Obligate heterodimerization of Arabidopsis phytochromes C and E and interaction with the PIF3 basic helix-loop-helix transcription factor

Abstract

Phytochromes are dimeric chromoproteins that regulate plant responses to red (R) and far-red (FR) light. The Arabidopsis thaliana genome encodes five phytochrome apoproteins: type I phyA mediates responses to FR, and type II phyB-phyE mediate shade avoidance and classical R/FR-reversible responses. In this study, we describe the complete in vivo complement of homodimeric and heterodimeric type II phytochromes. Unexpectedly, phyC and phyE do not homodimerize and are present in seedlings only as heterodimers with phyB and phyD. Roles in light regulation of hypocotyl length, leaf area, and flowering time are demonstrated for heterodimeric phytochromes containing phyC or phyE. Heterodimers of phyC and chromophoreless phyB are inactive, indicating that phyC subunits require spectrally intact dimer partners to be active themselves. Consistent with the obligate heterodimerization of phyC and phyE, phyC is made unstable by removal of its phyB binding partner, and overexpression of phyE results in accumulation of phyE monomers. Following a pulse of red light, phyA, phyB, phyC, and phyD interact in vivo with the PHYTOCHROME INTERACTING FACTOR3 basic helix-loop-helix transcription factor, and this interaction is FR reversible. Therefore, most or all of the type I and type II phytochromes, including heterodimeric forms, appear to function through PIF-mediated pathways. These findings link an unanticipated diversity of plant R/FR photoreceptor structures to established phytochrome signaling mechanisms.

Figure 1.

Yeast Two-hybrid Analysis of Binding Interactions among the Five Arabidopsis Phytochrome C Termini. (A) Illustration of the protein domains common to plant phytochromes and the regions of the phytochrome apoproteins used in yeast two-hybrid experiments. The chromophore attachment site within the GAF domain is indicated. Phytochrome domains are PAS (Per/Arnt/Sim), PLD (PAS-like), GAF (cGMP phosphodiesterase/adenyl cyclase/Fhl1), PHY (phytochrome-specific), and HKRD (histidine kinase-related domain). The indicated N-terminal or C-terminal phytochrome sequences were fused to the GAL4 AD and DNA BD. (B) Liquid β-galactosidase assay activities for all pairwise combinations of the C-terminal sequences of phyA through phyE. The amino acid sequence coordinates for the individual C-550 and C-200 regions are given in Methods. Error bars represent the se from three replicate experiments. Note the scale for the phyC panels is at background levels. (C) Immunoblot analysis of yeast strains expressing the phyC C-terminal fusions. Proteins were extracted from yeast strains expressing the phyC-C200 and phyC-C550 fusions. Extracts were separated by SDS-PAGE, blotted, and probed with anti-myc, anti-GAD, and anti-phyC antibodies.

Figure 2.

Overexpressed phyC Forms Heterodimers with phyB and phyD but Does Not Homodimerize. (A) Structures of the three epitope-tagged phyC overexpression (COE) transgenes. (B) Increased first primary leaf areas of epitope-tagged COE lines. Seedlings were grown for 10 d under continuous white light. Error bars represent the se of 15 to 20 leaves. (C) Immunoblot analysis of protein extracts prior to immunoprecipitation (extracts) and of proteins precipitated by the anti-myc antibody (immuno-ppt) from 7-d-old dark-grown seedlings expressing the m6-COE or COE-m6 transgenes. A phyD degradation band present in the extracts is marked with an asterisk. (D) Immunoblot analysis of 7-d-old light-grown seedling extracts and anti-myc immunoprecipitates from doubly transgenic lines expressing native phyC, myc₆-tagged phyC, and his₆-tagged phyC.

Figure 3.

