had FR light minima shifted to 728 nm. The alterations in FR light minima but not R light maxima of these mutants indicate alterations specifically in Pfr absorbance.

As another way to assess the conformational stability of the mutagenized phytochromes, we examined the rate of in vitro thermal reversion of Pfr back to Pr in the absence of light (dark reversion). The Pfr form of full-length oat phyA does not readily dark-revert, even in the presence of the reductant sodium dithionite (Fig. 7). In contrast, phyAs lacking the first 70 amino acids readily dark-reverted after treatment with sodium dithionite, with as much as 35% Pfr reverting to Pr after 2 h (Jordan et al., 1996). Figure 7 shows the thermal Pfr reversion rates of the immobilized deletion and Ala-scanning mutants, which were treated similarly. Of the mutant phyAs, only NTSA (2-18) was as stable as full-length oat phyA, with only 5% of Pfr dark-reverting to Pr after 2 h. The other mutants, NG (A2-22), NH (A22-47),

and ASM3 (50-62), showed various degrees of Pfr instabil- ity, with between 10 and 40% of Pfr dark-reverting after 2 h. This indicates that the region between amino acids 19 and 62 may play a role in maintaining the conformational stability of Pfr, but that the N-terminal Ser are not required.

NI (A22-30), NJ (A3147), ASMl (25-33), ASM2 (3547),

DlSCUSSlON

Regions within the N-Terminal 6-kD Domain of phyA Are lmportant for Activity

The results discussed here define three regions within the 6-kD domain of oat phyA involved in function through biological and biochemical characterizations of a series of deletion and Ala-scanning mutants within this region (Fig. 8). The biological efficacy of each mutant phyA was measured by its ability to elicit the light-exaggerated pheno-



N-Terminal Functional Regions of Phytochrome A 70 1

A

8 9 -E w 2 a

666

Wavelength (nm)

B 666

8

e w E cd

P

3

500 600 700 800

Wavelength (nm) 900

Figure 6 . FR light minus R light difference spectra of deletion (A) and Ala-scanning (€3) mutants of oat phyA. The mutants are described in Figure 1. Phytochromes were partially purified and immobilized by binding to Oat-22 antibody covalently coupled to an Affigel-1 O matrix. Difference spectra were determined directly with the immobilized phytochrome following saturating R and FR irradiation. The F R light minus R light difference spectrum of full-length oat phyA ( F L ) is shown for comparison in each panel.

type in seedlings and mature plants. The mutant chromoproteins were subsequently isolated from these transgenic plants and analyzed for alterations in both spectral properties and Pfr conformational stability.

Two regions important for phyA function are located within amino acids 25 to 33 and 50 to 62 (Fig. 8) and are necessary for biological activity. These regions were resolved by the success of Ala-scanning mutant ASM2 (35- 47), and the failure of mutants ASMl (25-33) and ASM3 (50-62) to elicit light-exaggerated responses in mature plants grown under white light or in seedlings irradiated with FRc or Rc light. The two regions necessary for biological activity lie within a larger region identified by deletion mapping between amino acids 19 and 62, which is essential for correct Pfr apoprotein/ chromophore interactions. Mu- tants that affected this larger region, including deletions NG (A2-22), NH (A2247), NI (A22-30), and NJ (A3147), were less conformationally stable than Pfr and had altered Pfr absorbance spectra.

The third functional region within the 6-kD domain, consisting of five Ser between amino acids 8 and 12, was located by characterizing deletion NF (A6-12) and the Ala- scanning mutant NTSA (2-18) (Jordan et al., 1996; this report). This region is not necessary for chromophore/ apoprotein interactions, but does affect phyA function. Seedlings expressing mutants NF (A6-12) and NTSA (2-18) were more sensitive to FRc and Rc light than seedlings expressing equivalent amounts of full-length oat phyA. Because both the Ala-scanning mutant NTSA (2-18) and deletion NF (A6-12) were equally effective at eliciting hyperactive responses in seedlings grown under low fluence rates of light (Fig. 5), the residues altered in common, Ser-8 to Ser-12, were likely to be responsible for this hyperactivity (Fig. 8). Stockhaus et al. (1992) found a similar hypersensitive effect in a rice phyA mutant containing Ala in place of the 10 N-terminal Ser (S/A). Together, these results suggest that the Ser-rich region (especially Ser-8 to

Ser-12) functions to attenuate phytochrome activity, such that its removal or alteration creates a hyperactive photoreceptor.

