UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The ATPase complex of Escherichia coli : studies on the DCCD-binding protein 1983

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1983_A1 L65_8.pdf
UBC_1983_A1 L65_8.pdf [ 13.95MB ]
Metadata
JSON: 1.0095885.json
JSON-LD: 1.0095885+ld.json
RDF/XML (Pretty): 1.0095885.xml
RDF/JSON: 1.0095885+rdf.json
Turtle: 1.0095885+rdf-turtle.txt
N-Triples: 1.0095885+rdf-ntriples.txt
Citation
1.0095885.ris

Full Text

THE ATPase COMPLEX OF E s c h e r i c h i a c o l i : STUDIES ON THE DCCD-BINDING PROTEIN by TIP WAH[LOO B . S c , The U n i v e r s i t y of B r i t i s h Columbia, 1977 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTORATE OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES Department of Biochemistry We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 1983 Q Tip Wah Loo, 1983 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department o r by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department o f BIOCHEMISTRY The U n i v e r s i t y o f B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 10 June 1983 DE-6 (3/81) ( i i ) ABSTRACT The ATPase complex of E_. c o l l consists of two functional units. E C F ^ i s an extrinsic membrane protein having the active site(s) for ATP synthesis and hydrolysis. F Q i s intrinsic and catalyzes the reversible transfer of protons across the membrane. E C F ^ consists of five polypeptides (<*-e) ranging i n molecular weight from 13 000 - 57 000. F Q has three polypeptides (9 000, 18 000, 24 000), the smallest of which i s the dicyclohexylcarbodiimide (DCCD)-binding protein postulated to be a transmembrane pathway for proton translocation. An E C F ^ F Q complex was solubilized from the membranes of E_. c o l i with N-lauroyl sarcosine and purified by chromatography on Phenyl-Sepharose CL-4B followed by sedimentation of the enzyme at 250 000 xg for 16-17 h. The purified E C F ^ F Q complex consisted of the eight polypeptides described above, as well as associated polypeptides of molecular weights 30 000, 28 000 and 14 000. Removal of E C F ^ from the membranes of the wild-type" IS. c o l i resulted i n the membranes becoming leaky to protons so that they could not be energized. The unc mutants, E . c o l i AN382, CBT-302 and could maintain a proton gradient across the membrane i n the absence of E C F ^ . A normal DCCD-binding protein was present i n the F Q complex of each mutant. However, the 18 000 dalton polypeptide of F Q was absent in the membranes of E_. c o l i N^^, suggesting that i t was required for a functional F Q . The involvement of the 18 000 dalton polypeptide in the proton-translocating activity was also suggested by the observation that this polypeptide was absent in the E C F ^ F Q complex immunoprecipitated from trypsin-treated "stripped" vesicles, which had been reconstituted with ( i i i ) ECF^. Although these trypsin-treated "stripped" vesicles could rebind ECF^, the membranes could not be energized during ATP hydrolysis. Leakiness of the membranes to protons could be repaired by the reaction of the ECF^stripped membranes with DCCD or ECF^. Similarly, antibody raised against the DCCD-binding protein prevented this leakage of protons. The antibody also inhibited the rebinding of ECF^ to the "stripped" everted membrane vesicles. These results indicated that the DCCD-binding protein was exposed on the cytoplasmic surface of the c e l l . Attempts to show whether the DCCD-binding protein was transmembranous were not successful. Radioimmunoassay techniques were used to show In vitro, the involvement of the arginyl residue(s) of the DCCD-binding protein i n the binding of ECF^. Binding of ECF 1 to the DCCD-binding protein appeared to involve the <* and/or B subunits of ECF^. Chemical modification of the methionyl residue(s) of the DCCD-binding protein did not alter i t s capacity to bind ECF^, but destroyed the antigenic site(s) of the polypeptide. In summary, these results are consistent with the I proposed "loop" arrangement of the( DCCD-binding protein in which the polar central region of this molecule i s at the cytoplasmic surface of the c e l l membrane. (iv) TABLE OF CONTENTS TITLE-PAGE U) ABSTRACT ( i i ) TABLE OF CONTENTS ( i v) LIST OF TABLES (xi) LIST OF FIGURES ( x ii) ABBREVIATIONS (xvi ) ACKNOWLEDGEMENTS ( x i x ) ( INTRODUCTION 1. Location of the F-^FQ Complex 3 . Role of the F^F^ Complex i n Energy Transduction 4. Properties of F^ 5. Subunit Composition of F^ 5. Tightly-Bound Nucleotides.. 7. Function of the Subunits of ECF 1 The Delta (6) Subunit 8. The Epsilon (e) Subunit 9. The Gamma (Y) Subunit 1 1 . The Alpha (a) and Beta (6) Sununits ' 12. Studies on the Active Site of ECF 1 1 3 . Cross-Reconstitution Studies 1 5 . The Arrangement of the Subunits of ECF.̂  1 6 . The F Q Complex 1 9 . Solubilization of the Fn F n Complex 19. (v) Criteria for Determining the Intactness and Purity of the F ^ F Q Complex 19 The' F 1 F Q Complex . 20 The Isolation of the F Q Complex 26 Biochemical Genetics 28 The DCCD-Binding Protein 33 Identification and Isolation 33 Reconstitution of Proton Translocating Activity 35 The Amino Acid Composition 36 The Amino Acid Sequence 38 DCCD-Resistant Mutants 40 Objectives of this Study 42 MATERIALS AND METHODS 44 Chemicals 44 Maintenance of Bacterial Strains 46 Growth of Cells 46 Media 48 Preparation of Membranes 48 Preparation of EDTA-Lysozyme Spheroplasts 49 Preparation of K -Loaded Spheroplasts 50 Isolation of ECF^. 50 Purification of ECF^ on AH-Sepharose 4B 51 Purification of ECF^ and TPCK-Trypsin Treated ECF^ by Sucrose Density Gradient Centrif ugation 1 52 Preparation of ECF^-Depleted Membranes 52 Preparation of Rat-Liver Mitochondrial Membranes 53 Preparation of the Subunits of ECF1 53 (vi) TPCK-Trypsin Treated ECF1. 53 a and J3 Subunits of ECF 1 54 Solubilization of Membrane Vesicles with Detergents 54 Purification of the E C F ^ Q Complex Solubilized with N-Lauroyl Sarcosine •• • 55 Gel F i l t r a t i o n on Sepharose 6B 55 Hydrophobic-Interaction Chromatography 56 Purification of the ECF^Fg Complex by Sucrose Density Gradient Centrif ugation 57 DEAE Ion-Exchange Chromatography 58 Preparation of DCCD-Binding Protein 58 Purification of DCCD-Binding Protein •. 60 Thin Layer Chromatography . 60 Chromatography on CM-Cellulose 60 Chromatography on Sephadex LH-60 62 SDS-Polyacrylamide Gel Electrophoresis • 62 Sample Preparation 62 Depolymerization of Samples 63 Slab-Gel Electrophoresis 64 (i) Preparation of Separating Gel i 64 ( i i ) Preparation of Stacking Gel 65 Gel Electrophoresis in Tubes... . 65 Gradient Gel Electrophoresis 66 Electrophoresis 68 (i) Slab Gels 68 ( i i ) Tube Gels 68 Two-Dimensional Isoelectric Focusing Gel Electrophoresis /68 Preparation of Sample 68 ( v i i ) First-Dimension Isoelectric Focusing 69 Determination of pH Gradient 70 Second-Dimension SDS-Polyacrylamide Gel 70 Crossed Immunoelectrophoresis 71 First-Dimension Gel Electrophoresis 71 Second-Dimension Gel Electrophoresis 72 Staining and Drying Gels 72 SDS-Polyacrylamide Gels .. 72 Crossed Immunoelectrophoresis Gels 73 Purification of Goat Anti-Rabbit Immunoglobulin by Affinity Chromatography 74 Preparation of Affi n i t y Column 74 Purification of Goat Anti-Rabbit Immunoglobulin 75 Radioiodination of Goat Anti-Rabbit Immunoglobulin 75 Peptide .Mapping of CNBr-Cleaved DCCD-Binding Protein 76 One-Dimensional Thin Layer Separation 76 Two-Dimensional Thin Layer Separation , 76 Detection of Peptides 77 Chemical Modification of Membranes 78 Labelling of Membrane Vesicles of E. c o l i with [l"C]DCCD 78 Treatment of Membrane Vesicles with Phenylglyoxal 78 Chemical Modification of the DCCD-Binding Protein X 79 Hydrolysis of the DCCD-Binding Protein 79 Treatment of the DCCD-Binding Protein with Cyanogen Bromide 80 Treatment of the DCCD-Binding Protein with Performic Acid 80 Treatment of the DCCD-Binding Protein with 2,3-Butanedione or Phenylglyoxal 81 Treatment with Proteases 82 ( v i i i ) ECFj^ and DCCD-Binding Protein 82 Preparation of Antigens f o r Immunization 82 ECF± 82 DCCD-Binding Protein. 83 Immunization of the Rabbit 83 Bleeding the Rabbit 84 Separation of Serum 85 P a r t i a l P u r i f i c a t i o n of Immunoglobulins 85 Binding of ECF 1 to Membrane Vesicles 86 Assays 87 Determination of Protein K. , 87 Determination of ATPase A c t i v i t y 88 ( i ) Rapid Assay 88 ( i i ) Slow Assay 88 Substrate Oxidation-Dependent Quenching of Fluorescence of 9-Aminoacridine 89 Measurement of Proton Conduction i n K +-loaded Membrane V e s i c l e s . . . 90 Determination of Cytochrome Content 90 Determination of Catalase A c t i v i t y 91 Determination of Ra d i o a c t i v i t y i n Gel S l i c e s 91 So l i d Phase Radioimmune Assays 92 RESULTS 95 Part I P u r i f i c a t i o n of the E C F ^ F Q Complex 95 Sel e c t i o n of an E. c o l i S t r a i n 95 S o l u b i l i z a t i o n of the E C F ^ Complex 99 Sel e c t i o n of Detergent 99 \ ( i x ) E f f e c t of Detergents on the Membrane-Bound ATPase A c t i v i t y . . . . 103 S t a b i l i t y of the S o l u b i l i z e d Enzyme 105 Molecular Size of the Detergent-Solubilized Enzyme 105 Gel F i l t r a t i o n on Sepharose 6B or Bio-Gel A-0.5m 105 P u r i f i c a t i o n of the E C F - ^ F Q Complex by Hydrophobic- Inte r a c t i o n Chromatography 113 Other Hydrophobic-Interaction Resins 118 Further P u r i f i c a t i o n of the E C F ^ F Q Complex 1 120 Intactness of the E C F ^ Q Complex 132 R e p r o d u c i b i l i t y of P u r i f i c a t i o n on Phenyl Sepharose CL-4B 137 Comparison of the Gel Electrophoresis and Protein-Detection Systems 139 Part II Studies on Mutants of E. c o l i Defective i n Proton-Translocating A c t i v i t y 143 S e n s i t i v i t y of the Membrane-Bound ATPase A c t i v i t y to I n h i b i t i o n by DCCD i n the PresenceJ^Cations 145 Measurement of the Proton Gradient using 9-Aminoacridine 145 L a b e l l i n g of Membranes of E. c o l i with [ ' " C J D C C D 151 P u r i f i c a t i o n of the DCCD-Binding Protein 152 Amino Acid Composition of the DCCD-Binding Protein 161 Analysis of the Membranes of E. c o l i by Two-Dimensional I s o e l e c t r i c Focusing Gel Electrophoresis 161 Part I I I Studies on the DCCD-Binding Protein of the Fo Complex of E_. c o l i . •> 168 E f f e c t of Antiserum to the DCCD-Binding Protein on the Energization of the Membrane of Urea-Stripped Everted Membrane V e s i c l e s 168 E f f e c t of Antiserum to the DCCD-Binding Protein on the Binding of ECF-^ to Urea-Stripped Everted Membrane Vesicles 170 E f f e c t of Antiserum to the DCCD-Binding Protein on the Proton Permeability of the Right-Side Out Vesicles of E_. c o l i . .171 I (x) Effect of Antiserum to the DCCD-Binding Protein on the Energization of the Membrane of Native Everted Membrane Vesicles 179 Energization of the Membrane of Trypsin-Treated Urea-Stripped Everted Vesicles 181 i Binding of ECF^ to Protease-Treated Membrane Vesicles 185 Effect of DCCD on the ATPase Activity of the ECFj^ Bound to Trypsin-Treated Vesicles 188 Immunoprecipitation of the ECF-^FQ Complex with Antiserum 191 Detection by Solid Phase Radioimmune Assay of the Reaction of Antibody with Membrane Vesicles 198 Reaction Site(s) for the Antibody on the DCCD-Binding Protein..... 203 Binding of ECF-̂  by Purified DCCD-Binding Protein 205 Reaction Site(s) on ECF-̂  for the DCCD-Binding Protein 209 Effect of Chemical Modification of the DCCD-Binding Protein on i t s Reaction with ECFj 214 Effect of Phenylglyoxal on the Binding of ECF-̂  to Urea- Stripped Everted Vesicles 219 f DISCUSSION. 225 Purification of the E C F ^ Q Complex ; 225 Some Mutants of E,. c o l i Defective in Proton Translocation 229 Orientation of the DCCD-Binding Protein in the Membrane 234 Interaction of the .DCCD-Binding Protein with ECF1 237 REFERENCES ; 243 f (xi) LIST OF TABLES Table 1. Polypeptide Composition of F^Fg-ATPase Complexes from Various Sources 22 2. Properties of Various Preparations of Bacterial ETJFQ Complexes 25 3. Properties of Various Preparations of Bacterial FQ Complexes... 27 4. Polypeptides Coded by the "unc" Genes 31 5. Amino Acid Composition of the DCCD-Binding Protein from Mitochondria, Chloroplast and Bacteria 37 6. Properties of DCCD-Resistant Mutants of E. c o l i 41 7. Bacterial Strains used in'this Study 47 8. Specific Activity of the Membrane-Bound ATPase of Different Bacterial Strains 96 9. Solubilization of the Membrane-Bound ATPase Activity of E_. c o l i with Sodium Cholate 102 10. Effect of Detergent on the Membrane-Bound ATPase Activity of E. c o l i . 104 11. Estimation of the Molecular Weight of the Solubilized E C F ^ Q Complex by Gel F i l t r a t i o n Chromatography I l l 12. Some Properties of the unc Mutants of IS. c o l i used in this Thesis 144 13. Amino Acid Composition of the DCCD-Binding Protein from Different Strains of E_. c o l i . 163 14. Energization of the Membrane of Trypsin-Treated Everted Membrane Vesicles of E. co l i 182 15. Subunit Composition of FQ from Various ECF^Fo and Fo Purifications 227 (xii) LIST OF FIGURES Figure — — — — \ t 1. Schematic representation of oxidative phosphorylation and generation of a proton gradient 2 2.. Three models for the arrangement of the subunits in the Fj-ATPase of E. c o l i . 18 3. Amino acid sequences of the DCCD-Binding Protein from Neurospora crassa bovine heart, Saccharomyces cerevisiae, spinach chloroplasts, Mastigocladus laminosus, Escherichia c o l i , and the thermophilic bacterium PS-3 39 4. Effect of cations on the membrane-bound ATPase activity of different strains of E_. c o l i . 97 5. Solubilization of the membrane-bound ATPase activity of E_. c o l i by various detergents.... 101 6. Stability of solubilized ATPase activity on storage at 4°C... 106 7. Chromatography of the detergent-solubilized ATPase complex on Sepharose 6B in the presence of various detergents 108 8. Effect of DCCD on the detergent-solubilized ATPase activity... 112 9. Chromatography of the detergent-solubilized ATPase complex on Phenyl-Sepharose CL-4B 115 10. Purification of the E C F ^ FQ complex by sucrose gradient centrif ugation 121 11. SDS-polyacrylamide gel electrophoresis of the E C F ^ F Q complex purified by chromatography on Phenyl-Sepharose CL-4B and sucrose gradient centrif ugation 125 12. SDS-polyacrylamide gel electrophoresis of the E C F ^ F Q complex obtained after chromatograpy on Phenyl-Sepharose CL-4B and sucrose gradient centrif ugation 128 13. SDS-polyacrylamide gel electrophoresis of the E C F ^ F Q complex obtained by chromatography on Phenyl Sepharose J CL-4B and sedimentation at 250 000 xg for 16-17 h 131 14. Effect of DCCD on the ATPase activity of the E C F ^ Q complex 134 15. SDS-polyacrylamide gel electrophoresis of ECF^ and ECF^Fo complex labelled with ['"CJDCCD 135 ( x i i i ) 16. SDS-polyacrylamide gel electrophoresis of the ECF^FQ complex obtained after chromatography on Phenyl-Sepharose CL-4B: Reproducibility of the purification 138 17. Comparison of the SDS-gel electrophoresis and protein- detection systems 140 18. Sensitivity of the membrane-bound ATPase activity to inhibition by DCCD in the presence of cations 146 19. Measurement of the proton gradient in everted membrane vesicles using the fluorescent dye, 9-aminoacridine 148 20. Effect of stripping everted membrane vesicles from the parent (E. c o l i WS1) and mutant (E. c o l i N J 4 4 ) on the energization of the membrane 150 21. SDS-polyacrylamide gel electrophoresis of,[x"C]DCCD- labelled membranes and of ether-precipitated proteins of chloroform-methanol extracts of the labelled membranes of E. c o l i 153 22. Thin layer chromatography of the DCCD-binding protein 155 23. Chromatography of the DCCD-binding protein on CM-cellulose.... 157 24. SDS-polyacrylamide gel electrophoresis of DCCD-binding protein obtained by chromatography on CM-cellulose 160 25. Chromatography of the CM-cellulose-purified DCCD-binding protein on Sephadex LH-60 162 26. Two-dimensional thin-layer chromatography of cyanogen ' bromide cleaved fragments of the DCCD-binding protein of E. c o l i CBT-302 164 27. Two-dimensional isoelectric focusing gel electrophoresis of membranes of parent (WS1) and mutant ( N J 4 4 ) strains of E. c o l i 166 28. Effect of the antiserum to the DCCD-binding protein and of ECF^ on the ascorbate-oxidation-dependent quenching of fluorescence of 9-aminoacridine by urea-stripped everted membrane vesicles 169 29/. Effect of antiserum to the DCCD-binding protein on the binding of ECF^ to urea-stripped everted membrane vesicles.... 172 30. Crossed Immunoelectrophoresis of antiserum to the DCCD- binding protein 175 31. Schematic representation of the proton-pathway provided by the "right-side out" vesicles of E_. c o l i DL-54 176 (xiv) , 32. Effect of DCCD and of antiserum to the DCCD-binding protein on the proton permeability of "right-side out" membrane vesicles 177. 33. Effect of DCCD and of antiserum to the DCCD-binding protein on the proton-permeability of "right-side out" membrane vesicles of E. c o l i DL-54 178. 34. Effect of antiserum to the DCCD-binding protein on the energization of untreated everted membrane vesicles 180. 35. Effect of antiserum to the DCCD-binding protein on the ,ascorbate-oxidation-dependent quenching of fluorescence of 9-aminoacridine by trypsin-treated everted membrane vesicles 184. 36. Binding of ECF^ to trypsin-treated everted membrane vesicles 186. 37. Binding of, ECF^ to Staphylococcus aureus Vs protease- treated everted membrane vesicles 189. 38. _ Effect xof DCCD on the ATPase activity of the ECF^ bound to trypsin-treated everted membrane vesicles 192. 39. SDS-polyacrylamide gel electrophoresis of the ECF^FQ complex immunoprecipitated with antiserum 193. 40. Two-dimensional gel electrophoresis of the ECF^Fo complex' obtained by immunoprecipitation with antiserum to ECF^ 196. 41. Titration of the DCCD-binding protein with antiserum to this polypeptide 199. 42. Inhibition of antibody binding to immobilized DCCD-binding protein by membrane vesicles of E_. c o l i , PS3 and rat liver mitochondria, and by phospholipid vesicles 200. 43. Inhibition of antibody binding to immobilized DCCD-binding protein by protease-treated or chemically-modified DCCD- binding protein......... 204. 44. Inhibition of antibody binding to immobilized DCCD-binding protein by protease-treated or chemically-modified everted membrane vesicles 206. 45. Schematic representation of the radioimmune binding assay 207. 46. Binding of ECFi to the DCCD-binding protein ... 208. 47. Effect of ECFi on the binding of anti-DCCD-binding protein serum to the DCCD-binding protein and the effect of DCCD- binding protein on the binding of anti-ECF^ serum to ECF^ 210. (ix) Effect of Detergents on the Membrane-Bound ATPase Activity.... 103 Stability of the Solubilized Enzyme 105 Molecular Size of the Detergent-Solubilized Enzyme 105 Gel F i l t r a t i o n on Sepharose 6B or Bio-Gel A-0.5m 105 Purification of the ECF^FQ Complex by Hydrophobic- Interaction Chromatography 113 Other Hydrophobic-Interaction Resins 118 Further Purification of the ECF^FQ Complex 120 Intactness of the ECF-JFQ Complex 132 Reproducibility of Purification on Phenyl Sepharose CL-4B 137 Comparison of the Gel Electrophoresis and Protein-Detection Systems 139 Part II Studies on Mutants of E_. c o l i Defective in Proton-Translocating Activity 143 Sensitivity of the Membrane-Bound ATPase Activity to Inhibition by DCCD in the Presence of Cations 145 Measurement of the Proton Gradient using 9-Aminoacridine 145 Labelling of Membranes of E. c o l i with [^CJDCCD 151 Purification of the DCCD-Binding Protein 152 Amino Acid Composition of the DCCD-Binding Protein 161 Analysis of the Membranes of E_. c o l i by Two-Dimensional Isoelectric Focusing Gel Electrophoresis 161 Part III Studies on the DCCD-Binding Protein of the FQ Complex of E_. co l i 168 Effect of Antiserum to the DCCD-Binding Protein on the Energization of the Membrane of Urea-Stripped Everted Membrane Vesicles 168 Effect of Antiserum to the DCCD-Binding Protein on the Binding of ECF^ to Urea-Stripped Everted Membrane Vesicles 170 Effect of Antiserum to the DCCD-Binding Protein on the Proton Permeability of the Right-Side Out Vesicles of E_. c o l i 171 (xv) 48. SDS-polyacrylamide gel electrophoresis of subunits of ECF^.... 211 49. Binding of ECF^ to the DCCD-binding protein: Effect of protease treatment 212 50. Binding of DCCD-binding protein to subunits of ECF^ 215 51. Effect of chemical modification of the DCCD-binding protein on the binding of ECF-̂  216 52. Effect of modification of the arginyl residue(s) of the DCCD-binding protein on the binding of ECF-L 217 53. Binding of ECF^ to phenylglyoxal-treated vesicles 220 54. SDS-polyacrylamide gel electrophoresis of the DCCD-binding protein of E. c o l i labelled with [7 1"C]phenylglyoxal 222 55. Amino acid sequence of the DCCD-binding protein of E. c o l i 238 (xvi) ADP AH-Sepharose 4B Aminoxid WS-35 ATP Ammonyx Lo Brij 35 BSA Chloramine T CMC CM-cellulose DCCD DEAE-Sepharose CL-6B DNase DTT EDTA EGTA F-L -ATPase F-^Fg-ATPase complex HEPES ABBREVIATIONS Adenosine-51-diphosphate. Amino Hexyl Sepharose 4B. Acyl aminopropyldimethylaminoxide. Adenosine-5'-triphosphate. Lauryl dimethylaminoxide. Polyoxyethylene (23) lauryl ether. Bovine serum albumin. N-chloro-4-methylbenzenesulphonamide, sodium salt. C r i t i c a l micellar concentration. Carboxymethyl cellulose. N,N'-dicyclohexylcarbodiimide. Diethylaminoethyl-Sepharose CL^6B. Deoxyribonuclease. Dithiothreitol. (Ethylenedinitrilo)-tetraacetic acid. [Ethylenebis(oxyethylenenitrilo)]-tetraacetic > acid. ATP phosphohydrolase (catalytic portion of the proton-translocating adenosine triphosphatase); CF l 9 chloroplast F1-ATPase; ECFj, E_. c o l i Fj^-ATPase; MF-̂ , mitochondrial Fj-ATPase; TF 1, F-j-ATPase from the thermophile, PS3. Proton-translocating adenosine triphosphatase; ECF-J^FQ, _E. c o l i F-LFQ complex; TFJFQ, F^FQ complex from the thermophile, PS3. N-2-hydroxyethylpiperazine N'-2-ethanesulphonic acid. HPLC - High performance (pressure) liquid chromatography. (xvii) Lubrol 17A-10 Lubrol PX Lubrol WX MOPS NADH NAP4-ADP NAP4-ATP Nbf-Cl NEM-Hg Ninhydrin Nonidet P-40 Pi PBS PEG PMS PMSF psi RNase SDS Sephadex LH-20 Sephadex LH-60 TAMM TCA Polyoxyethyleneglycol (n=?) cetyl-stearyl alcohol. Polyoxyethyleneglycol (n=?) cetyl-stearyl alcohol. Polyoxyethyleneglycol (17) cetyl-stearyl alcohol. Morpholinopropanesulphonic acid Reduced nicotinamide adenine dinucleotide. 3'-0-(4-[N-(4-azido-2-nitrophenyl) amino] butyryl) ADP. 3'-0-(4-[N-(4-azido-2-nitrophenyl) amino] butyryl) ATP. 4-chloro-7-nitrobenzofuran. Mercuriated N-pyrrolo-isomaleinimide. 1,2,3-triketohydrindene hydrate. Polyoxyethyleneglycol (9) p-t-octylphenol. inorganic phosphate. Phosphate-buffered saline (0.137 M NaCl, 2 mM KC1, 1.47 mM KH2P04, 8.09 mM Na2HP04, pH 7.5). Polyethyleneglycol. Phenazine methosulphate. Phenylmethysulphonylfluoride. J Pounds per square inch. Ribonuclease. Sodium dodecyl sulphate. Hydroxypropyl Sephadex G-25. Hydroxypropyl Sephadex G-50. Tetrakis (acetoxymercuri) methane. Trichloroacetic acid. ( x v i i i ) TEMED TPCK-trypsin Tris Triton X-100 Triton X-114 Tween 60 Tween 80 A pH - N,N,N',N'-tetramethylethylenediamine. - Trypsin treated with L-(tosylamido 2-phenyl) ethyl chloromethyl ketone. - Tris (hydroxymethyl)-aminoethane. - Polyoxyethyleneglycol (9-10) p-t-octylphenol. - Polyoxyethyleneglycol (7-8) p-t-octylphenol. - Polyoxyethyleneglycol (20) sorbitol monostearate. - Polyoxyethyleneglycol (20) sorbitol monooleate. - Membrane potential. - Difference i n pH across the membrane. - Electrochemical potential difference of protons across the membrane, proton-motive force. (xix) ACKNOWLEDGEMENT S I thank Dr. P.D. Bragg for his advice, patience and understanding throughout my graduate studies and during the writing of this thesis. The completion of this thesis would not have been possible without the help of the many members of the Department of Biochemistry. In particular, I thank Drs. R.S. Molday, R. Barton and J.C. Brown (Physiology) for access to the equipment in their laboratories. I especially thank Mr. D.J. Mackenzie for help i n setting up the radioimmunoassay experiments and for the photography. I am very grateful to my colleagues, Mr. David M. Clarke, Ms. Cynthia Hou and Dr. Helga Stan-Lotter for providing a pleasant atmosphere which was conducive to research and collaboration. I am also ' grateful to the Medical Research Council of Canada for supporting this work. The typing of this manuscript by Ms. Judith Smith is greatly appreciated. Finally, I would like to acknowledge the support of my family during this project. 1. INTRODUCTION The r e v e r s i b l e , p r o t o n - t r a n s l o c a t i n g adenosine triphosphatase (F^F^-ATPase complex) plays a major r o l e i n the energy-transduction r e a c t i o n s of the c e l l . S t r u c t u r a l l y s i m i l a r forms of the enzyme are found i n the membranes of eukaryotes and prokaryotes (1-4). The enzyme c a t a l y z e s the s y n t h e s i s of ATP by o x i d a t i v e or photo-phosphorylation, and the h y d r o l y s i s of ATP (5) according to the r e a c t i o n . nH?" + ATP + H o0 ADP + P. + nH + i n 2 l out The ATPase complex c o n s i s t s of two f u n c t i o n a l u n i t s , F^ and FQ. F^ i s a complex of e x t r i n s i c membrane p r o t e i n s c o n t a i n i n g the a c t i v e s i t e ( s ) of ATP s y n t h e s i s and h y d r o l y s i s (6-8). The F i n t e r a c t s w i t h FQ, which i s a complex of i n t e g r a l membrane p r o t e i n s not having ATPase a c t i v i t y . The FQ complex i s thought to extend through the membrane and f u n c t i o n s as a pathway f o r the r e v e r s i b l e t r a n s l o c a t i o n of protons through the membrane (9,10,14). The e x i s t e n c e of r e v e r s i b l e , p r o t o n - t r a n s l o c a t i n g pumps i s an important p o s t u l a t e of M i t c h e l l ' s chemiosmotic hypothesis on the mechanism of o x i d a t i v e phosphorylation (11). According to t h i s hypothesis, the components of the e l e c t r o n t r a n s p o r t c h a i n are arranged i n the membrane such that there i s v e c t o r i a l t r a n s l o c a t i o n of protons across the membrane during r e s p i r a t i o n or light-dependent e l e c t r o n f l o w ( F i g 1) (5,12,13). The v e c t o r i a l t r a n s l o c a t i o n of protons generates a proton-motive f o r c e (Au +), which c o n s i s t s of two components: the membrane p o t e n t i a l (A\|>) , generated as a r e s u l t of charge s e p a r a t i o n , and the chemical c o n c e n t r a t i o n g r a d i e n t , (ZApH). The r e l a t i o n i s gi v e n by the f o l l o w i n g equation: 2. FIG 1: Schematic representation of oxidative phosphorylation and generation of a proton gradient. "mH+" and "nH +" represent protons which are translocated, but whose exact number has not been agreed on. Ajfjjf = Aip-ZApH (where Z = ^ — = 59 mV at 25°C) The energy stored within this proton gradient i s used to drive other energy-requiring processes. For example, the synthesis of ATP i s coupled to the return of protons down the gradient through,the F^Fg-ATPase complex. Similarly, the proton gradient can be regenerated through the hydrolysis of ATP. A stoichiometry of two or three protons translocated per molecule of ATP hydrolyzed or synthesized has been measured (5,15). Some microorganisms such as Streptococcus faecalis, Streptococcus lactis (2) and the st r i c t anaerobe Clostridium pasteurianum (16), growing in the presence of a limiting amount of glucose or in the absence of oxy- gen or other terminal electron acceptors, cannot carry out oxidative phosphorylation. This i s because their respiratory chains are either non-functional or non-existent. These organisms rely solely on the ATP produced by substrate-level phosphorylation (glycolysis) and the f-^FQ complex to generate a proton-motive force, which i s capable of driving other energy-requiring processes. LOCATION OF THE FJFQ COMPLEX ATPase activity was shown to be localized i n the plasma membrane of Streptococcus faecalis (17) , Bacillus cereus and Escherichia c o l i (18). Ferritin-labelled antibody against purified F^-ATPase was used to show that the F^ was located on the inner surface of the bacterial plasma membrane (19). Negatively-stained preparations of everted membrane vesi- cles of E_. c o l i revealed the presence of "knobs", each with a diameter of about 100 A, attached to the inner surface of the plasma membrane via a "stalk". A similar morphology exists i n the submitochondrial particles and 4. in the thylakoid membranes (6,20). ROLE OF THE FjFp COMPLEX IN ENERGY TRANSDUCTION The role of the F^F^ complex of 15. c o l i i n energy-transduction reactions has been demonstrated through reconstitution studies and the isolation of mutants. The F^ component of the F^F^ complex can readily be released from the everted membrane vesicles of E_. c o l i by washing them in buffers of low ionic strength containing EDTA (21-23,33). The ECF^depleted mem- branes no longer exhibit energy-transducing properties such as respiration- induced proton uptake, ATP-driven and respiration-driven transhydrogenation of NADP+ by NADH, or oxidative phosphorylation. Restoration of these activities to normal levels can be achieved by the addition of purified 2+ ECF^-ATPase to the depleted membranes i n the presence of Mg or high ionic strength buffer (21-25,33). These results suggest that the loss of the energy-transducing properties of ECF^-depleted membranes i s due to an increased permeability of the membranes to protons upon the removal of ECF^, such that the proton-gradient cannot be built up or maintained by respiration. Proton impermeability can also be restored i n the ECF^-depleted membranes by treatment with either ECF^, isolated from an unc A mutant, which has no ATP hydrolytic activity or with DCCD. DCCD i s a potent inhibitor of the ATPase activity of the F ^ F Q complex and under appropriate conditions i t reacts with a specific component of F Q and inhibits proton-translocating activity (discussed under "The DCCD-binding protein"). In the E. c o l i mutants, DL-54 and NR-70, the F, i s either defective or absent (26,27). Membrane vesicles or whole c e l l preparations of these mutants show decreased levels of respiration-driven transhydrogenation and transport of proline