Were there any "misassignments" among iron-sulfur clusters N4, N5 and N6b in NADH-quinone oxidoreductase (complex I)?

Abstract

NADH-quinone oxidoreductase (complex I) in bovine heart mitochondria has a molecular weight of approximately 1 million Da composed of 45 distinct subunits. It is the largest energy transducing complex so far known. Bacterial complex I is simpler and smaller, but the essential redox components and the basic mechanisms of electron and proton translocation are the same. Over the past three decades, Ohnishi et al. have pursued extensive EPR studies near liquid helium temperatures and characterized most of the iron-sulfur clusters in complex I. Recently, Yakovlev et al. [G. Yakovlev, T. Reda, J. Hirst, Reevaluating the relationship between EPR spectra and enzyme structure for the iron-sulfur clusters in NADH:quinone oxidoreductase, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 12720-12725] challenged Ohnishi's group by claiming that there were EPR "misassignments" among clusters N4, N5 and N6b (in order to prevent confusion, we used current consensus nomenclature, as the nickname). They claimed that we misassigned EPR signals arising from cluster N5 to cluster N4, and signals from cluster N6b to cluster N4. They also proposed that cluster N5 has (4Cys)-ligands. Based on the accumulated historical data and recent results of our site-specific mutagenesis experiments, we confirmed that cluster N5 has (1His+3Cys)-ligands as we had predicted. We revealed that E. coli cluster N5 signals could be clearly detected at the sample temperature around 3 K with microwave power higher than 5 mW. Thus Hirst's group could not detect N5 signals under any of their EPR conditions, reported in their PNAS paper. It seems that they misassigned the signals from cluster N4 to N5. As to the claim of "misassignment" between clusters N4 and N6b, that was not a possibility because our mutagenesis systems did not contain cluster N6b. Therefore, we believe that we have not made any "misassignment" in our work.

Fig. 1

Arrangement of redox centers in the hydrophilic domain of T. thermophilus complex I reported in [4] with minor modifications. These authors used consensus nomenclature, which is based on EPR spectroscopy [2] and recent site-directed mutagenesis studies [8]. Cluster N1a is in subunit NuoE; cluster N3 and FMN in NuoF; clusters N1b, N4, N5, and N7 in NuoG; clusters N6a and N6b in NuoI; cluster N2 in NuoB in E. coli complex I. The edge-to-edge distances between the redox centers are given in Å. Subunits circled by the solid line reside in the NADH dehydrogenase fragment, and subunits circled by the dotted line are located in the connecting fragment of the E. coli complex I.

Fig. 2

(A) Comparison of EPR spectra between wild-type and mutants of the reconstituted MBP-NuoG subunit at 6 K in the presence of 10 mM sodium dithionite, 5 μM of benzyl viologen and methyl viologen. (B) EPR spectra of the reconstituted MBP-NuoGΔN4 and MBP-NuoGΔN5 were recorded at 4 K with different microwave power at 100 mW (thick line) and 4 mW (thin line), and 0.25 mW (dotted line). The signals were normalized to 1 mg/ml of protein concentration. The spectra were recorded under the following conditions: microwave frequency, 9.44 GHz; microwave power, 5 mW; modulation amplitude, 10.115 G; modulation frequency, 100 kHz; time constant, 164 ms. Principal g values were indicated. Mutant subunits, designated ΔN4 and ΔN5 (in which each set of the iron–sulfur binding motifs for cluster N4 and N5 in the E. coli NuoG subunit were individually inactivated by the substitution of all four ligands with Ala), were expressed as maltose-binding protein fusion proteins and were purified [8]. After in vitro reconstitution, wild-type and mutant subunits were characterized by EPR.

Fig. 3

EPR spectra of the E. coli NuoCDEFG subcomplex from various cluster N5 mutant strains, W221A (a), H101C (b), H101A (c) and cluster N5 knockout (ΔN5) (d), measured at 40 K (A), 6 K (B), and 3 K (C). Each sample was reduced with 10 mM sodium dithionite in the presence of 5 μM of benzyl viologen and methyl viologen. The signals were normalized to 1 mg/ml protein concentration. EPR spectra were recorded under the following conditions: microwave frequency, 9.45 GHz; microwave power, 5 mW; modulation amplitude, 8.0 G; time constant, 82 ms. Principal g values were indicated. The cluster N5 mutant strains were obtained by introducing point mutation(s) into the genomic DNA with homologous recombination. These genomic mutations drastically affected the stability of complex I, and consequently, NADH dehydrogenase subcomplex (NuoEFG subcomplex) were dissociated from the membrane. However, we successfully obtained these mutant NuoCDEFG subcomplexes by the overexpression of the His-tagged-NuoCD subunit. We deliberately employed the W221A mutant strain as a control to obtain the NuoCDEFG subcomplex carrying wild-type clusters.

Fig. 4

EPR spectra of cluster N6a and N6b of the overexpressed, truncated P. denitrificans Nqo9 (NuoI) subunit (A), which contains only clusters N6a and N6b, cited from [30], and the resolved connecting fragment (composed of Nuo B, C–D, I) of the E. coli complex I (B), which contains clusters N2, N6a, and N6b, cited from [31].

Fig. 5

Schematic structure of two Fe/S clusters in the NuoI subunit. The numbering is in accordance with the E. coli sequences. The alignment of eight highly conserved cysteine residues is similar to that of 2×[4Fe–4S] bacterial ferredoxin [32].