Roles of the 6-kD Domain in Pfr Structure and Function

Amino acids 25 to 33 and 50 to 62 may be important for chromoprotein structure for severa1 reasons. These regions may play indirect or direct roles in phytochrome function by maintaining the appropriate conformation of the Pfr chromophore and or polypeptide to allow other parts of the photoreceptor to initiate signaling. For example, alterations within these domains may prevent the entire 6-kD domain from folding into its correct conformation, thus precluding an adjacent "response" domain from transduc- ing the light signal. Alternatively, these two functional regions may interact with other molecules to initiate light- mediated signaling events directly. The involvement of the 6-kD domain in signal transduction is supported by the observation that FRc light-grown Arabidopsis seedlings expressing an N-terminal deletion of oat phyA missing the first 52 amino acids (AN52) were taller than wild-type tobacco seedlings (Boylan et al., 1994). This dominant negative phenotype suggests that the 6-kD domain may be necessary for productive interactions with downstream signaling molecules.

CD-spectra analyses of phyA showed that as Pr, the 6-kD domain is mostly randomly coiled and loosely associated with the chromophore, but as Pfr, assumes an a-helical structure positioned close to the chromophore (Furuya and Song, 1994). Based on these and additional biophysical data and sequence-based modeling studies, Parker et al. (1991) proposed that the 6-kD domain forms amphiphilic a- helices in Pfr, which contain hydrophilic faces that interact with the solvent and hydrophobic faces that interact with the chromophore. Furthermore, they proposed that one or more of these amphiphilic a-helices are essential for the




100

.- 2

2

.s 90 Q E

Q 80 PI

701

o FL b NTSA 0 NG m ASMZ

8 ASMl ASM3

O N I

A NJ V NH

60 1 O 0.5 1.0 1.5 2.0

Time (h) Figure 7. Pfr to Pr dark-reversion of deletion and Ala-scanning mutants of oat phyA. The mutants were partially purified and immobilized to Oat-22 antibody covalently coupled to an Affigel-1 O matrix. The suspensions were made 5 mM sodium dithionite and at t = O, the phytochromes were converted to Pfr by saturating R irradiation. Reversion of Pfr to Pr in the dark was detected by the R light minus FR light difference spectroscopy. The dark-reversion rate of full- length oat phyA (FL) is shown for comparison. The data represent the average of two independent experiments. Error bars are omitted for clarity. Representative SDS (at t = 60) are as foilows: FL, 22.1%; NG (A2-22), 25.7%; N H (A22-47); NI (A22-30), ?0.7%; NJ (A31-47), 54.2%; NTSA (2-18), 20.0%; ASMl (25-33), 55.3%; ASMZ (35- 47), 26.0%; and ASM3 (50-62), 20.7%. O, FL; O, NG (A2-22); V,

ASMl (25-33); t, ASMZ (35z47); and O, ASM3 (50-62). N H (822-47); O , NI (A22-30); A, NJ (A31-47); b, NTSA (2-18); C.,

biological function of phytochromes by providing structures that stabilize the Pfr chromoprotein to allow signal initiation. Using helical hydrophobic moment analysis ((pH); Eisenberg et al., 1982), amino acids 16 to 75 (for oat phyA) were calculated to have the highest probability of forming these amphiphilic structures (Parker et al., 1991).

