Allocation of Heme Is Differentially Regulated by Ferrochelatase Isoforms in Arabidopsis Cells

Abstract

Heme is involved in various biological processes as a cofactor of hemoproteins located in various organelles. In plant cells, heme is synthesized by two isoforms of plastid-localized ferrochelatase, FC1 and FC2. In this study, by characterizing Arabidopsis T-DNA insertional mutants, we showed that the allocation of heme is differentially regulated by ferrochelatase isoforms in plant cells. Analyses of weak (fc1-1) and null (fc1-2) mutants suggest that FC1-producing heme is required for initial growth of seedling development. In contrast, weak (fc2-1) and null (fc2-2) mutants of FC2 showed pale green leaves and retarded growth, indicating that FC2-producing heme is necessary for chloroplast development. During the initial growth stage, FC2 deficiency caused reduction of plastid cytochromes. In addition, although FC2 deficiency marginally affected the assembly of photosynthetic reaction center complexes, it caused relatively larger but insufficient light-harvesting antenna to reaction centers, resulting in lower efficiency of photosynthesis. In the later vegetative growth, however, fc2-2 recovered photosynthetic growth, showing that FC1-producing heme may complement the FC2 deficiency. On the other hand, reduced level of cytochromes in microsomal fraction was discovered in fc1-1, suggesting that FC1-producing heme is mainly allocated to extraplastidic organelles. Furthermore, the expression of FC1 is induced by the treatment of an elicitor flg22 while that of FC2 was reduced, and fc1-1 abolished the flg22-dependent induction of FC1 expression and peroxidase activity. Consequently, our results clarified that FC2 produces heme for the photosynthetic machinery in the chloroplast, while FC1 is the housekeeping enzyme providing heme cofactor to the entire cell. In addition, FC1 can partly complement FC2 deficiency and is also involved in defense against stressful conditions.

Keywords: biotic stress; cytochromes; ferrochelatase; heme allocation; photosynthesis; plastid.

FIGURE 1

Characterization of T-DNA insertional mutants of ferrochelatase isoforms. (A) Schematic representation and T-DNA tagging of FC1 (At5g26030) and FC2 (At2g30390) loci. Exons (black boxes) and untranslated regions (white boxes) are shown. Location and orientation of T-DNA insertions in each line are indicated. Arrows represent the primers used for genotyping of FC1 and FC2 (see Supplementary Table 1). (B) qRT-PCR analysis of FC1 and FC2 mRNA transcripts extracted from 7-day-old wild type, fc1-1, and fc2-2 seedlings. Values are presented as the fold difference from the wild-type after normalizing to the control gene ACTIN8. Bars indicate standard error of the mean (SEM) from three independent experiments. Asterisks indicate a significant difference from the wild-type (^∗P < 0.05, Student t-test). (C) Phenotypic and genomic analysis of fc1-2. (Upper) Photographs of small (s1) and middle (m1 and m5) size seedlings germinated from heterozygous fc1-2 seeds. (Lower) PCR-based genomic analysis of these seedlings, showing small size seedling (s1) is actually homozygous fc1-2 seedling.

FIGURE 2

Phenotypic characterization of fc1-1 and fc2-2. (A) Photographs of 7-day-old wild-type, homozygous fc1-1, and homozygous fc2-2 seedlings. Seedlings of fc2-2 show pale-green phenotype. (B) Fresh weights of wild-type, fc1-1, and fc2-2 of 7- and 14-day-old seedlings. (C) Chlorophyll contents of wild-type, fc1-1, and fc2-2 of 7- and 14-day-old seedlings. Chlorophyll a (white bars) and b (gray bars) contents are indicated. Numbers on the bars are chlorophyll a/b ratio. (D) Total endogenous heme contents in 7- and 14-day-old seedlings in wild-type, fc1-1, and fc2-2. Asterisks indicate a significant difference from the wild-type (^∗P < 0.05, Student t-test). In all figures, bars indicate standard error of the mean (SEM) from three independent experiments.

