Functional Relationships of Two NFU Proteins in Maintaining the Abundances of Mitochondrial Iron-Sulfur Proteins

Abstract

Iron-sulfur clusters are involved in many biological processes, including photosynthetic electron transport in the chloroplast and respiratory electron transport in the mitochondrion. Iron-sulfur cluster biosynthesis requires iron-sulfur carriers such as nitrogen-fixation-subunit-U [NFU]-type proteins. The Arabidopsis thaliana nuclear genome encodes two mitochondrion-targeted NFU proteins: NFU4 and NFU5, previously reported to have a primary role in the biosynthesis of the lipoate cofactor, mediated by the 4Fe-4S enzyme lipoyl synthase. Through in vitro reconstitution and spectroscopic analysis, we found that recombinant NFU4 and NFU5 proteins had UV-visible features characteristic of 4Fe-4S clusters. In addition, we confirmed that double homozygous, complete loss-of-function nfu4 nfu5 mutants had an embryo-lethal phenotype. To investigate the functional relationship between NFU4 and NFU5, we generated sesquimutants that were homozygous loss-of-function for one gene and heterozygous for the other, which appeared slightly smaller than nfu4-2, nfu4-4, and nfu5-1 single mutants. This suggests that the simultaneous decrease in levels of NFU4 and NFU5 proteins may have an additive effect on plant growth. Quantitative reverse transcription PCR showed that the NFU4 transcript was absent in mutants homozygous for nfu4-2 and nfu4-4 and that the NFU5 transcript level was substantially reduced in the nfu5-1 single mutant or sesquimutants. Consistent with the transcript data, the abundances of NFU4 and NFU5 proteins were either virtually absent or substantially reduced in the corresponding single mutants and sesquimutants. Immunoblot analysis showed that most nfu4 and nfu5-1 single, double, and sesquimutants had significant reductions in the levels of mitochondrial 4Fe-4S proteins, such as aconitase (ACO) and biotin synthase 2 (BIO2; note that BIO2 also contains a 2Fe-2S cluster). In addition, nfu4 nfu5 sesquimutants showed substantial reductions in the protein level of the 75-kDa subunit of respiratory complex I (CI75), which contains one 2Fe-2S cluster and two 4Fe-4S clusters. These observations indicate that NFU4 and NFU5 are important in maintaining the levels of mitochondrial 4Fe-4S proteins. Such observations are also consistent with the hypothesis that NFU4 and NFU5 may serve as iron-sulfur carriers and may play a role in the transfer of 4Fe-4S clusters to recipient apoproteins, such as ACO and CI75, during the biogenesis and maturation of mitochondrial 4Fe-4S clusters.

Keywords: Iron–sulfur cluster; NFU4; NFU5; iron–sulfur carrier; mitochondrion; nitrogen‐fixation‐subunit‐U (NFU).

FIGURE 1

Domain composition and absorption spectra of as‐purified, reduced, and reconstituted recombinant NFU4 and NFU5 proteins. (a) Domains in full‐length NFU4 and NFU5 proteins. Orange boxes represent the mitochondrial transit peptide; yellow boxes represent the N‐terminal domain unique to mitochondrial‐targeted NFU proteins; cyan boxes represent the redox‐active NFU domain with the conserved CXXC motif. (b–c) Absorption spectra of as‐purified, reduced, and reconstituted recombinant NFU4 (b) and NFU5 (c) proteins. Recombinant NFU4^{55–283 AA} and NFU5^{77–275 AA} were purified aerobically, and an absorption spectrum was recorded (black line). The blue curve represents the absorption spectrum of recombinant NFU4 and NFU5 proteins after reduction with 10‐mM sodium dithionite. The black and red arrow points to the absorption peak at 410 nm, a typical feature of 4Fe–4S clusters, which disappears upon reduction by 10‐mM sodium dithionite. The red curve represents the absorption spectrum of the recombinant proteins after reconstitution with ammonium ferrous sulfate and sodium sulfide. aa, amino acids.