than i n the corresponding parent-strains. K +-loaded spheroplasts of E_. c o l i DL-54 also show a larger change in the pH of the external medium upon addition of valinomycin, than i s seen with the wild- type strain (28). The increased permeability of these mutant membranes to protons i s due to an exposed FQ since the addition of DCCD or ECF^ causes the membranes to become less permeable to protons. These results suggest (24) that (i) the ECF^ is responsible not only for the energization of the membrane through ATP hydrolysis, but also for maintaining the impermeability of the membranes to protons by inter- acting with FQ, and ( i i ) the FQ acts as a proton-specific channel or pore. ( The concept of FQ being a proton-specific pore i s complicated by the observations that i t s proton-conducting properties in E. c o l i DL-54 and NR-70 varies significantly when the growth conditions are varied i n aerobic culture (29-32). PROPERTIES OF F Subunit Composition of F^ The molecular weight of ECF^ has been determined by a variety of methods, with the following values: gel f i l t r a t i o n , 360 000 - 390 000 (33); sedimentation equilibrium, 360 000 - 390 000 (34); laser light scattering, 362 000 (35); small angle X-ray scattering, 35 8000 (36); sedimentation coefficient, diffusion coefficient and partial specific volume, 350 000 (36); light scattering, 345 000 (36) and small angle neutron scattering, 315 000 (37). Three problems e x i s t which make i t d i f f i c u l t to determine the correct molecular weight of ECF^ (6). F i r s t , there i s the tendency of the enzyme to d i s s o c i a t e , r e s u l t i n g i n the loss of some subunits of ECF^. Secondly, the extent of the loss of the S subunit during p u r i f i c a t i o n of ECF^ i s variable and t h i s i s dependent to some extent on the choice of the p u r i f i c a t i o n procedure. F i n a l l y , the amount of $ subunit i n ECF^ preparations i s also dependent on the source, f o r the amount of 6 subunit i n F^ preparations i s o l a t e d from _E. c o l i s t r a i n s i s more variable than that from the E_. c o l i ML s t r a i n s . These factors cause the molecular weight of the ECF^ to be underestimated. The actual molecular weight of ECF^ may be c l o s e r to the molecular weight of the more stable TF^, which has been reported to be 380 000 (38). The F^ from a l l sources (with the exception of Clostridium pasteurianum and L a c t o b a c i l l u s casei) generally contain f i v e d i f f e r e n t polypeptides (a, J3, Y, 6 and e) (4). MF^ has an a d d i t i o n a l subunit (Table 1) which i s a natural i n h i b i t o r of the ATPase a c t i v i t y , but i s unnecessary f o r ATP synthesis (24). The F^ from C_. pasteurianum (16) consists of only three d i f f e r e n t subunits with molecular weights of 65 000, 57 000 and 43 000, while that from L_. casei (39) contains only one poly- peptide of 43 000 daltons. The molecular weights of each of the subunits of ECF^, TF^, CF^ and MF^ are very s i m i l a r (40), with those of ECF^ reported to be a, 56 800; B, 51 800; Y, 32 000; 6, 20 700; and e, 13 200 (7,41). These are i n good agreement with the molecular weights determined from the DNA sequence (42) of each polypeptide: a , 55 264; 6, 50 157; Y, 34 100; 6, 19 310; and e, 14 194. The r a t i o of these subunits i n F., remains c o n t r o v e r s i a l . Most of the data on F^ from bacteria and yeast mitochondria support an a^B^ySe stoichoimetry (8,41,43-44,72,155). Although the data on the F̂ ^ from mammalian mitochondria and chloroplast suggest a2^2^2^2 e2 a n c * a 2 ^ 2 Y ^ 6 ( l - 2 ) stoichiometries, respectively (45-48); more recent estimates support the a 3 J 3 3 Y 5 e s t o i c h i o m e t r y (4"9). Tightly-Bound Nucleotides CF^, ECF^, MF^ and TF^ a l l contain non-covalently bound nucleotides, which cannot be removed even a f t e r extensive p u r i f i c a t i o n of the enzyme (50,51,73). In E_. c o l i , three molecules of tightly-bound nucleotides per molecule of F^ were detected. The only nucleotides found to be present were ATP and ADP. D i f f e r e n t molar r a t i o s of these nucleotides were reported, but t h i s could be a t t r i b u t e d to the d i f f e r e n t procedures used f o r the preparation of ECF^ and f o r nucleotide detection. The function(s) of these tightly-bound nucleotides has not been established. The turnover rate of the tightly-bound ATP i s too slow f o r i t to be involved as an intermediate i n oxidative phosphorylation (52). It i s more l i k e l y that these tightly-bound nucleotides are present In the "regulatory" s i t e rather than i n the a c t i v e s i t e . At present, there i s very l i t t l e information about the regulation of the enzyme In vivo. FUNCTION OF THE SUBUNITS OF ECF^ Understanding the function and arrangement of the subunits of the ECF^ i s e s s e n t i a l to the determination of the mechanism of oxidative phosphorylation. Several approaches to the study of the function and properties of the subunits of ECF^ have been used. These have included chemical modification of s p e c i f i c residues, a f f i n i t y l a b e l l i n g , 8. immunological techniques, genetics, and subunit isolation and holoenzyme reconstitution. Recently, methods have been developed for the dissociation and isolation of the subunits of ECF^ in non-denatured form such that a functional ATPase could be reconstituted from i t s subunits (6). The Delta (6) Subunit ECF^ preparations can be obtained which are deficient in the 6 subunit (53-57). This 6-deficient, four-subunit enzyme (i.e. a^ByrO, has an ATPase activity equal to that of the native, five-subunit enzyme. However, the former i s incapable of reconstituting respiration-driven and ATP-driven transhydrogenase activity in F^-depleted membranes. Also, ATPase activity was not detected i n the depleted membranes, which had been reconstituted with the 6-deficient ATPase. Smith and Sternweiss (58,59) have purified the 6 subunit, by treatment of ECF 1 with 50% pyridine followed by chromatography of the fraction containing the 6 and e subunit on Sephadex G-75. Addition of the purified <5 subunit to the <S-deficient enzyme restored the membrane- binding capability of the enzyme. Besides rebinding, ATP-driven transhy- drogenase activity and the formation of ATP by oxidative phosphorylation were restored in these depleted membranes (57). These results suggest that the 6 subunit i s involved in the binding of ECF^ to the membrane and that the binding of 6 restores the ATPase to i t s native state. The role of the <5 subunit i n attaching the F^ to the membrane has also been confirmed for (60), TF.̂  (61) and F 1 from Streptococcus faecalis (62). 9. The E p s i l o n (e) Subunit i By passing the $ - d e f i c i e n t , four-subunit enzyme ( i . e . a^Q^ye) through an a f f i n i t y column c o n t a i n i n g immobilized a n t i b o d i e s to the e subunit, a three-subunit enzyme ( i . e . ot^B^Y) which i s d e f i c i e n t i n the e subunit has been obtained (63). This three-subunit enzyme had an ATPase a c t i v i t y which was 10-15% g r e a t e r than that of the n a t i v e , f i v e - subunit enzyme. The p u r i f i e d 6 or e subunit could bind to the t h r e e - subunit enzyme to form four-subunit complexes ( i . e . a^B^Y^ o r a^B^Ye). Neither of these complexes could bind ECF^-depleted membranes and r e c o n s t i - tute o x i d a t i v e phosphorylation a c t i v i t y unless both 6 and e were present. This suggests that both 6 and e are i n v o l v e d i n binding ECF^ to the membrane and that the " s t a l k s " seen-, i n n e g a t i v e l y - s t a i n e d preparations of everted E. c o l i membrane v e s i c l e s may be composed of the 6 and e subunits ( 6 ) . Thus, i t might be expected that the 6 and e subunits should i n t e r a c t d i r e c t l y w i t h the FQ component i n the membrane. However, the b i n d i n g of e i t h e r p u r i f i e d 6 or e subunit to ECF^-depleted membranes could not be detected by using a n t i b o d i e s r a i s e d against e i t h e r of these subunits. Furthermore, a mixture of p u r i f i e d Y, $ and e subunits d i d not reduce the proton p e r m e a b i l i t y of the depleted membranes as measured by r e s p i r a t i o n - d r i v e n transhydrogenase (6,63). By c o n t r a s t , both the 6 and e subunits of TF^ could bind to the TFQ which had been r e c o n s t i t u t e d i n t o liposomes (61) and to t h i s TFQ - 6e complex could be bound e i t h e r the p u r i f i e d Y subunit or a 3 subunit enzyme of TF^, c o n t a i n i n g a 3B 3Y. A second f u n c t i o n has been assigned to the e subunit. I t i s a non-competitive i n h i b i t o r of the ATPase a c t i v i t y of p u r i f i e d ECF-. 10. Addition of the purified e subunit to either the five-subunit enzyme (a^B^Y^e), t n e four-subunit enzyme (a^B^Y^) or the three-subunit enzyme ( a ^ Y ) resulted i n 70-90% inhibition of the ATPase activity (58,64). The inhibition of the activity of the five-subunit enzyme can be explained by the dissociable nature of the e subunit. The ATPase activity of an ECF^ preparation was increased by more than four-fold following dilution of the enzyme preparation. This suggested that by diluting the ECF^ preparation, the endogenous e subunit dissociated from the enzyme and was i n equilibrium with the ECF^-e complex. Addition of anti-e-serum to an ECF^ preparation also stimulated the activity of the enzyme by more than two-fold. Immunoprecipitation of a reconstituted five-subunit enzyme, in which the e subunit was labelled with 1 2 5 1 , with anti-e-serum, resulted i n quantitative precipitation of radioactivity (65). However, no ATPase activity was detected i n the precipitate. The anti-e-serum, therefore promoted the dissociation of e from the F^ either directly or by the removal of free e which was i n equilibrium with the ECF^-e complex. It appears that the relationship of the e subunit to ECF^ i s similar to that of the mitochondrial ATPase inhibitor protein to MF^, i n that both may be involved i n determining whether the ATPase i s operating i n the direction of ATP hydrolysis or synthesis. It has been suggested (66,67) that substrate oxidation or a low molar ratio of ATP/ADP tends to decrease the interaction between the inhibitor and MF^. The kinetic data (64) on ECF^ also suggest that ATP accelerates the release of the e subunit from ECF^, whereas ADP prevents this activation, perhaps by stabilizing the interaction of e with ECF^. The e subunit of ECF^ differs from the mitochondrial inhibitor protein (6) i n that (i) the mitochondrial ATPase inhibitor protein i s not 11. necessary for attaching MF^ to the membrane, and ( i i ) the mitochondrial ATPase inhibitor protein inhibits the ATPase activity of the membrane-bound enzyme, whereas an excess of e subunit has no effect on this activity i n J£. c o l i membranes. The Gamma (Y) Subunit The activity of purified ECT^ i s stimulated by up to 100% by brief treatment of the enzyme with trypsin (41). Examination of the trypsin- treated enzyme on SDS-polyacrylamide gels revealed that the 5 and e subunits, and to a small extent the Y subunit, were completely destroyed (53,63). The resulting enzyme following trypsin-treatment i s essentially a two subunit enzyme (i.e. a2Q3^ with three copies of each subunit. Dunn and coworkers (68) have shown that trypsin-treatment removes the f i r s t 15 residues from the NI^ terminus of each of the a subunits and that a fragment of the Y subunit of 10 000 dalton remained bound to the enzyme. Although the purified e subunit inhibited the ATPase activity of the three-subunit enzyme (i.e. a^B^Y), i t had no effect on the activity of the trypsin-treated enzyme. Incubation of the trypsin-treated enzyme following cold-dissociation with a Y - r i c h fraction prepared from native ECF^ restored the sensitivity of the trypsin-treated enzyme to inhibition by the e subunit (69). These results suggest that the Y subunit i s required to bind the e subunit to ECF^ and is supported by the observation (70) that there i s a high a f f i n i t y interaction between purified Y and e subunits, i n a 1:1 ratio. This interaction i s specific since interactions of the purified e with either purified a or Q subunits was not detected. Also, the addition of purified Y subunit to a purified e preparation prevented the latter from inhibiting the ATPase activity of a three-subunit 12. enzyme (a^B^)' Thus the y subunit not only binds the e subunit but i t , or a portion of i t , also holds the a and fi subunits together. The Alpha (a) and Beta (13) Subunits ECF^ has the property of being cold-labile and cold-dissociable. Taking advantage of this, Vogel and Steinhart (77) dissociated the enzyme by freezing i t in a solution of high ionic strength. Through ion-exchange chromatography, the dissociated enzyme was separated into three fractions, none of which possessed any catalytic activity. Fraction 1 contained the a, y and e subunits, fraction 2, the a, yt 6 and e subunits, and fraction 3, only the B-subunits. ATPase activity could be reconstituted by combining either fractions 1 and 3 or 2 and 3 but not 1 and 2. This was the f i r s t indication that the active site(s) for ATP hydrolysis was probably on the a and B subunits. Recently, Dunn and Futai (72) have been able to purify the individual subunits from the cold-dissociated enzyme. The purified subunits could be reconstituted to give enzyme activity. The highest specific activity was obtained when the molar ratio of the a, fi and y subunits was 3:3:1 (i.e. ot^fi^Y)., The Y subunit was essential for expression of ATPase activity since a reconstituted a and fi (1:1) preparation had only 0-10% of the maximum activity. ECF^ of the unc A mutant has no hydrolytic activity, but i s indistinguishable from the parent enzyme in subunit composition, bound nucleotides and possession of an inhibitor-sensitive (Nbf-Cl) tyrosine residue on the fi subunit (24,73). The activity of the mutant ECF^ could be restored by replacing the cold-dissociated enzyme fraction containing 13. the of , Y and e subunits with the corresponding fraction obtained from the wild-type strain (74). Similarly, the ATPase activity could be restored i n the from the unc A mutant, E_. c o l i AN120, by dialyzing the total cold-dissociated enzyme together with an excess of purified o-subunit from the wild-type strain. The 13 and Y subunits, purified from the wild-type ECF^ did not restore ATPase activity i n the mutant enzyme. These results indicate that the lesion responsible for the lack of ATPase activity in the unc A mutant resides i n the « subunit (75,76). STUDIES ON THE ACTIVE SITE OF ECFj^ It was previously mentioned that the tightly-bound nucleotides of ECF^ are unlikely to be involved as intermediates of oxidative phosphory- lation and are probably involved i n the regulation of enzyme activity. Therefore, the intermediates of oxidative phosphorylation must be at other nucleotide-binding sites. Detection of these binding sites have involved the use of photo-affinity labels as well as compounds which label specific amino acids and inhibit the ATPase activity of the isolated ECF^ (6). Under slightly acidic conditions (pH 6.5), the compound DCCD reacts with a carboxyl residue on the B subunit of ECF^ and inhibits the ATPase activity (77,78). The reaction with DCCD also affected the a b i l i t y of the ECF^ to bind to ATP, suggesting that there i s a loss of an ATP-binding site. The compound Nbf-Cl, which can react with cysteine and tyrosine residues, inhibits the ATPase activity of ECF^. Prolonged incubation of ECF 1 (24 h) with [^CJNbf-Cl (53), resulted in the B subunit being preferentially labelled, but after a shorter period of incubation (30 min) (79), most of the label was associated with the « subunit. The binding 14. of Nbf-Cl to ECF 1 also resulted in the loss of a nucleotide-bindlng site (80). Photo-affinity labelling of ECF 1 with arylazido analogs of ATP and ADP (NAP^-ATP and NAP^-ADP) showed that the inactivation of ATPase activity was associated with the incorporation of 2 mol NAP̂ -ADP per mol ECF 1 (80). At low concentrations (5 uM) of either analog, the a subunit was preferentially labelled, but at high concentrations (75 uM), both ot and fi subunits were labelled to the same extent. This suggests the presence of a high a f f i n i t y nucleotide-binding site(s) on the ct subunit and a lower a f f i n i t y binding site on the J3 subunit and that- the interaction of both sites may be essential for expression of activity. Other labels which also bind to the a subunit are [2,8-3H]-ATP, [2,8-3H]- ADP (72) and 8-azido ATP (79). In the ECF^ of the unc A mutant, E. c o l i AN120, the ct subunit was not labelled with 8-azido ATP. / Bragg et a l . (81,82) have used the 2',3'-dialdehyde derivates of ADP(oADP) and ATP(oATP) to probe for nucleotide-binding sites involved i n ATP hydrolysis i n ECF^. These compounds react covalently at their respective binding sites. Only the binding of oADP resulted in the inhibition of ATPase activity. Both the oADP and oATP binding sites were located on the ct subunit. The ct subunit of the unc A mutant, IS. c o l i AN120, did not possess the oADP binding site(s). These results suggest that the oADP-binding site on the ct subunit may be the active site(s) for ATP hydrolysis, while the oATP-binding site(s) may be regulatory sites. The compounds DCCD and Nbf-Cl also affected the binding of oADP to the ct subunit. The binding of oATP was influenced to a lesser extent. The inhibition of binding of oADP to the active site may be due to a conforma- tional change induced i n the ct subunit as a result of reaction of the Q subunit with DCCD or Nbf-Cl. However, i t is also possible that the 15. presence of these substituents on the fi subunit could sterically prevent the binding of oADP. The results suggest that the oADP-binding site on the a subunit must likely be adjacent to the active carboxyl, tyrosyl or cysteinyl residue on the B subunit. Therefore, the active site would likely be at the interface of the a and fi subunits. Such an arrangement would explain why individual a or fi or both subunits have low ATPase activity and that the maximum activity i s obtained when the a, fi and y subunits are present. Presumably the y subunit i s necessary to hold the a and fi subunits in the proper conformation, such that the active site i s preserved. CROSS-RECONSTITUTION STUDIES Reconstitution of energy-transducing reactions i n F^-depleted membranes with F^ i s not restricted to ATPases from the same species. Schatz et a l . (83) f i r s t reported hybrid reconstitution between F^ from Baker's yeast and F^-depleted submitochondrial particles of beef heart, with the restoration of oxidative phosphorylation activity. Respiration and ATP-driven transhydrogenase activity was also demonstrated i n Fj--depleted membranes of E. c o l i , which were reconstituted with F^ from Salmonella typhimurium (41). Similarly, hybrid reconstitution between rat liv e r MF^ and F^-depleted submitochondrial membranes from human l i v e r carcinoma has also been demonstrated (84). More recently (85), ATP-driven proton uptake was demonstrated in the membranes by cross-hybrid reconstitution between F^ and F^-depleted membranes of either 15. c o l i or rat li v e r mitochondria. Hybrid F^ has also been reconstituted by using subunits of F^ isolated from different sources (86,87). The following combinations, in 16. the molar ratios indicated were found to contain ATPase activity: E E T a^^Y (I.e. a and fl subunits from ECF^ and the Y subunit from TF^), X X E E X E agBgY and ̂ BgY • The Y subunit seems to be mutually interchangeable. These reconstituted hybrid enzymes had physical properties, such as thermostability and optimum pH, which were different from either ECF^ or ATPase activity could not be reconstituted by randomly combining the X E E E X X X E X subunits. For example, the combinations: "3^3^ and o^J^Y or "3^3^ d i d not reconstitute ATPase activity. THE ARRANGEMENT OF THE SUBUNITS OF EC^ The arrangement of the subunits of ECF^ has been examined by two general methods. The f i r s t method has involved the reconstitution with the , isolated subunits, or various forms of ECF^, and their binding to ECF^- depleted membranes as was discussed earlier. In this respect, one has to be f a i r l y cautious in interpreting the data because these reconstitution experiments do not take into consideration the possible conformational change induced i n the enzyme, when a purified subunit i s added. The second approach has been through the vuse of cleavable, cross- linking reagents, such as dithiobis (succinimidyl propionate) and cupric 1,10-phenanthroline, which w i l l cross-link suitably placed amino and sulphydryl groups, respectively. The results of such experiments (88,89) indicated cross-linking between aa, aJ3, a6, J3B, BY , B6, Be, Y E and possibly ay. There was no formation of Y 6 or pairs, but i t has been pointed out that the absence of cross-linked products does not exclude the possibility of other subunits being in close proximity. The results of these cross-linking experiments are often compatible with those obtained 17. through reconstitution studies. ctB, ay and BY interaction are suggested by the observations that the reconstitution of ATPase activity from the individual subunits require the Y subunit. A ye interaction has been demonstrated by Dunn (70) using purified y and e subunits. The ct6 interaction i s also suggested in the experiments of Dunn et a l . (68) i n which they have shown that the purified & subunit does not bind to trypsin-treated ECF^ because of the removal of 15 amino acids from the Nl^-terminal region of the ct subunits. Electron micrographs of purified ECF^ which has been negatively- stained with uranyl acetate suggest that the enzyme i s a planar hexagon of six subunits about a central core (7). The presence of three subunits of both OL and fi suggest that the six peripheral subunits of the hexagon are li k e l y to be composed of ct and B subunits. • It i s d i f f i c u l t to propose a unique arrangement of the subunits of ECF^ because the data do not completely support any one particular model. Three models (Fig. 2) have been proposed (7,89): i ) Model 1 consists of a slightly puckered hexagon of alternating a and B subunits around the central y subunit. The e and 6 subunits are placed on opposite sides of the hexagon. This arrangement of the <S and e subunits contradict the results from the reconstitution experiments in which both 6 and e were required to bind ECF^ to the membrane. In support of this arrangement i s the finding that anti-o"-serum detached the entire ECF^ from the membrane without affecting the ATPase activity, whereas anti-e- serum selectively removed only the e subunit (65). i i ) Model 2 i s similar to Model 1 except that the ot and B subunits are stacked as pairs, i i i ) Model 3 i s different from models 1 and 2 in that the ot and B do not alternate and the 6 and £ subunits are on the same side of the 18. Lower FIG 2: Three models for the arrangement of the subunits i n the F^-ATPase of E. c o l i (7, 89). 19. hexagon. This arrangement of the subunits is supported by the reconstitution experiments discussed for Model 1 and by results of cross-linking experiments (89). More recently, i t was shown that the 6 and e subunits contain additional antigenic sites in the isolated ECF^ which are inaccessible when the ECF^ is bound to the membrane (65). THE FQ COMPLEX Solubilization of the F^FQ Complex One approach to the study of the FQ polypeptides has been through the purification of the F^FQ complex. This requires the presence of high ionic strength buffers and the use of detergents to solubilize the F^FQ complex (4,24,25). Buffers of high ionic strength are needed during solubilization and purification of the F^FQ complex to keep the F^ attached to the FQ, as the activity of the F^ provides a means of following the presence of the F^FQ c o m P ^ e x * Non-ionic and ionic detergents, such as Triton X-100, sodium cholate and deoxycholate are most commoniy used to solubilize membrane proteins (90,91). Zwitterionic detergents such as Ammonyx Lo and Aminoxid WS 35 have recently become available and these are becoming the detergents of choice, since they have powerful solubilizing properties and often do not possess the denaturing effects shown by ionic detergents (92). Criteria for Determining the Intactness and Purity of the F^FQ Complex The f i r s t criterion often used to determine whether the F^ and the FQ have remained associated during the purification i s the retention of sensitivity to inhibitors of ATPase activity (5). Membrane-bound ATPase activity i s sensitive to DCCD in prokaryotes, while i n eukaryotes, i t i s 2 0 . sensitive to both DCCD, and oligomycin. Both inhibitors bind to a specific polypeptide of FQ (discussed under DCCD-binding.protein) to inhibit the ATPase activity when F^ i s coupled to FQ . Ryrie ( 9 3 ) has shown that the sensitivity of the purified F^FQ to these inhibitors i s often a poor guide to determine i t s intactness. The sensitivity of the F^FQ complex from yeast mitochondria to these inhibitors varied considerably and depended on the type of activity being measured (e.g. ATPase or ATP- 3 2Pi J exchange) and whether the complex was f i r s t reconstituted into phospholipid vesicles. Similar results were also found by Foster and Fillingame ( 9 4 ) with the E C F 1 F Q complex. The second criterion used i s to reconstitute the purified F^FQ complex into phospholipid vesicles and to demonstrate energy-transducing reactions such as ATP- 3 2Pi exchange activity, ATP-driven proton-uptake, and A'vf-f -driven ATP synthesis. These are properties shown by the n membrane-bound ATPase ( 5 , 2 5 ) . The criterion for purity of the functional F^FQ complex i s through SDS-polyacrylamide gel electrophoresis. However, this method of determining the purity of a F^FQ preparation can also be misleading, since the number of polypeptides present i n the gel w i l l depend on the resolving power of the gel system used and on the choice of the protein- detecting reagents. For example, the yeast mitochondrial F^FQ complex which was originally determined to contain 9 different subunits, was resolved into 1 2 bands on a more resolving gel system ( 9 5 ) . The F J F Q Complex The F^FQ complexes have been purified from various organisms. The numbers of polypeptides i n each of these preparations are summarized in 21. Table 1. The mitochondrial F-^ F0 P r e P a r a t i o n s contain more polypeptides either because the enzyme is more complex than in other organisms or because these preparations are contaminated with other proteins or protease digestion has occurred. The polypeptides l i s t e d i n Table 1 are those which the authors have determined to be the major components of the F^FQ complex. In many of these preparations, varying numbers of minor bands were present to which no function was assigned. ^ The least complex preparation of the F ^o c o m P l e x w a s obtained from the thermophilic bacterium, PS3 (Tables 1 and 2 ) . Densitometric tracing of the purified F^FQ complex, analyzed on SDS-polyacrylamide gels which had been stained with Amido Black, revealed the presence of eight major polypeptides. Five of these polypeptides were from TF^. The remaining three polypeptides with molecular weights of 19 000, 13 500 and 5 400 were designated as TFQ polypeptides. The TF-^FQ complex was capable of the energy-transducing reactions i summarized in Table 2. A , J g + -driven ATP synthesis could only be demonstrated using phospholipids from PS3 for reconstitution. Soybean phospholipids were ineffective. Furthermore, a of greater than 170 mV was necessary for the demonstration of r i appreciable formation of ATP (112,113). The ECF^FQ complex also appears to consist of eight subunits. Foster and Fillingame (94) solubilized the F^FQ complex from the membrane of E_. c o l i with deoxycholate i n the presence of 1M KC1. The solubilized fraction was subjected to ammonium sulphate precipitation and the F^FQ complex was purified by sucrose density gradient centri- fugation. The purified complex was capable of the energy-transducing reactions summarized in Table 2. Analysis of the f-^FQ preparation on SDS-polyacrylamide gels, revealed the presence of eight major subunits, TABLE 1 Polypeptide Composition of FiFn-ATPase Complexes from Various Sources Bacillus Source Beef Heart Rat Liver S_. cerevisiae N. crassa Chloroplast PS3 E_. c o l i Mitochondria Mitochondria Reference (96) (97) (95) (98) (99) (102) (94) F l a 54 000 62 500 52 000 59 000 59 000 56 000 55 000 48 000 1 48 000 000 48 000 56 000 55 000 53 000 50 000 Y 33 000 33 800 31 000 36 000 37 000 32 000 37 000 6 19 500 12 500 14 500 15 000 17 500 15 500 20 000 e 7 700 7 000 10 700 12 000 13 500 11 000 12 000 I.P* 12 300 13 000 7 000 a - - - OSCP** 20 800 22 000 23 000 20 000 - - - 25 000 46 800 28 500 21 000 17 500b 19 000b 24 000 20 500 31 600 24 500 19 000 15 500 13 500 19 000 20 000 26 000 21 500 16 000 13 500 5 400 8 400 14 700 9 200 16 700 8 000 7 500 12 300 12 700 11 800 9 000 8 500 6 500 * Inhibitor polypeptide ** Oligomycin-sensitivity conferring protein a Inhibitor polypeptide did not precipitate with purified enzyme b Recent studies suggest that this component may be an impurity (100, 101, 103) S3 23. five of which corresponded to the subunits of ECF^. The FQ polypeptides had molecular weights of 24 000, 19 000 and 8 400. The smallest subunit of FQ was identified as the DCCD-binding protein (discussed under "The DCCD- binding protein"). When the ECF^FQ complex was isolated from cells grown in a medium containing succinate, malate and acetate, rather than glucose as the carbon source, additional bands with molecular weights of 76 000, 68 000, 34 000, 26 000, 15 000 and 14 000 were also present in the preparation. Of these, only the 14 000 dalton polypeptide copurified with an invariant stoichiometry when different fractions of the sucrose gradient were analyzed on SDS-polyacrylamide gels. Thus, the 14 000 dalton poly- peptide cannot be excluded as a possible subunit of the F^FQ complex when the cells are grown on a mixture of succinate, malate and acetate. It has been observed i n _E. c o l i that some characteristics of the