FIGURE 3

Histochemical analysis of FC1pro::GUS line. (A) FC1-dependent GUS activities in 2-day-old seedlings. (B) Zoomed image of regions of root cap and elongation zone as indicated by a white circle in (A). Representative images of (C) 14-d-old seedling and (D) 21-day-old seedling. Zoomed image of regions of (E) primordial leaves, (F) root veins, and (G) veins as indicated by white circles in (C). (H) Zoomed image of primordial bolting stem as indicated by a white circle in (D).

FIGURE 4

Western blot analysis of heme-associated proteins and hemoproteins in ferrochelatase-deficient mutants. (A) CBB staining of SDS-PAGE. Samples of total (T), soluble (S), and membrane (M) fractions (40 μg each) were loaded on the gel. (B) Western blot analysis of glutamyl-tRNA reductase (HEMA1), FC2, cytochrome proteins, core subunits of PSII complexes, and peripheral LHC proteins. Note that all cytochromes and FC2 were detected in membrane fraction, while HEMA1 was also detected in soluble fraction. “LHC” means polyclonal antibodies that recognize multiple LHC proteins were used for detection.

FIGURE 5

Blue-native PAGE analysis of photosystem complexes in wild-type and ferrochelatase deficient mutants. Thylakoid membrane fractions obtained from 14-day-old seedlings are solubilized. Proteins containing 5 μg chlorophyll were loaded to 4–14% linear gradient gel. Molecular size markers are indicated on the left.

FIGURE 6

Chlorophyll fluorescence parameters of wild-type and ferrochelatase deficient mutants. Light response curves of (A) Φ_II, (B) F_v′/F_m′, (C) qP, (D) ETR, (E) Φ_NPQ, and (F) Φ_NO are shown. Φ_II, PSII quantum yield/operating quantum efficiency of PSII photochemistry; F_v′/F_m′, efficiency of open PSII reaction centers; qP, fractions of PSII centers in open states based on puddle model for the photosynthetic unit; ETR, electron transfer rate in PSII; Φ_NPQ, quantum yield of light-induced non-photochemical quenching (NPQ)/NPQ efficiency; Φ_NO, non-regulated energy dissipation. PAR means photosynthetic active radiation. Bar indicates SEM from five independent experiments for wild type and fc2-2. In addition, two biological replicates of the fc1-1 data were included as comparison.

FIGURE 7

Heme staining of ER-enriched microsomal light membrane (LM) fraction. (A) CBB and heme staining of the LM fraction (30 μg/well) of wild type, fc1-1, and fc2-2. Blue staining represents the heme-dependent peroxidase activity. (B) Heme levels in LM fractions were determined by highly sensitive heme assay. Bars indicate SEM from three independent experiments.

FIGURE 8

flg22-dependent induction of FC1. (A) qRT-PCR analysis of FC1 and FC2 in wild-type (WT), fc1-1, and fc2-2 treated with 1 μg ml^-1 flg22 for 6 h after 5 days of culture. Inset is photograph of GUS staining of FC1::GUS line treated with or without flg22. (B) qRT-PCR analysis of stress-responsive genes (HEMA2, CYP78, CYP81, and MYB41) in wild-type treated with 1 μg ml^-1 flg22 for 6 h after 5 days of culture. Values are presented as the fold difference from the flg22-untreated wild-type ACTIN8 gene. Bars indicate SEM from three independent experiments. (C) flg22-dependent changes of peroxidase activities. Rosette leaves of 27-day-old seedlings were vacuum infiltrated with or without 0.5 μg ml^-1 flg22 in the presence of 0.001% Triton X-100. After 24 h incubation, peroxidase activities were measured by using guaiacol as substrate. (D) Quantification of stress-inducible lignin of wild-type (WT), fc1-1, and fc2-2 seedlings treated with or without 1 μg ml^-1 flg22 for 3 days after 3 days of culture.