Biological Activity and Heterodimerization of Epitope-Tagged phyC Expressed from Its Native Promoter. (A) Structure of the m6-PHYC transgene and complementation of the phyC mutant hypocotyl elongation response under continuous R light. A 3.2-kb region upstream of the start codon of the PHYC gene was fused to the myc₆-tagged phyC coding sequence and transformed into the wild-type and phyC backgrounds. WT(m6-PHYC) and phyC(m6-PHYC) seedlings were incubated for 3 h white light/21 h dark and then grown for 3 d under R (30 μmol m⁻² s⁻¹) at 21°C. Error bars represent the se of 20 to 25 seedlings. (B) Immunoblot analysis of seedling protein extracts and anti-myc antibody immunoprecipitates from 7-d-old dark-grown wild-type and phyC lines expressing the m6-PHYC gene. WT(m6-PHYC) extracts contain both native phyC and the higher molecular weight myc₆-tagged phyC indicated by arrows. The phyC(m6-PHYC) extracts contain myc₆-tagged phyC and a degradation band marked with an asterisk.

Figure 4.

Biological Activity and Heterodimerization of Epitope-Tagged phyE. (A) Structure of the PHYE-m6 transgene and complementation of the phyE mutant early flowering response. The phyE-myc₆ coding sequence was fused to a 1.8-kb PHYE promoter region and transformed into the Ler wild-type, phyE, and phyBE genetic backgrounds. Plants were grown under short days (8 h light/16 h dark) at 21°C. Error bars represent the se of 12 plants. (B) Immunoblot analysis of protein extracts of 7-d-old dark-grown seedlings of the lines from (A) prior to immunoprecipitation and of proteins precipitated from those extracts by the anti-myc antibody. Lines expressing the PHYE-m6 transgene in the wild-type background contain both native phyE and the higher molecular weight myc₆-tagged phyE. (C) Phytochrome dimer contents of wild-type and phy mutant lines. The phytochrome complements that have been demonstrated by in vivo co-IP in the wild-type and phyB lines, and those projected for monogenic and selected multiply phy mutant lines are summarized. The symbol (±) indicates a very low amount of that dimer form.

Figure 5.

Quantitative Immunoprecipitation Assay for phyB-Containing Dimers. (A) Immunoblot analysis of anti-myc IP fractions from the phyB(myc-PHYB) line and standard curves of purified phy apoproteins. Samples of tissue extracts containing 833 μg (sample 1) or 1000 μg (sample 2) of total protein from 7-d-old dark-grown phyB(myc-PHYB) seedlings were immunoprecipitated with the anti-myc antibody. The immunoprecipitates were separated by SDS-PAGE along with standard curves of purified Escherichia coli–expressed phyB–phyE apoproteins. Gels were blotted and probed with the indicated anti-phy antibodies. (B) Quantification of phytochrome levels in immunoprecipitates. Chemiluminescence signals on autoradiography film images of the immunoblots from (A) were scanned and analyzed by densitometry. Arrows labeled 1 and 2 in each panel indicate the values for tissue extract samples 1 and 2 (A).

Figure 6.

Reduced Overall Levels of phyC but Increased phyC/D Heterodimerization in the phyB Mutant. (A) Immunoblots of the phyA-phyE protein levels in wild-type and phy mutant lines. Seedlings were grown for 7 d in the dark, proteins were extracted and separated by SDS-PAGE, and immunoblots were prepared and probed with the indicated anti-phy antibodies. (B) Immunoblots of extracts and myc antibody IPs of phyB mutant lines expressing the m6-COE transgene. Seedlings of the indicated lines were grown for 7 d in the dark, extracts were prepared and immunoprecipitated with anti-myc antibody, and immunoblotting was performed with the anti-phy antibodies. The asterisk indicates a presumed phyC degradation product that does not comigrate with native phyC on gels with higher separation.

Figure 7.

