Department of Biology, Plant Science Institute, University of Pennsylvania, Philadelphia, PA 19104.

# both authors have contributed equally to this manuscript

*To whom correspondence should be addressed

Phone: (215) 898-4394

fax: (215) 898-8780

email: fdaldal@sas.upenn.edu

Key words: bc₁ complex, cytochrome b mutants, Rhodobacter capsulatus, Saccharomyces cerevisiae, energy transduction, photosynthesis and respiration.

Abbreviations: cyt, cytochrome; bc₁ complex, ubihydroquinone cytochrome c oxidoreductase; HQNO, 2-heptyl-4-hydroxyquinoline N-oxide; Q_o, ubihydroquinone oxidation site; Q_i, ubiquinone reduction site; b_L, low potential cyt b heme; b_H, high potential cyt b heme; Ps-, photosynthesis incompetent; Gly^-, respiratory deficient; Inh^R, inhibitor resistant.

ABSTRACT

In anticipation of the structure of the bc₁ complex which is now imminent, we present here a preliminary compilation of all available cytochrome b mutants that have been isolated or constructed until now both in prokaryotic and eukaryotic species. We have briefly summarized their salient properties in respect to the structure and function of cytochrome b and to the Q_o and Q_i sites of the bc₁ complex. In conjunction with the high resolution structure of the bc₁ complex, this database is expected to serve as a useful reference point for the available data and help to focus and stimulate future experimental work in this field.

The bc₁ complex (or its plant counterpart b₆f complex) is an integral membrane multi-subunit enzyme which catalyzes the oxidation of ubihydroquinone and the reduction of its physiological electron carrier partners of c-type cytochromes (cyt). This reaction takes place during photosynthesis in the facultative phototrophic bacteria (as part of the cyclic electron transfer in conjunction with the reaction center) and also during the aerobic respiration in these bacteria and in the mitochondria of facultative (yeast) or obligatory (mammals) aerobic eukaryotic organisms. The redox reactions catalyzed by the bc₁ complex are tightly coupled to proton translocation, and result in the formation of an electrochemical gradient subsequently used for ATP production by the ATP synthetase.

The bc₁ complex is composed of at least three subunits in some bacteria like Rhodobacter capsulatus, and may contain as many as ten proteins in Saccharomyces cerevisiae and mammalian mitochondria. All bc₁ complexes have three redox proteins: the diheme cyt b containing two [a low potential (b_L) and a high potential (b_H)] not covalently bound hemes, the "Rieske" Iron Sulfur protein with a [2Fe-2S] cluster, and the monoheme cyt c₁ with a covalently bound heme c. These subunits form two ubihydroquinone/ubiquinone processing domains, named Q_o (Qp) and Q_i (Qn), located on the positive and negative side, respectively, of the cytoplasmic membrane in prokaryotes or the inner mitochondrial membrane in eukaryotes. This overall scheme constitutes the basic tenet of the Q cycle model initially proposed by Mitchell [1], and is widely supported by a large body of data including those obtained using inhibitor resistant and non-functional mutants, as well as by the emerging three-dimensional structure of the beef heart enzyme [2]. The ubihydroquinone oxidation (Q_o) site is formed of the cyt b (in the vicinity of heme b_L) and the FeS protein, and is inhibited by specific Q_o site inhibitors like myxothiazol, stigmatellin and mucidin. The ubiquinone reduction (Q_i) site is confined to the cyt b (in the vicinity of heme b_H on the other side of the membrane) and is inhibited by specific Q_i site inhibitors like antimycin, funiculosin, HQNO and diuron. Thus, the integral membrane protein cyt b, which contains at least eight transmembrane a helices (A to H) and a transversal helix (cd) (Fig. 1), constitutes the catalytic heart of the bc₁ complex. It is directly responsible for the binding of the substrate and product (ubihydroquinone and ubiquinone, respectively) and the inhibitors, and for the electronic communication between the Q_o and Q_i centers. Note that in eukaryotes cyt b is the only subunit of the bc₁ complex encoded by the mitochondrial genome whereas the remaining subunits are nuclear encoded and imported into the mitochondria.

