Eur. J. Biochem. 195, 731 -734 (1991) FEBS 1991

001429569100088K

Structural and functional characteristics of polypeptide subunits of the bovine heart ubiquinol - cytochrome-c reductase complex Tiziana COCCO I , Michele LORUSSO ', Anna Maria SARDANELLI', Michele MINUTO ', Severino RONCHI ', Gabriella TEDESCHI and Sergio PAPA'

Institute of Medical Biochemistry and Chemistry, University of Bari, Italy Centre for the Study of Mitochondria and Energy Metabolism, C.N.R., University of Bari, Italy Institute of Veterinary Physiology and Biochemistry, University of Milan, Italy

(Received June 21/September 12, 1990) - EJB 90 0713

Structural and functional characteristics of subunits of bovine heart cytochrome-c reductase have been investigated by controlled digestion of soluble and membrane-reconstituted purified bcl complex and direct amino acid sequencing of native and digested protein subunits.

The results obtained show that the N-terminal segments of core protein I1 and the 14-kDa protein extend at the periphery of the complex, protruding into the inner matrix space. The Fe-S protein, located at the outer C-periphery of the complex, is shown to be anchored to other subunits of the complex by the amphipathic N- terminal region. Proteolytic cleavage of 7 - 11 residues from the N-terminal segment of the 14-kDa protein is apparently associated with decoupling of redox-linked proton pumping. Partial digestion of core protein 11, the 6.4-kDa protein, and the C-terminal region of the 9.2-kDa protein, is without effect on the redox and proton- motive activity of the complex.

The bc, complex (proton-motive ubiquinol- cyto- chrome-c reductase) of bovine heart mitochondria is composed of 11 protein subunits [l, 21. The complex has eight supernumerary subunits in addition to the apoproteins of b cytochromes, cytochrome c1 and the Rieske Fe-S center, which are common to both eukaryotic and prokaryotic enzymes [3 - 51. This abundance of protein components in the eukaryotic bcl complex has prompted studies on the quaternary structure and membrane topology [2, 6, 71, biogenesis and assembly of the complex [8 - 91, and possible functional and/or regulatory involvement of supernumerary subunits in the redox and pro- ton-motive activity [3, 10- 131.

In previous work from our laboratory it was shown that papain digestion of the isolated soluble bovine heart bcl com- plex resulted in partial cleavage of core protein 11, the 14-kDa subunit and Rieske Fe-S protein [2 , lo]. Cleavage of core protein 11 and the 14-kDa protein was apparently associated with decoupling of proton pumping. Digestion of the Fe-S protein resulted in inhibition of electron flow and destabiliza- tion of the protein-bound antimycin-sensitive semiquinone species [2].

In this paper further studies are presented in which proteo- lytic digestion was carried out with papain and trypsin, both on the soluble and liposome-reconstituted bcl complex, and the cleavage products were unambigously identified by direct sequence analysis of the subunits. The results show that diges- tion of the N-terminal region of core protein 11 and of the

6.4-kDa protein is without effect on the redox and proton motive activity of the bcl complex. Proteolytic removal from the 14-kDa protein of an N-terminal segment of 7 - 11 resi- dues which is exposed at the matrix side of the membrane, appears to be associated with decoupling of redox-linked pro- ton pumping. It is also shown that the Fe-S protein is func- tionally anchored to the complex at the cytosolic side by the amphipathic N-terminal region.

MATERIALS AND METHODS

Preparation of bc, complex and bc,-complex vesicles

The bcl complex was isolated from bovine heart mitochon- dria according to Rieske [14] and characterized as described in [15]. Reconstitution of the bcl complex in phospholipid vesicles was performed by the cholate-dialysis method of Leung and Hinkle [I61 in 100 mM Hepes (potassium salt) (pH 7.8). Oxidation by ferricyanide of bcl complex which had been reduced by duroquinol, showed that bcl-complex vesicles were more than 90% right-side-out oriented.

Papain treatment

Papain treatment was carried out as described in [2]. The reaction was stopped by the addition of 10% trichloroacetic acid.