F^FQ complex do change depending on the growth conditions (29-32). Rosen and Hasan (105) have also purified the F-^FQ complex from E_. c o l i . Following solubilization of the membrane with deoxycholate i n the presence of 140 mM KC1, the F^FQ was purified by chromatography on DEAE-cellulose and glycerol gradient centrifugation. This preparation consisted of six subunits and was deficient i n the 6 subunit. Although the ATPase activity of this preparation was inhibited by DCCD, other energy-transducing activities could not be demonstrated with this preparation (Table 2). The polypeptides with molecular weights of 10000 and 8300 were designated as the subunits of FQ. Recently, the membranes of E_. c o l i were solubilized with the zwitterionic detergent, Aminoxid WS 35, and the F^FQ complex purified by chromatography on DEAE-Sepharose CL-6B (106). The purified complex was capable of the energy-transducing reactions summarized in Table 2 and 24. consisted of eight different polypeptides. Five of these polypeptides were subunits of F^ and the FQ polypeptides had molecular weights of 28 000, 19 000 and 8 500. About 4% of the total protein of the F-^FQ preparation was residual contamination, the majority of which had spectral characteristics of cytochrome b^ and a molecular weight of 65 000. Minor components with molecular weights of 24 000 and 14 000 were also present and these were concluded to be possible degradation products of the larger subunits. A similar subunit composition for the ECF^FQ complex was obtained by Schneider and Altendorf (107) who used the same method of purification, but included centrifugation of the active ATPase fractions from the DEAE- Sepharose CL-6B column at 220 000 xg for 15 h (Table 2). However, differences were s t i l l present. It was observed that the ECF^FQ prepa- ration obtained after chromatography on DEAE-Sepharose CL-6B contained the 28 000 and 19 000 dalton polypeptides i n approximately equal amounts, and the 8 500 dalton subunit was present in only a third of the amount of the e subunit. By including the centrifugation step, the 28 000 dalton subunit was present in about twice the level as the 19 000 dalton subunit, and the 8 500 daIton/and e subunits were present i n about equal amounts. Further modification of this purification procedure (108) has included the precipitation of the ECF^FQ complex, obtained after DEAE-Sepharose CL-6B, with polyethylene glycol 6000 and 400. Differences which were observed (Table 2) i n this E C F ^ Q preparation were: i ) the 28 000 dalton subunit so prominent in the previous preparations was present in almost negligible amounts, i i ) the 24 000 dalton subunit was/ one of the major polypeptides of F Q, and i i i ) the complete absence of the 14 000 dalton polypeptide. Table 2 l i s t s only the major subunits of F.F n which were present TABLE 2 Properties of Various Preparations of Bacterial F. F. Complexes Source Detergent Method of Puri f i c a t i o n of Subunit composition F j F 0 A c t i v i t y b F F 1 0 F o of F Q (M^ x 10 - 3). S e n s i t i v i t y of ATPase a c t i v i t y to DCCD. A T P - 3 2 P I Exchange. ATP - driven proton uptake. Ref. E. c o l i Deoxycholate Sucrose gradient n.d. 6 29; 9 a f s n.d. 6 n.d. 104 E. c o l i Deoxycholate DEAE-Cellulose and sucrose gradient n.d. 10; 8.3 ' * s n.d. - 105 Z. c o l i . Deoxycholate Sucrose gradient n.d. 24; 19; 8.4a s > + 94 Treatment of FJFQ with EDTA 24; 19; 8.4a N.A.d M.A. N.A. 9 E. c o l i . Aminoxid WS-35 DEAE-Sepharose CL-6B and centrifugation n.d. 28; 19; 8.3 a s + + 106," 107 Treatment of F J F Q with 7M Urea 19; 14; 8.3 M.A. N.A. . N.A. 107 E. c o l i Aminoxid WS-35 DEAE-Sepharose CL-6B and PEG p r e c i p i t a t i o i n.d 24; 19; 8.3 s + + 108 Treatment of F-F with EDTA and 0 KSCN. 24; 19; 8.3 N.A. N.A. • 108 PS3 T r i t o n X-100 DEAE-Cellulose and Sepharose 6B n.d. 19; 13.5; 5.4a s + + 102 Treatment of F,F 1 0 with 7M Urea 19; 13.5; 5.4a c s 109 Treatment of F Q with 4M Urea and CM-cellulose 13.5; 5.4a c s + c 103 M. p h l e i Triton X-100 Sucrose gradient n.d. 24; 18; 8 s n.d. n.d. 110 S. f a e c a l i s Deoxycholate Ammonium sulphate p r e c i p i t a t i o n and non-denaturing g e l . n.d. 27; 15; 6 a s n.d. n.d. 111 C. pasteurianum Triton X-100 DEAE-Sepharose CL-6B and Sephadex LH-60 Treatment of F^FQ with EDTA 15 a s n.d. '+ 16 a. This subunit i d e n t i f i e d as the DCCD-binding protein using [ c]DCCD e. n.d., not done. b. Measured after reconstitution into phospholipid v e s i c l e s . f . s , se n s i t i v e . c. Demonstrated when the appropriate Fj - ATPase was added. g . + ] detected, -, not detected. d. N.A., not applicable - see Table 3. N3 N 2 6 . on the SDS-acrylamide gels. In a l l cases, small amounts of minor contaminants were also present and these could have been proteolytic degradation products. Ryrie and Gallagher (95) have demonstrated the existence of proteases i n the F^FQ complex from yeast mitochondria. Thus, i t would do well to keep in mind the presence of proteases during the purification of the F^FQ complex as this could affect the identification of the FQ subunits. It is also clear from Table 2 that, although i t i s possible to purify the ECF^FQ complex and reconstitute energy-transducing reactions, the authenticity and the minimum number of subunits of FQ required to reconstitute these reactions are s t i l l controversial. The Isolation of the FQ Complex, The second approach to the study of the polypeptides of FQ has been through the isolation of the intact F Q complex from the purified or partially purified F-^FQ complex. ' Isolation of the FQ complex has involved the extraction of the F^FQ complex with either urea, EDTA, or EDTA and KSCN i n order to remove the F 1 subunits. As with the F^FQ complex, the intactness of the FQ complex can be determined by demonstrating proton-translocating activity in reconstituted vesicles. These results are summarized in Table 3. The FQ complex was f i r s t isolated in bacterial systems from the thermophile, PS3 ( 1 0 9 ) . It consisted of three subunits with molecular weights of 19 000, 13 500 and 5 400. Treatment of the F Q complex with urea and subsequent purification on CM-cellulose (103) resulted in an FQ preparation containing two subunits with molecular weights of. 13 500 and 5 400. This FQ complex was capable of mediating proton conduction i n reconstituted vesicles. Energy-transducing reactions could also be T a b l e 3 P r o p e r t i e s o f V a r i o u s P r e p a r a t i o n s o f B a c t e r i a l F „ Complexes " S o u r c e D e t e r g e n t Method o f P u r i f i c a t i o n o f S u b u n i t c o m p o s i t i o n o f F Q ( M r x 1 0 " J ) FQ A c t i v i t y 1 * F 1 F 0 F 0 P r o t o n t r a n s l o c a t i o n S e n s i t i v i t y o f P r o t o n C o n d u c t i o n t o DCCD Re f . _E. c o l i D e o x y c h o l a t e S u c r o s e g r a d i e n t n . d . 6 29; 9 a N.A. d N.A. 104 _E. c o l i D e o x y c h o l a t e D E A E - C e l l u l o s e and s u c r o s e g r a d i e n t n.d. 10; 8.3 N.A. N.A. 105 E . c o l i D e o x y c h o l a t e S u c r o s e g r a d i e n t n.d. 24; 19; 8 . 4 a N.A. N.A. 94 Treatment o f F J F Q w i t h EDTA 24; 19; 8 . 4 a +e f s 9 J2. c o l i A m i n o x i d WS-35 DEAE - S e p h a r o s e CL-6B and c e n t r i f u g a t i o n n.d. 28; 19; 8 . 3 a N.A. N.A. 106,107 Tre a t m e n t o f F J F Q w i t h 7M Urea 19; 14; 8.3 + s 107 E. c o l i A m i n o x i d WS-35 DEAE - S e p h a r o s e CL-6B and PEG p r e c i p i t a t i o n n.d. 24; 19; 8.3 N.A. N.A. 108 T r e a t m e n t o f F J F Q w i t h EDTA and KSCN 24; 19; 8.3 ' + s 108 PS3 T r i t o n X-100 DEAE - C e l l u l o s e and S e p h a r o s e 6B n.d. 19; 13.5; 5.4 a N.A. N.A. 102 T r e a t m e n t o f F J F Q w i t h 7M Ur e a 19; 13.5; 5.4 a + s 109 T r e a t m e n t o f F Q w i t h 4M Urea and C M - c e l l u l o s e 13.5; 5.4 a + s 103 M . p h l e i T r i t o n X-100 S u c r o s e g r a d i e n t n.d. 24; 18; 8 N.A. N.A. 110 S. f a e c a l i s D e o x y c h o l a t e Ammonium s u l p h a t e p r e c i p i t a t i o n and non- d e n a t u r i n g g e l . n.d. 27; 15; 6 a N.A. N.A. 111 C. p a s t e u r i a n u m T r i t o n X-100 DEAE-Sepharose CL-6B and Sephadex LH-60 Trea t m e n t o f F . F . w i t h EDTA 1 5 3 + 3 16 a . T h i s s u b u n i t i d e n t i f i e d a s t h e DCCD - b i n d i n g p r o t e i n u s i n g [ c]DCCD d . N.A., n o t a p p l i c a b l e - s e e T a b l e 2 b. Measured a f t e r r e c o n s t i t u t i o n i n t o p h o s p h o l i p i d v e s i c l e s e. +, d e t e c t e d ; -, not d e t e c t e d . c . n.d., not done f . S j s e n s i t i v e 2 8 . reconstituted by the addition of purified TF^ to the TF Q-reconstituted vesicles. It was concluded that the TFQ consisted of two subunits and that the 1 9 0 0 0 dalton polypeptide was most lik e l y a contaminant.^ When the ECF^FQ complex, which contained the 2 8 0 0 0 , 1 9 0 0 0 and 8 3 0 0 dalton polypeptides as the major subunits of FQ , and the 2 4 0 0 0 and 1 4 0 0 0 dalton polypeptides as the minor contaminants, was extracted with urea, an FQ preparation containing essentially the 1 9 0 0 0 , 1 4 0 0 0 and 8 3 0 0 dalton polypeptides was obtained ( 1 0 6 , 1 0 7 ) . This FQ complex could reconstitute proton translocating activity in reconstituted vesicles (Table 3 ) . It was suggested that the 2 8 0 0 0 dalton subunit was probably a dimer, such that conversion to the monomer was obtained after urea treatment. In contrast, treatment of the ECF^FQ complex, which contained the 2 4 0 0 0 , 1 9 0 0 0 and 8 3 0 0 dalton polypeptides as the major subunits of F Q, with EDTA or EDTA and KSCN resulted i n an F Q preparation consisting of the same three polypeptides ( 9 , 1 0 8 ) . The 2 4 0 0 0 , 1 9 0 0 0 and 8 3 0 0 dalton polypeptides were shown to be genuine subunits of FQ , through the use of the defective transducing phage, * _ which carries the genes for d.sn J / the ECF^FQ polypeptides ( 1 1 4 , 1 1 5 ) . This i s also confirmed by the results obtained from the genetic studies. Biochemical Genetics The third approach to the study of the subunits of FQ and therefore F^FQ, has been through the isolation and generation of mutants of E_. c o l i which are defective i n oxidative phosphorylation ( 2 5 , 1 1 6 , 1 1 7 ) . These mutants can be generated by the use of mutagens such as N-methyl-N'- nitrosoguanidine, ethyl methyl sulphonate, hydroxylamine, ultra-violet irradiation and the bacteriophage, Mu. The mutants defective In oxidative 29. y phosphorylation cannot grow on non-fermentable substrates such as acetate, malate or succinate, and have low yields when grown on a limiting amount of glucose. The membranes of these mutants also cannot be energized with ATP and are therefore referred to as "unc" ("uncoupled") mutants. I n i t i a l l y , two classes of mutants were described. The f i r s t , designated unc A (e.g. unc A401) did not possess ATPase activity, whereas the second group, which retained activity was designated, unc B (e.g. unc B402). The lesions i n these mutants were further characterized as affecting either the F^ or FQ portion of the ECF^FQ complex, through reconstitution experiments; Oxidative phosphorylation activity could be reconstituted in ECF^-depleted membranes from the unc A strain with purified ECF^ either from the wild-type strain or from the unc B strain. The ECF^-depleted membranes from the unc B strain were also relatively impermeable to protons upon energization of the membrane through substrate oxidation. Thus i t was established that the unc A mutation affected the subunit(s) of ECF^, whilst the unc B mutation affected the FQ portion of the ECF 1FQ complex (116,118-120). A l l of the known unc mutations map at the 83.5 min region i n the E_. c o l i chromosome (25,121). Since the ECF^FQ complex probably consists of 7-9 different polypeptides, and a l l the mutations map at the same locus, i t was not possible to distinguish mutations in the different genes by simple genetic-mapping. The exact composition of the FQ complex and the characterization of the order and number of genes responsible for oxidative phosphorylation ' required more refined biochemical and genetic experiments. A genetic complementation system was developed to characterize the defective gene in the unc mutants. This involved the construction of partial diploids which / 30. contained two different unc alleles : one on the host chromosome and the other on a plasmid (F-plasmid) (25,121-124). When the mutation affected the same gene on the plasmid as on the host chromosome, the resulting partial diploid remained defective i n oxidative phosphorylation and expressed the phenotype of being unable to grow in the presence of non-fermentable carbon sources. On the other hand, when the mutation affected different genes, there was a normal copy of each of the affected genes present in the c e l l . The resulting partial diploid was similar to the wild-type strain. Biochemical tests, such as ATP-driven membrane energization and two-dimensional isoelectric focusing gels were used to confirm the presence or absence of genetic complementation. Seven distinct complementation groups were identified (121) i n the unc region of the chromosome and these were designated unc A, B, C, D, G, F and G coding for the polypeptides shown i n Table 4. Since the unc genes a l l mapped in the same region of the E_. c o l i chromosome, the bacteriophage Mu was used to determine the gene inter-relationship and whether these genes formed an operon (125) . The bacteriophage Mu has a polar effect on an operon, in that the insertion of the phage in an early gene prevents the transcription of a l l subsequent genes. Genetic complementation tests on these Mu-induced mutants demonstrated that the unc genes did form an operon. Together with the information obtained from cloning experiments, the gene order was postulated by Downie et a l . (126,127) to be unc BFEAGDC. However, the subsequent DNA sequencing of the unc operon not only showed that the order unc BFE was incorrect, but also that the gene coding for the 6 subunit (unc H) was located between unc F and unc A (42, 128-133). Therefore, the correct order of the genes in the unc operon i s unc BEFHAGDC, with the unc TABLE 4 Polypeptides Coded by the "uric" Genes Gene Polypeptides of Molecular Molecular Weight b Weight c Gene 1 1 14 183 ? unc B a 30 258 24 000 E c 8 365 8 300 F b 17 233 19 000 H 6 19 310 20 700 A a 55 264 56 800 G Y 34 100 32 000 D B 50 157 51 800 C e 14 194 13 200 a The general nomenclature for the FQ polypeptides are a, b and c with being the DCCD-binding protein b Determined from the DNA sequence c Determined by SDS-polyacrylamide gel electrophoresis 32. B gene being closest to the promoter (Table 4). In vitro protein synthesis experiments with plasmids (126,127,134) or specialized X transducing phages (135) carrying the unc genes confirmed that the unc B, E and F genes code for polypeptides of FQ with molecular weights of 24 000, 8-9 000 and 18 000, respectively. The molecular weight of the unc B gene product determined by SDS-polyacrylamide gel electrophoresis has been underestimated. This appears to be common to integral membrane proteins (136,137). Steffens et a l . (138) have purified the 24 000 and the 18 000 dalton polypeptides from the purified ECF^FQ complex and found that the amino acid compositions agreed with those deduced from the corresponding DNA sequences. The DNA sequence of the unc operon also indicates the existence of another gene (Gene 1) coding for a polypeptide of molecular weight 14 183 (128). This gene l i e s between the promoter and the unc B gene. Whether gene 1 i s transcribed in vivo is not clear, but the presence of a 14 000 dalton polypeptide i n some ECF^FQ preparations may suggest that i t is (94,106,107). In vitro coupled transcription-translation experiments with plasmids containing the whole unc operon also resulted in the production of a 14 000 dalton polypeptide (126). The function of this polypeptide i s unknown, although i t has been postulated/that i t may function in regulating the assembly of the FQ polypeptides (128). In vitro studies (139) with plasmids containing the unc genes, but not the promoter or Gene 1, showed that the membrane-association of the FQ polypeptides occurred via the insertion of the proteins into the membranes and this process was independent of the synthesis of each FQ polypeptide or of other F^ polypeptides. Whether the polypeptides of FQ are inserted into the membrane in the correct orientation, or whether a functional FQ complex , 3 3 . was formed, was not clear. By contrast, Cox et a l . ( 1 4 0 ) have shown that the insertion of some of the FQ polypeptides into the membrane, and therefore the assembly of a functional FQ complex, required the presence of a and J3 subunits of ECF 1. THE DCCD-BINDING PROTEIN The function of each subunit of FQ has not been characterized because of the d i f f i c u l t i e s encountered in the isolation of the individual i subunits i n non-denatured form. However, there is considerable evidence i n prokaryotic and eukaryotic systems suggesting that the smallest subunit of FQ is intimately involved i n the proton-translocating properties of the FQ complex. This polypeptide i s the best characterized subunit of FQ i n mitochondria, chloroplasts, and bacteria with respect to i t s structure and function. Identification and Isolation The compound DCCD i s able to react non-specifically with amino, carboxyl, hydroxyl and sulfhydryl residues to form stable adducts ( 1 4 1 ) . But under basic conditions (pH 8 . 5 ) , i t reacts specifically with a carboxyl residue of a subunit of FQ and Inhibits the ATPase activity of either the membrane-bound ATPase or of the purified E C F ^ Q complex ( 7 8 , 1 4 2 ) . The reaction of DCCD with this subunit of FQ also decreases the permeability of the F^-depleted membranes to protons, as discussed before. Beechey et a l . ( 1 4 3 ) were the f i r s t to demonstrate that the inhibition of the ATPase activity in beef-heart mitochondria was associated with the specific incorporation of [1"C]DCCD into a 9 0 0 0 dalton polypeptide. In E. c o l i 34. (141,144), the label was also associated exclusively with a 8 000 - 9 000 dalton polypeptide. This polypeptide was termed the "DCCD-binding protein". DCCD also inhibits the energy-transducing reactions of the F-JFQ and FQ complexes which have been reconstituted into liposomes (Table 2). Inhibition of these reactions was associated with the labelling of the smallest polypeptide of F Q with ['"CJDCCD. The DCCD-binding protein i s unique i n that i t can be extracted from the membrane with organic solvents (145,146). It i s due to this property that i t was originally called a "proteolipid". The DCCD-binding protein from the mitochondria of Neurospora crassa and yeast can be extracted i n almost pure form with a mixture of chloroform-methanol (2:1) i f the mitochondrial membranes are prewashed several times with a mixture of chloroform:methanol:diethylether (2:1:12). Similarly, the DCCD-binding protein isolated from beef heart and'rat l i v e r mitochondria, lettuce or pea chloroplast and the sarcoplasmic reticulum, by extraction with n-butanol followed by precipitation with diethyl ether, i s also i n pure form (146). i By contrast, the extraction of membranes or whole cells preparations with a mixture of chloroform:methanol (2:1) without prior washing with solvent, also extracts together with the DCCD-binding protein, several other hydrophobic proteins and phospholipids. Purification of the DCCD-binding protein i s achieved by repeated precipitation of the extract with diethyl ether followed by either: thin layer chromatography (147), adsorption chromatography on Sephadex LH-20 or LH-60 (145), reversed-phase HPLC (148) or In the case of E. c o l i (149,150), by ion-exchange chromato- graphy on either DEAE or CM-cellulose followed by adsorption chromatography on Sephadex LH-60. The purified DCCD-binding protein of 15. c o l i (149,150) migrates with 35. an apparent molecular weight between 8 000 and 9 000 on SDS-polyacrylamide gels. Reconstitution of Proton Translocating Activity The DCCD-binding protein, isolated from chloroplasts (151) or beef heart mitochondria (152) by extraction with n-butanol was capable of proton-translocating activity when reconstituted into liposomes. With the DCCD-binding protein from beef-heart mitochondria, the rate of proton influx was dependent on the amount of the protein incorporated into the liposome. In both cases, proton-translocating activity was sensitive to DCCD but oligomycin-sensitivity was not observed with the protein from beef heart mitochondria. Similarly, the protein isolated by extraction of yeast mitochondria with a mixture of chloroformrmethanol (2:1) was also capable of proton-translocating activity when reconstituted into liposomes (153,154). The rate of proton influx was proportional to the amount of protein incorporated. Oligomycin, which i s a specific inhibitor of oxidative phosphorylation i n eukaryotes and which also binds to the DCCD- binding protein, inhibited proton conduction. The proton-translocating activity was insensitive to oligomycin when the DCCD-binding protein from an oligomycin-resistant strain was used. Similar types of reconstitution experiments with the DCCD-binding protein from bacterial systems, did not result in any proton-translocating activity (5). This may be due to the method of isolation and purification of this protein, which may result in the loss of the native oligomeric structure and subsequent loss of function. Evidence for the existence of this protein i n an oligomeric form in E. c o l i or PS3 comes from the determination of the stoichiometry of the subunits of the F,F-. .complex. 3 6 . The F-jFfj c o m P l e x isolated from cells grown in the presence of either 3 5S0 4 2", [U-^CJD-glucose (155) or L-[U-ll,C] amino acid mixtures (156) result i n stoichiometrics of a^Q^y&eab2C10 a n (* a2^2(&ea^2C5 E C F 1 F 0 and TF^FQ, respectively. In addition, the inhibition of membrane-bound ATPase activity of E. c o l i was complete when only one third of the total DCCD-binding protein was labelled with [^CJDCCD, suggesting that the protein existed as a trimer (144). The second reason for the lack of protein conduction upon reconsti- tution may be due to the requirement of additional polypeptides to direct the proper insertion of the DCCD-binding protein into the membrane (103,140). The Amino Acid Composition The DCCD-binding protein has been purified from various sources and the amino acid composition of each i s listed i n Table 5. In a l l cases, the amino acid composition is derived from the amino acid analysis and confirmed by either complete or partial sequence analysis (146). The DCCD-binding proteins from the various sources contain unusually high amounts of nonpolar amino acids and therefore are very hydrophobic in nature. The protein from E. c o l i i s the most hydrophobic of those which have been studied (Table 4). Only 16.5% of i t s amino acids are polar. The predominating nonpolar amino acids are alanine (15%), leucine (16%), glycine (13%) and methionine (10%). The DCCD-binding protein from a l l sources except Aspergillus nidulans (157) lack histidine. Similarly, tryptophan is present only i n the protein from the chromatophbres of Rhodospirillum rubrum (158). In addition to histidine and tryptophan, the protein from E_. c o l i (149) also lacks the T a b l e 5 Amino A c i d C o m p o s i t i o n of the DCCD-Binding P r o t e i n from M i t o c h o n d r i a , C h l o r o p l a s t and B a c t e r i a 3 (146) Amino a c i d Aspergillus nidulans Neuro- spora Saccharo- myaes Dovine Heart S p i n a c h Mastigo- cladus Escherichia coli PS-3 Bacillus acidocal- darius Halobac- teriwn Rhodospi- rillum L y s i n e 2 2 2 2 1 1 1 - 3 1 2 H i s t i d i n e 1 - - - - - - - - T r a c e - . A r g i n i n e 3 2 1 1 2 2 2 4 2 1 1 A s p a r t i c a c i d 6 4 3 3 2 3 5 1 3 4 4 T h r e o n i n e 1 ' 2 3 3 3 3 1 3 1 7 3 S e r i n e 7 5 5 5 3 4 ' • - 3 4 3 3 G l u t a m i c a c i d 6 5 2 3 7 7 4 5 5 5 3 P r o l i n e 5 1 2 1 4 4 3 3 2 6 1 G l y c i n e 12 11 10 11 11 10 10 11 12 10 10 A l a n i n e .12 14 10 13 17 16 13 9 14 18 16 V a l i n e 5 6 6 4 7 4 6 8 9 6 6 M e t h i o n i n e 4 4 3 3 2 2 8 2 4 1 3 I s o l e u c i n e 5 6 9 7 6 8 8 9 5 5 9 L e u c i n e 8 11 12 9 12 13 12 10 9 10 8 T y r o s i n e 3 2 1 2 1 1 2 1 2 1 1 P h e n y l a l a n i n e 4 6 6 7 3 3 4 3 4 3 3 C y s t e i n e - 1 1 - - - - - - - T r y p t o p h a n - - - - - - - - • - ND C 2 T o t a l r e s i d u e s 83 . 81 . 76 75 81 81 79 72 79 81 75 End group f-Met Tyr f-Met Asp f-Met f-Met. f-Met f-Met f-Met ND f-Met P o l a r i t y 30 24.7 21.7 22.7 22.2 24.7 16.5 22.2 22.8 25.9 25 ND, not done. V a l u e s a r e g i v e n i n u n i t s of moles per mole b f r o m (157) 38. amino acids serine and cysteine. In most cases, the amino acid content is i n agreement with the apparent molecular weight determined by SDS-polyacrylamide gel electro- phoresis. The exception is the DCCD-binding protein from PS3, whose molecular weight of 5 400 determined by SDS-gel electrophoresis was underestimated (102). The results from the amino analysis suggests that the molecular weight is 7 300. The Amino Acid Sequence If the native DCCD-binding protein i s indeed a proton channel, then the knowledge of i t s amino acid sequence i s a pre-requisite for the understanding of i t s structure and function (146). The advantages of knowing i t s primary structure are: i ) theoretical calculations can be used to predict the arrangement of the polypeptide in the membrane and i i ) a better understanding of how mutations i n the polypeptide can affect i t s reactivity with DCCD (discussed under DCCD-resistent mutants) and oligomycin, or affect the proton permeability of the membrane, can be obtained. The complete amino acid sequence of the DCCD-binding protein has been determined from a variety of organisms and these are summarized in Fig. 3. The amino acid sequence of the protein from E. c o l i has been confirmed by i t s DNA sequence (128). In almost a l l organisms ['''CJDCCD reacted with a specific glutamyl residue (Fig. 3). But i n E_. c o l i , i t reacted with an aspartyl residue (asp 6 1). The proteins from distantly related organisms have a high degree of homology in their amino acid sequence and this reflects their being involved i n a similar function i n the c e l l . 39. N. c r . l y r - S c r - S c r - C 1 u - n c - A U - C l n - A U - M c t - V < t - C l v - V < 1 - . S « r - l y s - A s n - l c u - C l y - r 1 t t - C 1 y - S < r > A l t - A l < - n c - C l / - l c u - B o v i n e A i p - I W - A j p - T h r - A l a - A l a - t y i - P h c - J l t - G l y - A l a - C l y - A l a - A l a - ' l h r - V a l - G l y - Y a l - J . t e r . f - H e t - C l n - U v - m - l « u - A U - A 1 < - l r s - t y r - ! 