Given the possibility that amphiphilic structures in the 6-kD domain are essential for biological activity, we compared the (pH) values of the first 70 amino acids of intact

Figure 8. Summary of functional regions of the N-terminal region of oat phyA based on biochemical and biological properties of deletion and Ala-scanning mutants (Jordan et al., 1996; this report). The shaded box indicates the Ser-rich region involved in regulating ..

oat phyA with those of the Ala-scanning mutants ASMl (25-33), ASM2 (35-47), and ASM3 (50-62). The data suggest that phyA activity is not linked to the probability of amino acids 25 to 62 forming amphiphilic structures.

As shown in Figure 9, amino acids 16 to 49 and 60 to 68 of full-length oat phyA had the highest probabilities to form amphiphillic structures, with average (pH) values of 0.39 and 0.51, respectively. The Ser-rich region between amino acids 5 and 15 had lower (& values (average (pH) = 0.03), as did the region between amino acids 50 and 59 (average (pH) = 0.20). The (pH) values of mutants ASMl (25-33) and ASM3 (50-62) were very similar to those of the full-length oat phyA, with average (pH) values of 0.37 between amino acids 16 and 49. This predicted that both Ala-scanning mutants had similar capacities for forming amphiphilic structures, even though they were both biologically inactive. In contrast, the (pH) values for active mutant ASM2 (35-47) were greatly reduced between amino acids 16 and 49 (average (pH) = 0.13). The ASM2 (3547) mutation altered the (pH) profile of most of the 6-kD domain, even though the amino acid substitutions were lim- ited to residues 35 to 47. The lower (pH) values indicated that a-helices formed by the 6-kD domain of mutant ASM2 (35-47) are potentially less amphiphilic than those of intact phyA, even though both photoreceptors were equally active. Therefore, the predicted amphiphilic structures formed by amino acids 35 to 47 may not be necessary for photoreceptor activity.

Potential Roles of the Ser-Rich Domain in Down- Regulating phyA Activity

In contrast to the region between amino acids 19 to 62, the Ser-rich domain (amino acids 8-12) is not required for the structural integrity of Pfr. However, this region does appear to be involved in down-regulating phyA activity. It is likely that this regulation is biologically meaningful, considering the high conservation of the Ser-rich tract in all known phyAs (Mathews et al., 1995). For example, it could provide a way to rapidly control the number of active photoreceptors available to initiate signaling events, thus improving the response of plants to fluctuating light envi-

23-62 Pfr Spectral

Integrity

1 I 4? :o lu phyA activity. The hatched boxes represent the

two regions necessary for biological activity, located within a larger region necessary for maintaining the spectral integrity and conformational stability of Pfr.

~ S S R P n S S S S S R M i Q S S ~ Q ~ ~ ~ E S G D S F D Y S ~ ~ Q ~ P ~ Q ~ R S E ~ ~ Q

I 25-33 50-62 Regulation Biological Biological

Activity Activity 8-12 I

19-62 Pfr Conformational

Stability www.plantphysiol.orgon May 25, 2019 - Published by Downloaded from

Copyright © 1997 American Society of Plant Biologists. All rights reserved.



FL ASMl(25-33) 0.8 - 0.6 - -

2 0.4 - - 0.2 - -

A

V

O 1 1 1 1 1 1 1 I I I I I I I

O 10 20 30 40 50 60 70 O 10 20 30 40 50 60 70

ASMZ(35-47) ASM3 (50-62)

I - I :::m1 -

A

V 9 0.4

0.2 - O I I I I I I I

O 10 20 30 40 50 60 70 O 10 20 30 40 50 60 70 Residue Number Residue Number

Figure 9. Helical hydrophobic moment analysis of the 6-kD domain of full-length oat phyA (FL) and Ala-scanning mutants ASMl (25-33), ASM2 (35-47), and ASM3 (50-62). The helical hydrophobic moment values ( (pH) ) for each amino acid between 1 and 73 were averaged over a window of eight residues. Amino acids located between the dotted lines on the ASMl (25-33), ASM2 (35-47), and ASM3 (50-62) plots reflect regions directly altered in each mutant. Plots were derived from the PepPlot program from the Cenetics Computer Croup.