____ Correspondence to S. Papa, Institute of Medical Biochemistry and Trypsin treatment

Chemistry, University of Bari, Piazza Giulio Cesare, 1-70124 Bari, Italy Soluble bc, complex in 10 mM Hepes (pH 7.8) was incu-

Abbreviations. bc, complex, ubiquinol -cytochrome-c reductase, bated at 18 "c with trypsin under the conditions described in Enzymes. Ubiquinol- cytochrome-c reductase, bcl complex the legends to the figures. The reaction was stopped by adding

a fivefold excess of trypsin inhibitor. 15 - 20 p1 trypsinized bc, (EC I .10.2.2).

732

A B

Coren ~ K V A P K V K A T E - Corel\ ~ ~ T E A P A G V P P H P Q D L E F R R L P N G - - - - ~ ~ ~ H C y t . b \

FeS iHTDIKVPDFSD--- Cyt.c, -

64 ~SADVLAMAKIEIKLSDIPEG----CCOH

14kDa AGRPAVSASSRW--- ~ S S R W L E G I R K W Y Y N A A G F N K L G L M - - - - ~ ~ ~ H 11 kDa- 9.2kDa - 8kDa-

Z2kDa' 6.4 kDa /-

Fig. I . SDS/PAGE of'soluhle bc, complex after digestion with papain. (A) Control : papain added to the trichloroacetic-acid-treated bc, complex. (B) hc, complex digested for 30 min by papain. The band just below that of the Fe-S protein corresponds to papain. In the SDS/ PAGE of 2-mercaptoethanol-treated bc, complex, cytochrome c1 ap- pears as a doublet (Figs 2, 4 and 5 ) [6]. The N-terminal sequences of core protein 11, Fe-S protein and the 14-kDa protein, and of their proteolytic products, are prcsented

complex was diluted with 0.4 ml 15% sucrose in 0.1 M Tris/ C1 (pH S), and 0.3 ml saturated ammonium sulphate solution, at 0 'C. The mixture was centrifuged prior to SDSjPAGE to remove soluble peptides. Aliquots of trypsin-digested bcl complex were added directly to a sonicated phospholipid sus- pension for measurement of reductase activity and proton translocation. Controls showed that trypsin treatment did not alter the orientation of the vesicles.

bel-complex-containing vesicles (2 mg protein/ml) were directly incubated with trypsin at 18 'C under the conditions specified in the legends to the figures; the reaction was stopped as described abovc. The cleaved enzyme was separated from phospholipids prior to SDSjPAGE as described in [17].

Measurement of cjstochronze-c reductase activity cmd proton translocation

The reductase activity was measured spectrophotometri- cally with a dual-wavelength spectrophotometer at 550 - 540 nm in 100 mM Hepes (pH 7.2), 10 mM KCl, 1 mM so- dium azide and 8 pM ferricytochrome c, with 33 pM duroquinol as substrate. Measurement of redox-linked proton translocation was carried out as described in 11.51.

Electrophoretic unulvsis qf the bc, complex

SDSjPAGE analysis of the bcl complex was performed according to Schiigger et al. [l]. Coomassie-blue-stained gels were scanned at 590 nm with a Camag TLC scanner I1 densi- tometer. The areas corresponding to density bands were integrated with a D-2000 Chromato-integrator (Merck- Hi tachi) .

Amino acid sequenc~ analysis

Homogeneous protein bands, resolved by SDS/PAGE, were electrotransferred to poly(viny1idene difluoride) mem- branes (Immobilon transfer) following the procedure de-

A

Core II-

FeS-

14kDa AGRPAVSASSRWL---

6.4 kDa-

B

-WLEGIRKWYYNAAGFNKLGLM---COOH 12

Fig. 2. SDSjPAGE of soluble bc, complex ufter digestion With trypsin. bcl complex (40 KM cytochrome el) was incubated with trypsin at a concentration of 1 mg/200 nmol complex for 60 min. For other experimental details, see Materials and Methods. (A) Control; (B) trypsin-digested bc, complex. The sequences of the N-terminus of the 14-kDa protein and of its proteolytic product is presented. The bands just below that of the Fe-S protein are the proteolytic products of the Fe-S protein