1 c - C 1 y - A U - C l / - n « - W r - ' t h r - l l c - C t y - l t u - S p i n a c h f - r V t - A j n - P r o - U u - i l t - A l a - A l a - A l a - S « r - V a l - n r - A l a - A l a - C ! y - U u - A l a - » a I - G l y - U u - A l a - S t r - H. l a » . f - M c t - A » p - P r o - U u - l l t - W - A l a - A l a - S « r - V a l - l t u - A l a - A l a - A W - l e u - A l a - l l t - G l y - l t u - A l a - A l a - ( . c o l l f - M c l - C l u - A s n - l » u - A j n - r V t - A i p - t e u - l « u - T y r - h e t - A l » - A U - A U - V j l - r V t - M e t - C I ) - t t u - A l « - A W - P S - 3 f - h c l - S t r - l l u - G l y - V a l - l r u - A l a - A l a - A l a - l l t - A l a - Y a l - C l y - l t u - G l y - A l a - 3 0 4 0 SO N. c r . l h r - C ) j r - A U - C l / - l l < - C l / - ] l < - C l y - L c u - V < l - P h c - A l 4 - A l < - : « u - l e u - A s n - C 1 y - V < l - A l < - A r 9 - A s n - P r o - A l « - l c u - A r s - B o v i n e A n - C 1 y - S « r - C I / - A l i - C l ) r - l l » - C l / - I h r - V l 1 - P h « - C ) y - S « r - l * u - H e - l t « - C l y - , t / r - A U - A r 9 - A i i i - P r c - S e r - t » u - l y » - 5 . c e r . l r u - C l j - » n - C l y - l U - C I / - l 1 c - A l j - l U - » » l - P h t - A l « - A ) » - l t u - H « - A 5 n - C l y - V » l - S t r - A r 9 - A s n - P r o - S t r - l U - l y » - S p i n a c h I U - C l j - P r o - C l / - V i l - C 1 / - G l n - C l y - l h r - A l j - A U - C l / - C ) r > - A ) j - V « l - C l u - C l y - I U - A r « - A r 9 - C l n - P r o - G l u - A U - C l u - K. I n . l l c - C l y - P r o - C l y - n t - C l y - C l n - C l y - A s n - A n - A l a - C l y - C l n - A l < - V < l - C l u - C l y l l c - A l < - A r 9 - C l n - P r o - C l u - A U - C l u - t . c o M n » - 0 1 / - A I ; - A U - l U - C I J ' - r . t - C I / - l l e - l t u - C l / - C l y - l / » - P h t - l t u - C l n - C l y A l l - A l j - A r 9 - C l n - P r o - A s p - l t u - l U - P S - 3 l t u - C l y - A U - C l y - 1 l t - C l y - A s n - C l y - l c u - l l f V l l - S r r - A r f - l h r - I U ' C l u - C l y - n t - A U - A r g - C l n - P r o - G l u - L t u - A r g - eo ;o H. c r . C l y - G l n - l ( u - P h c - S c r - T y r - A U - t l f l t u - G l y - P r « - A 1 < - P h c - V < l - G l u - A l < - U c - G l y - l e u - P h c - A s p - l c u - H e t - V a l - A l a - B o v i n e G l n - G l n - l c u - P h r - S c r - i y r - A i a - n t - l e u - G l y - r h t - A l a - l t u - S c r - G l u - A U - r v t - G l y - l t u - P h r - C y s - l e u - r V l - V a l - A U - S . e r r . A i p - l t i r - v < l - P h c - P r o - K c t - A l a - l l r - l r u - G 1 y - P h f - A l a - L ( u - $ < r - G l u - A l a - l h r - G 1 y - l r u - P h e - C y s - l « u - H r t - V a l - S e r - S O . W K G l y - l y l - l I c - A r j - G l y - l h r - l c u - l e u - l e u - S r r - l e u - A I a - P h t - M t l - G l u - A l a - l t u - l h r - 1 l c - T y r - G l y - U u - V a l • V a l - A l a - H. l a m . G l y - l y s .1 I r - A r g . G l y - l ( i r - l r u - l r u - L e u - 1 h r - l e u - A l a - P » i t - H c t - G l u - S e r - l c u - l h r - H e - t y r - G ) y - l c u - V a l - I I t - A l a - t . c o l i P r o - t e u - l t u - A r j - l t . r - G l n - P h e - P h t - l i e - V a 1 - K c l - G l y - l c u - V a l - A s p - A l a - 1 l e - P r o - K e l - I l e - A I a - V a l - G l y - l f u - G l y - P S - 3 P r o - V a l - l « u - G l n - l h r - I h r - « e l - r h c - I U - G l y - V a l - A ' l a - l * u - V a l - G l u - A l a - l c u - P r o - l U - I l * - G l y - V a l - V a l - P h c - S c r - 8 0 * . c r . I c u - K c t - A l a - l y i - P h c - l h r B o v i n e P h c - l c u - I l e - l e u - T h e - A l a - H o t S . c c r . P h r - l c u - l c u - t e u - r h c - G l y - V a l S p i n a c h l c u - A l a - l c u - l c u - P h c - A l a - A * n - P r o - P h e - V a l M. l a m . I c o - V a l - l c u - L c u - r h c - A l a - A i n - P r o - P h c - S c r I. c o l i l e u - l y r - V a l - n i - r t c - A l a - V a l - A l a PS -3 P h c - I l c - l y r - l c u - G l y - A r g FIG. 3 Amino acid sequences of the DCCD-binding protein from Neurospora crassa (N. cr.) bovine heart (Bovine), Saccharomyces cerevisiae, (S. cer.), spinach chloroplasts (Spiiiach), Mastigocladus laminosus (M. lam.), Escherichia c o l i (E. c o l i ) , and the thermophilic bacterium PS-3. The numbering Is according to the Neurospora sequence. Reprinted from Sebald and Hoppe (146). • 40. DCCD-RESISTANT MUTANTS DCCD-resistant mutants have been isolated from E_. c o l i . Two types of these mutants (25) have been identified: i ) Mutants of the unc B phenotype. These mutants have a functional ECF^, with activities comparable to those of the wild-type strain, but the activity of the membrane-bound ATPase i s insensitive to DCCD. Also, the membranes of these mutants cannot be energized through ATP hydrolysis suggesting a defect in the FQ component, i i ) Mutants i n which the alteration causes an insensitivity of the membrane- bound ATPase activity to DCCD. These mutants can further be divided into two classes: class I and class II (159). The wild-type membrane-bound ATPase activity i s inhibitied half-maximally at 3-5 nmol DCCD per mg membrane protein, but the mutants of class I and II are inhibited half- maximally, at 30 and 200 nmol DCCD per mg membrane protein, respectively.. These mutants are distinct from the unc B?phenotypes, i n s t h a t the removal of ECF^ from the membrane results-in the membrane, becoming leaky to protons. Sequencing studies on the-DCCD-binding protein^isolated from these DCCD-resistant mutants, revealed that point-mutations in the polypeptide are responsible for DCCD-insensitivityand proton*impermeability. These results are summarized in Table 6. In the unc B phenotypesj i t i s observed that the replacement of the residue responsible for reacting with DCCD (i.e. asp) with either glycine or asparagine caused the loss of DCCD-binding and the membrane became impermeable to protons. The membrane-bound ATPase activity was not affected. This suggests that ATP hydrolysis and proton-translocation are uncoupled and that the mutation causes a conformational change i n F Q or in the DCCD-binding protein, which does not favour proton conduction. The TABLE 6 Properties of DCCD-Resistant Mutants of :E. co l i Strain Sensitivity of the Membrane-bound ATPase Activity to DCCD Proton-Permeability of F^-depleted Membranes Binding of [l-C]DCCD to DCCD-binding Protein Amino Acid affected in the DCCD-binding „ b References Protein Control sensitive permeable binds none 159 "Unc B" DG 7/1; DG 7/10 DG 18/3; DG 3/2 insensitive insensitive impermeable impermeable does not bind does not bind Asp 6 1 * Gly Asp 6 1 * Gin 159, 160 161 Class I DC 1; DC 13 DC 19; DC 24 sensitive a permeable binds l i e 2 8 - Val 160, 162 Class II DC 25; DC 54 sensitive 3 permeable binds 3 l i e 2 8 * Thr 162 a only at very high concentrations of DCCD ) 42. importance of this acidic residue (aspartic acid) i n proton translocation i s also seen in the DCCD-binding proteins from other organisms in which this acidic residue i s conserved (glutamic acid). Also, the mutants of classes I and II, i n which the aspartic acid at position 61 i s conserved, have membranes which are permeable to protons upon removal of ECF^. In the mutants of classes I and II, DCCD-resistance i s due to the replacement of isoleucine at position 28 with either value or threonine. Since a change at i l e 2 8 can cause such a large change in DCCD-sensitivity, i t has been proposed that this residue interacts at the DCCD-binding site (asp 6 1). Although the mutations are 33 residues away from the DCCD- binding site, these two residues could be I n close proximity to each other i f the DCCD-binding protein exists as a "hairpin" structure (150), similar to that proposed for bacteriorhodopsin (163). OBJECTIVES OF THIS STUDY From the information presented in the INTRODUCTION, i t is evident that more i s known about ECF^ than about the FQ complex. The elucidation of the subunit composition of FQ, of the mechanism of proton translocation through FQ, and of the interaction of FQ with ECF^ has been hindered by the lack of a purified, intact ECF^FQ complex. Therefore, one of the aims of this thesis was to purify the ECF^FQ complex in order to identify the subunits of the FQ complex. Secondly, mutants of E_. c o l i were available in which the membranes were relatively impermeable to protons. Since, an ECF^FQ complex with an altered or missing subunit(s) i n the FQ complex could give an insight into the function of the i n d i v i - dual subunits of the FQ complex, these mutants were characterized by a variety of biochemical methods. Finally, understanding the orientation 43. and/or organization of the FQ polypeptide(s) i n the membrane i s a prerequisite for studying the mechanism of proton-translocation. The DCCD-binding protein of the FQ complex has been implicated in the protonophoric activity of FQ. Therefore, the orientation and/or organization of this polypeptide in the membrane, as well as i t s interaction with ECF, were studied by immunological techniques. 44. MATERIALS AND METHODS CHEMICALS A l l chemicals were of reagent-grade purity and were purchased from the following sources: Aldrich Chemical Company: 9-aminoacridine hydrochloride. Amersham Corporation: ACS (Aqueous counting s c i n t i l l a n t ) , NCS tissue solubilizer, [7- 1*C]phenylglyoxal. BioRad Laboratories: Acrylamide, agarose, ammonium persulphate, Bio-Gel P6DG, Bio-Gel HTP (hydroxylapatite), Cellex-CM, 2-mercaptoethanol, N,N'-methylene-bis-acrylamIde, SDS, TEMED, urea. Calbiochem Corporation: Complete and incomplete Freund's adjuvant, ct-chymotrypsin (bovine pancreas), octyl B-D-glucopyranoside, PMS, pronase, valinomycin. ' Chemical Dynamics Corporation: e-aminocaproic acid, p-aminobenzamidine dihydrochloride. Difco Laboratories: Bacto-agar, Bacto-Penassay broth, Bacto-peptone, Bacto-tryptone, casamino acids, nutrient broth, yeast extract^ E. Merck. AG (Germany): Amido Black 10B. Eastman Kodak Company: Chloramine T, DCCD, Merthiolate. Fisher Scientific Company: Silicotungstic acid. Goldschmidt (Essen): Aminoxid WS-35. LKB-Produkter AB (Sweden): Ampholytes. Mann Research Laboratories: Tween 60, Tween 80, thyroglobulin (Pig). Mandel Scientific Company: Whatman CM-32. i 45. i Matheson, Coleman and Bell Manufacturing Chemists: Cyanogen bromide, sodium dithionite. Miles-Yeda Ltd.: w-Amino Butyl Agarose, Butyl Agarose, Decyl Agarose. Miles Laboratories: Staphylococcus aureus Vg protease. Onyx Chemical Company: Ammonyx Lo. ) Particle Data Laboratories Ltd.: Nonidet P-40. Pharmacia Fine Chemicals, Inc.: AH-Sepharose 4B, Blue Dextran, DEAE-Sepharose CL-6B, Octyl-Sepharose CL-4B, Phenyl-Sepharose CL-4B, Sepharose CL-6B, Sephadex G-50, Sephadex LH-60, Molecular weight markers for gel electrophoresis. Pierce Chemical Company: Amino acid standards for analyzer, ninhydrin. Research Products International Corporation: ['"CpCCD. Sigma Chemical Company: ATP, bovine serum albumin, Brig 35, 2,3-butanedione, Coomassie B r i l l i a n t Blue G, Coomassie B r i l l i a n t Blue R, catalase (bovine), DNase 1 (bovine pancreas), DTT, guanidine hydrochloride, HEPES, L-amino acids, lysozyme (egg), Lubrol PX, Lubrol WX, Lubrol 17A-10, L-a-phosphatidylcholine (egg), MOPS, NADH, phenylglyoxal, poly-L-lysine (Mr, 400 000), PMSF, phosphatidylcholine (soybean, 22%), RNase (bovine pancreas), sodium cholate, sodium deoxycholate, sodium N-lauroyl sarcosine, sodium succinate, sucrose, Tris base, Triton X-100, Triton X-114. Worthington Biochemical Corporation: TPCK-trypsin, soybean trypsin inhibitor. Rabbit y-globulin and Na( 1 2 SI) were generous gifts from Dr. R.S. Molday. TPCK-trypsin treated ECF was a kind g i f t from Cynthia Hou. 46. MAINTENANCE OF BACTERIAL STRAINS Table 7 l i s t s the strains of bacteria used in this study. These strains were maintained as slants and stabs containing Penassay broth-agar. The slants were prepared by boiling 1.75 g Bacto-Penassay broth and 1.5 g agar i n 100 ml d i s t i l l e d water. The mixture was then dispensed in 5 ml portions into screw-cap tubes and autoclaved for 20 min at a pressure of 15 psi. The tubes were allowed to cool such that the nutrient broth sol i d i f i e d as slants. Exponentially growing bacterial cultures were streaked onto the slants, incubated overnight at 37°C and then stored at 4°C u n t i l needed. These strains were subcultured every 4-6 months after checking for nutritional markers and then transferred to fresh slants. Cultures to be stored for longer periods of time (9 months) were inoculated into nutrient broth prepared as for slants, but containing 0.7% (w/v) agar and dispensed i n 2.5 ml aliquots into screw-cap vi a l s . GROWTH OF CELLS A 10 ml volume of Penassay broth was inoculated with a culture from a slant and grown overnight at 37°C. This was then transferred to 400 ml of -the appropriate medium i n a 2 1 conical flask and grown overnight at 37°C with shaking (at 250 rpm) i n a New Brunswick Rotary Incubator. The cells were either harvested and used immediately or the growing culture was used as 10% (v/v) innocula for growing larger quantities of cells (4.5 1). Batches of 4.5V1 were grown at 37°C (except for PS3 which was grown at 60°C) with vigorous aeration (at 25 1/min) in a Lab-Line/S.M.S. Hi-Density Fermentor. Cell growth was monitored by measuring the absorbance of the culture at 420 nm. The c e l l s were harvested at the late exponential phase of growth by TABLE 7 Bacterial Strains used in this Study Strain Genotype Origin E_. c o l i ML 308-225 DL-54 AN 180 AN 120 AN 382 WS1 NI44 CBT-1 CBT-302 Thermophile PS3 i " , Z", y+, a+ i ~ , Z~, y +, a +, unc F~, arg, t h i , mtl, xyl, s t r R F~, arg, t h i , mtl, xyl, s t r R , unc A401 F~, arg, t h i , mtl, xyl, s t r R , tfr-3, X-, unc B402 F~, pro, lac, J^g R, gal, ara, his, xyl, man, t h i , s t r R same as WS1 but neo R and unc thi t h i , unc H.R. Kaback R.D. Simoni F. Gibson F. Gibson B.J. Bachmann D.L. Gutnick D.L. Gutnick T.C.Y. Lo T.C.Y. Lo Y. Kagawa 48. centrifugation at 4 500 xg for 15 min, then washed twice in 0.9% (w/v) NaCl and sedimented by centrifugation at 17 600 xg for 10 min. The cells were stored at -20°C. MEDIA A l l the strains listed in Table 7, except PS3, were grown on a minimal salts-glucose media (164) containing 0.7% (w/v) K2HP04.3H20, 0.3% (w/v) KH2P04, 0.05 % (w/v) sodium citrate .2H20, 0.02% (w/v) MgS04.7H20, 0.1% (w/v) (NH 4) 2S0 4, 0.4% (w/v) glucose and supplemented with 0.1% (w/v) casamino acids. In some experiments, the glucose concentration was increased to 0.8% (w/v). Ferric citrate (12 PM) was added to media of volumes exceeding 1 l i t r e . Where required, amino acids and thiamine were added at 50 pg per ml and 1 Pg per ml, respectively. In order to obtain a higher yield of c e l l mass when growing mutants, the media was also supplemented with 0.1% (w/v) yeast extract, 0.1% (w/v) Bacto-tryptone, 3.3 uM CaCl 2, 6 PM CaS04, 6 PM MnCl2, 7 PM CoCl 2, 0.6 UM ZnCl 2 and 64 PM EDTA. In experiments requiring large quantities of E. c o l i ML 308-225, the cells were purchased from a commercial source (University of Alberta). These cells were grown to the late exponential phase on minimal -salts glucose (0.4% (w/v)). PS3 was grown on media (165) containing 13.3% (w/v) Nutrient broth, 6.7% (w/v) Bacto-peptone, 5% (w/v) NaCl and 5% (w/v) K2HP04.3H20 at pH 7.4. PREPARATION OF MEMBRANES The method described by Bragg et a l . (164) was used. Cells were suspended in buffer containing 50 mM Tris-H ?S0 4, 10 mM MgCl 2 pH 8.0 49. (TM buffer) at a ratio of 1 g wet weight/3 ml. About 5-10 pg of DNase was added and the c e l l s broken by, one or, i n some cases, two passages through an Aminco French Pressure c e l l precooled to 0°C, at a pressure of 2 1400 Kg/cm . Cell debris and unbroken cells were removed by centrifuga- tion at 12 000 xg for 10 min and the membranes sedimented by centrifugation at 250 000 xg for 2 to 2.5 h. Unless indicated otherwise, a l l centrifuga- tion steps were carried out at 0-5°C. The membranes were washed twice by resuspension in TM buffer and resedimented as before. They were processed in a number of ways to be described below. PREPARATION OF EDTA-LYSOZYME SPHEROPLASTS These were prepared by the method of Singh and Bragg (166). When possible, freshly-grown cells were used. E_. c o l i cells were grown to the late exponential phase and harvested by centrifugation at 4 500 xg for 15 min. The cells were washed three times by resuspension i n 50 mM Tris-HCl buffer, pH 8.0 and centrifugation at 12 000 xg for 10 min. The washed cells (2.75 g) were suspended at 20°C i n 37 volumes (w/v) of 50 mM Tris-HCl buffer, pH 8.0 containing 0.5 M sucrose. EDTA and lysozyme were added to f i n a l concentrations of 2.5 mM and 0.1 mg/ml, respectively. I The rate of spheroplast formation was. followed by removing a 50 ul sample of the c e l l suspension at various' intervals, diluting i t to 2 ml with d i s t i l l e d water, and then measuring the absorbance at 420 nm. After 30 min of incubation, the suspension was centrifuged at 12 000 xg for 10 min. The pellet was suspended in 275 ml of 20 mM potassium phosphate buffer, pH 8.0 and MgSÔ  was added to a f i n a l concentration of 1 mM. This was followed by addition of RNase and DNase at 20 yg/ml and 50. the suspension stirred for 20 min and then centrifuged at 6 400 xg for 15 min. The supernatant was centrifuged at 125 000 xg for 45 min. When K +-loaded vesicles were not required, the spheroplasts were washed twice by resuspension i n a buffer containing 0.4 M Sucrose and 10 mM MgC^, and resedimented as before. The spheroplasts were f i n a l l y taken up i n the same buffer, stored at 4°C and were used within 12 h. PREPARATION OF K+-L0ADED SPHEROPLASTS The pellet containing the spheroplasts was taken up in 10 volumes (w/v) of 0.5 M potassium phosphate buffer, pH 7.0 and gently heated at 40°C for 0.5 h (28,167). The suspension was then rapidly chilled i n an ice-bath and "after 10 min of incubation, MgSO^ was added to a f i n a l concentration of 10 mM. After a further 10 min incubation, the spheroplasts were sedimented by centrifugation at 125 000 xg for 45 min. The spheroplasts were washed twice by resuspension i n 0.4 M Sucrose, 10 mM MgC^ and centrifuged as before. The pellet containing K+-loaded spheroplasts was fi n a l l y taken up in the same buffer, and stored at 0°C. It was used within 12 h. ISOLATION OF EC^ ECF^ was isolated from 60 g of cells as described by Bragg and Hou (89). Membrane vesicles from 60 g of E_. c o l i ML 308-225 were prepared in TM buffer as previously described. They were washed twice by suspension i n a buffer (200 ml) containing 5 mM Tris-HCl, pH 7.3, 10% (v/v) glycerol, 20 mM e-aminocaproic acid, 6 mM p-aminobenzamidine, and 0.5 mM DTT, followed by centrifugation at 250 000 xg for 2 h. The washed vesicles were sus- pended in 60 ml of 1 mM Tris-CHl buffer pH 7.5 containing N0.5 mM EDTA, 0.1 51. mM DTT and 10% (v/v) glycerol ("dialysis buffer") and dialyzed against 3 1 dialysis buffer for 16 h at 4°C. The dialyzed material was diluted to 200 ml with dialysis buffer and then centrifuged as above. The supernatant was recentrifuged at 250 000 xg for 3 h. After this step, the supernatant was removed and concentrated to 18 ml by u l t r a f i l t r a t i o n using an Amicon XM-100A f i l t e r . Methanol (4.5 ml) was then added drop-wise to the concentrated solution followed by 0.46 ml of 1M CaCl 2. After incubation of the mixture for 20 min at 20°C, i t was centrifuged at 12 000 xg for 20 min and the supernatant, containing ECF^, was applied to a column of AH-Sepharose 4B. PURIFICATION OF EC^ ON AH-SEPHAROSE 4B AH-Sepharose 4B was suspended in an excess of 0.5 M NaCl and e q u i l i - brated for 4 h at 20°C. The suspension was degassed and then poured into a 2 cm diameter column to a height of 7.5 cm. The column was washed with several volumes of 0.5 M NaCl, d i s t i l l e d water and f i n a l l y equilibrated with dialysis buffer containing 1 mM Tris-HCl, pH 7.5, 10% (v/v) glycerol, 0.5 mM EDTA and 0.1 mM DTT. The sample containing ECF^ was adsorbed onto the column at 20°C which was then washed with 2-3 volumes of dialysis buffer. Elution of the enzyme was carried out with a linear gradient (total volume, 200 ml) of 0.25 to 0.75 M KC1 i n dialysis buffer. The fractions containing ATPase activity were pooled and concentrated to one or two ml by passage through an Amicon XM-100A f i l t e r . The concentrated material was either used immediately or divided into small aliquots, rapidly frozen, and stored at -70°C. The column was regenerated between experiments by washing with several volumes of 1 M KC1 in dialysis buffer, re-equilibrated with dialysis buffer, and stored at 4°C. 52. In some cases, the ECF 1 obtained after AH-Sepharose 4B was further purified by sucrose density gradient centrifugation. PURIFICATION OF ECFj^ AND TPCK-TRYPSIN TREATED ECF 1 BY. SUCROSE DENSITY GRADIENT CENTRIFUGATION A discontinuous gradient (23) consisting of successive 1 ml layers of 25%, 22.5%, 20%, 17.5% and 15% (w/v) sucrose i n 50 mM Tris-H 2S0 4 buffer, pH 7.8 containing 0.5 mM EDTA and 0.1 mM DTT was poured. The sample (0.4 ml, 3-5 mg) protein was, loaded and the gradient centrifuged at 260 000 xg for 16 h. Twenty drop fractions were collected by using a Beckman fractionator attached to a fraction collector. PREPARATION OF ECF^-DEPLETED MEMBRANES Several methods were used to strip the membranes of ECF^. a) The membrane fraction remaining after dialysis against 1 mM Tris-HCl buffer, pH 7.5 containing 0.5 mM EDTA, 0.1 mM DTT and 10% (v/v) glycerol ("dialysis buffer") as described under "Isolation of ECF^", was essen- t i a l l y devoid of ECF^. The membranes were washed once by resuspension i n dialysis buffer and centrifuged as before, and then stored at 4°C. b) More drastic treatments with urea, guanidine hydrochloride, s i l i c o - tungstic acid, or proteases were, also used. Dialyzed membranes (described- i n (a)), or membranes prepared in TM i buffer, pH 8.0, suspended i n dialysis buffer, pH 7.5 at a protein concen- tration of 10 mg/ml were incubated at 20°C for 30 min with either 2 M urea, 2 M guanidine hydrochloride, 1% or 2% (w/v) silicotungstic acid, Staphylo- coccus aureus Vg protease or TPCK-trypsin at an enzyme to protein ratio of 1:40 and 1:15, respectively. In the case of TPCK-trypsin, the reaction 53. was stopped by the addition of soybean trypsin inhibitor at a ratio to trypsin of 3:5 with incubation for another 10 min. The incubated mixtures were diluted 4-fold with buffer and then centrifuged at 250 000 xg for 2.5 h. The sedimented vesicles were washed twice by resuspension in dialysis buffer followed by resedimentation as above. i The ECF^-depleted membranes prepared by either method (a) or (b) were f i n a l l y suspended i n the appropriate buffer at 4°C. The pH of the vesicle suspension was readjusted to pH 7.5 with dilute KOH when s i l i c o - tungstic acid was used. PREPARATION OF RAT-LIVER MITOCHONDRIAL MEMBRANES Phosphate-washed purified inner mitochondrial membranes were prepared from rat-liver mitochondria as described by Soper and Pedersen (168). The purified membranes at 5 mg protein/ml of PBS-10 mM KgCl^ were sonicated at a power of 50 W in a Branson W185D sonifier for 15 s periods for a total of 2.5 min. The temperature was kept at 0°. Large fragments were removed by centrifugation at 12 000 xg for 10 min. The supernatant was either centrifuged at 250 000 xg for 2 h to sediment "sonicated washed mitochon- d r i a l membranes" or incubated with 4 M urea for 30 min at 22°C prior to centrifugation as above. The "urea-treated washed mitochondrial membranes" were washed by recentrifugation of a suspension in PBS-10 mM MgC^. A l l membranes were resuspended in PBS-10 mM MgC^. PREPARATION OF THE SUBUNITS OF ECF^ TPCK-Trypsin Treated ECFj^ ECF lacking the 6 and e, and perhaps part of the Y subunit, was 54. prepared by c o n t r o l l e d treatment of the ? 1 w i t h TPCK-trypsin, as described by Bragg and Hou (89). ' 1.25 mg of sucrose g r a d i e n t - p u r i f i e d EZY^ i n 0.385 ml of 22.5% (w/v) sucrose i n 20 mM triethanolamine-HCl b u f f e r , pH 7.5 c o n t a i n i n g 0.5 mM EDTA was t r e a t e d w i t h 37 yg TPCK-trypsin i n 10 y l b u f f e r at 37°C f o r 10 min. Bovine pancreas t r y p s i n i n h i b i t o r (146 yg i n 10 y l b u f f e r ) was added and a f t e r 5 min at 20°C, the s o l u t i o n was d i l u t e d w i t h an equal volume of b u f f e r and the t r y p s i n - t r e a t e d ATPase r e i s o l a t e d by sucrose d e n s i t y i gradient c e n t r i f u g a t i o n . a and B Subunits of ECF^ These were prepared by Dr. Helga Stan-Lotter ( U n i v e r s i t y of B r i t i s h Columbia). ECF.^ obtained a f t e r p u r i f i c a t i o n on AH-Sepharose 4B was d i s s o c i a t e d i n t o i t s subunits w i t h high i o n i c s t r e n g t h b u f f e r as described by Dunn and F u t a i (72). The d i s s o c i a t e d ECF.^ subunits were a p p l i e d to a f r e s h l y made column of h y d r o x y l a p a t i t e . The subsequent separation of the a and £ subunits was c a r r i e d out on a column of DEAE-Sepharose CL-6B as described by Bragg et a l . (82). SOLUBILIZATION OF MEMBRANE VESICLES WITH DETERGENTS - The s o l u b i l i z a t i o n c h a r a c t e r i s t i c s of various detergents were determined as f o l l o w s . Membranes of E_. c o l i ML 308-225 were prepared i n TM b u f f e r as described p r e v i o u s l y . They were suspended i n 0.5 M Tris-H2S0^ b u f f e r , pH 8.0 c o n t a i n i n g 0.25 M Na^O^ and 10% (v/v) g l y c e r o l at a membrane p r o t e i n c o n c e n t r a t i o n of 20-25 mg/ml. The suspension was d i v i d e d i n t o s e v e r a l a l i q u o t s and to each was added, drop-wise, d i f f e r e n t amounts of 55. detergents from either a 10% (w/v) or 20% (w/v) stock solution i n d i s t i l l e d water, or i n some instances added neat. After incubation for 30 min at 20°C, they were centrifuged at 250 000 xg for 2.5 h and the ATPase activity and protein content determined in the supernatant and pellet fractions. PURIFICATION OF THE ECFJFQ COMPLEX SOLUBILIZED WITH N-LAUROYL SARCOSINE v Gel F i l t r a t i o n on Sepharose 6B Membranes of E_. c o l i ML 308-225 were prepared in TM buffer as described previously. They were suspended in 0.5 M Tris-^SO^ buffer, pH 8.0 containing 0.25 M Na^O^ and 10% (v/v) glycerol at a protein concentration of 20-25 mg/ml. Detergent was added drop-wise, from a 20% (w/v) stock solution i n d i s t i l l e d water, to give a detergent (mg) to protein (mg) ratio of 0.5. After incubation at 20°C for 30 min, the mixture was centrifuged at 250 000 xg for 2 h. The supernatant was recentrifuged, then concentrated to about one quarter of i t s original volume (20-25 mg protein per ml) by u l t r a f i l t r a t i o n using an Amicon XM-100 A f i l t e r . The concentrated material was loaded onto a column of Sepharose 6B. Pre-swollen Sepharose 6B was suspended in 50 mM Tris-^SO^ buffer, pH 8.0 containing 10% (v/v) glycerol at 4°C. A slurry was poured into a 2.5 cm diameter column to a height of 37 cm, and the column equilibrated with 50 mM Tris-H^O^ buffer, pH 8.0, containing 0.25 M Na^O^, 10% (v/v) glycerol and 0.5% (w/v) of detergent. The sample (2-5% (v/v) of total bed volume) was loaded onto the column and eluted with the appropriate buffer. The column was regenerated between experiments by washing with 2-3 bed volumes of the apropriate buffer. 56. Hydrophobic-Interaction Chromatography Membranes of E_. c o l i were prepared in TM buffer and then washed twice by. resedimentation at 250 000 xg for 2 h from 5 mM Tris-HCl buffer, pH 7.3 containing 10% (v/v) glycerol, 20 mM e-aminocaproic acid, 6 mM p-aminobenzamidine and 0.5 mM DTT. The washed membranes were suspended in 0.5 M Tris-H 2S0 4 buffer, pH 8.0 containing 0.25 M Na 2S0 4 and 10%' (v/v) glycerol at a protein concentration of 20-25 mg/ml. Detergent was added drop-wise, from a 20% (w/v) stock solution i n d i s t i l l e d water, to give a detergent (mg) to protein (mg) ratio of 0.25-0.5. After incubation at 20°C for 30 min, the mixture was centrifuged at 250 000 xg for 2 h. The supernatant was recentrifuged and then subjected to ammonium sulphate precipitation. Saturated ammonium sulphate (adjusted to pH 7.5 with NĤ OH) was added to the supernatant at 0°C to give 35% of saturation. After 30 min incubation, the solution was centrifuged at 30 000 xg for 20 min. The supernatant was next brought to 50% of saturation. The pellet precipitating at 35-50% (0.35-0.5 P) saturation of (NH^SO^ was taken up at 4°C, i n solubilization buffer at a protein concentration of 4-6 mg/ml and N-lauroyl sarcosine added to give a detergentrprotein ratio of 0.8-1.0. This was applied to a column of Phenyl-Sepharose CL-4B. Phenyl-Sepharose CL-4B (or other hydrophobic-exchange resins) i n 50 mM Tris-H 2S0^ buffer, pH 8.0 was poured into a 1.8 cm diameter column to a height of 25 cm and the column equilibrated at 4°C with 50 mM Tris-H^O^ buffer, pH 8.0 containing 20% (NH^SO^ (by saturation), 10 mM MgCl 2 and 10% (v/v) glycerol. The solubilized fraction was adsorbed onto the column which was then washed with several volumes of the same buffer. The enzyme was eluted f i n a l l y with the appropriate buffer as described i n the "RESULTS" section. 57. The resin was regenerated by washing with 2-3 bed volumes of 2% (w/v) Triton X-100 followed in succession by i) 2 bed volumes of d i s t i l l e d water i i ) 2 bed volumes of ethanol i i i ) 2 bed volumes of n-butanol iv) 2 bed volumes of ethanol N v) 2 bed volumes of d i s t i l l e d water and vi) equilibration with starting buffer. With Phenyl-Sepharose CL-4B, an additional washing with 2-3 bed volumes of 2% (w/v) SDS after the Triton X-100 wash was found to be necessary to remove tightly-bound proteins. PURIFICATION OF THE E C F ^ Q COMPLEX BY SUCROSE DENSITY GRADIENT CENTRIFUGATION The enzyme obtained after chromatography on Phenyl-Sepharose CL-4B was concentrated by u l t r a f i l t r a t i o n using an Amicon PM-10 f i l t e r , and purified by sucrose density gradient centrifugation. A linear 17.5% to 25% (w/v) sucrose gradient (12 ml, total volume) was poured. The sucrose was dissolved in 50 mM Tris-^SO^ buffer, pH 8.0 containing 0.5% (w/v) sodium cholate, 5 mM MgCl 2 and 0.25 mM DTT. In some experiments, 12 uM p-aminobenzamidine was included in the gradients as a protease inhibitor and, on occasion, 0.1% (w/v) L-ct-phosphatidylcholine (soybean) was also , added. The sample (1-1.2 ml, 5-10 mg protein per ml) was loaded and the gradients centrifuged at 280 000 xg for 23 h and ten-twenty drop fractions collected by using a Beckman fractionator connected to a fraction collector. i 58. DEAE ION-EXCHANGE CHROMATOGRAPHY This was based on the method of Friedl et a l . (106). A slurry of pre-swollen gel in 50 mM Tris-HCl buffer, pH 8.0 at 4°C was poured into a 1.5 cm diameter column to a height of 13 cm and the column equilibrated with 50 mM Tris-HCl buffer, pH 8.0 containing 1 mM MgCl 2, 0.2 mM DTT, 0.2 mM EGTA, 0.1 mM PMSF, 100 mM KC1, 20% (v/v) methanol, 50 mg/ml soybean phospholipid, 0.9% (w/v) Aminoxid WS-35, un t i l the conductivity of the effluent was the same as the buffer. The sample was adsorbed onto the column and washed with equilibration buffer. The enzyme was eluted with a linear gradient of 100-300 mM KC1 i n equilibration buffer. The resin was regenerated by washing with 1 M KC1 i n 50 mM Tris-HCl buffer, pH 8.0 and equilibrated with the appropriate buffer. PREPARATION OF DCCD-BINDING PROTEIN The method described by Fillingame (149) was used. Membrane vesicles or ECF^-depleted membranes at a protein concentration of 40-50 mg/ml i n either TM buffer, pH 8.0 or in d i s t i l l e d water, were mixed with 20 volumes of chloroform:methanol (2:1) and stored at 4°C for 24 h. The suspension was passed through a f r i t t e d glass funnel of medium porosity. To the f i l t r a t e , in a glass-stoppered bottle, was added 0.20 volume of d i s t i l l e d water and the contents mixed by gently inverting the bottle. After standing at 4°C overnight, the contents of the bottle were warmed to 20°C to completely resolve the two phases. The lower phase was washed twice with 0.05 volume (of original f i l t r a t e ) of a mixture of chloroform: methanol:water (3:47:48). Finally, the washed lower phase was c l a r i f i e d with a small amount of methanol and the mixture diluted with one volume of chloroform. The chloroform was added in small amounts and a requisite 59. amount of methanol was added whenever the mixture started to turn cloudy. It was then carefully taken to dryness by rotoevaporation at 20°C using the vacuum from a water-aspirator. On occasion, chloroformrmethanol (10:1) was added to the evaporation flask whenever the contents started to turn cloudy. The dried contents of the flask were taken up in a small volume of chloroform:methanol (2:1) and any traces i n insoluble material removed by passage through a medium or fine porosity f r i t t e d glass funnel. The f i l t r a t e was chilled to -20°C and 4 volumes of precooled (-20°C) diethyl ether added slowly with st i r r i n g . The mixture was stored at -20°C for 24 h and the precipitate removed by centrifugation for 45 min at 4 000 xg and -20°C i n a capped stainless steel tube. The pellet was again taken up in the original volume of chloroforro:methanol (2:1) and precipitation with 4 volumes of pre-cooled diethyl ether was repeated. The precipitate was removed by centrifugation and dried at 20°C with a stream of nitrogen gas. It was dissolved in a small volume of chloroform methanol (2:1) and - insoluble material removed by f i l t r a t i o n through a f r i t t e d glass funnel. The f i l t r a t e was then applied to a thin-layer plate or adsorbed onto a column of CM-cellulose as described below. For the isolation of large amounts of DCCD-binding protein, the method of Altendorf et a l . (150) was followed. In this procedure, the use of whole cells as starting material was more convenient than using membrane vesicles. 