ronments. In the dark, the phytochrome pool of newly germinated seedlings is predominantly phyA in the inactive Pr form. Immediately upon irradiation, most of these photoreceptors could convert to the biologically active Pfr form. If activity is not attenuated, plants could over- respond to light by eliciting too many photomorphogenic signaling events and / or become desensitized to subsequent changes in light quality or quantity. Although degradation of Pfr is one important route to remove active photoreceptors from the cell (Vierstra, 1994), it may not be rapid enough in a changing light environment, necessitat- ing other methods of posttranslational regulation.

How might the Ser-rich region regulate phyA function? This region could help modulate the number of photoreceptors available to initiate signaling by controlling the turnover rate of Pfr. This mechanism is not likely the predominant means of control, as the deletion NF (A6-12) appears to be degraded at a similar rate as full-length oat phyA (data not shown). However, because this turnover rate reflected that of the total NF (A6-12) pool, it does not rule out the possibility that a subpopulation of active NF (A6-12) photoreceptors are less rapidly degraded. Alterna- tively, the pool size of active photoreceptors could be controlled by the rapid aggregation or compartmentalization of phyA. In fact, phyA is known to aggregate after photoconversion to Pfr in many plant species (Pratt, 1994). Such changes in solubility or localization could provide a way to quickly reduce the pool of free Pfr available for signal transduction. Mutant phyAs that do not aggregate or com- partmentalize as rapidly as full-length oat phyA would thus be more effective.

A third way the Ser-rich domain could regulate phyA function is by controlling the specific activity of individual

phytochrome molecules. This type of regulation could oc- cur either by interactions with inhibitors or by a posttranslational modification that interferes with function. Phos- phorylation of the N-terminal Ser is one attractive possibility. In support, phytochromes are phosphoproteins in vivo, with at least one phosphorylation site mapped to the Ser-rich region adjacent to the N terminus (Quail et al., 1978; Hunt and Pratt, 1980; McMichael, 1991). Phosphory- lation at this site could reduce the rate of Pfr signaling or promote the association of phyA with inhibitors. Therefore, phyA mutants in which the Ser-rich region has been removed or replaced, e.g. NF (A6-12) and NTSA (2-18), may retain activity compared with intact phyA because they are poorly phosphorylated.

It is also possible that severa1 of the above-mentioned mechanisms function concurrently. For instance, light- dependent phosphorylation of the Ser-rich region may en- hance the degradation of active photoreceptors. This type of regulation has been seen for IKB-a, a mammalian regulatory protein that controls the cytoplasmic localization of the transcription factor NF-KB (Alkalay et al., 1995). Signal-induced phosphorylation of IKB-cu triggers its rapid degradation by the ubiquitin-degradation system, thus allowing NF-KB to enter the nucleus. Alternatively, phosphorylated phyA may interact specifically with inhibitors. A paradigm for this mechanism is seen with the interaction of the mammalian photoreceptor rhodopsin with the inhibitor arrestin (Yarfitz and Hurley, 1994). Light activation of rhodopsin increases its affinity for the heterotrimeric G-protein transducin, the next member of a light-signaling cascade. However, light also triggers phosphorylation of the photoreceptor by rhodopsin kinase, which increases the affinity of rhodopsin for arrestin. The rhodopsin / arrestin complex is inactive, thus effectively limiting the ability of the photoreceptor to trigger further light-mediated signaling.

Taken together, the biological and biochemical behavior of the phyA deletion and Ala-scanning mutants further illustrate the importance of the N-terminal6-kD domain of phyA in structure and function. In addition to its role in maintaining the chromoprotein structure needed for initiating signal transduction, the 6-kD domain also appears to be involved in the posttranslational control of photoreceptor activity, potentially by a mechanism involving Ser phosphorylation. A better understanding of the phosphorylation states of the Ser-rich region in vivo will be essential for elucidating this mechanism of phyA regulation.