Core II d-

20

HA- rat io

15

10 0 30 60

time (min)

Fig. 3. Effect of trypsin digestion of polypeptides ,from soluble bc, complex on proton-motive activity of liposome-reconstituted bcl com- plex. bcl complex was incubated with trypsin as described in Fig. 2, and reconstituted into liposomes for measurement of the proton- motive activity. The densitometric bands corresponding to the various subunits were integrated and the integrals presented as percentages of the control bands

scribed by Matsudaira [18]. Sequence analysis was performed on an Applied Biosystem sequencer (mod. 477A) equipped with an on-line phenylthiohydantoin analyser.

Chemicals

Papain and horse heart cytochrome c (type VI) were from Sigma Chemicals. Trypsin and trypsin inhibitor were from Boehringer, Duroquinol from K & K Laboratories. Poly- (vinylidene difluoride) membranes (0.45 pm pore size) (Immobilon transfer) were from Millipore and sequencing- grade reagents from Applied Biosystems. All other reagents were of the highest-purity grade commercially available.

733

RESULTS

Fig. 1 shows the SDSjPAGE pattern of soluble bcl com- plex before and after a 30-min digestion with papain. As previously reported [2, 101, papain partially digested core pro- tein 11, the Rieske Fe-S protein and the 14-kDa protein. Direct amino acid sequencing of the subunits after electroblotting from gel-slabs onto poly(viny1idene difluoride) membranes, showed that papain cleaved off the first nine N-terminal resi- dues from core protein I1 and the first seven residues of the N-terminal segment from the 14-kDa protein. Digestion of the Fe-S protein was more extensive, resulting in the removal of 63 residues from the N-terminal region. The residual C-terminal segment of the Fe-S protein was released from the complex, as shown by its separation from the other subunits

A B

1 14 kDa AGRPAVSASSRWL- -

92kDa-

12 -W LEGIRKWYY NAAGFNKLGLM---looH

Fig. 4. SDSjPAGE of soluble bcl complex after digestion with trypsin. The experimental conditions are those reported in the legend to Fig. 2, except for the trypsin concentration which was 1 mg/80 nmol bcl complex protein. (A) control; (B) trypsin-digested bcl complex

A a b

during exclusion chromatography on Sephadex G-25. Analy- sis of the N-terminal sequences showed that no other polypeptide subunits were digested either by papain or by trypsin.

In Fig. 2 the digestion pattern of soluble bcl complex by trypsin is presented. Trypsin, like papain, partially digested core protein 11, the Rieske Fe-S protein and the 14-kDa pro- tein. In addition, it also digested the 6.4-kDa protein. The cleavage products of the Fe-S protein were somewhat larger [6] than that produced by papain (compare Figs 1 and 2) and were not released from the complex. Trypsin cleaved off the first 11 residues from the N-terminal segment of the 14-kDa protein. The trypsin-digested bcl complex was then reconsti- tuted into liposomes and tested for its redox and proton- motive activity. The results of these measurements, presented in Fig. 3, show that the partial digestion ofcore protein I1 and the digestion of the 6.4-kDa protein, which were selectively digested during a short incubation time, were without effect on redox proton-motive activity of the bcl-complex vesicles. Both the 14-kDa protein and the Fe-S protein were, on the other hand, digested by trypsin. As already observed for papain [2, 101, trypsin digestion caused decoupling of proton pumping which was apparently associated with proteolysis of the 14-kDa protein. A substantial decrease ofthe H'je- ratio was, in fact, observed at incubation times where a significant digestion of the 14-kDa had occurred, whereas that of the Fe-S protein was negligible. Increasing the amount of trypsin also resulted in digestion of the 9.2-kDa subunit (hinge pro- tein) 1191 (see Fig. 4) without any further effect on the proton- motive activity (not shown). Reconstitution of the bcl com- plex into phospholipid vesicles prevented trypsin digestion of core protein 11, the Rieske Fe-S protein, the 6.4-kDa and the 14-kDa proteins (see Fig. 5A). Under these conditions, the 9.2-kDa protein was only digested in the C-terminal region (Fig. 5A) and this did not affect the redox or proton-motive activities of the complex (Fig. 5B). Kinetic measurements also showed that under these conditions, no change occurred in the apparent K,,, and V,,, for cytochrome c (not shown).