400 to 500 g (wet weight) of frozen IS. c o l i c e l l was mixed with 20 volumes (w/v) of chloroform:methanol (2:1) at 20°C and the suspension mixed i n a Waring blender at top speed for 1 min. It was stirred at 4°C for 24 h and then f i l t e r e d through a medium porosity f r i t t e d glass funnel. The i i f i l t r a t e was treated as described above. 60. PURIFICATION OF DCCD-BINDING PROTEIN Thin Layer Chromatography v A 20 x 20 cm glass plate, precoated with s i l i c a gel G (1500 pm, Analtech Inc.) was activated by placing i t into an oven at 100°C for 1 h and then allowed to cool to 20°C. The isolated DCCD-binding protein (2 mg protein) i n chloroformmethanol (2:1) was applied as a 0.5 cm wide streak, 9 cm in length and 2.5 cm from the bottom edge of the plate. The streak was thoroughly air-dried and the chromatogram developed in a covered tank at 20°C in a mixture of chloroformmethanol:water (65:25:4) containing 20 mM HC1 (147). The solvent was allowed to ascend to within 1 cm from the top of the plate (0.5,-0.75 h). The plate was air-dried and placed i n a tank equilibrated with iodine crystals for 3-5 min. The stained spots were scraped from the plate and treated with enough water to completely wet the powder. After 30 min at 20°C, the powder was extracted with 5-10 ml of chloroformmethanol (2:1) containig 20 mM HC1 for 30 min. The suspension was centrifuged at 2 500 xg for 15 min. The s i l i c a gel was re-extracted in the same manner, three times. The combined extracts were passed through a fine-porosity f r i t t e d glass funnel and the f i l t r a t e dried under a stream of nitrogen. The residue was taken up i n a small volume of chloroformmethanol (2:1) and chilled to -20°C. To this was added 4 volumes of precooled (-20°C) diethyl ether and the mixture stored at -20° C for 24 h. The preci- pitate containing the DCCD-binding protein was removed by centrifugation at 4 000 xg and -20°C for 45 min and then dried under nitrogen at 20°C. Chromatography on CM-Cellulose The methods described by Rouser and coworkers (169) and Altendorf and coworkers (150) were modified for our purposes. 61. CM cellulose (Whatman CM-32 or BioRad Cellex CM) was suspended in 15 volumes (w/v) of glacial acetic acid for 1.5 h at 20°C. The fines were removed and the resin washed with d i s t i l l e d water until the pH of eluent was between 5 and 7. Next, the resin was suspended in 10 volumes (v/v) of 25% (v/v) ammonium hydroxide. After 1.5 h at 20°C, the suspension was diluted several fold with d i s t i l l e d water and centrifuged at 4 500 xg for 10 min. The sedimented resin was washed repeatedly with d i s t i l l e d water un t i l the pH of the supernatant was between 7 and 9. Finally, the resin was washed sequentially with 3 volumes (v/v) each of methanol, chloroform: methanol (1:1), and chloroform methanol (2:1). A slury of the resin in chloroform:methanol (2:1) was poured into a solvent-resistant column and the column bed protected with glass wool held down with marbles. Two volumes (v/v) of chloroformmethanol (2:1) was passed through the column and the sample (in chloroformmethanol (2:1)) was then applied. The column was washed sequentially with 5-10 volumes of chlorof ormmethanol (2:1), 3-5 volumes of chlorof orm methanol (1:1), and the DCCD-binding protein eluted with 5'volumes of chlorof ormmethanol: water (5:5:1). In some experiments, the column was washed with 5 volumes of chlorof ormmethanol: water (10:10:1) prior to elution of the DCCD-binding protein. The fractions containing the DCCD-binding protein were pooled and chloroform and water added so that,the f i n a l ratio of chloroformmethanol: water was 8:4:3 and a two-phase system was obtained (149). The lower phase was diluted with 1 volume of chloroform and the requisite amount of methanol was added to keep the solution clear. The mixture was then taken to dryness by rotoevaporation at 20°C. The dried contents were immediately taken up i n a small volume of chlorof ormmethanol (2:1) containing 20 mM ammonium acetate and applied to a Sephadex LH-60 column. Chromatography on Sephadex LH-60 Thi s was based on the method of F i l l i n g a m e (149). Sephadex LH-60 was s t i r r e d i n t o 25 volumes (w/v) of chloroform:methanol (2:1) c o n t a i n i n g 20 mM ammonium acetate and l e f t at 20°C f o r 6 h. A s l u r r y of t h i s suspension was poured i n t o a s o l v e n t - r e s i s t a n t column and a sm a l l amount of washed reagent- grade sea sand c a r e f u l l y a p p l i e d to the top of the bed to f a c i l i t a t e sample a p p l i c a t i o n . The column was e q u i l i b r a t e d w i t h the same solvent at 20°C and the sample a p p l i e d . The DCCD-binding p r o t e i n was e l u t e d w i t h the same solvent system a t a flo w r a t e of 0.1 to 0.15 ml per min. The f r a c t i o n s c o n t a i n i n g the p u r i f i e d DCCD-binding p r o t e i n were pooled and stored a t -20°C. When the p r o t e i n was needed, samples were removed and taken to dryness under reduced pressure at 16-18°C i n a r o t a r y evaporator (Buchler Instruments, Rotary Evapo-mix). SDS-POLYACRYLAMIDE GEL ELECTROPHORESIS Sample P r e p a r a t i o n S a l t s and detergents were removed from the samples i n one of two ways: ( i ) Samples from f r a c t i o n s obtained a f t e r column chromatography were d i a l y z e d at 4°C f o r 20-48 h against 100 volumes of d i s t i l l e d water ( c o n t a i n i n g 0.01% (v/v) h i b i t a i n e ) and which was changed at 8 h i n t e r v a l s . The d i a l y z e d f r a c t i o n s were l y o p h i l i z e d and stored at -70°C. ( i i ) A more r a p i d method of removal of s a l t and detergent from the samples was by the centrifugation-column chromatography procedure of Penefsky (170). A s l u r r y of Sephadex G-50 e q u i l i b r a t e d w i t h 50 mM T r i s - H 2 S 0 4 b u f f e r , pH 8.0 c o n t a i n i n g 10 mM M g C l 2 was poured i n t o the b a r r e l of a 1 ml disposable t u b e r c u l i n syringe (Becton-Dickinson) which had been f i t t e d 63. with a porous polyethylene disk. It was placed in a 15 x 125 mm test tube and allowed to drain. The volume of the gel was 1 cm3. Excess buffer was removed by centrifuging at 900 xg for 2 min at 20°C (International Equipment Company, Model CL45436M). A 100 ul sample was carefully loaded onto the column which was then recentrifuged. The eluent was depolymerized before gel electrophoresis. Depolymerization of Samples The eluent was depolymerized by adding an equal volume of 125 mM Tris-HCl buffer, pH 6.8 containing 4% (w/v) SDS, 4% (v/v) 2-mercaptoethanol, 20% (v/v) glycerol and 0.002% (w/v) bromophenol blue, and the mixture heated at 100°C for 5 min. The sample was cooled to 20°C. Lyophilized samples were dissolved i n a solution containing 62.5 mM Tris-HCl buffer, pH 6.8, 2% (w/v) SDS, 2% (v/v) 2-mercaptoethanol, 10% (v/v) glycerol and 0.001% (w/v) bromophenol blue and heated at 100°C as above. The DCCD-binding polypeptide was depolymerized in the same buffer as described previously except that the f i n a l Tris concentration was 50 mM and the sample was heated at 45°C for 18 h. The protein concentration of the depolymerized sample was i n the range of 0.25-1 mg/ml. In some instances, centrifugation of the depolymerized sample in a desk-top centrifuge (International Equipment Company, model CL45436M) at f u l l speed for 5 min was needed to remove any insoluble material prior to gel electrophoresis. 64. Slab-Gel Electrophoresis (i) Preparation of Separating Gel SDS-polyacrylamide gels were run using the discontinuous buffer system of Laemmli (171). A 13% (w/v) acrylamide gel was prepared as follows: 15 ml of separating gel containing 0.375 M Tris-HCl buffer, pH 8.8, 0.1% (w/v) SDS, 13% (w/v) acrylamide and 0.35% (w/v) N,N'-methylene-bis- acrylamide was prepared (the acrylamide and N,N'-methylene-bis-acrylamide were made up as a stock solution of 30% (w/v) and 0.8% (w/v), respectively, i n d i s t i l l e d water, fi l t e r e d through a Whatman 1 MM f i l t e r and stored in the dark at 4°C). 75 ul of freshly-prepared 10% (w/v) ammonium per- sulphate was added to the mixture and the solution degassed by aspiration. Polymerization of the solution was initiated by addition of 15 ul of TEMED and the solution was quickly poured between two glass plates of a BioRad slab-gel former (172) separated by 0.75 mm spacers. A pocket-former consisting of a teflon comb, 0.75 mm thick, and with a row of ten or twenty teeth, was inserted between the upper edges of the plates and the separating gel was poured to within 1.0 - 1.5 cm from the bottom of the comb. A small amount of tertiary butanol was gently layered over the gel so as to obtain a straight boundary during polymerization. The gel was allowed to poly- merize at 20°C for 1-2 h or in some cases, overnight, after which the tertiary butanol was removed by means of a f i l t e r paper. The surface of J the gel was rinsed with d i s t i l l e d water and the excess water removed with a f i l t e r paper. The pocket-former was reinserted between the plates and the stacking gel poured to the top of the comb. In some experiments, the separating gel consisted of 10% (w/v), 15% (w/v) or a linear gradient of 7.5 to 16.5% (w/v) acrylamide. 65. ( i i ) Preparation of Stacking Gel A 5 ml stacking gel solution containing 0.125 M Tris-HCl buffer pH 6.8, 0.1% (w/v) SDS, 4% (w/v) acrylamide and 0.105% (w/v) N,N'-methylene- bis-acrylamide was prepared.^ To this was added 15 ul of a freshly- prepared solution of 10% (w/v) ammonium persulphate and the solution deaerated. Polymerization was initiated by adding 10 ul of TEMED and the solution layered over the separating gel. The gel was allowed to poly- merize for 1.5-2 h after which the pocket former was carefully removed and the pockets rinsed and f i l l e d with electrode buffer, pH 8.4 consisting of 25mM Tris base, 192 mM glycine and 0.1% (w/v) SDS. The depolymerized samples (25-100 ug protein) were applied to the gel beneath the electrode buffer by means of a Hamilton syringe. Gel Electrophoresis in Tubes Two types of gels were used. (i) A separating gel solution was prepared as described before under "Preparation of separating gel". After i n i t i a t i o n of polymerization, the gel solution was poured into a glass tube to a height of 10 cm. The tube had a length of 13 cm with an internal diameter of 6 mm. The lower end was sealed with a piece of parafilm. The surface of the gel was overlayed with tertiary butanol and the gel allowed to polymerize for 1.5-2 h. Following removal of the tertiary butanol, a 4% (w/v) acrylamide stacking gel, prepared as described above was. layered on top of the separating gel to a height of 1.0-1.5 cm. Tertiary butanol was again layered on top of the stacking gel solution which was allowed to stand at 20°C for 2-3 h. The gel surface was then rinsed with d i s t i l l e d water, then with electrode buffer. The tube was f i l l e d to the surface with electrode buffer, pH 8.4 66. and the sample applied to the surface of the gel beneath the buffer as before. ( i i ) The second method, described by Fillingame (144), is a modifi- cation of that of Laemmli (171) and includes 8 M urea. A gel solution consisting of 0.375 M Tris-HCl buffer, pH 8.8, 0.5% (w/v) SDS, 8 M urea, 10% (w/v) acrylamide, 0.5% (w/v) N,N'-methylene-bis-acrylamide and 0.075% T (w/v) ammonium persulphate was deaerated and polymerization initiated by addition of TEMED at a concentration of 125 y l per ml of gel solution. The solution was immediately poured into 13 cm x 0.6 cm (internal diameter) glass tubes, to a height of 10 cm. Tertiary butanol was carefully layered on top of the gel solution and allowed to stand at 20°C for 1.5-2 h. The surface of the gel was rinsed with d i s t i l l e d water, and then with electrode buffer, pH 8.4 which contained 25 mM Tris base, 192 mM glycine and 0.2% (w/v) SDS. The stacking gel was omitted. Samples were loaded onto the gel surface as before. Gradient Gel Electrophoresis v One of two procedures was used: ( i ) The f i r s t method used was developed by Cox and coworkers (173) and i s a modification of the Laemmli method (171). Two gel solutions were prepared simultaneously. 6 ml of a gel solution containing 0.375 M Tris-HCl buffer, pH 8.8, 0.1% (w/v) SDS, 7.5% (w/v) acrylamide, 0.2% (w/v) N,N'-methylene-bis-acrylamide and 0.033% (w/v) ammonium persulphate and a second 6 ml solution containing 0.375 M Tris-HCl buffer, pH 8.8, 10% (v/v) glycerol, 0.1% (w/v) SDS, 16.5% (w/v) acrylamide, 0.44% (w/v) N,N'-methylene-bis-acrylamide and 0.01% (w/v) ammonium persul- phate were deaerated. Polymerization was initiated by addition of 6 ul and 3 Ul of TEMED to the s o l u t i o n , r e s p e c t i v e l y . Both s o l u t i o n s were immediately t r a n s f e r r e d to a Buchler Instruments gradient-maker attached to a Buchler Instruments p o l y s t a l t i c pump by means of a Tygon tubing (2 mm i n t e r n a l diameter). The o u t l e t of the tubing was placed between the two p l a t e s of a s l a b - g e l former separated by two 0.75 mm t h i c k l u c i t e spacers. The g e l mixture was allowed to d r i p between the g l a s s p l a t e s such that a l i n e a r gradient of 7.5% (w/v) acrylamide on the top to 16.5% (w/v) a c r y l - amide a t the bottom was obtained. T e r t i a r y butanol was used to g i v e a s t r a i g h t boundary during p o l y m e r i z a t i o n , a f t e r which a s t a c k i n g g e l (4% (w/v) acrylamide) was c a s t . The pockets were washed and f i l l e d w i t h electode b u f f e r pH 8.4 and sample loaded onto the g e l surface as p r e v i o u s l y d e s c r i b e d . ( i i ) The second method was a m o d i f i c a t i o n of the procedure of Weber and Osborn (174) and i n v o l v e d a phosphate-buffered system. Two g e l s o l u t i o n s were prepared simultaneously. A 6 ml s o l u t i o n c o n s i s t i n g of 0.1 M sodium phosphate b u f f e r , pH 7.2, 0.1% (w/v) SDS, 7.5% (w/v) acrylamide, 0.2% (w/v) N,N'-methylene-bis-acrylamide and 0.033% (w/v) ammonium p e r s u l - phate and a second 6 ml s o l u t i o n c o n t a i n i n g 0.1 M sodium phosphate b u f f e r , pH 7.2, 10% (v/v) g l y c e r o l , 0.1% (w/v) SDS, 16.5% (w/v) acrylamide, 0.45% (w/v) N,N'-methylene-bis-acrylamide and 0.01% (w/v) ammonium persulphate were degassed. P o l y m e r i z a t i o n was i n i t i a t e d by a d d i t i o n of 6 P i and 3 ul of TEMED, r e s p e c t i v e l y . A 0.75 mm t h i c k gradient s l a b g e l was ca s t as i n the discontinuous b u f f e r system. The s t a c k i n g g e l c o n s i s t e d of 0.1 M sodium phosphate b u f f e r , pH 7.2, 0.1% (w/v) SDS, 4% acrylamide, 0.11% (w/v) N,N'-methylene-bis-acrylamide and 0.03% (w/v) ammonium persulphate. P o l y m e r i z a t i o n was i n i t i a t e d by a d d i t i o n of 2 Ml TEMED per ml of g e l s o l u t i o n . The sample pockets were r i n s e d and f i l l e d w i t h an el e c t r o d e 68. buffer consisting of 0.1 M sodium phosphate, pH 7.2 and 0.1% (w/v) SDS. Samples were then loaded onto the gel. Electrophoresis ( i ) Slab Gels The two glass plates containing the gel were attached to a gel electro- phoresis c e l l (BioRad, Model 220, dual vertical slab gel electrophoresis ce l l ) and the samples loaded onto the gel surface. The cathode and the anode reservoirs were then f i l l e d with the appropriate electrode buffer and electrophoresis was carried out at 20°C at a constant current of 30 mA and 40 mA for 0.75 mm and 1.5 mm thick gels, respectively. Electrophoresis was stopped when the bromophenol blue dye was within 0.5-1 cm from the anodic edge of the gel.. ( i i ) Tube Gels The tube containing the gel was placed in a BioRad Model 150A gel electrophoresis c e l l and sample loaded onto the gel surface. The appro- priate electrode buffer was placed in the cathode and anode chambers and electrophoresis was carried out at 20°C and at a constant current of 5 mA unt i l the dye front was within 0.5-1 cm from the anodic edge of the gel. TWO-DIMENSIONAL ISOELECTRIC FOCUSING GEL ELECTROPHORESIS Preparation of Sample (i) This was based on the method of Cox and coworkers (140). 10 ml of precooled (4°C) acetone was added to 15 mg of membrane protein (10 mg/ml). After 30 min at 4°C, the mixture was centrifuged i n a desk-top centrifuge at maximum speed for 10 min. The pellet was again extracted with 10 ml of acetone and then dried under a .stream of nitrogen gas at 20°C. The dried material was taken up in lysis buffer containing 9.5 M urea, 2% (w/v) Nonidet P-40, 1.6% (w/v) pH 5-7 ampholytes, 0.8% (w/v) pH 3.5-10 ampholytes and 5% (v/v) 2-mercaptoethanol, and centrifuged at 15 600 xg for 10 min to remove any insoluble material. Extraction with acetone was omitted when either immunoprecipitated or purified ATPase complex preparations were analyzed. ( i i ) The second method was that developed by Merril et a l . (175). To 1 ml of membrane protein (5 mg/ml) was added 1 volume of denaturation solution containing 4% (w/v) SDS, 10% (v/v) 2-mercaptoethanol, 40% (v/v) glycerol, 4.5% (w/v) Nonidet P-40, 3.2% (w/v) pH 5-7 ampholytes and 0.16% (w/v) pH 3.5-10 ampholytes. The mixture was sonicated at 50 W and 0°C in a Branson W185D Sonifier, for 15 second periods for a total of 1 min. It was then heated at 95°C for 5 min, after which i t was cooled to 20°C and centrifuged at 15 600 xg for 10 min. To the supernatant was added urea at a concentration of 56 mg per 100 y l (6.6 M urea) and then applied to the gel. First-Dimension Isoelectric Focusing The first-dimension gel was prepared as described by O'Farrel (176) with the following modifications. The first-dimension gel contained 9.5 urea, 3.77% (w/v) acrylamide, 0.215% (w/v) N,N'-methylene-bis-acrylamide, 2% (w/v) Nonidet P-40, 1.6% (w/v) pH 5-7 ampholytes, 0.8% (w/v) pH 3.5-10 ampholytes and 0.03% (w/v) ammonium persulphate. Polymerization was initiated by addition of 1 y l TEMED per ml of gel solution and the gel cast in tubes of 2 mm internal diameter and to a height of 10 cm. Tertiary butanol was layered over the gel which was allowed to stand at 20°C for 6-8 h. The surface of the gel 70. was then rinsed with lysis buffer. 50-250 yg of the prepared sample was loaded onto the gel and overlayed with a solution of 8 M urea i n water. The gels were placed i n a BioRad Model 150A gel electrophoresis c e l l and the anode and cathode chambers f i l l e d with freshly prepared 0.01 M Ĥ PÔ and deaerated 0.02 M NaOH, respectively. The first-dimension was run at 20°C and 400-450 V for 16-18 h, followed by 800 V for 1-2 h, to give a total of 9000 to 9500 volt-hours. Determination of pH Gradient In the first-dimension, a gel loaded with sample preparation buffer rather than the prepared sample was run simultaneously. After electro- phoresis, the acidic or basic end of the gel was identified and the gel cut into 1 cm segments. Each segment was incubated with 1.5 ml of deaerated d i s t i l l e d water for 1.5-2 h at 20°C after which the pH was measured. Second-Dimension SDS-Polyacrylamide Gel The second-dimension gel was either a 13% (w/v) or a linear 7.5% to 16.5% (w/v) acrylamide gel prepared in the discontinuous buffer system of i Laemmli (171). The slab gels were cast as described before but modified as. follows. The gel was 1.5 mm thick rather than 0.75 mm and the pocket former was not used with the stacking gel. Instead, the stacking gel was cast to a height of 1.5-2 cm and was about 3 mm from the upper edge of the glass plate. Prior to running the second dimension, the acidic or basic end of the gel from the f i r s t dimension was identified with India ink'. The gel was .incubated for 3 h in 5 ml of 125 mM Tris-HCl buffer, pH 6.8 containing 10% 71. (v/v) glycerol, 4% (w/v) SDS, 4% (v/v) 2-mercaptoethanol and 0.001% (w/v) bromophenol blue. When the gel from the first-dimension was not to be run immediately in the second dimension, the gel in the above solution was frozen rapidly i n a dry-ice ethanol mixture and stored at -70°C. The gel from the f i r s t dimension was mounted above the second- dimension gel in 1% (w/v) agarose in 62.5 mM Tris-HCl buffer, pH 6.8, 10% (v/v) glycerol, 2% (w/v) SDS, 2% (v/v) 2-mercaptoethanol and 0.001% (w/v) bromophenol blue. . ' i Electrophoresis was carried out as previously described except that i t was continued for 10-15 min after the dye front had run off the lower (anode) end of the gel. The gel could then be stained immediately without removal of ampholytes. CROSSED IMMUNOELECTROPHORESIS This was based on the methods of Nowotny (177), Bjerrum and Lundhal (178) and Mayer and Walker (179). First-Dimension Gel Electrophoresis 2.4 ml of 1% (w/v) agarose in Bjerrum buffer, pH 8.8 containing 100 mM glycine, 38 mM Tris and 1% (w/v) Triton X-100 at 55°C was poured onto a 50 x 50 mm glass plate surrounded by a plastic mold. The agarose was allowed to cool to 20°C and the mold carefully removed such that a 40 x 40 x 1.5 mm gel was obtained. Agarose i n the same buffer was layered around the periphery of the cast gel to the edges of the plate (agarose bridges) to give a more uniform conduction of the current. A row of four wells (3 mm diameter), about 1 cm from the cathodic edge of the gel, and about 1 cm apart, was cut out with a BioRad gel-puncher. The wells were f i l l e d with various levels of antigen and the gel placed in a Pharmacia flat-bed electrophoresis unit at 4°C. Wicks, 4 cm wide (Ultrawicks, BioRad) were placed onto the agarose bridges in such a manner that they did not overlap , onto the wells. The chambers were f i l l e d with Bjerrum buffer, pH 8.8, without any detergent and electrophoresis was carried out for 1.5-2 h at 4°C and 100 V. After electrophoresis, the gel was cut into strips (5 mm width) such that each contained a sample well at one end. The individual strips were stained immediately or run in the second dimension. Second-Dimension Gel Electrophoresis 1% (w/v) agarose in Bjerrum buffer, pH 8.8 and various levels of antiserum (f i n a l volume, 2.4 ml) were'mixed at 55°C and cast into a 40 x 40 x 1.5 mm gel as above. A strip of the gel from the first-dimension was placed at the cathodic end of the gel and agarose bridges constructed around the gel. The electrode buffer was the same as in the f i r s t dimension. Electrophoresis was carried out perpendicular to the direction of the f i r s t dimension for 16 to 18 h at 10 V and 4°C. STAINING AND DRYING GELS SDS-Polyacrylamide Gels After electrophoresis, these gels were either stained immediately or fixed prior to staining. Gels were fixed in either a solution of: (i ) 5% (w/v) TCA, 5% (w/v) sulfosalicylic acid and 10% (v/v) methanol for 30 min at 60°C (173) or ( i i ) 50% (w/v) TCA, overnight at 20°C (171). The staining solutions were fi l t e r e d through a Whatman 1 MM f i l e r before use. The gels were then stained in one of the following staining systems for different periods at 20°C. (i) 0.12% (w/v) Coomassie Blue ( B r i l l i a n t Blue R) i n ethanolracetic acid:water (25:8:67) (171) ( i i ) 0.05% or 0.1% (w/v) Coomassie Blue in isopropanol:acetic acid:water (25:10:65) (180) ( i i i ) 0.25% (w/v) Coomassie Blue in methanol:acetic acid:water (45:10:45) (174) (iv) 1% (w/v) Amido Black 10B i n 7% (v/v) acetic acid (165). Gels were then destained in 10% (v/v) acetic acid un t i l a f a i r l y clear background was obtained. The tube gels were scanned at 545 nm and 620 nm when Coomassie Blue and Amido Black were used, respectively. Slab gels were incubated i n a solution of 10% (v/v) acetic acid and 2% (v/v) glycerol for 4-6 h and then dried onto a piece of Whatman 3 MM paper i n a Gel Slab Dryer, Model 224 (BioRad), equipped with a heating unit and connected to a vacuum pump. Crossed Immunoelectrophoresis Gels The gels were processed by the methods of Weeke (188) and Mayer and Walker (179). After electrophoresis, the glass plate containing the gel was taken and the sample well f i l l e d with 0.9% (w/v) NaCl. A 5 x 5 cm piece of Whatman 3 MM paper was carefully placed on top of the gel, followed by a pad of absorbent paper towels, a glass plate and f i n a l l y a weight of about 0.5 Kg. After 10 min at 20°C, the pressed gel was washed thrice by incubating each time with 0.9% (w/v) NaCl for 20 min. The gel was then soaked i n d i s t i l l e d water for 20 min and then a piece of 5 x 5 cm Whatman 3 MM paper gently placed on top of the gel. This was placed i n an oven at 70°C and allowed to dry. After the gel had dried onto the glass 74. plate, the paper was gently removed and the cooled (20°C) plate was incubated for 30 min in a solution of 0.25% (w/v) Coomassie Blue in methanolracetic acid:water (45:10:45). The plate was rinsed with d i s t i l l e d water and destained in a solution containing 10% (v/v) acetic acid and 45% (v/v) methanol for 5-10 min or un t i l the stained proteins (rockets) could be seen against a sufficiently colourless background. Finally, the plate was rinsed with d i s t i l l e d water and dried with a hair-dryer. It was kept i n the. dark. PURIFICATION OF GOAT ANTI-RABBIT IMMUNOGLOBULIN BY AFFINITY CHROMATOGRAPHY Preparation of Affinity Column The a f f i n i t y column was prepared by L. Molday (University of British Columbia) using the method of Cuatrecasas (181). 750 mg of CNBr was dissolved in 10 ml of d i s t i l l e d water and mixed with 20 ml of washed Sepharose 4B (equal volumes of resin and water) which had previously been adjusted to pH 11 with NaOH. The suspension was stirred at 0°C for 25 min and the pH kept between 10 and 11 by addition of 10 M NaOH. The gel was fil t e r e d and washed with 200 ml of d i s t i l l e d water, then with 0.1 M sodium borate buffer pH 9.0. The resin was washed for 2-3 min since CNBr-activated agarose i s unstable. 