[4Fe-4S] cluster trafficking mediated by Arabidopsis mitochondrial ISCA and NFU proteins

Abstract

Numerous iron-sulfur (Fe-S) proteins with diverse functions are present in the matrix and respiratory chain complexes of mitochondria. Although [4Fe-4S] clusters are the most common type of Fe-S cluster in mitochondria, the molecular mechanism of [4Fe-4S] cluster assembly and insertion into target proteins by the mitochondrial iron-sulfur cluster (ISC) maturation system is not well-understood. Here we report a detailed characterization of two late-acting Fe-S cluster-carrier proteins from Arabidopsis thaliana, NFU4 and NFU5. Yeast two-hybrid and bimolecular fluorescence complementation studies demonstrated interaction of both the NFU4 and NFU5 proteins with the ISCA class of Fe-S carrier proteins. Recombinant NFU4 and NFU5 were purified as apo-proteins after expression in Escherichia coliIn vitro Fe-S cluster reconstitution led to the insertion of one [4Fe-4S]²⁺ cluster per homodimer as determined by UV-visible absorption/CD, resonance Raman and EPR spectroscopy, and analytical studies. Cluster transfer reactions, monitored by UV-visible absorption and CD spectroscopy, showed that a [4Fe-4S]²⁺ cluster-bound ISCA1a/2 heterodimer is effective in transferring [4Fe-4S]²⁺ clusters to both NFU4 and NFU5 with negligible back reaction. In addition, [4Fe-4S]²⁺ cluster-bound ISCA1a/2, NFU4, and NFU5 were all found to be effective [4Fe-4S]²⁺ cluster donors for maturation of the mitochondrial apo-aconitase 2 as assessed by enzyme activity measurements. The results demonstrate rapid, unidirectional, and quantitative [4Fe-4S]²⁺ cluster transfer from ISCA1a/2 to NFU4 or NFU5 that further delineates their respective positions in the plant ISC machinery and their contributions to the maturation of client [4Fe-4S] cluster-containing proteins.

Keywords: Arabidopsis thaliana; ISCA proteins; NFU proteins; Raman spectroscopy; circular dichroism; iron-sulfur cluster trafficking; iron-sulfur protein; mitochondria; protein-protein interaction.

Figure 1

Binary Y2H interaction between Arabidopsis NFU4 and NFU5 and other late-acting ISC components.A, the yeast strain CY306 was cotransformed with sequences encoding the mature forms of NFU4 and NFU5 proteins fused at the C terminus of the Gal4 DNA-binding domain (BD) as indicated and other ISC components fused at the C terminus of the Gal4 activation domain (AD). The cotransformed yeast cells were plated on a control medium containing histidine (+ HIS) and interactions were tested on medium without histidine (− HIS) in the presence of 2 mm or 5 mm 3-AT as indicated. Yeast growth was analyzed after 5 days. Only ISCA1b exhibited a slight autoactivation in the – His medium which disappeared in the presence of 3-AT. No autoactivation was observed for BD-NFU4 and NFU5 (not shown). B, the yeast strain CY306 was cotransformed with sequences encoding the N- or C-terminal domains of NFU4 and NFU5 proteins fused at the C terminus of the Gal4 DNA-binding domain (BD) and ISCA1a or ISCA1b fused at the C terminus of the Gal4 activation domain (AD). The NFU4-N and -C domains correspond respectively to amino acids 80–168 and 187–283 of NFU4, whereas the NFU5-N and -C domains correspond respectively to amino acids 75–163 and 182–275 of NFU5. The conditions are as in (A).

Figure 2

BiFC interactions between Arabidopsis mitochondrial ISCAs and NFUs.Arabidopsis protoplasts isolated from 4-week-old plantlets were transfected with combinations of two vectors expressing either NFU4 or NFU5 fused to the N-terminal region of the YFP protein (NFUs-N in panels) and ISCAs fused to the C-terminal region of YFP (ISCA-C in panels). The YFP fluorescence was recorded 24 h posttransfection by confocal microscopy. Negative controls verifying that none of the NFU proteins tested alone with the empty partner vector can restore YFP fluorescence are shown as Empty-C/NFUs-N in panels; those verifying ISCA-C/Empty-N combinations are shown in Fig. S1. Protoplast cotransfections using opposite protein chimera (ISCA-N with NFUs-C) provided similar patterns but also showed a strong tendency to form cytosolic aggregates (not shown). All images were captured using the LAS X software at confocal plans at selected Z dimensions and processed using the Adobe Photoshop software. Results are representative of three independent bombardment experiments including the analysis of 10–20 cells per transformation event. MitoTracker® Orange CMXRos (Invitrogen) was used at 100 nm to label mitochondria within cells. Bars = 10 µm.

Figure 3

Room temperature UV-visible absorption spectra and CD spectra of At NFU4 and At NFU5. As-purified apo-NFU4 and NFU5 are shown as black lines and reconstituted NFU4 (A) and NFU5 (B) are shown as blue lines. All ε and Δε values are based on NFU4 and NFU5 protein monomer concentration.