Lack of Dimerization of phyE in the Absence of phyB and phyD and Accumulation of Overexpressed phyE as a Monomer. (A) Immunoprecipitation of phyE-m6 and myc₃-phyE from seedling extracts of WT(PHYE-m6) and phyBD(m3-PHYE) lines. Seven-day-old dark-grown seedlings of the indicated lines were extracted. Extracts were immunoprecipitated with the myc antibody, and IP samples were fractionated on SDS-PAGE, blotted, and probed with the anti-phy antibodies. Very long exposures of the blots in this experiment did not reveal evidence for dimerization of m3-phyE protein with any potential partner phytochrome in the phyB phyD mutant background. (B) Native gel analysis of nondenatured extracts of seedlings expressing myc-tagged phyA–phyE. 35S promoter–driven overexpression constructs for m6-phyA, m6-phyC, m6-phyD, and m6-phyE were in the No-0 wild-type genetic background. The m6-phyB protein was expressed from the PHYB promoter in the No-0 phyB(myc-PHYB) line, and the phyE-m6 protein was expressed from the PHYE promoter in the Ler phyE(PHYE-m6) line. Samples from two independent lines of the epitope-tagged phyE-expressing genotypes were analyzed. Nondenatured extracts were prepared from 7-d-old dark-grown seedlings and fractionated on 4 to 20% native PAGE gels. Blots of the gels were probed with the anti-myc antibody (d, dimer; m, monomer). A small amount of monomeric phyA is seen in the m6-AOE extract. The bottom panel shows a myc antibody-probed immunoblot of an SDS-PAGE gel of the extracts used in the native gel analysis.

Figure 8.

phyC Is Stabilized by the Presence of Chromophoreless phyB but Is Inactive. (A) Immunoblot analysis of phytochrome levels in phyB(PHYB^C357S) lines. Extracts of 7-d-old dark-grown seedlings were fractionated by SDS-PAGE, blotted, and probed with anti-phy antibodies. (B) Fluence-response curve for hypocotyl length in 6-d-old phyB(PHYB^C357S) lines grown under continuous R. Mean and se values are representative of at least 20 seedlings for each light treatment.

Figure 9.

Light Responses of Mutants Lacking Selected Combinations of Phytochrome Dimers. Flowering times under an 8-h-light/16-h-dark photoperiod and hypocotyl lengths of 5-d-old seedlings grown under continuous R (30 μmol m⁻² s⁻¹) were measured for the indicated lines. Error bars represent the se for 12 plants and 20 to 25 seedlings, respectively.

Figure 10.

R/FR-Reversible Co-IP of Phytochromes with PIF3-myc₆ from Seedling Extracts. (A) Rapid binding of phyA, phyB, phyC, and phyD to PIF3-m6 following a pulse of R light. Five-day-old dark-grown wild-type and 35S:PIF3-myc₆ seedlings were kept in the dark (D) or exposed to a 30-s pulse of R (30 μmol m⁻² s⁻¹) and returned to the dark. The R-pulsed 35S:PIF3-m6 seedlings were harvested over the indicated 10-min time course, and one R-pulsed wild-type sample was harvested at 3 min following return to the dark. Protein extracts were prepared and IP performed with the anti-myc antibody. Samples of the immunoprecipitates were fractionated by SDS-PAGE, blotted, and probed with the anti-myc antibody to detect PIF3-m6 or with the anti-phy antibodies. Arrows to the right of the blot panels indicate the positions of the respective antigens. The two arrows to the right of the PIF3-m6 panels indicate the positions of presumed phosphorylated (top arrow) and unphosphorylated (bottom arrow) PIF3-m6 (Al-Sady et al., 2006). (B) FR reversibility of R-induced phy-PIF3 binding. Five-day-old dark-grown 35S:PIF3-myc₆ seedlings were kept in the dark, exposed to 15 s of R (30 μmol m⁻² s⁻¹) or 15 s R followed by 15 s FR (39 μmol m⁻² s⁻¹), returned to the dark, and harvested 3 min later. Proteins were extracted and immunoprecipitated with anti-myc antibody. IP samples were fractionated by SDS-PAGE, and immunoblots were probed with the anti-myc or anti-phy antibodies. The IP blots on the right include parallel blots of the dark extract (Ext.) that were fractionated on the same PAGE gels as the IP samples to provide size standards for the phy proteins.