80 mg of rabbit Y-globulin in 4 ml of 0.1 M sodium borate buffer pH 9.0 was added to 20 ml of the CNBr-activated gel (equal volumes of gel and buffer) in borate buffer and then stirred at 20°C for 16-18 h. The amount of protein bound to the resin was determined from the amount s t i l l remaining unbound i n the supernatant following sedimentation of the gel. This was calculated from absorbance at 280 nm and using the.extinction of 0.1% . £280 = 1 ' 3 5 - The r e s i n was then washed w i t h 0.1 M sodium borate b u f f e r pH 9.0 and the r e s i d u a l a c t i v e s i t e s quenched by i n c u b a t i n g the r e s i n i n s e v e r a l volumes of 0.1 M g l y c i n e f o r 6-8 h at 20°C. F i n a l l y , the r e s i n was washed w i t h phosphate-buffered s a l i n e , pH 7.5 (PBS: 0.137 M NaCl, 2.68 mM KC1, 1.47 mM KH 2P0 4, 8.09 mM Na^PO^) and stored i n the same b u f f e r i n the presence of 20 mM NaN^ at 4°C. P u r i f i c a t i o n of Goat An t i - R a b b i t Immunoglobulin The r a b b i t Y-globulin-Sepharose 4B a f f i n i t y column i n PBS was poured i n t o a 10 x 1 cm BioRad column to a height of 4 cm and e q u i l i b r a t e d w i t h PBS. 5 ml of goat a n t i - r a b b i t serum was a p p l i e d to the column a t 20°C and washed w i t h s e v e r a l volumes of PBS. Goat a n t i - r a b b i t immunoglobulin was el u t e d from the column w i t h 3M NaSCN i n d i s t i l l e d water (pH 7.5). F r a c t i o n s c o n t a i n i n g goat a n t i - r a b b i t immunoglobulin, as determined by SDS-polyacrylamide g e l e l e c t r o p h o r e s i s , were pooled and d i a l y z e d against 50 volumes of 0.1 M NH 4HC0 3 b u f f e r pH 7.4 c o n t a i n i n g 20 mM NaN^ f o r 24 h a t 4°C w i t h exchanges of the e x t e r n a l b u f f e r at i n t e r v a l s of 6 h. The d i a l y z e d m a t e r i a l was l y o p h i l i z e d and stored a t -70°C. RADI0I0DINATI0N OF GOAT ANTI-RABBIT IMMUNOGLOBULIN To 1 mg of a f f i n i t y - p u r i f i e d goat a n t i - r a b b i t immunoglobulin (1 mg/ml) i n phosphate bu f f e r e d s a l i n e , pH 7.4 (PBS) was added 1 mCi of N a ( l 2 5 I ) fo l l o w e d by 100 u l of a 0.4% (w/v) Chloramine T s o l u t i o n . The mixture was incubated i n the fume-hood at 20°C f o r 20 min, a f t e r which, i t was loaded onto a 19 x 1.5 cm column of BioGel P6DG which had been e q u i l i b r a t e d w i t h PBS. The column was e l u t e d w i t h the same b u f f e r at a f l o w r a t e of 0.1-0.15 ml per min and 0.9 ml f r a c t i o n s c o l l e c t e d . F r a c t i o n s c o n t a i n i n g 76. protein (determined by absorbance at 280 nm) and the highest radioactivity ) were pooled, made up to 20 mM NaN^, and stored at 4°C. A specific radio- activity of 0.5 to 2 x 106 dpm/ug was obtained PEPTIDE MAPPING OF CNBr-CLEAVED DCCD-BINDING PROTEIN One-Dimensional Thin Layer Separation A 20 x 20 glass plate precoated with s i l i c a gel G (250 um, Analtech Inc.) was activated by placing i t in an oven at 100°C for 1 h. The plate was cooled to 20°C and then the sample (in 80% (v/v) formic acid) was applied as a spot of 2-3 mm in diameter. The sample was applied repeatedly on the same spot and f i n a l l y air-dried at 20°C. The plate was placed in a covered glass tank, which had been equilibrated with the appropriate solvent, and developed at 20°C. The solvent system consisted of one of the following: (1) chloroformmethanol (9:1 (v/v)). Run time, 40 min. (2) n-butanol:acetic acid:water (4:1:5). Run time, 3 h. (3) chloroformmethanol:water (65:25:4) containing 20 mM HC1. Run time, 65 min. (4) n-butanol:pyridine:acetic acid:water (60:40:12:48). Run time 4.5-5 h. After the plates were developed, they were dried overnight at 20°C and the positions of the peptides determined. Two-Dimensional Thin Layer Separation The f i r s t dimension was thin-layer electrophoresis. Prior to electrophoresis, a 20 x 20 cm cellulose MN300 plate (CEL 300-10, 0.1 mm, Macherey-Nagel Co.) was developed in n-butanolrpyridineracetic acid:water (60:40:12:48) perpendicular to the direction of electrophoresis, for 6 h at 20°C and then thoroughly air-dried. The plate was then sprayed with thin-layer electrophoresis (TLE) buffer, pH 6.45 consisting of a mixture of pyridine:acetic acidrwater (20:0.7:180) and electrophoresis carried out at 400 V for 1 h i n a Camag TLE apparatus. The sample (in 80% (v/v) formic acid) was spotted (2-3 mm diameter) onto the dried plate at the midpoint between the electrodes and 2.5 cm from the edge of the plate. The spot was air-dried and protected by gently inverting a clean v i a l over the spot. The plate was resprayed with TLE buffer and through diffusion, the area under the v i a l was dampened with buffer. Electrophoresis was carried out at 400 V for 65 min, after which the plate was thoroughly dried at 22°C, overnight. Separation i n the second dimension was by ascending chromatography. A covered glass chromatography tank was equilibrated at 20°C with a solvent mixture of n-butanol:pyridine:acetic acid:water (60:40:12:48) by lining the tank with sheets of Whatman 3 MM paper soaked with( the solvent mixture. The plate from the f i r s t dimension was then developed in the solvent system such that the direction of migration of the solvent front was perpendicular to the direction of electrophoresis. After 5.5-6 h, at which point the solvent front was 2-3 cm from the top edge of the plate, the plate was removed, thoroughly dried at 20°C, and the peptides detected. Detection of Peptides i The dried thin-layer plate was sprayed with a solution containing 1% (w/v) ninhydrin and 4% (v/v) acetic acid i n acetone, and then developed i n an oven at 100°C for 15 min. An outline of the peptide spots was made 78. immediately. In some cases, the spots were fixed by spraying the plate with solution containing 1 ml saturated Cu(NC>3)2, 0.2 ml 10% (v/v) HN03 and 100 ml 96% (v/v) ethanol. The plate was air-dried and stored in the dark. CHEMICAL.MODIFICATION OF MEMBRANES Labelling of Membrane Vesicles of E. c o l i with ^"CjDCCD [l"C]DCCD (50 mCi/mMol) was purchased in ether i n a sealed ampoule. It was opened and the contents transferred to a v i a l . The ether was evaporated at 20°C in the fume-hood and the dried contents taken up in absolute ethanol and stored at -20°C. The method of labelling was based on that described by Fillingame (144). Membranes prepared i n TM buffer, pH 8.0 were suspended i n 10 mM Tris-H 2S0 4 buffer, pH 7.6 containing 0.25 M sucrose, 5 mM MgS04 and 0.2 mM DTT at a protein concentration of 20 mg/ml. To 0.5 ml of membrane suspen- sion was added 0.02 volumes (10 ul) of 5 mM [l"C]DCCD i n absolute ethanol and the suspension stirred at 4°C for 24 h. The labelled membranes were washed three times by resuspension in buffer following centrifugation at 250 000 xg for 2 h. The membranes were used immediately or stored at -20°C. Treatment of Membrane Vesicles with Phenylglyoxal A 200 mM phenylglyoxal solution was prepared by dissolving 53.8 mg in 1.8 ml of 50 mM HEPES-K0H buffer, pH 7.5 containing 10 mM MgCl 2 and 10% (v/v) glycerol. The pH was readjusted to 7.5 by addition of KOH and the f i n a l volume made up to 2 ml with buffer. Two ml of urea-stripped membrane vesicles in 50 mM HEPES-K0H buffer, pH 7.5 containing 10 mM MgCl 2 and 10% (v/v) glycerol at a protein concen- tration of 12.5 mg/ml was incubated with 1 volume of 200 mM phenylglyoxal in the dark at 20°C for 3 h. The suspension was then diluted 8-10 fold with buffer and centrifuged at 250 000 xg for 2.5 h. The membranes were washed once by resedimentation in buffer and finaly suspended at a protein concentration of 10 mg/ml. They were used immediately or rapidly frozen and stored at -70°C. CHEMICAL MODIFICATION OF THE DCCD-BINDING PROTEIN Hydrolysis of the DCCD-Binding Protein This was based on the method of Moore and Stein (182). Heavy-walled 13 x 100 mm Pyrex culture tubes were washed sequentially i n chromic acid, alcoholic KOH and 1 M HC1. They were rinsed with d i s t i l l e d water after each step and f i n a l l y dried at 100°C. 100-150 ug of purified DCCD-binding protein i n chloroformmethanol (2:1) containing 20 mM ammonium acetate, was taken to dryness in a 13 x 100 mm Pyrex tube at 16-18°C by using a rotary evaporator (Buchler Instruments, Rotary Evapo-mix). The protein i n the tube was taken up in 1.5-2 ml of 6 M HC1. With an oxygen flame, a section of the tube about 3 cm from the top was partially constricted. The lower half of the tube was immersed in a dry ice-ethanol mixture for 5-10 min, at which point the solution in the tube started to become viscous. The tube was then connected to a vacuum pump through a piece of Tygon tubing and the system evacuated to a pressure of less than 50 microns. During this period, gently warming the lower half of the tube by hand caused bubbles to rise from the viscous solution. Re-immersing the tube into the dry ice mixture broke the bubbles. The tube was sealed at the constriction, when bubble formation had almost ceased. The vacuum seal 80. was checked for leaks with an ionizing gun. Hydrolysis was then conducted at 110°C ± 2°C for 24, 31, 42 and 60 h after which the tubes were cooled to 20°C. Any liquid on the inner walls of the tube was spun down by centrifugation in a desk-top centrifuge (International Equipment Company, model CL 45436 M) at 900 xg for a few minutes. The,tube was scored with a f i l e near the constriction and cracked open by means of a hot glass rod. Hydrochloric acid was removed under reduced pressure at 70-80°C by placing the sample tube i n an 18 x 150 mm Pyrex ignition tube and attaching i t to the rotary evaporator. Almost a l l of the HC1 was removed in 45-60 min. The dried contents were taken up in the appropriate buffer for analysis of i t s amino acid composition. The amino acid composition was determined by using the Durram (model D-500) amino acid analyzer. Treatment of the DCCD-Binding Protein with Cyanogen Bromide The method of Sebald et a l . (183) was followed. 2 mg of purified DCCD-binding protein was taken to dryness and the residue dissolved in 0.8 ml of 1 M cyanogen bromide in 98% (v/v) formic acid. 0.2 ml d i s t i l l e d water was added to the mixture which was incubated i n the dark at 20°C for 17-18 h. The solvent was then removed by rotary evaporation under reduced pressure at 30-40°C as previously described. ^The dried residues were dissolved in the appropriate buffer. Control experi- ments were simultaneously carried out with DCCD-binding protein, which were treated in the same manner but in the absence of cyanogen bromide. Treatment of the DCCD-Binding Protein with Performic Acid This was done according to the method of Hirs (184). Performic acid was prepared by mixing 9 ml of 90% (v/v) formic acid 81. with 1 ml of 30% (v/v) H 20 2 and incubating the mixture at 20°C for 1 h. The solution was cooled to 0°C and used immediately. DCCD-binding protein in chloroformmethanol (2:1) containing 20 mM ammonium acetate was taken to dryness and performic acid added to the dry protein such that the protein concentration was 1-2% (w/v). The tubes were capped and incubated'in the dark at 0°C for 20 h, after which the performic acid was removed by evaporation under reduced pressure at 35-40°C. Evapora- tion to dryness took about 45 min. The residue was either subjected to hydrolysis i n the same tube to determine i t s amino acid composition as described before or taken up in the appropriate buffer. In some cases, control experiments were simultaneously carried out with DCCD-binding pro- tein, which were treated in the same manner with 80% (v/v) formic acid only. Treatment of the DCCD-Binding Protein with 2,3-Butanedione or Phenylglyoxal A 200 mM solution of 2,3-butanedione and a 150 mM solution of phenylglyoxal were made up in 100 mM sodium borate buffer containing 2% (w/v) Triton X-100, and the pH ad jus-ted to 8. 8 withNaOH. 0.3 mg of dried DCCD-binding protein was dissolved in I ml of borate- Triton buffer, pH 8.8 and incubated with an equal volume of either 200 mM 2,3-butanedione or 150 mM phenylglyoxal i n the dark at 20°C for 3 h. 1 ml of 330 mM L-arginine hydrochloride in 100 mM sodium borate buffer, pH 8.8 was then added. The f i n a l mixture was incubated for another 30 min and then lyophilized. The dried residue was dissolved in borate-Triton buffer at a protein concentration of 0.15 mg per ml. Control experiments were simultaneously carried out with DCCD-binding protein, which were treated i n the same manner but in the absence of either 2,3-butanedione or phenylglyoxal. 82. TREATMENT WITH PROTEASES ECFj^ and DCCD-Binding Protein ECF^ obtained after purification on AH-Sepharose CL-4B was suspended i n 50 mM HEPES-KOH buffer, pH 7.5 containing 10 mM MgCl 2 and 10% (v/v) glycerol and treated with one of the following: a-chymotrypsin, Staphylococcus aureus Vg protease or pronase at a protein:protease ratio of 10:1. The f i n a l protein concentration of ECF^ was 0.75 mg/ml. Similarly, DCCD-binding protein in 100 mM sodium borate buffer, pH 8.8 containing 2% (w/v) Triton X-100 was treated with one of the above proteases at a protein:protease ratio of 10:1. The DCCD-binding protein was also treated with TPCK-trypsin at a protein:protease ratio of 10:1. The reaction mixtures were incubated at 37°C for 3 h after which TPCK-trypsin was inhibited by addition of an equivalent amount of soybean trypsin inhibitor followed by incubation for 15 min at 37°C. The f i n a l concentration of DCCD-binding protein in the reaction mixtures was 0.13 mg protein/ml. 25 pi of these mixtures were incubated i n flex-vinyl micro- t i t r e wells, which had been previously coated with 0.1% (w/v) polyrL-lysine. For controls, 25 Pi of these mixtures containing no DCCD-binding protein was also incubated i n polylysine-treated microtitre wells. The appropriate radioimmune assays were then carried out as described in the legends to the Figures i n the RESULTS' section. PREPARATION OF ANTIGENS FOR IMMUNIZATION ECF^ ECF^ was prepared for injection by Helga Stan-Lotter (University of Br i t i s h Columbia) as follows: 0.7 mg of ECF (obtained after AH-Sepharose 4B chromatography) at 83. 1.4 mg/ml was mixed with 1 volume of complete Freund's adjuvant and the mixture emulsified by repeated passing between two syringes connected through a 22 gauge needle. The mixture was ready for injecting into the rabbit when a drop of the emulsion placed on the surface of the water did not disperse when the beaker was slightly agitated. DCCD-Binding Protein The DCCD-binding protein was dissolved in a solution containing 0.9% (w/v) NaCl and 2% (w/v) SDS at a protein concentration of 2 mg/ml. It was mixed with two volumes of complete Freund's adjuvant and emulsified. In some preparations, incomplete Freund's adjuvant was used. IMMUNIZATION OF THE RABBIT / A female, New Zealand rabbit of about 3 Kg was obtained from the University of British Columbia Animal Care Unit, several months prior to immunization and maintained as recommended (185). A month after i t s arr i v a l , the rabbit was bled at 4-week intervals u n t i l enough preimmune serum was obtained. Three weeks after the last bleeding, the animal was immunized a follows. The rabbit was wrapped in a blanket leaving only the head and the site of injection exposed. A small amount of hair was clipped off at the area to be injected and the injection site cleaned with 70% (v/v) ethanol. A tent of skin was raised with the thumb and index finger and with the other hand, the needle (22 gauge) was inserted underneath the skin and parallel to the underlying muscle. The antigen was injected immediately, the needle withdrawn, and the site of injection wiped with ethanol. The rabbit was generally injected at two sites on the thigh of the hind legs with a total of either 0.6 mg ECF1 or 0.3 mg DCCD-binding 84. protein. Subsequent injections were at 4-week intervals, with a total of either 0.3 mg ECF.. or 0.2 mg DCCD-binding protein and incomplete Freund's adjuvant was substituted for the complete Freund's adjuvant in the case of ECF^. For the DCCD-binding protein, incomplete Freund's adjuvant wis used after the f i f t h injection as continued use of the complete adjuvant resulted in the development of hard pea-size lumps at the sites of injection. BLEEDING THE RABBIT The method of Herbert (186) was used. The rabbit was wrapped in a blanket, leaving only the head exposed. A small area over the outer marginal vein of the ear was shaved and cleaned with soap and water. The shaved area was dried and lightly smeared with petroleum j e l l y . A gauze soaked in xylene was pressed onto the shaved area for a few seconds, un t i l the marginal ear vein was visibly dilated. With a st e r i l e razor blade, an incision of about 2 mm was made into the vein i n the longitudinal direction and about 2 inches from i t s distal end. The ear was held in the horizontal position and the thumb and index finger was used to occlude the venous return. A centrifuge tube was placed under the ear to collect the blood. When the flow of blood stopped prematurely, the wound was wiped with a piece of gauze to open the wound. Usually 25 to 30 ml of blood was collected at each bleeding. The bleeding was stopped by placing a dry gauze on the wound and securing i t with a paper c l i p . On occasion, the bleeding was persistent and a piece of Gelfoam sponge (The Upjohn Co., Michigan) was used to stop the bleeding. After the bleeding had stopped, any xylene remaining was removed with 70% (v/v) ethanol and the ear wiped with soap and water. The rabbit was bled alternately from each ear at 7-10 days post-injection. / 85. SEPARATION OF SERUM This was based on the method of Garvey (187). The blood on collection clotted to some degree. The clot was separated from the wall of the centrifuge tube with a wooden applicator stick and the blood allowed to stand at 20°C for 2 h, then at 4°C for 18-24 h. The serum was removed and centrifuged at 2 000 xg for 30 min. Centri- fugation was repeated when the serum was not clear. On occasion, the serum was turbid due to the presence of l i p i d s (or haemoglobin) and could not be cleared by centrifugation at 2 000 xg. In these cases, the serum was centrifuged at 25 000 xg for 30 min. A film of l i p i d formed on the surface which was easily removed with a tissue paper. The serum was passed through a sterile 0.22 ym Millipore f i l t e r and stored at -70°C. PARTIAL PURIFICATION OF IMMUNOGLOBULINS This was based on the methods of Nowotny (177) and Mayer and Walker (179). 3 ml of saturated (NH^SO^ (neutralized to pH 7.0 with NĤ OH) was added dropwise to 5 ml of serum and the mixture stirred at 20°C. The pH of the mixture was monitored during the addition of (NH^^SO^. After 30 min of incubation, the mixture was centrifuged at 17 600 xg for 30 min. The pellet was resuspended to the original volume (5 ml) in 10 mM potassium phosphate buffer, pH 7.0 and to this was added 2.6 ml of saturated (NH^^SO^ and the precipitate removed by centrifugation. It was then taken up in 5 ml of 10 mM potassium phosphate buffer, pH 7.0 and dialyzed at 4°C against 500 ml of d i s t i l l e d water for 20 h followed by dialysis against 250 ml of 0.4 M Sucrose and 10 mM MgSÔ  for 12 h. The external medium was changed at 6 h intervals. The dialyzed material was centrifuged to remove 86. insoluble material and the supernatant concentrated to half i t s original volume (2.5 ml) by u l t r a f i l t r a t i o n using an Amicon XM-50 f i l t e r . It was stored at -70°C. BINDING OF ECFj^ TO MEMBRANE VESICLES The measurement of the extent of rebinding of ECF^ to urea-treated membrane vesicles which^had been treated with proteases or phenylglyoxal was carried out as follows. ( i ) 2-4 mg of the treated membrane vesicles i n 50 mM HEPES-KOH buffer, pH 7.5 containing 10 mM MgCl 2 and 10% (v/v) glycerol were incubated with various levels of ECF^ at 4°C for 45 min i n a f i n a l volume of 0.5 ml. The mixture was then diluted 8-10 fold with buffer and the membranes sedimented at 250 000 xg for 2.5 h. The membranes were resuspended in buffer and the ATPase activity measured. ( i i ) Urea-treated membrane vesicles were also treated with antiserum prior to rebinding of ECF^. Urea-treated membranes (15 mg/ml) in 50 mM HEPES-KOH buffer, pH 7.5 containing 10 mM MgCl 2 and 10% (v/v) glycerol were incubated at 4°C for 20 h with different amounts of antiserum in a f i n a l volume of 4.4-4.5 ml. After a 4-fold dilution i n buffer, the mixture was centrifuged at 250 000 xg for 2.5 h. The sedimented vesicles were washed once by suspension i n buffer followed by resedimentation as before. The washed vesicles were i suspended i n buffer at 6.5 mg protein per ml and reconstituted with ECF^ as described above. 87. ASSAYS Determination of Protein One of several procedures was used. (i) The protein i n most experiments was determined by the method of Lowry et a l . (189). Bovine serum albumin was used as the standard over a range of 0-300 yg. Interference by components in the sample buffer was corrected for by incorporating in the standard curve the appropriate volume of sample buffer equivalent to that assayed. Triton X-100 in high concentrations gave a greenish-yellow precipitate. In these cases, the precipitate was removed by centrifugation at 12 000 xg for 10 min and the absorbance of the supernatant measured at 500 nm. ( i i ) The protein content i n samples obtained during the isolation and purification of the DCCD-binding protein was determined by a modified procedure of Lowry et a l . (189), as done by Fillingame (144). Membrane samples were dissolved i n 0.5 M NaOH whereas DCCD-binding protein samples were taken to dryness under reduced pressure to remove organic solvents and then solubilized by adding 2.5% (w/v) SDS in 0.5 M NaOH. Both were incubated at 37°C for 2 h prior to assay. In addition, 1% (w/v) SDS was included in the 2% (w/v) Na2C03 Lowry reagent used in these assays. Bovine serum albumin was used as the standard protein over a range of 0-300 yg. ( i i i ) The third method was that of Bradford (190). This method was used for soluble enzyme preparations (e.g. ECF^ and i t s subunits) where interfering substances i n the buffer were minimal and the amount of protein to be assayed was low. Bovine serum albumin was used as the standard over a range of 0-50 yg. If an interfering substance was present in the sample buffer, an 88. equivalent volume was,incorporated i n the samples used for the standard curve. 3 ml of fi l t e r e d reagent containing 0.01% (w/v) Coomassie B r i l l i a n t Blue G-250, 5% (v/v) 95% (v/v) ethanol and 0.1% (v/v) 85% (v/v) H 3P0 4 was added to 0.1 ml of sample or standard protein and the absorbance at 595 nm measured after 15 min at 20°C. Determination of ATPase Activity Measurement of ATPase activity was done according to the method of Davies and Bragg (21). 1 The inorganic phosphate released during the reaction was determined by a modified method of Ames (122). (i) Rapid Assay This assay was used when interfering substances such as phospholipids and detergents were not present i n the sample and was generally used to determine the activity of the ECF^ preparations. The sample was incubated with 0.5 ml of 100 mM Tris-HCl buffer, pH 8.3 containing 5 mM ATP and 2.5 mM CaCl 2 at 37°C. After a period of Incubation, 2.5 ml of "inorganic phosphate assay mixture" containing 1.8 ml ammonium molybdate (0.42% (w/v) ammonium molybdate i n 2.86% (v/v) H^O^), 0.3 ml of 10% (w/v) ascorbic acid, 0.25 ml of 10% (w/v) TCA and 0.15 ml water, was added. The reaction mixture was incubated at 37°C for 15 min and the reaction stopped by placing the mixture at 0°C. The absorbance at 660 nm was measured against a reagent blank and the amount of inorganic phosphate liberated was calculated from a standard curve. MgCl,, was substituted for CaCl 2 when the activity of TPCK-trypsin-treated ECF^ was measured. ( i i ) Slow Assay This assay was used when interfering substances in the samples were • - 89. present. The procedure was essentially the same as described for the "rapid" assay except that after incubation of the sample with ATP, the reaction was stopped by addition of 0.5 ml of 10% (w/v) TCA and the reaction mixture centrifuged at 12 000 xg for 10 min 0.5 ml of the supernatant was removed and the inorganic phosphate which had been released was determined by addition of 2.5 ml of "inorganic phosphate assay mix" as before. One unit of ATPase activity was defined as the amount of enzyme which liberated 1 nmol of phophate per min at 37°C. Substrate Oxidation-Dependent Quenching of Fluorescence of 9-Aminoacridine This was based on the method of Singh and Bragg (191). 0.1 ml of everted membrane vesicles (10 mg/ml) in 50 mM HEPES-KOH buffer, pH 7.5 containing 5 mM MgCl 2 was mixed in a fluorescence cuvette in a f i n a l volume of 2 ml with 10 mM HEPES-KOH buffer pH 7.5 containing 300 mM KC1, 5 mM MgCl 2 and 2-10 UM 9-aminoacridine. The energy sources were NADH, ascorbate i n the presence of PMS or ATP. Each was prepared fresh in 10 mM HEPES-KOH buffer, pH 7.5 containing 300 mM KC1 and 5 mM MgCl 2. In the case of ascorbate, PMS was added to the membrane vesicles before 9-aminoacridine. DCCD, when required, was prepared as a stock solution in absolute ethanol and stored at -20°C. The maximum level of ethanol allowed in the reaction cuvette was 2% (v/v). Fluorescence was excited by light at 420 nm and emission was measured at 500 nm (192). At these wavelengths, there was no interference from the fluorescence of NADH. Fluorescence was measured at 20°C with a Turner Model 420 Spectrofluorometer connected to a Varian strip chart recorder. 90. Measurement of Proton Conduction in K -loaded Membrane Vesicles K +-loaded vesicles i n 0.4 M sucrose, 10 mM MgSÔ  (0.85 ml) were stirred in a glass cuvette at 20°C. Valinomycin was added to the vesicle and the pH change which followed was measured with a glass pH electrode connected to a Fisher Accumet Model 325 expanded scale pH meter. The output of the meter was amplified so that 0.2 pH units gave a full-scale deflection on the recorder (John's Scientific Linear Co.). In some experiments, the vesicles were preincubated with DCCD (in ethanol) or with ammonium sulphate purified antiserum for 45 min. at 20°C. Each assay was internally calibrated by addition of an aliquot of known concentration of acid (H 2S0 4 or HC1). Determination of Cytochrome Content This was used i n a semi-quantitative manner to detemine whether any form of separation was achieved between the ATPase complex and the cytochromes during ion-exchange, hydrophobic-exchange or gel exclusion chromatography. 1.5 ml of sample at 20°C was placed i n each of two cuvettes of 1 cm light path. 25-50 ul of 0.3% (v/v) H 20 2 was added to the reference cuvette, while the other sample was reduced with an excess of sodium dithionite. The reduced versus the oxidized (difference) spectrum was then scanned between 500 and 670 nm in a Perkin Elmer model 356 Double Beam Spectrophotometer equipped with a recorder. Baseline readings were obtained by scanning two cuvettes of untreated samples in the same range. Cytochome d content was calculated by measuring one-half the height in absorbance units from the trough at 648 nm to the peak at 628 nm i n the -1 -1 difference spectra and using the extinction coefficient of 8.51 mM cm 91. (193). The content of cytochrome was determined from the absorbance above the baseline at 594 nm and using the extinction coefficient of cytochrome a^ of 4.6 mM "*"cm (193). Cytochrome b^ content was determined from the height of the peak above the baseline at 558 nm and using the extinction coefficient of 16.0 mM "*"cm ^ (193). Determination of Catalase Activity Catalase was used as a standard of molecular weight 232 000 for calibrating g e l - f i l t r a t i o n columns. Activity was determined by the method of Beers and Sizer (194). 1.5 ml of 4% (v/v) 1^02 was mixed with 25 ml of 50 mM potassium phosphate buffer, pH 7.0. To 1.99 ml of this mixture in a cuvette was added 10 vl of the enzyme solution to be measured. The change i n absorbance at 240 nm was measured with a spectrophotometer equipped with a recorder. One unit of activity was expressed as the rate of change in absorbance at 240 nm per min at 37°C. Determination of Radioactivity in Gel Slices The tube gel was frozen at -70°C for 30-45 min and then sliced into 1 mm thick disks. Each slice was incubated with 0.5 ml of a mixture of NCS:water (9:1) at 50°C for 2 h, after which i t was cooled to 4°C and 5-10 ml of ACS added. These were chilled to 4°C i n the dark and the radio- activity determined i n a Packard Model 3000 s c i n t i l l a t i o n counter. The samples were corrected for quenching by using standard 1 *C-quenched series under identical conditions. In some experiments, the gel was sliced into 2 mm thick disks, i n which case the volume of NCSrwater (9:1) used was 92. increased to 1.0 ml. Solid Phase Radioimmune Assays The procedures of Mackenzie and Molday (195) were used. In the "competitive inhibition assay" flex vinyl microtitre plate wells were pretreated with 25 y l portions of 0.1% (w/v) polylysine at 22°C for 4 to 6 hours. The wells were subsequently washed with water to remove unbound polylysine. DCCD-binding protein (25 pi) at a concentration of 0.13 mg/ml in 100 mM sodium borate buffer, pH 8.8, containing 2% (w/v) Triton X-100 was incubated in the wells for 8 to 10 hours at 22°C. Unbound poly- peptide was removed by extensive washing with phosphate-buffered saline (PBS) (0.13 M NaCl, 2.68 mM KC1, 1.47 mM KH2P04, 8.09 mM Na2HP04, pH 7.5) containing 10 mM MgCl2» Nonspecific binding sites were then quenched by incubation overnight at 4°C with radioimmune assay (RIA) buffer consisting r of 2% (w/v) bovine serum albumin, 2% (v/v) fetal calf serum and 0.1% (w/v) NaN^ i n PBS-10 mM MgCl 2. The assay wells were then incubated with 25 yl portions taken from a mixture of varying concentrations of the free antigen (vesicles, modified DCCD-binding protein, etc.) which had been preincubated with appropriately diluted antiserum to the DCCD-binding protein (1:500 in RIA buffer) for 60 min. After 1.5 h at 22°C, the wells were washed with PBS-10 mM MgCl 2, and then incubated with 25 y l of a f f i n i t y purified 1 2 5 1 - l a b e l l e d goat anti-rabbit immunoglobulin (15-40 yg/ml RIA buffer: 1-2 x 106 dpm/yg) for 60 min. The wells were rinsed extensively with PBS-10 mM MgCl 2, cut out, and the bound radioactivity determined in a Beckman Gamma 8000 counter. Two basic versions of the "binding assay" were used which differed in the nature of the free antigen. This was either E. c o l i F or the DCCD-binding protein. Binding of the 93. free antigen to the fixed antigen was then detected by using an antiserum to the free antigen. For example, polylysine-treated micro-titer wells were incubated with 25 ul portions of the antigen to be fixed (ECF^, 0.75 mg protein/ml in 50 mM HEPES-KOH buffer, pH 7.5 containing 10 mM MgCl 2 and 10% (v/v) glycerol; DCCD-binding protein, 0.13 mg protein/ml i n 100 mM sodium borate buffer, pH 8.8, containing 10 mM MgCl 2 and 2% (w/v) Triton X-100). Following quenching of non-specific binding sites as in the competitive inhibition assay, the wells were incubated with EGF^ (0.75 mg/ml in 50 mM HEPES-KOH buffer, pH 7.5, containing 10 mM MgCl^ 10% (v/v) glycerol and 3% (w/v) bovine serum albumin) or DCCD-binding protein (0.13 mg/ml i n 100 mM sodium borate buffer, pH 8.8, containing 10 mM MgCl 2, 3% (w/v) bovine serum albumin and 2% (w/v) Triton X-100) depending on the nature of the fixed antigen. The wells were washed with PBS-10 mM MgCl 2 and then reacted with 25 ul of serial dilutions in RIA buffer of the antiserum to the free antigen. The extent of binding of the rabbit antiserum was measured with 1 2 5 I - l a b e l l e d goat anti-rabbit immunoglobulin as described above. Controls for the non-specific binding of ECF^ or DCCD-binding protein to the wells were run by omitting the fixed antigen in the procedure. Controls for potential cross-reactivity between ECF^ and the antiserum to the DCCD-binding protein, or between the DCCD-binding protein and the antiserum to ECF^, were run by omitting the free antigen in the procedure. For experiments i n which the fixed antigen (ECF^, chemically-modified or normal DCCD-binding protein, subunits of ECF^) was titrated with the free antigen (DCCD-binding protein or ECF^), the experiments were carried out as described above except that the concentration of free antigen was varied with a constant amount of fixed antigen. (The amounts are given i n 94. the legends to the Figures in the "Results" section). The antisera against ECF1 and the DCCD-binding protein were diluted 1:300 i n RIA buffer. 25 ul was used in each well. "Net binding" of free to fixed antigen was always corrected for any non-specific binding of the free antigen or for cross-reactivity with heterologous antiserum. 95. RESULTS PART I PURIFICATION OF THE ECFJFQ COMPLEX SELECTION OF AN E. c o l i STRAIN The ATPase activity of the membrane-bound enzyme from several strains of bacteria was measured at 37°C and pH 8.3 using the "slow ATPase" assay procedure as described i n MATERIALS AND METHODS. Table 8 shows the average specific a c t i v i t i e s for the ATPase enzyme in the membrane vesicles of E_. c o l i AN180, WS1 and ML308-225. Under these conditions, E. c o l i ML308-225 had a higher average specific activity (560 nmol per min per mg protein) than either E. c o l i AN180 (250 nmol per min per mg protein) or E_. c o l i WS1 (370 nmol per min per mg protein). From the results i n Table 8, i t was not known i f the different specific a c t i v i t i e s were due to each strain of E_. c o l i having a different cation to ATP ratio for optimal activity. In order to determine this, membrane vesi- cles were prepared from each strain of E_. c o l i l i s t e d i n Table 8 and suspen- ded i n 50 mM Tris-^SO^ buffer, pH 8.0. ATPase activity was then measured as previously described, except that CaC^ was replaced with different amounts of one of the following cations: MgC^, CaC^, MnC^ or ZnC^. 2+ The results of such an experiment are shown in Fig. 4. Mg had the highest capacity for stimulating the membrane-bound ATPase activity. For 2+ each strain, maximum activity was obtained at a [Mg ] to [ATP] ratio of 0.3 to 0.5; above and below which, the activity decreased significantly. These results are consistent with the findings of others (196,197) in which 2+ the maximum activity was obtained at a [Mg ] to [ATP] ratio of 0.5. 