Loss of a pyridoxal-phosphate phosphatase rescues Arabidopsis lacking an endoplasmic reticulum ATP carrier

Abstract

The endoplasmic reticulum (ER)-located ATP/ADP-antiporter (ER-ANT1) occurs specifically in vascular plants. Structurally different transporters mediate energy provision to the ER, but the cellular function of ER-ANT1 is still unknown. Arabidopsis (Arabidopsis thaliana) mutants lacking ER-ANT1 (er-ant1 plants) exhibit a photorespiratory phenotype accompanied by high glycine levels and stunted growth, pointing to an inhibition of glycine decarboxylase (GDC). To reveal whether it is possible to suppress this marked phenotype, we exploited the power of a forward genetic screen. Absence of a so far uncharacterized member of the HaloAcid Dehalogenase (HAD)-like hydrolase family strongly suppressed the dwarf phenotype of er-ant1 plants. Localization studies suggested that the corresponding protein locates to chloroplasts, and activity assays showed that the enzyme dephosphorylates, with high substrate affinity, the B6 vitamer pyridoxal 5'-phosphate (PLP). Additional physiological experiments identified imbalances in vitamin B6 homeostasis in er-ant1 mutants. Our data suggest that impaired chloroplast metabolism, but not decreased GDC activity, causes the er-ant1 mutant dwarf phenotype. We present a hypothesis, setting transport of PLP by ER-ANT1 and chloroplastic PLP dephosphorylation in the cellular context. With the identification of this HAD-type PLP phosphatase, we also provide insight into B6 vitamer homeostasis.

Figure 1

Suppression of the er-ant1 phenotype by ser-ant1 mutations. A, Suppressor mutants selected for further characterization and mapping. Photographs show representative 28-d-old third-generation (M₃) plants for each selected mutant line, plants were grown on soil at ambient air. Images were digitally extracted for comparison. B, Total chlorophyll content in leaves of er-ant1 and M₃  ser-ant1 plants relative to the wt. Mean values of three individual replicates ± se. Asterisks indicate the significance level between wt and mutant plants according to a Student’s t test (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 2

Maximum quantum yield of PSII (F_v/F_m) of wt, er-ant1, and two selected M3 ser-ant1 plants. A, F_v/F_m at 2,000 parts per million (ppm) CO₂. B, F_v/F_m at ambient air. Data represent mean values of seven individual replicates, ±se. Asterisks indicate the significance level between wt and mutant plants according to a Student’s t test (**P < 0.01, ***P < 0.001).

Figure 3

Glycine content of leaves from er-ant1 and ser-ant1 mutants relative to the wt level. Plants were cultivated at ambient CO₂ conditions. The glycine content of the wt was set to 1. Data represent mean values of five individual replicates ± se. Asterisks indicate the significance between the glycine levels of wt and mutant plants according to a Student’s t test (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 4

Absence of the At2g33255 transcript suppresses the growth defect of er-ant1. A, Phenotypic comparison of 4 weeks old plants grown at ambient CO₂. Images were digitally extracted for comparison. B, At2g33255 gene expression analyzed by RT-qPCR. RNA for cDNA synthesis was extracted from leaves of 3 weeks old plants. Data were normalized to the SAND (At2g28390) housekeeping gene. Data represent mean values of three biological replicates ± se.

Figure 5

ClustalW alignment of the amino acid sequences of the PLPP from S. cerevisiae (PLPP_Sc; YOR131C) and the At2g33255 gene product. Residues identical among the two sequences or with similar properties are indicated by black shading. Residues not identical are highlighted by different shading (gray/white). Red boxes mark conserved HADSF motifs I–IV (I: DxD; II, T/S; III: K/R; IV: E/DD, GDxxxD, or GDxxxxD). The putative mitochondrial targeting sequence (according to TargetP) is shaded in red. The predicted chloroplast targeting sequence (according to ChloroP) comprises also the green shaded residues. Dashes represent introduced gaps for alignment improvement. Numbers at the right indicate amino acid positions. The amino acid sequence alignment also demonstrates that the At2g33255 encoded protein is N-terminally extended by 20 residues when compared with the cytosolically located PLPP YOR131C from S. cerevisiae. Such extensions often act as targeting sequences signifying the import of proteins into mitochondria or chloroplasts. Notably, the At2g33255 encoded protein does not share substantial sequence similarity (25%) to a recently identified chloroplastic PLP phosphatase from N. tabacum, named NtPLPP1 (ShuoHao et al., 2019; Supplemental Figure S5).

Figure 6

Localization of the HAD-type hydrolase in N. benthamiana and Arabidopsis protoplasts. Pictures were taken with a confocal laser-scanning microscope 16 h after transient transformation. Scale bars = 10 μm. Chl, chlorophyll fluorescence.

Figure 7

Enzyme assays verified the PLPP function of the At2g33255 encoded protein. A, Time course of phosphate release from PLP (0.2 mM). B, Phosphate release is dependent on the active enzyme (+E), presence of PLP (0.2 mM), and Mg²⁺. Each assay time was 30 min. C, Michaelis–Menten curve indicating substrate saturation of enzyme activity, the inset shows a corresponding Lineweaver–Burk analysis indicating an apparent K_m for PLP of 30 µM and a V_max of 0.123 µmol mg protein⁻¹ min⁻¹. Each assay time was 30 min. Shown are mean values of four replicates ± se.

Figure 8

Effect of PN feeding on growth of er-ant1 and wt plants. Twenty plants were cultivated on agar plates (0.5 MS including GA5 vitamins, 0.8% agar) at ambient CO₂ conditions (10-h light/14-h dark) for 25 d. A, wts and er-ant1 plants on agar plates without additional PN (control) and with different concentrations of additional PN. B, FW of er-ant1 (light gray) and corresponding wt plants (dark gray) grown on agar plates containing different concentrations of PN. The FW is given as a percentage of the respective control FW. Shown are mean values of at least five individual plates ± se. Asterisks indicate the significance level between wt and mutant plants according to a Student’s t test (***P < 0.001).

Figure 9

Analysis of the B₆ vitamer contents in the different plant lines. A, Analysis of the individual B₆ vitamers PMP, PLP, PM, PN, and PL profile of the lines as indicated. B, Total vitamer contents (sum of PMP, PLP, PM, PN, and PL). Plants were grown at high CO₂ (2,000 ppm) for 33 d to avoid pleiotropic effects on B₆ vitamers in dwarf type er-ant1 plants. Data represent the mean of four replicates per line that was used for two extractions each. SEs are given. Statistical relevance was determined by a Student’s t test using the wt of the respective growth condition as a reference: ^*P ≤ 0.05, ^**P ≤ 0.005, and ^***P ≤ 0.0005.