ACKNOWLEDCMENTS

We thank Drs. Lee Pratt and Marie-Michèle Cordonnier-Pratt for supplying the Oat-22 cell line, Dr. Joel Cherry for creating deletion NF (66-12) and re-engineering the N-terminal coding sequence of the oat PHYA gene, Dr. Howard Hershey (DuPont) for supplying oligonucleotides, and Seth Davis, Paul Bates, Dr. Tanya Falbel, and Dr. Jed Doelling for helpful suggestions concerning the manuscript.

Received April 22, 1997; accepted June 24, 1997. Copyright Clearance Center: 0032-0889 / 97/ 115/ 0693/ 12.



704 Jordan et al. Plant Physiol. Vol. 11 5, 1997

LITERATURE ClTED

Alkalay I, Yaron A, Hatzubai A, Orian A, Ciechanover A, Ben- Neriah Y (1995) Stimulation-dependent I K B ~ phosphorylation marks the NF-KB inhibitor for degradation via the ubiquitin- proteasome pathway. Proc Natl Acad Sci USA 92: 10599-10603

Bevan M (1984) Binary Agrobacterium vectors for plant transformation. Nucleic Acids Res 12: 8711-8721

Boylan M, Douglas N, Quail PH (1994) Dominant negative sup- pression of Arabidopsis photoresponses by mutant phytochrome A sequences identifies spatially discrete regulatory domains in the photoreceptor. Plant Cell 6 449-460

Cherry JR, Hershey HP, Vierstra RD (1991) Characterization of tobacco expressing functional oat phytochrome. Domains responsible for the rapid degradation of Pfr are conserved between monocots and dicots. Plant Physiol 96: 775-785

Cherry JR, Hondred D, Walker JM, Keller JM, Hershey HP, Vierstra RD (1993) Carboxy-terminal deletion analysis of oat phytochrome A reveals the presence of separate domains required for structure and biological activity. Plant Cell5: 565-575

Cherry JR, Hondred D, Walker JM, Vierstra RD (1992) Phyto- chrome requires the 6-kDa N-terminal domain for full biological activity. Proc Natl Acad Sci USA 89: 5039-5042

Cherry JR, Vierstra RD (1994) The use of transgenic plants to examine phytochrome structure/ function. Zn RE Kendrick, GHM Kronenberg, eds, Photomorphogenesis in Plants, Ed 2. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp

Cordonnier M-M, Greppin H, Pratt LH (1984) Characterization by enzyme-linked immunosorbent assay of monoclonal antibodies to Pisum and Avena phytochrome. Plant Physiol 74: 123-127

Eisenberg D, Weiss RM, Terwilliger TC (1982) The helical hydrophobic moment: a measure of the amphiphilicity of a helix. Nature 299: 371-374

Eisenberg D, Weiss RM, Tenvilliger TC (1984) The hydrophobic moment detects periodicity in protein hydrophobicity. Proc Natl Acad Sci USA 81: 140-144

Finer-Moore J, Stroud RM (1984) Amphipathic analysis and possible formation of the ion channel in an acetylcholine receptor. Proc Natl Acad Sci USA 81: 155-159

Furuya M, Song P-S (1994) Assembly and properties of holophy- tochrome. Zn RE Kendrick, GHM Kronenberg, eds, Photomor- phogenesis in Plants, Ed 2. Kluwer Academic Publishers, Dor- drecht, The Netherlands, pp 105-140

Hahn T-R, Song P-S, Quail PH, Vierstra RD (1984) Tetranitrometh- ane oxidation of phytochrome chromophore as a function of spectral form and molecular weight. Plant Physiol 74: 755-758

Hunt RE, Pratt LH (1980) Partia1 characterization of undegraded oat phytochrome. Biochemistry 19: 390-394