2+ In the presence of Mg , the maximum activity of E. c o l i ML308-225 Table 8 Specific A c t i v i t y of the Membrane-Bound ATPase of Different Bacterial Strains 96. Strain Specific Activity (units/mg protein) Mean ± S.D* Range E. c o l i AN 180 (6)** E. c o l i WS1 (16) E. c o l i ML 308-225 (40) 0.25 ± 0.07 0.37 ± 0.095 0.56 ± 0.117 0.12 - 0.31 0.14 - 0.56 0.39 - 0.85 * Standard deviation ** Figures in parenthesis indicate the number of membrane vesicle preparations assayed. Membrane vesicles from each strain of E. c o l i were suspended in either 50 mM T r i s - ^SO^ buffer, pH 8.0 containing 10 mM MgCl 2 or in 0.5 M Tris-H 2S0 4 buffer, pH 8.0 containing 0.25M Na^O^ and 10% (v/v) glycerol. ATPase a c t i v i t y was assayed by the "slow assay" as described in MATERIALS AND METHODS. 97. F i g . 4 Effect of cations on the membrane-bound ATPase act iv i ty of different strains of IS. c o l i . Membranes from E. c o l i ML308-225, WS1 and AN180 were prepared in 50 mM T r i s - ^ S O ^ buffer, pH 8.0 containing 1 mM MgCl2- The membrane vesicles were suspended at a protein concentration of 3.3 mg/ml i n 50 mM Tris-r^SO^ buffer, pH 8.0. ATPase ac t iv i ty was measured by the "slow ATPase" assay as described in MATERIALS AND METHODS, except that CaCl 2 was replaced with various levels of either MgCl 2 (•-•) , C a C l 2 (O-O), MnCl 2 (A~A) or Z n C l 2 ( A - A ) . The concentration of ATP in the assay mixture was 5 mM and the amount of membrane protein per assay was 16.5 Pg. Enzyme ac t iv i ty i s expressed as units per mg protein. ML 308-225 (980 nmol per min per mg protein) was present in about twice the amount of that found i n jS. c o l i AN180 (460 nmol per min per mg protein) and IS. c o l i WS1 (420 nmol per min per mg protein). The results suggested that the ATPase enzyme was li k e l y to be present in a higher amount in E_. c o l i ML308-225 than in E_. c o l i AN180 and WS1 and thus this strain would be a good source of the ECF^FQ complex. SOLUBILIZATION OF THE ECF-^Q COMPLEX Selection of Detergent In contrast to water-soluble proteins, membrane-bound proteins must be released from the membrane prior to purification. Numerous detergents have been used for this purpose (90,91). An effective detergent should satisfy certain conditions. It should not inactivate the enzyme. As some purification procedures were carried out over a period of 3 to 5 days, i t was important to select a detergent which would preserve the enzyme activity. Second, the detergent should selectively solubilize a significant amount of the enzyme without solubilizing a l l other membrane-bound proteins. On this basis, a number of detergents were tested, so that a suitable detergent could be selected for the extraction of the ATPase complex from the membranes of IS. c o l i . High ionic strength buffers have often been used to keep the F-^-ATPase associated with the FQ. Sone et a l . (102) used a buffer containing 0.5 M Tris-H 2S0 4, pH 8.0, 0.25 M Na 2S0 4 and 10% (v/v) glycerol to keep the TF^FQ complex together during solubilization of the membrane of PS3. The same buffer was used in my experiments for studying the solubilization characteristics of different detergents. 100. Membrane vesicles of E_. c o l i ML308-225 were suspended in high ionic strength buffer and treated with various levels of detergents as described in MATERIALS AND METHODS. Membrane proteins and ATPase activity were considered as being "solubilized" i f they were not sedimented following centrifugation at 250 000 xg for 2 h. The results of the treatment with various detergents are illustrated in Fig. 5. Optimal solubilization of the ATPase activity was obtained with the detergent, N-lauroyl sarcosine. At a detergent to protein ratio of 0.25, 80% of the activity was solubilized by N-lauroyl sarcosine, whilst less than 60% was solubilized by the other detergents. N-lauroyl sarcosine also exhibited some selectivity, i n that only 38% of the membrane protein was solubilized at this detergentrprotein ratio. This selectivity was lost at higher ratios. At a detergentrprotein ratio of 0.96, almost a l l the activity was solubilized (96%), while at the same time, the amount of protein solubilized increased to 80%. Ammonyx Lo and sodium deoxycholate also solubilized significant amounts of ATPase activity at higher detergent to protein^ratios. At at . ratio of 0.4, Ammonyx Lo and deoxycholate solubilized 80% and 50% of the activity, respectively. With sodium cholate, only 50% of the activity was solubilized at a ratio of 0.8. Sodium cholate and deoxycholate required the presence of high ionic strength buffers, in order for significant amounts of membrane proteins to be solubilized (94,198). As shown In Table 9, sodium cholate at a detergentrprotein ratio of 0.69 in the presence of low ionic buffer (50 mM Tris-H 2S0 4, pH 8.0, 10 mM MgCl2 and 10% (v/v) glycerol), solubilized 3.9% of the activity and 8.4% of the membrane proteins. But at the same ratio, i n the presence of high ionic strength buffer, sodium cholate Detergent -- mg P r o t e i n -- mg Fig. 5 Solubilization of the membrane-bound ATPase activity of E_. c o l i by various detergents. Membrane vesicles of E. c o l i ML308-225 were suspended at a protein concentration of 15-20 mg/ml i n 0.5 M Tris-H 2S04 buffer, pH 8.0 contain- ing 0.25 M Na2S04 and 10% (v/v) glycerol, and solubilized, as described i n MATERIALS AND METHODS, with one of the following detergents: PANELS A and B: N-lauroyl sarcosine ( A - A ) , Ammonyx Lo ( A - A ) , sodium deoxycholate (•-•), sodium cholate (O-O), octyl-B-D-glucopyranoside (•-•), B r i j 35 (•-•). PANELS C and D: Triton X-100 (•-•), Triton X-114 (O-O) , Lubrol WX (*-A) , Lubrol 17A-10 ( A - A ) , Tween 60 or Tween 80 (•-•), Lubrol PX (•-•). Table 9 Solubilization of the Membrane-Bound ATPase Activity of E. c o l i with Sodium Cholate Detergent: Protein Total Protein (mg) Total Units Percent Solubilized ATPase Ac t i v i t y Protein A 0 37 33.5 1.6 2.7 0.34 34.2 2.1 4.5 0.69 37.2 3.9 8.4 1.42 39.7 11.3 13.1 B 0 35.5 37.2 11.8 30 0.22 37.2 34.8 32.9 0.46 37.6 43 36.5 . 0. 69 36.9 45 40.5 0.94 37.7 50.8 48 Membranes vesicles of E. c o l i ML308-225 were suspended at a protein concentration of 15-20 mg/ml in buffer containing either (A) 50mM T r i s - H 2S0 4 pH 8.0, lOmM MgCl 2 and 10% (v/v) glycerol or (B) 0.5 Tris-H^O^ pH 8.0, 0.25M Na 2S0 4 and 10% (v/v) glycerol. The vesicles were solubilized with various amounts of sodium cholate as described in the MATERIALS AND METHODS section. 103. solubilized 45% and 40.5% of the activity and protein, respectively. The non-ionic detergents, Triton X-100, Triton X-114 and Lubrol WX showed some potential i n extracting the membrane-bound ATPase activity (Fig. 5). At a detergent to protein ratio of 0.96, Triton X-100, Triton X-114 and Lubrol WX solubilized 70%, 55% and 45% of the ATPase activity, while the corresponding amounts of protein solubilized were 42%, 45% and 55%, respectively. At the same detergentrprotein ratio, B r i j 35, Tween 60, i S Tween 80, Lubrol PX and Lubrol 17A-10 extracted less than 30% of the activity and protein from the membrane. Effect of Detergents on the Membrane-Bound ATPase Activity The detergents also stimulated the membrane-bound ATPase activity (Table 10). In general, greater stimulation of the activity was observed with higher detergent to protein ratios. N-lauroyl sarcosine, at ratios of 0.12 and 0.48, stimulated the ATPase activity by 1.3- and 4.5-fold, respectively. There also appeared to be some correlation between the extent of stimulation and the solubilizing capacity of the detergent. Comparing the ionic detergents, sodium cholate and deoxycholate, i t was found that at a ratio of 0.98, the activities were increased by 1.1- and 2.45- folds, respectively. Similarly, the non-ionic detergent, Triton X-100, at a ratio of 0.96 stimulated the activity by 2.5 fold whereas Lubrol PX at a ratio of 2.48 stimulated the activity by only 1.1 times. The activities of the solubilized fractions could also be stimulated by up to 100%, when assayed in the presence of L-a-lysolecithin (data not shown). It was not known i f the increase in activity in the presence of detergent lyso-phospholipid was due to activation of latent ATPase enzymes 104. Table 10 E f f e c t of Detergent on the Membrane-Bound ATPase A c t i v i t y of E. c o l i I o n i c Detergents Non-Ionic Detergents 1 Percent Percent C o n t r o l C o n t r o l Detergent D:P* Value Detergent D:P Value - 0 100 - 0 100 Ammonyx Lo 0.34 340 B r i j 35 0.52 120 0.68 410 1.06 125 1.4 425 L u b r o l PX 0.59 105 N - l a u r o y l 0.12 130 2.48 110 sar c o s i n e 0.24 270 0.48 450 L u b r o l WX 0.14 160 0.28 165 Octyl-fr-D- 0.21 115 glucopyrano- 0.42 170 L u b r o l 17A-10 0.14 150 s i d e 0.84 260 0.28 180 Sodium 0.48 105 T r i t o n X-100 0.47 200 c h o l a t e 0.98 110 0.96 250 2.08 170 T r i t o n X-114 0.47 280 Sodium 0.48 145 0.96 325 deoxycholate 0.98 245 2.08 350 Tween 60 0.33 135 0.67 140 Tween 80 0.33 140 0.67 155 * Detergent to p r o t e i n r a t i o Membrane v e s i c l e s of JE. c o l i ML308-225 were suspended at a p r o t e i n l e v e l c o n c e n t r a t i o n of 15 - 20 mg/ml i n 0.5M Tris-H^SO^ b u f f e r , pH 8.0 c o n t a i n i n g 0.25 M Na 2S0, and 10% (v/v) g l y c e r o l . D i f f e r e n t l e v e l s of v a r i o u s d e t e r - gents were added to a l i q u o t s of the membrane v e s i c l e suspension. A f t e r 30 min. at 20 C, samples were withdrawn and assayed f o r ATPase a c t i v i t y by the "slow assay" as des c r i b e d i n MATERIALS AND METHODS. 105. as i n Mycobacterium phlei (199) and Micrococcus lysodeikticus (200) or to proteolytic digestion of ECF1 (6,7,24). Stability of the Solubilized Enzyme Since the detergents stimulated the ATPase activity, i t was of interest to determine whether the activity of the solubilized fraction was stable over the period of time spanned by some of the purification procedures used i n this study. The solubilized- fractions of membrane vesicles of E. c o l i ML308-225, solubilized with either Ammonyx Lo, N-lauroyl sarcosine or sodium cholate (at detergent to protein ratios of 0.68, 0.48 and 0.48, respectively) were stored at 4°C. Samples were removed at timed intervals and the ATPase i activity measured. In the presence of Ammonyx Lo, the activity decreased by 30% after 140 h (Fig. 6). This decrease In activity could be due to denaturation or solubilization of the other, membrane components, necessary for the functional conformation of the enzyme. By contrast, the activity i n the presence of sodium cholate had increased to 112% forty hours after solubilization, and decreased to i t s original level (100%) only after another 200 h. Similarly, the activity in the presence of N-lauroyl sarcosine or in control membranes decreased by only 5-10% over 10 days. MOLECULAR SIZE OF THE DETERGENT-SOLUBILIZED ENZYME Gel F i l t r a t i o n on Sepharose 6B or Bio-Gel A-0.5m The solubilized enzyme was chromatographed on gel f i l t r a t i o n columns in order to estimate the molecular weight of the solubilized ATPase enzyme. In addition, gel f i l t r a t i o n was thought to be suitable as an 106. Fig. 6 Stability of solubilized ATPase activity on storage at 4°C. Membrane vesicles of IS. c o l i ML308-225 were suspended in 0.5 M T r i s - H2SO4 buffer, pH 8.0 containing 0.25 M Na 2S04 and 10% (v/v) glycerol at a protein concentration of 15-20 mg/ml. The vesicles were solubilized with either Ammonyx Lo ( A - A ) , N-lauroyl sarcosine (O-O), or sodium cholate (•-•) at detergent:protein ratios of 0.68, 0.48 and 0.48, respectively. The conditions of solubilization are described in MATERIALS AND METHODS. The ATPase activity of the solubilized fraction was measured immediately and at various times following storage at 4°C. A control, consisting of a suspen- sion of untreated membrane vesicles of E_. c o l i ML308-225 (•-•) , was also stored at 4°C and the activity measured at timed intervals. 107. i n i t i a l purification step. Since the ATPase complex (molecular weight 380 000) i s much larger than most other membrane-bound proteins in J £ . c o l i , i t should be separated from the proteins with smaller molecular weights on a gel f i l t r a t i o n column. Membrane particles of E. c o l i ML308-225 in 0.5 M Tris-H 2S0 4 buffer, pH 8.0 containing 0.25 M Na 2S0 4 and 10% (v/v) glycerol were solubilized with N-lauroyl sarcosine at a detergent to protein ratio of 0.5. The solubilized material was concentrated and then applied to a Sepharose 6B column, which had been equilibrated with 0.5%, (w/v) of various detergents. Profiles of the separations on Sepharose 6B i n the presence of 0.5% (w/v) Lubrol WX and 0.5% (w/v) N-lauroyl sarcosine are illustrated in Fig. 7, panels A and B. Fig. 7C i s the separation profile on Sepharose 6B, i n the presence of 0.5% (w/v) N-lauroyl sarcosine, of the solubilized material which was made up to 3% (w/v) N-lauroyl sarcosine and 2% (w/v) sodium cholate, prior to loading onto the column (201). In a l l three cases, the ATPase activity migrated as a single peak, close to the void volume (V = 75 ml). Cytochromes a-̂ , b^ and d comigrated with the ATPase activity peak as did the majority of the protein which was applied to the column. Similar elution profiles were obtained i n the presence of the other detergents lis t e d in Table 11. When the solubilized fraction was passed through a column of Bio-Gel A-0.5 m, which was equilibrated with 0.5% (w/v) N-lauroyl sarcosine, the enzyme activity also co-eluted with a l l the cytochromes and the majority of the protein close to the void volume (V = 85 ml). r o The recovery of activity depended on the detergent used during chroma- tography (Table 11). Recoveries ranged from 55 - 60% with N-lauroyl sarcosine to 125% with sodium cholate. Although the recovery was low i n 108. Fig. 7 Chromatography of the detergent-solubilized ATPase complex on Sepharose 6B i n the presence of various detergents. Membrane vesicles of E_. c o l i ML308-225 were solubilized with N-lauroyl sarcosine at a detergentrprotein ratio of 0.5. The solubilized fraction was concentrated by u l t r a f i l t r a t i o n through an Amicon XM-100A f i l t e r as described i n MATERIALS AND METHODS. The concentrated material (100-150 mg protein) was applied to a column of Sepharose 6B (2.5 x 37 cm) equilibrated with 50 mM Tris-H 2S04 buffer, pH 8.0, containing 0.25 M Na2S04, 10% (v/v) glycerol and 0.5% (w/v) of one of the following detergentsr PANEL A, Lubrol WX; PANELS B and C, N-lauroyl sarcosine. In PANEL C, the concen- trated material was made up to 3% (w/v) N-lauroyl sarcosine and 2% (w/v) sodium cholate prior to loading onto the column. The columns were then eluted with the same buffer containing the appropriate detergent. Fractions (10 ml) were collected and assayed for ATPase activity (•-•), protein (O-O) and total cytochromes (aj_, b^ and d) ( A - A) as previously described. ' 109. Volume-ml 110. the presence of N-lauroyl sarcosine, i t was higher (105%) i f the solubilized material was made up to 3% (w/v) N-lauroyl sarcosine and 2% (w/v) sodium cholate before applying to the column. The profiles i n a l l cases showed that the enzyme activity eluted as a broad peak. This suggested that the solubilized enzyme most lik e l y existed i n different aggregated forms. The molecular weight of the solubilized enzyme was estimated in the presence of different detergents. These values are summarized i n Table 11. The molecular weights were calculated to be in the range of 450 000 to 890 000. In general, chromatography i n the presence of non-ionic detergents resulted i n higher molecular weights for the ATPase enzyme than i n the presence of ionic detergents. With the non-ionic detergents, the highest molecular weight for the enzyme was obtained with either Lubrol WX or Triton X-114 (890 000 daltons), and the lowest with Triton X-100 (640 000 daltons). With the ionic detergents, the highest molecular weight was obtained i n the presence of Ammonyx Lo (580 000 daltons) and the lowest with N-lauroyl sarcosine (450 000 to 510 000 daltons). The molecular weights estimated by gel f i l t r a t i o n suggested that the ECF^ and FQ remained associated during solubilization of the membrane and subsequent chromatography. In order to test for this, the sensitivity to inhibition by DCCD of the fraction from the Sepharose 6B column contain- ing the highest ATPase activity was determined for two preparations of the enzyme. The results are shown in Fig. 8. The ATPase activity i n the presence of either 0.01% (w/v) B r i j 35 or 0.025% (w/v) N-lauroyl sarcosine was inhibited by 50% at 0.4 mM and 1 mM DCCD, respectively. By contrast, ECF 1 was inhibited by 5-10% at 1 000 yM DCCD. These results suggested that the solubilized ATPase enzyme was the intact ECF F n complex. Table 11 Estimation of the Molecular Weight of the Solubilized ECF.F Complex by Gel F i l t r a t i o n Chromatography 111. 1 Gel f i l t r a t i o n resin Detergent in equilibration buf f er Detergent Concentration [% (w/v)] Estimated Molecular Weight ,- (M x 10 ) r Recovery of activ i t y (%) Sepharose 6B Brij 35 0.5 8.5 110 Sepharose 6B Lubrol WX 0.5 8.9 80 Sepharose 6B Triton X-100 0.5 6.4 64 Sepharose 6B Triton X-114 0.5 8.9 69 Sepharose 6B Ammonyx Lo 0.5 5.8 73 Sepharose 6B Sodium cholate 0.5 5.2 125 Sepharose 6B N-lauroyl sarcosine 0.5 5.1 61 Sepharose 6Ba N-lauroyl sarcosine 0.5 4.7 107 Bio-Gel A 0.5m N-lauroyl sarcosine 0.5 4.5 - 4.8 55 aThe solubilized material was made up to 3% (w/v) N-lauroyl sarcosine and 2% (w/v) sodium cholate before loading onto the column. The details of the experiment are described in the legend to Fig. 7. Thyroglobulin (669 00 daltons), catalase (232 000 daltons) and haemoglobin (64 500 daltons) were used to calibrate the gel f i l t r a t i o n columns under the various conditions. 112. Fig. 8 Effect of DCCD on the detergent-solubilized ATPase activity. Membrane vesicles of E. c o l i ML308-225 were solubilized with N-lauroyl sarcosine at a detergentrprotein ratio of 0.5. The solubilized fraction was chromatographed on a column of Sepharose 6B i n the presence of 0.5% (w/v) detergent, as described in the legend to Fig. 7, PANELS A and B.. Samples of the most active fraction from the Sepharose 6B column, i n the presence of 0.5% (w/v) B r i j 35 (O-O) or N-lauroyl sarcosine (•-•) were incubated with various levels of DCCD in 0.1 M Tris-HCl, pH 8.0 for 45 min at 37°C. ECF1 of E. c o l i ML308-225 (O-O), obtained after dialysis of the membranes i n low ionic strength buffer, was also treated with DCCD. Following incubation with DCCD, ATPase activity was measured as described i n MATERIALS AND METHODS. The specific a c t i v i t i e s of the ATPase enzyme i n the presence of Bri j 35 ( f i n a l concentration, 0.01% (w/v)), N-lauroyl sarcosine ( f i n a l concentration, 0.025% (w/v)) and of ECFj^ were 2.06, 2.52 and 3.04 units per mg protein and the amount of protein per assay was 3, 13.5 and 14 wg, respectively. 113. Since, the ATPase activity was not separated from other solubilized proteins (cytochromes) by chromatography on Sepharose 6B, other methods of purifying the ECF^FQ complex were investigated. These are discussed i n the following sections . Purification of the ECF^FQ Complex by Hydrophobic-Interaction Chromatography Hydrophobic-interaction chromatography has been used mainly for the purification of water-soluble proteins (202). Membrane-bound proteins interact very strongly with these resins and can only be released from the resin with the use of detergents. Usually l i t t l e separation i s achieved. However, hydrophobic-interaction chromatography on Phenyl Sepharose CL-4B was used successfully to purify the membrane-bound enzyme, fumarate reductase (203). This enzyme consists of a membrane-bound and a water- soluble component and was eluted from the resin by lowering the ionic strength. The enzyme obtained was essentially i n a pure form. Since the ECF^FQ complex resembles fumarate reductase i n having a water-soluble (ECF^) and a membrane-bound (FQ) component, experiments were carried out to test the f e a s i b i l i t y of using Phenyl-Sepharose CL-4B as a means of purifying the ECF^FQ complex. Membrane vesicles of E. c o l i ML308-225 were solubilized with N-lauroyl sarcosine at a detergent to protein ratio of 0.5, as described earlier. The solubilized material was applied to a column of Phenyl-Sepharose CL-4B, which was equilibrated with 50 mM Tris-H 2S0 4 buffer, pH 8.0 containing 0.25 M Na 2S0 4, 10 mM MgCl 2, 10% (v/v) glycerol and 0.5% (w/v) sodium cholate. The column was washed with the same buffer and then a linear decreasing gradient of 0.25 M to 0 M Na„S0, applied. Finally, the 114. column was washed with 2%, (w/v) Triton X-100. The elution profile i s shown in Fig. 9A. The ATPase activity eluted as two peaks during the i n i t i a l washing of the column. The f i r s t activity peak was eluted during the application of the f i r s t two column volumes of wash buffer and represented 37% of the total activity recovered. The second activity peak was eluted during the last two column volumes of wash buffer and 63% of the total activity was recovered i n this peak. No enzyme activity was detected in the fractions from the gradient elution. Only a small amount of cytochrome d co-eluted with the ATPase activity. Most of the cytochromes (a^, b^ and d) were tightly bound to the resin and were eluted with 2% (w/v) Triton X-100. An attempt was made to increase the interaction of the ATPase enzyme with the resin, by omitting sodium cholate from the equilibration buffer only. The experiment was repeated as described above and the elution profile i s shown in Fig. 9B. Again, the ATPase activity was detected only i n the fractions obtained during the i n i t i a l washing. As before, the enzyme activity eluted as two peaks. However, i n contrast to Fig. 9A, 96% of the total activity was associated with the second peak. Although most of the cytochromes (a^, b^ and d) bound very tightly to the resin, some co-eluted with the two ATPase activity peaks. Further attempts to improve the interaction of the ATPase enzyme with the resin was achieved by. using ions which had "salting-out" properties. This was achieved by replacing the ̂ £80^ with 20% ammonium sulphate (by saturation) i n the buffers described i n the legend to Fig. 9B. The gradient consisted of a linear decreasing concentration of 20% to 0% ammonium sulphate. The experiment was repeated as described above and the elution profile i s shown in Fig. 9C. A l l the cytochromes bound very 115. Fig. 9 Chromatography of the detergent-solubilized ATPase complex on Phenyl-Sepharose CL-4B. Membrane vesicles of Ê . c o l i ML308-225 were solubilized with N-lauroyl sarcosine at a detergent:protein ratio of 0.5. The solubilized fraction (100-125 mg protein) was applied to a column of Phenyl-Sepharose CL-4B (1.8 x 25 cm) equilibrated with 50 mM Tris-H 2S04 buffer, pH 8.0 containing 10 mM MgCl 2, 10% (v/v) glycerol and the following: PANEL A, 0.25 M Na 2S04 and 0.5% (w/v) sodium cholate; PANEL B, 0.25 M Na 2S04; PANELS C and D, 20% (NH-4)2S04 (by saturation). The columns were then washed with 4-5 column volumes of equilibration buffer but with the inclusion of 0.5% (w/v) sodium cholate i n PANELS B and C. The columns were then developed with a decreas- ing linear gradient (8-10 column volumes) of either 0.25 M to 0 M Na 2S04 (PANELS A and B) or 20% to 0% (NH4)2S04 (by saturation) (PANELS C and D) in 50 mM Tris-H 2S0 4 buffer, pH 8.0 containing 10 mM MgCl 2, 10% (v/v) glycerol and 0.5% (w/v) sodium cholate. Following elution with the gradient, the columns were washed with 50 mM Tris-H 2S04 buffer, pH 8.0, containing 2% (w/v) Triton X-100. The arrows indicate the position of the buffer changes. Fractions (10 ml) were collected and assayed for ATPase activity (•-•), protein (O-O), total cytochromes (a^, b^ and d) ( A _ A ) , and conductivity ( A - A ) a s described i n MATERIALS AND METHODS. 116. F r a c t i o n Number 117. tightly to the resin and were eluted with 2% (w/v) Triton X-100. ATPase activity was again detected only in the fractions from the wash. However, the activity eluted as a single peak during elution with the last two column volumes of buffer. It appeared that the presence of sodium cholate in the buffer was causing the enzyme to be released from the resin. To test for this possibility, sodium cholate was omitted in the wash buffer and the experiment performed as described for Fig. 9C. The elution profile i s shown in Fig. 9D. Again, a l l the cytochromes were completely separated from the ATPase activity. The activity eluted as a single peak during the gradient elution. The fraction with the highest activity eluted at an ammonium sulphate concentration of 15-16% (34-36 mmho). Thus, the presence of sodium cholate decreased the interaction of the ATPase enzyme with the resin, whilst ions with "salting-out" properties caused greater binding of the protein to the resin. The latter was also suggested by the elution profile of the cytochromes. When bound in the presence of Na2S0^, the cytochromes were subsequently eluted as a broad peak (Figs. 9A and 9B). In contrast, they were eluted as a sharp peak (Figs. 9C and 9D) when bound in the presence of ammonium sulphate. As It was possible to bind the solubilized enzyme to the Phenyl Sepharose CL-4B resin, i t was of interest to determine how ECF.̂  behaved under identical conditions. ECF^ of E_. c o l i ML308-225 obtained after purification on a sucrose gradient or after dialysis of membranes in low ionic strength buffer, as described i n MATERIALS AND METHODS, was chromatographed onto a Phenyl-Sepharose CL-4B column under conditions identical to those i n Fig. 9D. In both cases, the ATPase activity eluted as a single peak during the elution with the gradient. The elution profiles were similar to Fig. 9D. However, the fractions containing the 118. highest ac t i v i t i e s eluted at an ammonium sulphate concentration different to that required to elute the detergent-solubilized enzyme. Purified ECF^ or ECF^ obtained after dialysis of the membranes were eluted at ammonium sulphate concentrations of 9.5-10% (22-24 mmho) and 10.5-11.5% (24-26 mmho), respectively. Therefore, i t appeared that the detergent- solubilized ATPase enzyme was of a different composition than soluble ECF-j^. The elution of the solubilized ATPase at a higher ionic concentration than purified ECF^ suggested that the former must have a different conformation. This might be expected of an intact ECF^FQ complex. The presence of an intact ECF-^FQ complex was confirmed when fractions from the Phenyl-Sepharose CL-4B column were found to be sensitive to DCCD. The activity was inhibited by 50% at a DCCD concentration of 600 yM (not shown). Other Hydrophobic-Interaction Resins In addition to Phenyl-Sepharose CL-4B, a number of other hydrophobic resins were tested for their potential use i n the purification of the ECF^FQ complex. The resins, w-Amino Butyl Agarose, w-Amino Hexyl agarose (AH-Sepharose CL-4B), Butyl Agarose, Octyl Sepharose CL-4B and Decyl Agarose were selected for this purpose. Membrane vesicles of E. c o l i ML308-225 in 0.5 M Tris-H 2S0 4 buffer, pH 8.0 containing 0.25 M Na 2S0 4 and 10% (v/v) glycerol were solubilized with N-lauroyl sarcosine at a detergent to protein ratio of 0.5. The solubilized fraction was applied to a column of hydrophobic-interaction resin, under conditions identical to those described in the legend to Fig. 9D. Under these conditions, only a small amount (15%) of the protein was 119. bound to either w-Amino Butyl Agarose or w-Amino Hexyl Agarose (AH-Sepharose CL-4B). In both cases, the ATPase activity co-eluted as a single peak with the cytochromes (a^, b^ and d) during the i n i t i a l column wash step. Therefore, both of these resins were considered to be unsuitable for the purification of the ATPase complex. It i s probable that both functioned predominantly as ion-exchange resins. The identical experiment was repeated with Butyl Agarose, Octyl Sepharose CL-4B and Decyl Agarose. In a l l three cases, the ATPase enzyme bound very tightly to the resin and was only released during elution with the gradient. The elution profiles were almost identical to that in Fig. 9D. The ATPase activity eluted as a single peak at an ammonium sulphate concentration of 15-16%. With Butyl Agarose and Octyl Sepharose CL-4B, the cytochromes (a^, b^ and d) could be eluted from the resin with 2% (w/v) Triton X-100. However, cytochromes were not detected in the fractions when Decyl Agarose was eluted with buffer containing either 2% (w/v) or 5% (w/v) Triton X-100. It was not known If the cytochromes were eluted in a denatured form or were s t i l l tightly bound to the resin. The specific activity of the eluted ATPase enzyme was determined i n each case. The highest specific a c t i v i t i e s i n the fractions from Butyl Agarose, Decyl Agarose and Octyl Sepharose CL-4B were 2.25, 4.7 and 5.2 units per mg protein, and represented 4.7-, 6.1- and and 8.7-fold purifications, respectively. Although some purification was observed in each case, i t was not as high as that obtained in the fraction eluted from Phenyl-Sepharose CL-4B (24.8-fold purification). When the most active ( fractions from these columns were analyzed on SDS-polyacrylamide gels, the fewest number of polypeptide bands was observed i n the fraction from the Phenyl-Sepharose CL-4B column. Therefore, Phenyl-Sepharose CL-4B was 120. chosen as the resin for the purification of the ECF^FQ complex. Further Purification of the ECF^FQ Complex The ATPase complex was partially purified on Phenyl-Sepharose CL-4B as described earlier. Analysis of the fraction with the highest specific activity on SDS-polyacrylamide gel revealed the presence of approximately 14 major protein-staining bands (including the subunits of ECF^) and many minor bands (the gel was similar to Fig. 11A, lane c). The ATPase complex of the thermophile, PS3, was purified by Sone et a l . (102) and shown to consist of eight different polypeptides. It was thought that the composi- tion of the ECF^FQ complex should be similar. Therefore, an attempt was made to remove some of the polypeptides (contaminants) of the partially purified ECF^FQ complex by sucrose gradient centrifugation. Membrane vesicles of E_. c o l i ML308-225 were solubilized with N-lauroyl sarcosine at a detergent to protein ratio of 0.5 as described i n MATERIALS AND METHODS. The solubilized fraction was applied to a column of Phenyl- Sepharose CL-4B and chromatography carried out as described i n the legend to Fig. 9D. The active fractions eluted during the gradient were pooled and concentrated by u l t r a f i l t r a t i o n using an Amicon PM-10 membrane. The concentrated material was applied on a 15 to 25% (w/v) sucrose density gradient and centrifuged at 280 000 xg for 23 h. The results of the separation are shown i n Fig. 10. The ATPase activity peak was well J separated from the majority of the protein, which remained at the top of the gradient. Approximately 10% of the total activity applied to the sucrose gradient was recovered. Attempts to stimulate the ATPase enzyme by addition of L-a-lysolecithin, prior to assaying for activity, were not successful. As shown in Fig. 10, the added lyso-phospholipid inhibited the 121. 