Reduced Arogenate Dehydratase Expression: Ramifications for Photosynthesis and Metabolism

Abstract

Arogenate dehydratase (ADT) catalyzes the final step of phenylalanine (Phe) biosynthesis. Previous work showed that ADT-deficient Arabidopsis (Arabidopsis thaliana) mutants had significantly reduced lignin contents, with stronger reductions in lines that had deficiencies in more ADT isoforms. Here, by analyzing Arabidopsis ADT mutants using our phenomics facility and ultra-performance liquid chromatography-mass spectrometry-based metabolomics, we describe the effects of the modulation of ADT on photosynthetic parameters and secondary metabolism. Our data indicate that a reduced carbon flux into Phe biosynthesis in ADT mutants impairs the consumption of photosynthetically produced ATP, leading to an increased ATP/ADP ratio, the overaccumulation of transitory starch, and lower electron transport rates. The effect on electron transport rates is caused by an increase in proton motive force across the thylakoid membrane that down-regulates photosystem II activity by the high-energy quenching mechanism. Furthermore, quantitation of secondary metabolites in ADT mutants revealed reduced flavonoid, phenylpropanoid, lignan, and glucosinolate contents, including glucosinolates that are not derived from aromatic amino acids, and significantly increased contents of putative galactolipids and apocarotenoids. Additionally, we used real-time atmospheric monitoring mass spectrometry to compare respiration and carbon fixation rates between the wild type and adt3/4/5/6, our most extreme ADT knockout mutant, which revealed no significant difference in both night- and day-adapted plants. Overall, these data reveal the profound effects of altered ADT activity and Phe metabolism on secondary metabolites and photosynthesis with implications for plant improvement.

Figure 1.

Layout of phenomics experiments and growth characteristics of Arabidopsis wild-type (WT) plants. A, Example of a block of nine plants representing the nine genotypes. It was assumed that the microclimate in this block is uniform. B, Examples of how individual blocks (numbered) are arranged in the phenomics facility. The numbers in B indicate the order of functional measurements. C, For this study, a total of four sets were averaged, leading to 40 plants per genotype. D, Example of false-color chlorophyll fluorescence images (F_o parameter) of the plants shown in A. E, Growth curve indicating plant size (for the wild type) as a function of DAS expressed as total leaf area deduced from chlorophyll images (see D); the latter was first measured when the plants were moved to the phenomics facility (i.e. 20 DAS). Data for the four examined data sets are depicted. F, The data from E were averaged and presented as means ± se. Photographs at right are examples of the different growth phases (indicated by blue arrows). Bar = 2 cm.

Figure 2.

Comparison of growth characteristics and photosynthetic parameters in wild-type (WT) and adt3/4/5/6 plants. A, Growth curve indicating plant size as a function of DAS expressed as total leaf area deduced from chlorophyll images, as explained in Figure 1. Data are means ± se, with significant differences indicated by red stars (Student’s t test, P < 0.05). B, Linear electron transport expressed as the ΦII parameter. Data are means ± se calculated from the four measuring sets. C and D, qE (C) and 1-qL (D). Data are derived as in A. For details, see text. In B to D, histograms at right depict means ± se averaged over the entire measuring period, with significant differences indicated by red stars (Student’s t test; one star, P < 0.05; three stars, P < 0.001).

Figure 3.

Analysis of metabolites in wild-type (WT) and adt3/4/5/6 plants. A, Starch contents. B, Flavonoid contents. C, Phenylpropanoid contents. D, Lignan contents. E, Phenolic glucoside contents. All values are relative to the internal standard naringenin. Data are means ± se, collected for DA and NA plants. nd, Not detected.

Figure 4.

Photosynthetic parameters of NA wild-type (WT) and adt3/4/5/6 plants in the laboratory. Measurements were performed using the DIRK approach (see “Materials and Methods”). A, ΦII. B, NPQ. C, Linear correlation between ΦII and NPQ. Data are means ± se, with significant differences indicated by red stars (Student’s t test, P < 0.05).

Figure 5.

Energization of thylakoid membranes measured by ECS. A, Examples of ECS kinetics for different light intensities (indicated by color-associated numbers in μmol quanta m^–2 s^–1). ECS relaxations from NA plants illuminated with the various light intensities were induced by a dark pulse (black bar). The arrow indicates the amplitude of ECS signal (ECS total). B, Dependency of total ECS signal on light intensity. Data are means ± se derived from seven measurements as shown in A. WT, Wild type.

Figure 6.

Metabolite contents in DA and NA wild-type (WT) and ADT mutant plants. A, Flavonoids. B, Phenylpropanoids. C, Lignan glucosides. D, Phenolic glucosides. E, Glucosinolates. F, Gluconasturtiin. G, Putative galactolipids. All values are relative to the internal standard naringenin. *, 0.01 < P ≤ 0.05; **, P ≤ 0.01.

Figure 7.

Relationship between starch content and ΦII in wild-type (WT) and ADT mutant plants. A, Starch content (red outlined bars, right axis) and ΦII (gray bars) of the various genotypes. The red horizontal line indicates wild-type values for comparison. B, Linear correlation between starch content and ΦII. All data are means ± se.

Figure 8.

CO₂ levels in wild-type (WT) and adt3/4/5/6 plants following the depletion (NA) or accumulation (DA) of starch reserves. Measurements were made after a steady-state adjustment period of 50 min in the dark (for DA plants) and 50 min in the light (for NA plants). Data are means ± se. No significant differences exist between wild-type and adt3/4/5/6 values (Student’s t test, P = 0.194 and 0.993 for the NA and DA data sets, respectively).

Figure 9.

Metabolite profiling in the wild type (WT) and ADT mutants. A, Metabolite heat map generated using MeV version 4.8 from data normalized using naringenin as an internal standard and transformed to have zero mean and unit variance. The heat map displays major trends observed in the levels of annotated metabolites, with clustering based on feature intensity patterns across samples. B, PLSDA analysis of wild-type and adt3/4/5/6 DA and NA plants based on annotated metabolites, with clear separation of sample groups and time of harvest. The loadings show those metabolic features that are more characteristic of the sample groups.