Bottom-up Assembly of the Phytochrome Network

Abstract

Plants have developed sophisticated systems to monitor and rapidly acclimate to environmental fluctuations. Light is an essential source of environmental information throughout the plant's life cycle. The model plant Arabidopsis thaliana possesses five phytochromes (phyA-phyE) with important roles in germination, seedling establishment, shade avoidance, and flowering. However, our understanding of the phytochrome signaling network is incomplete, and little is known about the individual roles of phytochromes and how they function cooperatively to mediate light responses. Here, we used a bottom-up approach to study the phytochrome network. We added each of the five phytochromes to a phytochrome-less background to study their individual roles and then added the phytochromes by pairs to study their interactions. By analyzing the 16 resulting genotypes, we revealed unique roles for each phytochrome and identified novel phytochrome interactions that regulate germination and the onset of flowering. Furthermore, we found that ambient temperature has both phytochrome-dependent and -independent effects, suggesting that multiple pathways integrate temperature and light signaling. Surprisingly, none of the phytochromes alone conferred a photoperiodic response. Although phyE and phyB were the strongest repressors of flowering, both phyB and phyC were needed to confer a flowering response to photoperiod. Thus, a specific combination of phytochromes is required to detect changes in photoperiod, whereas single phytochromes are sufficient to respond to light quality, indicating how phytochromes signal different light cues.

Fig 1. The effect of single phytochrome photoreceptors on the regulation of germination and their synergistic interactions.

Germination of the genotypes indicated on the abscissas at 23°C under different light regimes (for clarity, phytochromes present in each line are indicated above, in capital letters, and genotypes below, in italics). Light regimes: continuous FR (60 μmol m^-2 s^-1), continuous red (R) (30 μmol m^-2 s^-1), a 15-min R pulse, a 15-min FR pulse, a 15-min R pulse followed immediately by a 15-min FR pulse (R/FR), a 15-min FR pulse followed immediately by a 15-min R pulse (FR/R), and darkness. Data are averages ± SE of 16 independent plates with 20 seeds each and 4 independent seed pools (collected from independently grown plants).

Fig 2. A network of phytochrome interactions is necessary to regulate flowering in response to photoperiod and temperature.

Plants bearing the indicated phytochromes were grown under long days (LD, 16 h light/8 h dark) (A) or LD and short days (SD, 8 h light/16 h dark) (B), at temperatures ranging from 18 to 24°C. LD data in (B) are the same as in (A) and included for the purpose of direct comparison. The total leaf number at the time of flowering was recorded. Data points represent the mean ±SE of at least 18 plants from two independent experiments for each genotype and condition.

Fig 3. phyB is sufficient for a full hypocotyl response to R.

Plants bearing the indicated phytochromes were stratified for 3 days at 4°C in the dark in a solution of 100 μM GA₄₊₇ and then plated on MS salts agar plates and incubated at 23°C either under continuous red (R) light (20 μmol m^-2 s^-1) or kept in darkness (control) for 5 days. The values obtained under R are given relative to the corresponding dark control in each independent experiment. Data are averages ± SE of four independent plates.

Fig 4. The individual roles of phyB, phyD and phyE depend on protein sequence rather than expression level or pattern.

Germination rates (A) and flowering time (B) of independent transgenic lines harboring each phytochrome under the 35S promoter in a background devoid of other phytochromes. Plants harboring the indicated phytochromes were grown under LD conditions at 18°C (B) or under white light at 23°C (A) and total leaf number and germination rates were determined as in Figs 1 and 2. WT, quadruple and quintuple phytochrome mutants were compared to transgenic lines bearing HA tagged versions of phyB, phyD or phyE in the quintuple phytochrome mutant background and the empty vector control lines. Data points represent the mean ±SE of 4 independent transgenic lines for the vector control and 6, 8 and 12 independent lines for the constructs bearing, 35S:PHYB, 35S:PHYD and 35S:PHYE respectively. Quantification of protein levels and the germination and flowering responses of individual lines are shown in (S8 Fig and S2 Table).