FIGURE 2

Phenotype of nfu4 and nfu5‐1 single, double, and sesquimutants. (a) Structure of NFU4 and NFU5 genes and locations of nfu4 and nfu5 mutations. White boxes, untranslated regions; black boxes, exons; lines, introns; and triangles, T‐DNA insertions. Red arrows 1 and 2, left and right qRT‐PCR primers for NFU4; red arrows 3 and 4, left and right qRT‐PCR primers for NFU5. The nfu4‐2, nfu4‐3, nfu4‐4, and nfu5‐1 mutants are in the Columbia (Col) background. The nfu5‐3 mutant is in the Landsberg erecta (Ler) background, which is morphologically different from Col. Therefore, nfu5‐3 was only used to create sesquimutant plants for seed segregation analysis. (b) Images of 4‐week‐old plants grown on a 12‐h light/12‐h dark photoperiod with an irradiance of 150‐μmol photons m⁻² s⁻¹ during the light period. (c) Above‐ground fresh weight of 4‐week‐old plants. Values are presented as mean ± SE (n = 4 individual plants/genotype). Values not connected by the same uppercase letter are significantly different (Student's t‐test, p < 0.05). bp, base pair; WT, wild type.

FIGURE 3

Representative siliques from nfu4‐2/4 ^−/− nfu5‐1/3 ^+/− plants. (a–d) Siliques from (a) nfu4‐2 ^−/− nfu5‐1/ ^+/− , (b) nfu4‐2 ^−/− nfu5‐3 ^+/− , (c) nfu4‐4 ^−/− nfu5‐1 ^+/− , and (d) nfu4‐4 ^−/− nfu5‐3 ^+/− plants. In (a), seven of the 32 seeds in this silique were aborted. In (b), seven of the 27 seeds in this silique were aborted. In (c), three of the 12 seeds in this silique were aborted. In (d), six of the 28 seeds were aborted. White arrows indicate aborted seeds.

FIGURE 4

Pigment contents in leaf tissues of 4‐week‐old plants. (a) Chlorophyll a, (b) chlorophyll b, (c) carotenoid, and (d) total chlorophyll contents in leaf tissues of 4‐week‐old plants. Values are presented as mean ± SE (n = 4 individual plants/genotype). Values not connected by the same uppercase letter are significantly different (Student's t‐test, p < 0.05). FW, fresh weight; WT, wild type.

FIGURE 5

Relative (a) NFU4 and (b) NFU5 transcript levels in leaf tissues of 4‐week‐old plants. Total RNA was extracted from rosette leaves and analyzed with qRT‐PCR. ACTIN2 (At3g18780) was used as an internal reference for normalization. The average ratio of NFU4 or NFU5 transcripts to ACTIN2 transcripts was normalized to 1.0 in the Col WT. Values are presented as mean ± SE (n = 6 individual plants/genotype). Values not connected by the same uppercase letter are significantly different (Student's t‐test, p < 0.05).

FIGURE 6

Amounts of representative mitochondrial iron–sulfur proteins in leaf tissues of 4‐week‐old plants. (a) Representative immunoblots of mitochondrial iron–sulfur proteins in leaf tissues of 4‐week‐old plants. (b–i) Relative amounts of (b) NFU4, (c) NFU4 + NFU5, (d) 2Fe–2S protein mFDX1, (e) 3Fe–4S protein GOGAT, (f) 4Fe–4S protein ACO, (g) 2Fe–2S and 4Fe–4S proteins BIO2 and (h) CI75, and (i) non‐Fe–S mitochondrial protein MCCC1 in leaf tissues of 4‐week‐old plants. In (b), blots were detected with the anti–NFU4 antibody. In (c), blots were detected with the anti–NFU4/NFU5 antibody, which cross‐reacts with both NFU4 and NFU5. The mitochondrial protein MCCC1 does not contain any iron–sulfur (Fe‐S) clusters and was used as a protein expected to be unaffected by changes in NFU4 and NFU5 abundances. Rubisco large subunit (RbcL) stained with Coomassie Brilliant Blue was used as the loading control. Densitometric analysis of immunoblots was performed with the Multi Gauge V3.0 software; the average relative amount of each protein was normalized to 1.0 in the Col wild type (WT). Values are presented as mean ± SE (n = 3 individual plants/genotype for NFU4 and NFU4 + NFU5 and n = 4 individual plants/genotype for other proteins). Values not connected by the same uppercase letter are significantly different (Student's t‐test, p < 0.05).