Figure 4

Comparison of the resonance Raman and EPR spectra of reconstituted At NFU4 and At NFU5.A, resonance Raman spectra of reconstituted NFU4 and NFU5 recorded at 17 K with 457.9-nm excitation. Each spectrum is the sum of 100 individual scans, with each scan involving photon counting for 1 s at 0.5-cm⁻¹ increments, with 7-cm⁻¹ spectral resolution. Bands due to frozen buffer solution have been subtracted from both spectra. B, X-band EPR spectra of reconstituted NFU4 and NFU5 reduced with one reducing equivalent of dithionite and frozen immediately in liquid nitrogen. EPR conditions: microwave frequency, 9.60 GHz; microwave power, 10 milliwatt; modulation amplitude, 0.63 mT; temperature, 10 K.

Figure 5

Cluster transfer from At [4Fe-4S]²⁺-ISCA1a/2 to At apo-NFU4 monitored by CD spectroscopy as a function of time.A, CD spectra of the cluster transfer reaction mixture that was initially 30 µm in ISCA1a/2 [4Fe-4S]²⁺ clusters and 60 µm in DTT-pretreated apo-NFU4 monomer. The thick red line corresponds to [4Fe-4S]²⁺-ISCA1a/2 recorded before addition of apo-NFU4 to the reaction mixture. The thin gray lines correspond to CD spectra recorded at 4, 7, 11, 20, 36, 40, 45, and 50 min after the addition of apo-NFU4. The thick blue line corresponds to complete [4Fe-4S]²⁺ cluster transfer to NFU4. The arrows indicate the direction of intensity change with increasing time at selected wavelengths and Δε values were calculated based on the initial concentration of [4Fe-4S]²⁺ clusters in the reaction mixture. The cluster transfer reaction was carried out under anaerobic conditions at room temperature in 100 mm Tris-HCl buffer at pH 7.8. B, kinetic simulation of cluster transfer from [4Fe-4S]²⁺-ISCA1a/2 to apo-NFU4 based on second-order kinetics and the initial concentrations of [4Fe-4S]²⁺ clusters on [4Fe-4S]²⁺-ISCA1a/2 and of apo-NFU4. Percent cluster transfer was assessed by the difference in CD intensity at 326 and 362 nm (black circles) and simulated with a second-order rate constant of 9.1 × 10³m⁻¹ min⁻¹ (black line). The residual [2Fe-2S]²⁺-ISCA1a/2 peak-to-trough CD intensity at 362 and 326 nm at zero time was added on to each data point as there is no evidence for any [2Fe-2S]²⁺ cluster transfer.

Figure 6

Cluster transfer from At [4Fe-4S]²⁺-ISCA1a/2 to At apo-NFU5 monitored by CD spectroscopy as a function of time.A, CD spectra of the cluster transfer reaction mixture that was initially 40 µm in ISCA1a/2 [4Fe-4S]²⁺ clusters and 80 µm in DTT-pretreated apo-NFU5 monomer. The thick red line corresponds to [4Fe-4S]²⁺-ISCA1a/2 recorded before addition of apo-NFU5 to the reaction mixture. The thin gray lines correspond to CD spectra recorded at 5, 9, 13, 18, 27, 40, and 60 min after the addition of apo-NFU5. The thick blue line corresponds to complete [4Fe-4S]²⁺ cluster transfer to NFU5. The arrows indicate the direction of intensity change with increasing time at selected wavelengths, and Δε values were calculated based on the initial concentration of [4Fe-4S]²⁺ clusters in the reaction mixture. The cluster transfer reaction was carried out under anaerobic conditions at room temperature in 100 mm Tris-HCl buffer at pH 7.8. B, kinetic simulation of cluster transfer from [4Fe-4S]²⁺-ISCA1a/2 to apo-NFU5 based on second-order kinetics and the initial concentrations of [4Fe-4S]²⁺ clusters on [4Fe-4S]²⁺-ISCA1a/2 and of apo-NFU5. Percent cluster transfer was assessed by the difference in CD intensity at 326 and 362 nm (black circles) and simulated with a second-order rate constant of 7.0 × 10³m⁻¹ min⁻¹ (black dots). The residual [2Fe-2S]²⁺-ISCA1a/2 peak-to-trough CD intensity at 362 and 326 nm at zero time was added on to each data point, because there is no evidence for any [2Fe-2S]²⁺ cluster transfer.

Figure 7

Activation of apo-ACO2 using [4Fe-4S] cluster-bound ISCA1a/2, [4Fe-4S] cluster-bound NFU5, and [2Fe-2S] cluster-bound ISCA1a/2. Apo-ACO2 (2.4 µm) was incubated with [4Fe-4S] cluster-loaded ISCA1a/2 (red data) or NFU5 (black data) (both 7.4 µm in [4Fe-4S] clusters) and [2Fe-2S] cluster-loaded ISCA1a/2 (blue data) (14.5 µm in [2Fe-2S]²⁺ clusters) at room temperature under anaerobic conditions. 10-µl aliquots of the reaction mixture were removed at selected time points and assayed immediately for aconitase activity. Residual aconitase activity of apo-ACO2, in the absence of a cluster donor, was assessed and subtracted from all measured activities. Aconitase specific activity as a function of incubation time with the cluster donor was expressed as a percentage of the maximal specific activity of [4Fe-4S]²⁺ cluster-replete ACO2. Solid lines are best fits to second-order kinetics, with the indicated rate constants, k, based on the initial concentrations of apo-ACO2 and [4Fe-4S]²⁺ clusters on ISCA1a/2 or NFU5 and half the initial [2Fe-2S]²⁺ cluster concentration of [2Fe-2S]-ISCA1a/2.

Figure 8

Summary scheme for iron-sulfur cluster trafficking between NFU4 and NFU5 and their partner proteins.