Here we have attempted to compile a preliminary list of the currently available cyt b mutations (Table 1). The first section (Table 1a) contains the single mutations while the second section (Table 1b) includes multiple mutations. In conjunction with the emerging three-dimensional structure of the bc₁ complex, we believe that this database provides a useful reference point for future experiments sharply directed at understanding how this sophisticated molecular machine functions at atomic scale. This list is possibly exhaustive and includes all cyt b mutants of prokaryotic (Rhodobacter species) and mitochondrial (Saccharomyces cerevisiae and others eukaryotes) origin. These mutants have been isolated mainly by three different approaches: first, by selecting inhibitor resistant (Inh^R) mutants from species that are naturally sensitive to these compounds, and functional revertants from incompetent photosynthetic (Ps-) bacteria or mitochondrial respiratory deficient (Gly^-) mutants; second by using in yeast the selective mitochondrial genome mutagenesis with ethidium bromide to obtain respiratory deficient mutants (Gly^-), and third by site directed mutagenesis currently carried out only in bacteria due to the difficulty of mitochondrial transformation. Fortunately, these methods are complementary of each other since for example while the spontaneous mutations are limited in target size, yet they yield functionally perturbed mutants of defined phenotype. On the other hand, site directed mutagenesis allows readily the substitution of every residue of a protein with all possible amino acids, but it requires a preconceived target site like a previously chemically labeled [3] or a highly conserved resiude [4] at a specific position. The large amount of data in the literature does not permit us to cite all related references, thus we wish to refer the reader to recent excellent reviews [4-8] where the related references and more detailed descriptions of the mutants can be found. Only recent references that have not been cited in these reviews are included here. In addition, two cartoons are used to visualize the mutated positions of cyt b. Fig. 1a indicates the amino acid residues of cyt b that affect inhibitor resistance or hypersensitivity, and Fig. 1b those that impair the function of the bc₁ complex.

Several observations derive from the compilation of cyt b mutants presented here. The rather large number of available cyt b mutants affecting the Q_o and Q_i sites of the bc₁ complex is striking. Out of the 437 amino acids of cyt b (R. capsulatus numbering) about one fifth (77, exclusive of the second site suppressors mutations) have already been substituted at least once. Further, if the second site suppressors are also considered this fraction becomes even larger and approaches almost what has been accomplished with the extensively mutated proteins like the lactose permease [9] or the lac repressor [10]. Moreover, in the case of cyt b if one adds to this collection of mutations yet another large database which has been compiled recently by Degli Esposti et al. [4] containing all available cyt b amino acid sequences from over one thousand phylogenetically different species (over 100,000 cyt b amino acids), one is truly impressed with the wealth of information currently available on this protein.

Yet, it should be stressed that the distribution of the cyt b mutations is far from being random. In fact, they are concentrated mainly in four regions (Q_iI and Q_iII delimiting the Q_i site and Q_oI and Q_oII the Q_o site) of cyt b. This bias is likely to reflect earlier intense studies aimed at defining the location and structure of the catalytic domains of the bc₁ complex [6-8]. Fig. 1 reveals that the Q_o site mutations are grouped on the positive side of the membrane, at the bottom end of helix C and on the transmembrane cd helix (Q_oI) as well as in the ef loop containing the conserved PEWY sequence located between the helices E and F (Q_oII). Conversely, those that affect the Q_i site are located on the negative side of the membrane, at the amino-terminal part of cyt b (Q_iI) and the de loop between the helices D and E (Q_iII). Several points are noteworthy: the first is the paucity of the bacterial Inh^R mutants affecting the Q_i site in part due to the natural resistance of Rhodobacter species to this class of inhibitors. Second, the Q_oI and the Q_iII portions have also been highlighted by biochemical crosslinking studies using quinone derivatives [11,12], and provide independent but concurrent experimental data for the location of the quinone processing domains. Third, in particular the Q_oI region thought to be in close proximity of cyt b_L has attracted most attention possibly due to its additional involvement in the protein-protein interactions between the FeS subunit and cyt b in forming the complete Q_o site of the bc₁ complex [13,14]. In support of this contention, specific cyt b mutations located in the cd loop of the Q_oI portion and in the ef loop of the Q_oII region of cyt b have been found to destabilize the FeS protein subunit [13,14], pointing out intricate inter-subunit protein-protein interactions in this region of the enzyme complex.

From a complementary perspective, the data presented in Fig. 1 also indicate that no mutation affecting the binding of either the Q_i or the Q_o site inhibitors has been found until now either in the ab or the bc loops of cyt b, respectively. Further, the carboxyl terminal portion of cyt b extending from the middle of helix F to its last amino acid residue appears not involved in the formation of the quinone processing domains. However, several R. sphaeroides mutations [15] affecting Q_i site catalysis have been found recently in the bc loop between the helices B and C of cyt b, indicating that additional portions of cyt b could well be part of the quinone processing domains. Equally, several mutations located in the helices F to H of cyt b are available both in bacterial and mitochondrial systems, but they have not yet been characterized (Table 1 and Fig. 1). Finally, a comparison of the distribution pattern of Inh^R mutants (Fig 1a) with that of the non functional mutants (Fig. 1b) indicates that these two groups of mutations are located in the same regions of cyt b, inferring that the inhibitor binding and the ubiquinone processing domains of the bc₁ complex overlap with each other.

