Mutant screen distinguishes between residues necessary for light-signal perception and signal transfer by phytochrome B

Abstract

The phytochromes (phyA to phyE) are a major plant photoreceptor family that regulate a diversity of developmental processes in response to light. The N-terminal 651-amino acid domain of phyB (N651), which binds an open tetrapyrrole chromophore, acts to perceive and transduce regulatory light signals in the cell nucleus. The N651 domain comprises several subdomains: the N-terminal extension, the Per/Arnt/Sim (PAS)-like subdomain (PLD), the cGMP phosphodiesterase/adenyl cyclase/FhlA (GAF) subdomain, and the phytochrome (PHY) subdomain. To define functional roles for these subdomains, we mutagenized an Arabidopsis thaliana line expressing N651 fused in tandem to green fluorescent protein, beta-glucuronidase, and a nuclear localization signal. A large-scale screen for long hypocotyl mutants identified 14 novel intragenic missense mutations in the N651 moiety. These new mutations, along with eight previously identified mutations, were distributed throughout N651, indicating that each subdomain has an important function. In vitro analysis of the spectral properties of these mutants enabled them to be classified into two principal classes: light-signal perception mutants (those with defective spectral activity), and signaling mutants (those normal in light perception but defective in intracellular signal transfer). Most spectral mutants were found in the GAF and PHY subdomains. On the other hand, the signaling mutants tend to be located in the N-terminal extension and PLD. These observations indicate that the N-terminal extension and PLD are mainly involved in signal transfer, but that the C-terminal GAF and PHY subdomains are responsible for light perception. Among the signaling mutants, R110Q, G111D, G112D, and R325K were particularly interesting. Alignment with the recently described three-dimensional structure of the PAS-GAF domain of a bacterial phytochrome suggests that these four mutations reside in the vicinity of the phytochrome light-sensing knot.

Figure 1. Hypocotyl Phenotypes of N651-GUS-NLS Mutants Carrying Missense Mutations.

(A) Locations of missense mutations found in the present (plain) and previous (italic) studies. For details, see Table 1. PLD, GAF and PHY were delimited as amino acid residues 103–219, 252–433 and 444–623, respectively, according to a sequence-based domain database, Pfam version 20.0 (http://www.sanger.ac.uk/Software/Pfam). The N-terminal extension was defined as amino acid residues 1–102. The closed triangle represents the chromophore binding site. (B) Hypocotyl lengths of mutants grown under different light conditions. For the hypocotyl measurement, plants were grown under weak cR (0.05 µmol m⁻² sec⁻¹) (shaded, upper panel), strong cR (5.5 µmol m⁻² sec⁻¹) (open, upper panel), cFR (10 µmol m⁻² sec⁻¹) (shaded, lower panel) or in darkness (closed, lower panel). The mean±SE (n = 25) is shown. (C) Immunoblot detection of the N651G-GUS-NLS proteins. For detection, 50 µg of total protein was loaded in each lane, blotted onto nitrocellulose membrane after SDS-PAGE, and probed with an anti-GFP monoclonal antibody (SIGMA) (upper panel). To confirm protein loading amount, the same samples were subjected to Coomassie Brilliant Blue (CBB) staining (lower panel).

Figure 2. Chromophore Ligation to Mutant N651 Fragments.

Results of zinc-blot (upper panels) and immunoblot (lower panels) analyses are shown. Crude extracts from Escherichia coli expressing the mutant N651 fragments were incubated with 5 µM PCB and separated by SDS-PAGE ,. Immunoblot detection was performed using antiserum against chitin binding domain (New England Biolabs).

Figure 3. Pr-Pfr Difference Spectra of Mutant N651 Fragments.

The mutated holoproteins were prepared as for Figure 2 and subjected to spectrophotometry. Blue and red lines indicate 650 and 700 nm, respectively.

Figure 4. Dark Reversion Rates in the Mutant N651 Fragments.

(A) In vitro dark reversion rates measured by spectrophotometry. Holoproteins were converted to Pfr by irradiation with saturating R (89 µmol m⁻² sec⁻¹) for 10 min, and then dark reversion from Pfr to Pr was monitored with a spectrophotometer. The level of Pfr at the beginning of the measurement was set to 100%. The incubation temperature was 22°C. Each value represents the mean of three independent measurements. (B) Hypocotyl responses in mutant lines to intermittent pR at 4 hr intervals. Plants were grown under intermittent pR (55.2 µmol m⁻² sec⁻¹ for 5 min) (open) or under cR (2.3 µmol m⁻² sec⁻¹) (shaded) for 5 days. Data are the mean±SE (n = 25).

Figure 5. Alignment of Arabidopsis and Bacterial Phytochrome Sequences.

The PLD-GAF region of phytochrome sequences are aligned. PHYB, Arabidopsis thaliana phyB; PHYA, Arabidopsis thaliana phyA; CPH1, Synechosystis PCC6803 Cph1; DrCBD, Deinococcus radiodurans BphP. Arrows and short bars on top of the sequences represent β-strands and α-helices, respectively. Blue and green lines at the bottom indicate PLD and GAF, respectively. Domains were delimited as for Figure 1A. A broken red line indicates the loop extended from GAF, which forms the light sensing knot together with the N-terminal end of PLD (broken blue). In the knot structure, the β1′, β2′ and β3′ strands (red arrows) form a small β-sheet. Amino acid residues at which mutations were found are indicated in red. The cysteine residues that bind the chromophore are indicated by green background. Amino acid residues that are in direct contact with the chromophore in DrCBD are indicated in green. The three dimensional structure is based on .

Figure 6. Regulatory Activity of Mutant Forms of PBG.

(A) Hypocotyl lengths (upper panel) in transgenic Arabidopsis grown under cR (shaded) or darkness (closed) and immunoblot detection of PBG proteins (lower). For the hypocotyl measurements, plants were grown under cR (5.5 µmol m⁻² sec⁻¹) or in darkness for 5 days. Data are the mean±SE (n = 25). For the immunoblot detection of PBG proteins, 50 µg of total protein was loaded in each lane and PBG proteins were detected with a mouse monoclonal anti-phyB antibody (middle panel). To comfirm protein loading amount, the same samples were subjected to Coomassie Brilliant Blue (CBB) staining (lower panel). (B) Real-time PCR using RNA from 4-day-old seedlings grown in the dark (D0), kept in the dark for an additional 12 hr (D12) or exposed to cR for 12 hr (R12). Data are shown for ELF4 (upper), SAUR-LIKE (middle) and AMYLASE (lower). Cycle threshold values were used to calculate fold-induction with Ler dark values set to 1. Values from three biological replicates are plotted with SE.

Figure 7. Subcellular Localization of Mutant Forms of PBG.

Confocal microscopic observation of GFP fluorescence in transgenic Arabidopsis seedlings. Hypocotyl epidermal cells of 3-day-old seedlings were observed. Dark-grown seedlings (upper), those treated with cW for 2 min (middle) and those treated with cR for 24 hr (lower) are shown. The bar indicates 10 µm.