Jordan ET, Cherry JR, Walker JM, Vierstra RD (1996) The amino- terminus of phytochrome A contains two distinct functional domains. Plant J 9: 243-257

Keller JM, Shanklin J, Vierstra RD, Hershey HP (1989) Expres- sion of a functional monocotyledonous phytochrome in transgenic tobacco. EMBO J 8: 1005-1012

Kendrick RE, Kronenberg GHM, eds (1994) Photomorphogenesis in Plants, Ed 2. Kluwer Academic Publishers, Dordrecht, The Netherlands

271-297

Mathews S, Lavin M, Sharrock RA (1995) Evolution of the phytochrome gene family and its utility for phylogenetic analyses of angiosperms. Ann Missouri Bot Gard 8 2 296-321

McMichael RW (1991) Phytochrome-mediated signal transduction: the role of protein phosphorylation. PhD thesis. University of California, Davis, CA

Nagatani A, Kay SA, Deak M, Chua N-H, Furuya M (1991) Rice type I phytochrome regulates hypocotyl elongation in transgenic tobacco seedlings. Proc Natl Acad Sci USA 88: 5207-5211

Parker W, Romanowski M, Song P-S (1991) Conformation and its functional implications in phytochrome. In B Thomas, CB John- son, eds, Phytochrome Properties and Biological Actions. Springer-Verlag, New York, pp 85-112

Parks BM, Quail PH (1993) hy8, a new class of Arabidopsis long hypocotyl mutants deficient in functional phytochrome A. Plant Cell 5: 39-48

Pratt LH (1994) Distribution and localization of phytochrome within the plant. In RE Kendrick, GHM Kronenberg, eds, Pho- tomorphogenesis in Plants, Ed 2. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp 163-185

Quail PH, Boylan MT, Parks BM, Short TW, Xu Y, Wagner D (1995) Phytochromes: photosensory perception and signal transduction. Science 268 675-680

Quail PH, Briggs WR, Pratt LH (1978) Zn vivo phosphorylation of phytochrome. In Carnegie Institution of Washington Annual Report. Department of Plant Biology, Stanford University, Stan- ford, CA, pp 342-344

Shanklin J, Jabben M, Vierstra RD (1987) Red light-induced formation of ubiquitin-phytochrome conjugates: identification of possible intermediates of phytochrome degradation. Proc Natl Acad Sci USA 84: 359-363

Smith H (1995) Physiological and ecological function within the phytochrome family. Annu Rev Plant Physiol Plant Mo1 Biol46:

Stockhaus J, Nagatani A, Halfter U, Kay S, Furuya M, Chua N-H (1992) Serine-to-alanine substitutions at the amino-terminal region of phytochrome A result in an increase in biological activity. Genes Dev 6 2364-2372

Vierstra RD (1993) Illuminating phytochrome functions. There is light at the end of the tunnel. Plant Physiol 103: 679-684

Vierstra RD (1994) Phytochrome degradation. Zn RE Kendrick, GHM Kronenberg, eds, Photomorphogenesis in Plants, Ed 2. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp

Vierstra RD, Quail PH (1983) Purification and initial characterization of 124-kilodalton phytochrome from Avena. Biochemistry

Vierstra RD, Quail PH, Hahn TR, Song P-S (1987) Comparison of the protein conformations between different forms (Pr versus Pfr) of native (124 kDa) and degraded (118/114 kDa) phytochromes from Avena. Photochem Photobiol45: 429432

Yarfitz S, Hurley, JB (1994) Transduction mechanisms of verte- brate and invertebrate photoreceptors. J Biol Chem 269: 14329- 14332

Zoller MJ (1991) New molecular biology methods for protein engineering. Curr Opin Struct Biol 1: 605-610

289-315

141-162

2 2 2498-2505



characterization of 6-kilodalton domain of biologi&d activity’ fileplant physiol. (1 997) 11...

Documents