In summary, crucial information has been obtained by studying the Inh^R and the non functional mutations located in cyt b. In particular, the eight a transmembrane helices topology of this protein, the identification of the histidine ligands of its two heme groups, the determination of the amino acid residues affecting the binding of inhibitors and quinone, and the subunit-subunit interactions have all been aided by the availability of these mutations. Obviously, similar studies will continue in greater detail after the availability of the structure of the bc₁ complex and of its cyt b subunit. Moreover, the emerging structural data will now able the workers in this field to address issues that have been more elusive previously, such as the definition of the parameters setting the physicochemical properties of the redox active prosthetic groups and the quinone intermediates, the determinants for the binding of quinone analogues as inhibitors, the specific amino acid side chains involved in intramolecular and intermolecular electron transfer between the various redox centers of this enzyme, the molecular basis of coupling of electron transfer and proton translocation, the amino acid chains involved in proton uptake and release at the Q_i and Q_o sites, respectively, as well as the possible proton "network" from the bulk water phase to the active sites of the enzyme as discussed recently by Brandt et al. [5]. Noteworthy in this respect is a recent work by Bruel et al. [16] that describes an uncoupled yeast mitochondrial cyt b mutant (G137E) still able to transfer electrons while being impaired to establish a proton gradient across the membrane.

In the near future, it would be desirable to expand the database initiated here for cyt b to the FeS and cyt c₁ mutants so that after the resolution of the structure of the bc₁ complex, a more complete retrospective look could be attained for the impact of the mutations thus far studied in order to better understand their effects on the structure and mechanism of function of the bc₁ complex. Then perhaps this fresher look could open up new pathways of research for structure-aided mechanistic studies, and enable us to design more incisive mutations to better probe the function of this enzyme complex. Clearly, with the emerging three-dimensional structure and the available collection of mutations, once again time seems ripe for new excitement in the field of the bc₁ complex as it has happened previously with the establishment of the 3D structure of the photochemical reaction center [17], and more recently with that of the cytochrome oxidase [18].

References

1 Mitchell, P. (1976) J. Theor. Biol. 62, 327-367

2 Xia, D., Yu, C.-A., Deisenhofer, J., Xia, J.-Z. and Yu, L. (1996) Biophys. J. 70, A253

3 Mather, M.W., Yu, L. and Yu, C.-A. (1995) J. Biol. Chem. 270, 28668-28675

4 Degli Esposti, M., De Vries, S., Crimi, M., Ghelli, A., Patarnello, T. and Meyer, A. (1993) Biochim. Biophys. Acta 1143, 243-271

5 Brandt, U. and Trumpower, B.L. (1994) Crit. Rev. Biochem. Molec. Biol. 29, 165-197

6 Colson, A.-M. (1993) J. Bioenerg. Biomembr. 25, 211-220

7 Gennis, R.B., Barquera, B., Hacker, B., Van Doren, S.R., Arnaud, S., Crofts, A.R., Davidson, E., Gray, K.A. and Daldal, F. (1993) J. Bioenerg. Biomembr. 25, 195-209

8 Gray, K.A. and Daldal, F. (1995) in Anoxygenic Photosynthetic Bacteria (Blankenship, R.E., Madigan, M.T., and Bauer, C., Eds.) Kluwer Academic Publ., The Netherlands, pp. 747-774

9 Kaback, H.R., Frillingos, S., Jung, H., Jung, K., Privé, G.G., Ujwal, M.L., Weitzman, C., Wu, J. and Zen, K. (1994) J. Exp. Biol. 196, 183-195