9.2kDa GDPKEEEEEEEELV-- 1

-GDPKEEEEEEEELV- - - 0 30 60

time (min)

Fig. 5. Trypsin digestion of reconstituted bcl complex. Redox and proton-motive activity. bcl vesicles (2 mg protein/ml) were incubated for 60 min with trypsin at a concentration of 1 mg/80 nmol complex. For SDSjPAGE, the cleaved enzyme was extracted from phospholipids as described under Materials and Methods. (A) SDSjPAGE of control (a) and trypsin-treated bcl complex vesicles (b). N-terminal sequences of the 9.2-kDa protein and of its proteolytic product are presented. (B) e- flow (0 ) and H+ translocation (M) in trypsin-treated bcl complex vesicles. For experimental procedure see under Materials and Methods. The densitometric bands corresponding to the various subunits were integrated and the integrals presented as percentages of the control bands

134

DISCUSS I 0 N The pattern of proteolytic digestion for the soluble purified

hc, complex shows that the N-termini of core protein 11, the 14-kDa protein and the Rieske Fe-S protein, protrude at the periphery of the complex. Proteolytic digestion of the soluble complex with papain results in the removal of the first nine residues from the N-terminus of core protein 11, and the first seven from the N-terminus of the 14-kDa protein (1 1 residues are cleaved off in the case of digestion by trypsin). The finding that cleavage of these two subunits is prevented when the hc, complex is reconstituted into phospholipid vesicles before exposure to trypsin, indicates that the N-termini of these two proteins protrude into the inner space of the vesicles. It can be noted that this topological arrangement of core protein I1 and the 14-kDa protein in the reconstituted vesicles (direct controls showed the hcl complex to be oriented in liposomes with greater than 90% right-side-out orientation) is similar to that exhibited by the enzyme in mitochondria where their N-termini appear to be located at the matrix side [6].

The pattern ofpapain digestion for the soluble bc, complex shows that the Fe-S protein, which has been located at the cytosolic side of the complex [6, 20, 211, is anchored to other subunits of the complex by an N-terminal segment. It is in- teresting to note that reconstitution of the complex in phospholipid vesicles prevents proteolytic cleavage of the N-terminus of the Fe-S protein. The 63-residue N-terminal segment of the Fe-S protein thus seems to represent an amphipathic anchor for the protein to the complex at the external cytosolic side.

Papain cleavage of this segment results in release from the complex of the remaining C-terminal fragment, which holds the Fe-S center [20, 221, inhibition of the electron-transfer activity of the complex and destabilization of the antimycin- sensitive semiquinone species present in stoichiometric amounts in the complex [2, 71. The N-terminal 63-residue anchor of the Fe-S protein, with 21 hydroxyl residues, nine positively charged residues and nine acidic residues [21], can provide binding sites for the quinone and constitute a hydro- gen-bond conducting wire [23]. This is possibly involved in proton pumping in the complex, mediating H' extraction from the quinone pocket to the outer space.

The digestion of the N-terminus of the 14-kDa protein with either papain [2, 101 or trypsin (Fig. 3), appears to be associated with decoupling of Hf pumping. It can be noted that proteolytic digestion of the Fe-S protein is negligible under conditions were the H'je- ratio is lowered by more than 50% (see Fig. 3, this paper and Fig. 2 in [2]). Digestion of the 14-kDa protein was prevented when the bc, complex was reconstituted in vesicles prior to exposure to trypsin. Under these conditions, only the C-terminal region of the 9.2- kDa protein was digested and electron transfer and proton pumping were unaffected by the action of the protease. Thus, the C-terminus of the 9.2-kDa subunit protrudes at the outer surface of the liposome-reconstituted complex, and is not involved in the function of the complex. The N-terminal seg-

ment of the 14-kDa protein, which apparently protrudes into the matrix space and has two arginine and three serine resi- dues, may contribute by binding H i from the inner-aqueous phase and promoting asymmetric proton conduction to the quinone pocket for protonation of reduced species of bound quinone. This apparently plays a critical role in the pump [7]. In conclusion, the observations presented appear to be consistent with the concept that the H f pumping activity of the complex depends critically on co-operative proton conduc- tion by specific segments of constituent proteins.