FIGURE 7

Correlation analysis among the relative NFU4 and NFU5 transcript levels and the relative NFU4 and NFU5 protein levels in 4‐week‐old plants. (a) Correlation of the relative NFU4 protein level on the relative NFU4 transcript level; (b) correlation of the relative NFU5 protein level on the relative NFU5 transcript level; (c) correlation of the total amount of NFU4 and NFU5 proteins on the relative level of NFU4 protein; and (d) correlation of the total amount of NFU4 and NFU5 proteins on the relative level of NFU5 protein. Relative transcript levels were determined by qRT‐PCR; the average ratio of NFU4 or NFU5 transcripts to ACTIN2 transcripts was normalized to 1.0 in the Col WT, as shown in Figure 5. The relative NFU4 protein level was determined by SDS‐PAGE and immunoblotting with the anti–NFU4 antibody; the total amount of NFU4 and NFU5 proteins was determined by SDS‐PAGE and immunoblotting with the anti–NFU5 antibody, which cross‐reacts with NFU4; the relative NFU5 protein level was calculated according to the relative levels of NFU4 and NFU4 + NFU5 proteins; the average relative amount of each protein (NFU4, NFU5, and NFU4 + NFU4) was normalized to 1.0 in the Col wild type (WT), as shown in Figure 6. Each blue point represents one genotype for the 10 genotypes shown in Figures 2 and 4, 5, 6. Error bars represent standard errors (n = 6 individual plants/genotype for transcript levels and n = 3 individual plants/genotype for protein levels). Dotted lines are linear regressions. The r‐squared value and correlation coefficient r are also shown for each chart.

FIGURE 8

Correlation analysis of the relative levels of mitochondrial iron–sulfur proteins with the relative NFU4 and NFU5 protein levels in 4‐week‐old plants. (a–f) Correlation analysis of the relative (a) mFDX1, (b) GOGAT, (c) ACO, (d) BIO2, (e) CI75, and (f) MCCC1 protein levels with the relative NFU4 protein level. (g–l) Correlation analysis of the relative (g) mFDX1, (h) GOGAT, (i) ACO, (j) BIO2, (k) CI75, and (l) MCCC1 protein levels with the relative NFU5 protein level. (m–r) Correlation analysis of the relative (m) mFDX1, (n) GOGAT, (o) ACO, (p) BIO2, (q) CI75, and (r) MCCC1 protein levels with the total NFU4 + NFU5 protein level. The relative NFU4 protein level was determined by SDS‐PAGE and immunoblotting with the anti–NFU4 antibody; the total amount of NFU4 and NFU5 proteins was determined by SDS‐PAGE and immunoblotting with the anti–NFU5 antibody, which cross‐reacts with NFU4; the relative NFU5 protein level was calculated according to the relative levels of NFU4 and NFU4 + NFU5 proteins; the average relative amount of each protein (NFU4, NFU5, and NFU4 + NFU4) was normalized to 1.0 in the wild type (WT), as shown in Figure 6. Each blue point represents one genotype for the 10 genotypes shown in Figures 2 and 4, 5, 6. Error bars represent standard errors (n = 3 individual plants/genotype for NFU4, NFU5, and NFU4 + NFU5; n = 4 individual plants/genotype for other proteins). Dotted lines are linear regressions. The r‐squared value and correlation coefficient r are also shown for each chart.