Fig 5. phyB requires phyC to regulate the photoperiodic response.

Flowering time of transgenic lines bearing phyC under the 35S promoter in a background containing only phyB or phyE. Plants harboring only the indicated phytochromes were grown under SD conditions at 23°C and flowering time was determined as in Fig 2. The numbers inside each bar represent the number of independent T1 lines used.

Fig 6. Light conditions differentially affect the intracellular localization patterns of phyB/phyB homodimers and phyB/phyC heterodimers.

(A) Nuclear/cytoplasmic partitioning of phyB homodimers differs from that of phyB/phyC heterodimers in Nicotiana benthamiana transient assays. Each pair or constructs bearing phyC-cEYFP and phyB-nEYFP (BC) or phyB-cEYFP and phyB-nEYFP (BC) were agroinfiltrated in LD-grown Nicotiana benthamiana leaves together with ECFP-NLS as a nuclear marker (Blue). The following day, plants were treated with a FR pulse and then grown for two more days in darkness. Leaves were collected under a green safe light and then fixed with formaldehyde in darkness before confocal microscopy examination. (B, C) PhyB/phyC heterodimers are unresponsive to light quality. Nicotiana benthamiana leaves from plants grown in LD were agroinfiltrated as in (A), grown for another 12 h in WL, dark adapted for 36 h and then received a R treatment for 3 h. After the R treatment, a set of leaves were collected (R control) and another set of plants received either a pulse of 15 min FR followed by dark or continuous FR, to revert phytochrome to the Pr form. Leaves were collected after either 6 or 24 h after the ending of the R treatment. Leaves were fixed and examined by confocal microscopy. Black arrows in the scheme (C) indicate treatments, while white arrows indicate harvesting points. For quantitative data shown in (C), randomly selected individual cells were used to quantify the fluorescence intensity in three randomly selected areas of the nucleus and the cytoplasm. Nuclear and cytoplasmic intensities were averaged for each cell and then averaged among independent leaves. Data are means ±SE of 6 independent leaves. (D, E) Coexpression of phyB and phyC changes their localization patterns. phyB-Cerulean was coexpressed with either GFP alone as a control, phyB-GFP or phyC-GFP (left set of panels in D) and phyC-Cerulean was coexpressed with either GFP alone as a control, phyB-GFP or phyC-GFP (right set of panels in D). Nicotiana benthamiana leaves from plants grown in LD were agroinfiltrated, grown for another 12 h in WL, treated with a pulse of 15 min FR to revert phytochrome to the Pr form and dark adapted for 36 h before confocal microscopy. (E) Images were quantitated as in (C) and data are means ±SE of 6 independent leaves. (F-G) phyC promotes the localization of phyB to large nuclear bodies in Arabidopsis during the night period. Transgenic lines bearing phyB-GFP in the quintuple phytochrome mutant background were crossed to lines either bearing only phyC (phyA phyB phyD phyE quadruple mutants) or the quintuple phytochrome mutant as a control. The F1 lines were grown in SD conditions and used to observe the effect of phyC on the localization of phyB-GFP at two time points, 1 h before lights-on (End of Day) and 1 h before lights-off (End of Night). (F) The phytochromes indicated above panels are the only phytochromes present in these F1 lines. Chloroplasts are observed in red, whereas phyB-GFP is observed as green dots within the nuclei (the three left panels) or diffuse green nuclei (the right panel) of Arabidopsis hypocotyl cells. (G) Quantification of large nuclear bodies from confocal images. Data points represent the mean ±SE of 12 nuclei, 4 nuclei from 3 seedlings for each genotype and condition.

Fig 7. Model of the phytochrome network.

Summary of the roles of phytochromes and the interactions between them during germination (A) and flowering (B). Positive interactions are depicted by arrows and negative interactions by lines. The thickness of each line indicates the strength of the action.