REFERENCES 1. Schagger, H., Link, T. A., Engel, W. D. & Von Jagow, G. (1986)

Methods Enzymol. 126,224-237. 2. Lorusso, M., Cocco, T., Boffoli. D., Gatti, D.. Meinhardt, S.,

Ohnishi. T. & Papa, S . (1989) Eur. J . Biochem. 179, 535-540. 3. Papa, S., Lorusso, M., Gatti, D. & Cocco, T. (1987) in Bioener-

getics, structure uiid function of energji transducing systems (Ozawa, T. & Papa, S., eds), pp. 305-320, Japan Science So- ciety Press.

4. Gabellini, N. (1988) J . Bioenerg. Biomenibr. 20, 59-83. 5. Hauska, G., Nitschke, W. & Hermann. R. G. (1988) J . Bioenerg.

Biomembr. 20,211 -288. 6. Gonzales-Halphen, D., Lindorfer, M. A. & Capaldi, R. A. (1988)

Biochemistry 27, 7021 - 7031. 7. Papa, S . , Lorusso, M., Cocco, T., Boffoli, D. & Lombardo, M.

(1990) in Highlights in ubiquinone research (Lenaz, G., Barnabei, O., Rabbi, A. & Battino, M., eds), pp. 122- 135, Taylor & Francis Ltd, London.

8. Crivellone, M., Gampel, A., Muroff, I., Wu, M. & Tzagoloff, A. (1987) in Cytochrome systems. Molecular biology and bioener- getics (Papa, S., Chance, B. & Ernster, L., eds), pp. 67-77, Plenum Press, New York.

9. Weiss, H., Leonard, K. & Neupert, W. (1990) TIBS 15, 178- 180.

10. Lorusso, M., Gatti, D., Marzo, M. & Papa, S. (1985) FEBS Lett.

11. Lorusso, M., Gatti, D., Marzo, M., Boffoli, D., Cocco, T. & Papa, S. (1987) Eur. J . Biochem. 162,231 -238.

12. Lorusso, M., Boffoli, D., Cocco, T. &Papa, S. (1988) in Molecular biology of human diseases. (Gorrod, J. W., Albano, T. & Papa, S . , eds.), vol. 2, pp. 412-418, Ellis Horwood Limited, West Sussex, England.

182, 370-374.

13. Kim, C. H. & King, T. E. (1987) Biochemistry 26, 1955-1961. 14. Rieske, J. S . (1967) Methods Enzymol. 10. 239-245. 15. Papa, S., Lorusso, M., Boffoli, D. & Bellomo, E. (1983) Eur. J .

16. Leung, K. H. & Hinkle, P. (1975) J . Biol. Chem. 250,8467-8471. 17. Nalecz, M. J., Casey, R. &Azzi, A. (1983) Biochim. Biophys. Acta

18. Matsudaira, P. (1987) J . Biol. Chem. 262, 10035-10038. 19. King, T. E. & Kim, C. H. (1983) J . Biol. Chem. 258, 13543-

20. Li, Y., De Vries, S . , Leonard, K. & Weiss, H. (1981) FEBS Lett.

21. Schagger, H., Borchart, U., Machleidt, W., Link, T. A. & Von

22. Gatti, D., Meinhardt, S. W., Ohnishi, T. & Tzagoloff, A. (1989)

23. Nagle, J. F. & Morowitz, H. (1978) Proc. Nut1 Acad. Sci. USA

Biochem. 137,405-412.

724, 75 - 82.

13551.

135,271-280.

Jagow, G. (1987) FEBS Lett. 219, 161 - 168.

J. Mol. Bid. 205, 421 -435.

75, 298 - 302.