10 Markiewicz, P., Kleina, L., Cruz, C., Ehret, S., Miller, J.H. (1994) J. Mol. Biol. 240, 421- 433

11 He, D.-Y., Yu, L. and Yu, C.-A. (1994) J. Biol. Chem. 269, 2292-2298

12 Frank, K. and Trebst, A. (1995) Photochem. Photobiol. 61, 2-9

13 Giessler, A., Geier, B.M., di Rago, J.-P., Slonimski, P.P. and Von Jagow, G. (1994) Eur. J. Biochem. 222, 147-154

14 Saribas, A.S., Tokito, M.K. and Daldal, F. (1996) Biophys. J. 70, A356

15 Hacker, B., Barquera, B., Gennis, R.B. and Crofts, A.R. (1994) Biochemistry 33, 13022-13031

16 Bruel, C., Manon, S., Guérin, M. and Lemesle-Meunier, D. (1995) J. Bioenerg. Biomembr. 27, 527-539

17 Deisenhofer, J., Epp, O., Miki, K., Huber, R. and Michel, H. (1985) Nature 318, 618-624

18 Iwata, S., Ostermeier, C., Ludwig, B. and Michel, H. (1995) Nature 376, 660-669

19 Brasseur, G. (1995) Ph.D. Thesis, University of Aix-Marseille I, France

20 Coppée, J.-Y., Tokutake, N., Marc, D., di Rago, J.-P., Miyoshi, H. and Colson, A.-M. (1994) FEBS Lett. 339, 1-6

21 Brasseur, G. and Brivet-Chevillotte, P. (1995) Eur. J. Biochem. 230, 1118-1124

22 Edderkaoui, B. and Colson, A.-M., article in preparation

23 Crofts, A.R., Barquera, B., Bechmann, G., Guergova, M., Salcedo-Hernandez, R., Hacker, B., Hong, S. and Gennis, R.B. (1995) in Photosynthesis: from Light to Biosphere (Mathis, P. Ed.) Kluwer Academic Publ., The Netherlands, vol II , pp. 493-500

24 Colson, A.-M., Edderkaoui, B. and Coppée, J.-Y. (1992) Biochim. Biophys. Acta 1101, 157-161

25 Crofts, A., Hacker, B., Barquera, B., Yun, C.-H. and Gennis, R. (1992) Biochim. Biophys. Acta 1101, 162-165

26 Ding, H., Daldal, F. and Dutton, P.L. (1995) Biochemistry 34, 15997-16003

27 Ding, H., Moser, C.C., Robertson, D.E., Tokito, M.K., Daldal, F. and Dutton, P.L. (1995) Biochemistry 34, 15979-15996

28 Saribas, A.S. and Daldal, F. (1995) in Photosynthesis: from Light to Biosphere (Mathis, P. Ed.) Kluwer Academic Publ., The Netherlands, vol II , pp. 501-506

29 Saribas, A.S., Ding, H., Dutton, P.L. and Daldal, F. (1995) Biochemistry 34, 16004-16012

30 Bruel, C. (1995) Ph.D. Thesis, University of Aix-Marseille II, France

31 Bruel, C. and Lemesle-Meunier, D. (1993) Biochem. Soc. Trans. 22, 61S

32 Mather, M.W., Yu, C.A. and Yu, L. personal communication

33 Bruel, C., di Rago, J.-P., Slonimski, P.P. and Lemesle-Meunier, D. (1995) J. Biol. Chem. 270, 22321-22328

34 Saribas, A.S., Tokito, M. and Daldal, F. unpublished results

35 Hacker, B., Barquera, B., Crofts, A.R. and Gennis, R.B. (1993) Biochemistry 32, 4403-4410

36 Gray, K.A., Dutton, P.L. and Daldal, F. (1994) Biochemistry 33, 723-733

37 Coppée, J.-Y., Brasseur, G., Brivet-Chevillotte, P. and Colson A.-M. (1994) J. Biol. Chem. 269, 4221-4226

38 Brasseur, G., Coppée, J.-Y., Colson, A.-M. and Brivet-Chevillotte, P. (1995) J. Biol. Chem. 270, 29356-29364

39 Brasseur, G. and Brivet-Chevillotte, P. (1993) Biochem. Soc. Trans. 22, 60S

40 Brasseur, G. and Brivet-Chevillotte, P. (1994) FEBS Lett. 354, 23-29

41 Uhrig, J.F., Jakobs, C.U., Majewski, C. and Trebst, A. (1994) Biochim. Biophys. Acta 1187, 347-353

42 di Rago, J.-P., Hermann-Le Denmat, S., Pâques, F., Risler, J.-L., Netter, P. and Slonimski, P.P. (1995) J. Mol. Biol. 248, 804-811

43 di Rago, J.-P., Macabre, C., Lazowska, J. and Slonimski, P.P. (1993) FEBS Lett. 328, 153-158

Fig. 1. Amino acid residues of cyt b conferring resistance and hypersensitivity to inhibitors (Fig. 1a) or affecting the function (Fig. 1b) of the bc₁ complex. A cyt b model with eight transmembrane a helices (A to H) was used to indicate them along the four histidine () ligands of the b_L and b_H hemes. Open circles (O) and filled triangles () correspond to bacterial and mitochondrial mutations, respectively. The numbering corresponds to R. capsulatus cyt b, and the Q_o (Q_oI and Q_oII) and Q_i (Q_iI and Q_iII) sites on the positive (Y+) and negative side (Y-), respectively, of the bacterial cytoplasmic membrane or the inner mitochondrial membrane are indicated.