0 5 10 15 2 0 25 (Bottom) Fraction (Top) Fig. 10 Purification of the ECF^FQ complex by sucrose gradient centrifugation. The ECF^FQ complex obtained after chromatography on Phenyl- Sepharose CL-4B was concentrated by u l t r a f i l t r a t i o n using an Amicon PM-10 f i l t e r . The concentrated material (13 mg protein i n a volume of 1.2 ml) was applied to a linear gradient of 15% - 25% (w/v) sucrose i n 50 mM Tris-H 2S04 buffer, pH 8.0 containing 0.5% (w/v) sodium cholate, 5 mM MgCl 2 and 0.25 mM DTT. The gradient was centrifuged at 280 000 xg for 23 h and ten drop fractions collected. Protein ( A - A ) , and ATPase activity determined in the absence (p-a) or presence (•-•) of 0.02% (w/v) L-a-lyso- phosphatidylcholine were measured as described i n MATERIALS AND METHODS. 122. enzyme activity. Similarly, the inclusion of 0.1% (w/v) soybean phospha- tidylcholine i n the sucrose gradients did not prevent the loss of activity. When the concentrated fraction, which was applied to the sucrose gradient, was stored at 4°C for the same length of time spanned by the sucrose gradient centrifugation step (24-28 h), about 70% of the ATPase activity was lost. These results indicated that the enzyme was unstable at 4°C. The highest specific activity i n the sucrose gradient fractions was 7.7 units per mg protein and this represented a 22-fold purification over the intact membranes. Because of the different activating properties of the detergents as well as the unstable nature of the ATPase enzyme during purification, a comparison of the purification with respect to specific activities i s meaningless. A better method for comparing the purification procedures i s the analysis of the fractions on SDS-polyacrylamide gels. The active fractions from the Phenyl-Sepharose CL-4B column, when analyzed on SDS-polyacrylamide gels, revealed that the fractions from the leading edge (the gel was again similar to Fig. 11A, lanes b to d) of the activity peak appeared to contain many more minor bands than those from the tra i l i n g edge. Many of these contaminants were subsequently removed by sucrose gradient centrifugation. In addition to the subunits of ECF^ (ct-e), polypeptides with molecular weights of 48 000, 35 000, 24 000, 18 000, 14 000 and 9000 were also present in invariant stoichiometry. Although the sucrose gradient removed many of the minor bands with molecular weight less than 30 000 as well as a major contaminant of 70 000 daltons, many minor bands of less than 50 000 daltons were s t i l l present. /Many more polypeptides than those shown in Fig. 11, were present in this preparation. In order to remove some of these minor bands, an ammonium 123. sulphate precipitation step was included prior to chromatography on Phenyl-Sepharose CL-4B. Membrane vesicles of _E. c o l i ML308-225 were solubilized at a detergent to protein ratio of 0.25 rather than 0.5, as described earlier. The fraction precipitating between 0.35 and 0.5 saturation of ammonium sulphate (0.35-0.5 P fraction) was suspended in the buffer used for solubilization at a protein concentration of 5-7 mg/ml. N-lauroyl sarcosine was added to give a detergent to protein ratio of 0.8 to 1.0. This fraction was chromatographed on a column of Phenyl-Sepharose CL-4B under conditions identical to those described i n the legend to Fig. 9D except that the column was eluted with a linear decreasing gradient of 20% to 12.5% ammonium sulphate. On occasion, the 0.35-0.5 P fraction did not pellet following centrifugation at 30 000 xg for 20 min, but floated to the surface. This problem was alleviated by using 10% (v/v) methanol rather than glycerol i n the solubilization buffer. The presence of 10% (v/v) methanol did not r appear to affect the purification and was subsequently used in the other experiments. Applying a shallower gradient (20-12.5%) did not result i n a change i n the elution profile of the enzyme from that seen in Fig. 9D. The activity eluted as a single peak at an ammonium sulphate concentration of 15-16%. The active fractions were concentrated and applied to a linear 17.5 to 25% (w/v) sucrose gradient as described in MATERIALS AND METHODS. As before, approximately 15% of the activity was recovered following sucrose gradient centrifugation. The separation on the sucrose density gradient was similar to Fig. 10. The majority of the protein (contaminants) remained at the top of the gradient, whereas the ATPase activity was associated with a smaller 124. i protein peak close to the bottom of the tube. i Inclusion of the ammonium sulphate fractionation step reduced the number of minor bands of less than 50 000 daltons in the fractions from the Phenyl-Sepharose CL-4B column and subsequently in the fractions from the sucrose gradient (Fig. 11). In the fractions from the sucrose gradient, the subunits of ECF^ (a - e) as well as major-protein staining bands with molecular weights of 48 000, 35 000, 24 000, 18 000, 14 000 and 9 000 were present i n invariant stoichiometry. Relative to the other subunits of ECF^, the <5 subunit (molecular weight, 21 000) was present in f a i r l y low amounts in the active fractions (Fig. 11B) from the sucrose gradient. It has been reported that this subunit i s very susceptible to protease digestion (6,24). Foster and Fillingame (94) have also reported the purification of the 1 ECF^FQ complex. The FQ complex consisted of three subunits of molecular weight 24 000, 18 000 and 8 500. They reported that dialysis of the detergent-solubilized fraction followed by resolubilization with detergent removed many of the minor protein-staining bands. Membrane vesicles of E. c o l i ML308-225 were solubilized at a detergent to protein ratio of 0.25 a described earlier. The solubilized fraction was dialyzed against buffer for 24-30 h. The reaggregated material i n the dialysate was collected by centrifugation and resolubilized with N-lauroyl sarcosine at a detergent to protein ratio of 0.8 to 1.0. Only 60% of the activity was recovered following dialysis. The resolubilized material was subjected to ammonium sulphate precipitation, chromatography on Phenyl- Sepharose CL-4B and sucrose gradient centrifugation as described i n the previous experiment. Again, only 15% of the total activity applied to the sucrose gradient was recovered. However, the activity peak was well r 125. Fig. 11 SDS-polyacrylamide gel electrophoresis of the ECF]FQ complex purified by chromatography on Phenyl-Sepharose CL-4B and sucrose gradient centrifugation. Membrane vesicles of E_. c o l i ML308-225 were solubilized with N-lauroyl sarcosine at a detergent to,protein ratio of 0.25. The solubilized fraction was subjected to ammonium sulphate precipitation and the fraction precipi- tating between 35 and 50% of saturation was suspended in the solubilizing buffer at a protein concentration of 5-7 mg/ml. N-lauroyl sarcosine was added to give a detergent to protein ratio of 0.8-1.0, as described i n MATERIALS AND METHODS. The resolubilized material was applied to a column of Phenyl-Sepharose CL-4B (1.8 x 25 cm) under conditions identical to those described in the legend to Fig. 9D. The active fractions from the Phenyl- Sepharose CL-4B column were concentrated and applied to a linear 17.5 to 25% (w/v) sucrose gradient and centrifuged at 280 000 xg for 23 h as described i n the MATERIALS AND METHODS section. The active fractions from the Phenyl-Sepharose CL-4B column and the sucrose gradient were analyzed by SDS-gel electrophoresis. The separating gel (Tris-buffered system) consisted of a linear 7.5 - 16.5% (w/v) acrylamide gradient. The stacking gel consisted of 4% (w/v) acrylamide. Following electrophoresis, the gels were stained with 0.1% (w/v) Coomassie Blue. PANEL A: Fractions from the Phenyl-Sepharose CL-4B column. Lane a represents the material applied to , the column. Lanes b-f represent the fractions eluted from the column, with the fraction highest in ATPase activity in lane c. Lane g i s purified ECF^. The position of migration of the subunits of ECF^ (a-e) are i n d i - cated. PANEL B: Fractions from the sucrose gradient. Lane a represents the material loaded on the sucrose gradient. Lanes b-i represent the fractions from the sucrose gradient. The fraction containing the highest ATPase activity i s i n lane e. Lane j is purified ECF^. The position of migration of the subunits of ECF^ ( a-e) are also indicated. 126. a b c d e f g 127. separated from the majority of the protein (as in Fig. 10). Inclusion of the dialysis step appeared to decrease the number and intensity of the minor protein-staining bands with molecular weights greater than 20 000, in the fractions from the Phenyl-Sepharose CL-4B J column (Fig. 12A), when compared to those i n Fig. 11A. However, those minor bands with molecular weights less than 20 000 were increased. Nevertheless, the inclusion of the dialysis step did improve the overall purification as seen in the active fractions from the sucrose gradient (Fig. 12B). The subunits of ECF 1 (ct - e), as well as major protein- staining bands with molecular weights of 35 000, 24 000, 18 000, 14 000 and 9 000, appeared to be present i n invariant stoichiometry. In contrast to the results i n Fig. 11B, the 48 000 dalton polypeptide was almost completely absent in this preparation. In addition, the staining-intensity of the 18 000 and 14 000 dalton polypeptides were greater i n this prepara- tion. Furthermore, the 6 subunit was more distinguishable, suggesting that this preparation was perhaps less contaminated with proteases. The inclusion of the protease inhibitors PMSF (0.1 mM) and p-aminobenzamidine (6 mM) during the concentration step, as well as in the sucrose gradient, did not result i n any change in the pattern of bands seen in Fig. 12B. It was not known i f these protease inhibitors were capable of inhibiting the activity of the protease(s) which may have been present i n the preparation. The fraction from the sucrose gradient which was richest i n ATPase activity was found to be sensitive to DCCD. The ATPase activity was inhibited by 30% at a DCCD concentration of 200 uM. More recently, Schneider and Altendorf (107) have also purified the ECF^FQ complex by chromatography on DEAE-Sepharose CL-6B in the presence of Aminoxid WS-35, followed by centrifugation of the active fractions at 128. Fig. 12 SDS-polyacrylamide gel electrophoresis of the ECF^FQ complex obtained after, chromatography on Phenyl-Sepharose CL-4B and sucrose gradient centrifugation. Membrane vesicles from E_. c o l i ML308-225 (25-30 g wet weight) were solubilized with N-lauroyl sarcosine at a detergent to protein ratio of 0.25. The solubilized material was dialyzed (1:100) against 50 mM Tris- H2SO4 buffer, pH 8.0 containing 1 mM DTT, 0.1 mM EGTA, 25 mM Na 2S0 4 and 10% (v/v) methanol for 24-30 h at 4°C, with changes of the external buffer at 6-8 h intervals. The reaggregated material was collected by centrifugation at 250 000 xg for 3 h and resolubilized with N-lauroyl sarcosine at a detergent to protein ratio of 0.8-1.0. The resolubilized fraction was then subjected to ammonium sulphate precipitation, as described in MATERIALS AND METHODS. The fraction precipitating between 35 and 50% of saturation was applied to a column of Phenyl-Sepharose CL-4B (1.8 x 25 cm) under conditions identical to those described i n the legend to Fig. 9D, except that the enzyme was eluted with a linear decreasing gradient of 20% - 12.5% ammonium sulphate (by saturation). The active fractions from the column were concentrated and applied to a linear 17.5 to 25% (w/v) sucrose gradient and centrifuged at 280 000 xg for 23 h as described previously. The active fractions were analyzed by SDS-gel electrophoresis. The separating gel (Tris-buffered system) consisted of a linear 7.5-16.5% (w/v) acrylamide while the stacking gel was 4% (w/v) acrylamide. The gels were fixed i n a solution containing 5% (w/v) TCA, 5% (w/v) sulfosalicylic acid and 10% (v/v) methanol, and then stained with 0.1% (w/v) Coomassie Blue. PANEL A: Lane a represents the material applied to the column of Phenyl-Sepharose CL-4B. Lanes b-i represent the active fractions eluted from the column, with the material i n lane e containing the highest activity. Lane j i s purified ECF^. The migration positions of the subunits of ECF^ (a-e) are indicated. PANEL B: Lane a represents the material loaded onto the sucrose gradient. Lanes b-j represent the active fractions from the sucrose gradient, with the material i n lane d containing the highest ac t i v i t y . Lane k i s purified ECF^. The migration positions of the subunits of ECF^ (a-e) are also shown. ® a b c d e f g h i j k I f f i s s t s - 130. 220 000 xg for 15 h. Therefore, centrifugation of the active fractions from the Phenyl-Sepharose CL-4B column was used to overcome two of the problems associated with the purification on a sucrose gradient: the poor recovery of activity and the small quantity of the purified enzyme which could be obtained with a sucrose gradient. The experiment described previously (legend to Fig. 12) was repeated except that the sucrose gradient step was omitted. Instead, the active fractions from the Phenyl-Sepharose CL-4B column were made up to 0.1 mM PMSF and 6 mM p-aminobenzamidine prior to centrifugation at 250 000 xg for 16-17 h. The pellet was taken up in 50 mM MOPS-KOH buffer, pH 7.5 containing 10 mM MgCl 2, 20% (v/v) glycerol and 0.2% (w/v) Triton X-100. Approximately 40% and 10% of the total activity centrifuged was recovered in the pellet and supernatant fractions, respectively. A recovery of 40% was a substantial improvement over.the 8-10% yield usually obtained following sucrose gradient centrifugation. The preparation of the samples for SDS-polyacrylamide gel electro- phoresis was modified at this stage. Detergent and salts, which interfered with the migration of the polypeptides in the gels were removed by the column chromatography-method of Penefsky (170) as described in MATERIALS AND METHODS. This i s i n contrast to the previous preparations, in which the samples were dialyzed against water for 1-2 days and then lyophilized. The fraction from Phenyl-Sepharose CL-4B richest i n ATPase activity revealed the presence of at least 15 major protein-staining bands and several minor ones on SDS-polyacrylamide gels. The gel resembled that in Fig. 12A. Analysis of the sedimented ATPase enzyme on SDS-polyacrylamide gels i s shown in Fig. 13 (lane a). Centrifugation of the active fractions resulted in the enrichment of 11 polypeptides. Of these, five polypeptides 131. Fig. 13 SDS-polyacrylamide gel electrophoresis of the E C F ^ FQ complex obtained by chromatography on Phenyl Sepharose CL-4B and sedimentation at 250 000 xg for 16-17 h. The experiment described in the legend to Fig. 12 was repeated except that purification of the E C F ^ FQ complex by sucrose gradient centrifugation was omitted. Instead, the active fractions from the Phenyl-Sepharose CL-4B column were pooled and made up to 0.1 mM PMSF and 6 mM-pr-aminobenzamidine, and centrifuged at 250 000 xg for 16-17 h. The sedimented enzyme (lane a) was subjected to further purification on a column of DEAE-Sepharose CL-6B, as described in MATERIALS AND METHODS. It was taken up in 50 mM Tris-HCl buffer, pH 8.0, containing 1 mM MgCl 2, 0.2 mM DTT, 0.2 mM EGTA, 0.1 mM PMSF, 100 mM KC1, 20% (v/v) methanol, 50 mg/ml soybean phospholipid and 0.9% (w/v) Aminoxid WS-35 prior to loading onto the column. The fraction containing the highest activity was analyzed by SDS-gel electrophoresis (lane b). The composition of the SDS-gel was the same as that described in the legend to Fig. 12 except that the concentration of SDS in the gel was 0.5% (w/v). Lane c represents purified ECF^. The migration positions of the subunits of ECF^ (<*-e) as well as of the DCCD-binding protein (identified using ["cpCCD, see Fig. 15), the 28 000 and 18 000 dalton polypeptides of FQ, are also indicated. 132. were the subunits of ECF^ and the remaining had molecular weights of 30 000, 28 000, 18 000, 14 000, 9 000 and 7 500. Minor bands of 85 000, 71 000 and 24 000 daltons were also present: Some of these polypeptides were thought to be contaminants. Attempts were made to remove these contaminat ing polypeptides by subjecting the sedimented ATPase enzyme to further fractionation. The f i r s t method involved the treatment of the enzyme with PEG 6 000 and 400, as described by Friedl et a l . (108). This was not successful. Precipitation of the ATPase complex with PEG 6000 and 400 resulted in almost complete (98%) inactivation of the enzyme activity. In addition to being -y-deficient, the precipitated enzyme also contained very low amounts of the a and B subunits of ECF^ The second method involved ion-exchange chromatography on DEAE- s Sepharose CL-6B in the presence of Aminoxid WS-35, as described by Friedl and Schairer (106). The sedimented ATPase complex dissociated during ion-exchange chromatography. The active fractions from the DEAE-Sepharose CL-6B column were analyzed by SDS-polyacrylamide gel electrophoresis and found to contain only 6-def icient ECF^ (Fig. 13", lane b). The absence of intact ECF^FQ complex was confirmed, by determining the sensitivity of the ATPase activity of this preparation to DCCD (Fig. 14). INTACTNESS OF THE ECF^^FQ C O M P L E X In this study, centrifugation of the active fractions from the Phenyl-Sepharose CL-4B column, at 250 000 xg for 16-17 h resulted in optimal purification of the ATPase enzyme. The intactness of the purified ECF^FQ complex was determined by using DCCD. The results are shown i n Fig. 14. 133. DCCD — LI M Fig. 1 4 Effect of DCCD on the ATPase activity of the ECF JFQ complex. The ATPase enzymes obtained after chromatography on Phenyl-Sepharose CL-4B and sedimentation at 250 000 xg for 16-17 h (Fig. 13, lane a), or after further purification of the sedimented enzyme on DEAE-Sepharose CL-6B (Fig. 13, lane b), were incubated with various levels of DCCD in 0.1 M Tris-HCl buffer, pH 8.0, containing 5 mM MgCl 2 ( f i n a l volume, 0.1 ml) at 37°C for 45 min. ECF^ obtained after AH-Sepharose 4B chromatography was also treated with DCCD. Following incubation with DCCD, a sample of the mixture was removed and the ATPase activity measured as described in MATERIALS AND METHODS. The specific a c t i v i t i e s of the sedimented enzyme (O-O), DEAE-Sepharose CL-6B-purified enzyme (•-•) and ECF]_ (*-*) were 18, 31.4 and 26.3 units per mg protein and the amount of protein per assay was 1.75, 0.85 and 0.8 ng, respectively. 134. Over a DCCD concentration range of 500 uM, the activity of the purified ECF^ was inhibited by only 5-10%. Similar results were obtained with the DEAE-Sepharose CL-6B-purified enzyme. This was expected since this preparation appeared to be 6-defIcient ECF^ and not the ECF^F^ complex. By contrast, the enzyme sedimented by centrifugation at 250 000 xg for 16-17 h, was sensitive to DCCD. The ATPase activity was inhibited by 30% and 45% at DCCD concentrations of 100 uM and 500 uM, respectively. Treatment of a l l three enzyme preparations with [llfC]DCCD (136 UM) at pH 8.0 in the presence of 5 mM MgC^, under conditions identical to those described in the legend to Fig. 14, resulted i n the labelling of the B subunit (52 000 daltons). In addition, a polypeptide of about 9 000 daltons was labelled i n the enzyme sedimented by centrifugation at 250 000 xg for 16-17 h (Fig. 15). Other evidence that the sedimented ATPase enzyme was present in intact form was otained by using [l - 1*CJiodoacetamide and the fluorescent label, 5-iodoacetamidofluorescein. Recently, Paradies et a l . (35) have shown that the compounds TAMM andjNEM-Hg labelled only the fl subunit of purified ECF^. However, the labels were associated only with the a subunit when the F^FQ complex was used. [ l l*C]- iodoacetamide was found to label both the a and B subunits of the sedimented ATPase complex but only the a of the purified ECF^. With the fluorescent probe 5-iodoacetamidofluorescein, both the a and B subunits of the intact ATPase complex were labelled, with a higher amount of the label on the a subunit. But with purified ECF^ only the B subunit was labelled (these experiments were done in collaboration with Drs. Helga Stan-Lotter and P.D. Bragg). These results suggested that the ECF1 remained associated with F n 135. Fig. 15 SDS-polyacrylamide gel electrophoresis of ECF^ and ECFJFQ complex labelled with ['"CJDCCD. ECF^ (0.34 mg), obtained after chromatography on AH-Sepharose 4B, and the purified ECF-^FQ complex (0.35 mg, Fig. 13, lane a) were suspended i n 0.25 ml of 50 mM Tris-HCl buffer, pH 8.0 containing 5 mM MgCl 2. To each was added 7 H i of 5 mM ['"CJDCCD (Specific activity, 50 mCi/nmol) and the mixtures incubated at 37°C for 45 min. The unreacted label was removed by the centrifugation column-chromatography method described by Penefsky (170), The eluents were depolymerized and 82.5 ug of the ECF^ (PANEL A) and 70 yg of the ECF^Q complex (PANEL B) were applied to the SDS-gel. The slab gel (Tris-buffered system) consisted of a separating gel of 13% (w/v) acrylamide and a stacking gel of 4% (w/v) acrylamide with the SDS concentration i n the gel being 0.5% (w/v). After electrophoresis, the gel was stained with 0.1% (w/v) Coomassie Blue, the separating gel sliced into 1 mm segments, and the radioactivity of the slices determined as described in MATERIALS AND METHODS, a - e refer to the migration positions of the subunits of ECF^ in the gel. The position of the tracking dye, bromophenol blue (BP) i s also indicated. 136. 0 40 80 S l i c e 137. following centrifugation at 250 000 xg for 16-17 h, but dissociated during subsequent ion-exchange chromatography. > REPRODUCIBILITY OF PURIFICATION ON PHENYL SEPHAROSE CL-4B Problems were encountered when attempts were made to use Phenyl- Sepharose more than once. Although the position of elution of the activity (15-16% ammonium sulphate concentration) was almost identical when the resin was re-used, analysis of the active fractions from the column on SDS-polyacrylamide gels revealed that a different pattern of polypeptide bands was obtained each time. The experiment was repeated as described for Fig. 12 except that the fraction was applied to a column of Phenyl-Sepharose CL-4B which had been used and regenerated under different conditions. The fractions richest in ATPase activity were analyzed on SDS-polyacrylamide gels. About 14-16 protein straining bands were found when the ATPase enzyme was chromatographed on a column of freshly-prepared resin (Fig. 16, lane a). When the enzyme was chromatographed on a column of resin which had been used several times and regenerated as recommended by the manufacturer (Pharmacia), more than 22 different polypeptides were found to be present. These extra protein bands ranged i n molecular weight from 10 000 to 100 000 (Fig. 16, lane b). A more complete regeneration of the resin was obtained when the previously used resin was washed with 2% (w/v) SDS and then regenerated as recommended. The most active fraction contained 13-15 major polypeptides (Fig. 16, lane c). However, the resin continued to deteriorate with repeated use, even with inclusion of a 2% (w/v) SDS washing step (Fig. 16, lane d). \ 138. Fig. 1 6 SDS-polyacrylamide gel electrophoresis of the E C F ^ F Q complex obtained after chromatography on Phenyl-Sepharose CL-4B: Reproducibility of the purification. The experiment described in the legend to Fig. 12 (PANEL A) was repeated except that the fraction was applied to a column of Phenyl- Sepharose CL-4B (1.8 x 25 cm) which had been regenerated under several different conditions. The most active fraction from the column was analyzed on SDS-gels. The gel consisted of 13% (w/v) acrylamide with a 4% (w/v) acrylamide stacking gel, as described by Laemmli (171). Protein bands were visualized with 0.1% (w/v) Coomassie Blue as described i n MATERIALS AND METHODS. Lane a, freshly-prepared Phenyl-Sepharose CL-4B; lane b, resin was used and regenerated according to the manufacturer's (Pharmacia) recommendations; lane c, resin used for experiment in lane b was regenerated as before but an extra wash with 2% (w/v) SDS was included; lane d, the resin used for the experiment in lane c was regenerated as for the experiment in lane c; lane e consists of molecular weight marker proteins (Mr, 94 000 - 14 400). 139. Fairly reproducible purifications could be obtained each time with a column of freshly-prepared resiri. COMPARISON OF THE' GEL ELECTROPHORESIS AND PROTEIN-DETECTION SYSTEMS The main determinant in assessing the purity of the ATPase complex was by SDS-polyacrylamide gel electrophoresis. The number of polypeptide bands observed has been reported to depend on the resolving power of the gel electrophoresis system (5,95). In this study, a comparison of the resolving power of the phosphate-buffered (174) and the Tris-buffered (171,173) gel electrophoresis systems as well as the protein-detection systems was made. A preparation of the ATPase complex purified on a sucrose gradient was subjected to the different electrophoresis systems. The results are shown in Fig. 17. In each case, the gels were stained with one of the staining systems described in MATERIALS AND METHODS. The Tris-buffered system (Fig. 17A) was more resolving than the phosphate-buffered system (Fig; 17B). In the former, the subunits of ECF^ (q - e) were highly resolved, whereas i n the latter, the a and fl subunits overlapped. The e subunit could barely be identified with the phosphate system. In addition, many of the minor protein bands seen in the Tris system were not resolved i n the phosphate system. In general, the polypeptide bands were more distinct in the Tris-buffered system. The number of protein-staining bands seen on the gel also depended on the staining properties of the dye used. A comparison of four different staining systems i s shown i n Fig. 17. The most intensely-stained prep- aration and the greatest number of bands was obtained with 0.05% (w/v) Coomassie Blue in 25% (v/v) isopropanol and 10% (v/v) acetic acid (Fig. 17A, I ) . A slightly weaker staining-intensity, but equally as resolving, 140. i Fig. 17 Comparison of the SDS-gel electrophoresis and protein-detection systems. * Partially purified ECF-^FQ complex obtained after chromatography on Phenyl-Sepharose CL-4B and sucrose gradient centrifugation, and sucrose gradient-purified ECF^, were subjected to SDS-gel electrophoresis on a linear 7.5% - 16.5% (w/v) acrylamide gradient gel in the Trisrbuffered system (PANEL A), or i n the phosphate-buffered system (PANEL B), as described in MATERIALS AND METHODS. Following electrophoresis, the gels were fixed with a solution containing 5% (w/v) TCA, 5% (w/v) sulfosalicylic acid and 10% (v/v) methanol for 30 min at 60°C. The gels were then stained for 9 h at 20°C with (I), 0.05% (w/v) Coomassie Blue i n isopropanol:acetic acid:water (25:10:65); (II), 0.25% (w/v) Coomassie Blue i n methanol:acetic acid:water (45:10:45); (III), 0.12% (w/v) Coomassie Blue i n ethanol:acetic acid:water (25:8:67); (IV), 1% (w/v) Amido Black i n 7% (v/v) acetic acid. Lanes a and b are partially purified ECF^FQ complex, with lane b containing twice the amount of ECF^FQ complex of that in lane a. Lane c i s purified ECF-̂ . The migration position of the subunits of ECF^ (a - e) are indicated. i 141, ® 1 II III IV 1 " I I 1 a b c a b c i ; i a b c * — i a b c WW ST'*"} IHj . ' . w - • m n : St 11 «- £ i i • i ® 1 _M_ IN IV b a b a b a b MM M • ffi r 142. was obtained with 0.25% (w/v) Coomassie Blue in 45% (v/v) methanol and 10% (v/v) Acetic acid (Fig. 17A, II). The number of minor bands observed decreased considerably when the gel was stained with either 0.12% (w/v) Coomassie Blue i n 25% (v/v) ethanol and 8% (v/v) acetic acid or 1% (w/v) Amido Black in 7% (v/v) acetic acid. The least number of polypeptide bands as well as the weakest staining was observed with Amido Black. 143. PART II STUDIES ON MUTANTS OF E. c o l l DEFECTIVE IN PROTON-TRANSLOCATING ACTIVITY The ECF^FQ complex purified in the preceding section consisted of eleven major polypeptides. Other workers have shown that the complex consists of eight subunits (Tables 2 and 3). Five of these polypeptides were subunits of ECF^ (a - e). The remaining polypeptides of 24 000, 18 000 and 8 400 daltons were considered to be subunits of FQ. However, the mechanism of proton conduction through FQ and the relationship of these polypeptides to each other are not known. One approach to this problem has been through the isolation of mutants of E_. c o l i which are defective i n oxidative phosphorylation (unc mutants). It i s conceivable that a defect in any of the three subunits of FQ would alter i t s proton-translocating properties. Several of these mutants have been isolated and the properties of those studied i n this thesis are listed i n Table 12. E_. c o l i WS1 i s a wild-type strain. Both of the unc mutants, E_. c o l i N I 4 4 and CBT-302 did not exhibit ATPase activity. Analysis of membrane preparation from these two mutants on O'Farrell gels (176) revealed the absence of any subunits of ECF 1 # ATPase activity was detected i n the mutant, E_. c o l i AN382. In contrast to the wild-type strain, the activity of E_. c o l i AN382 was relatively Insensitive to DCCD. At a DCCD concentration of 63 uM, the ATPase ac t i v i t i e s of E. c o l i WS1 and AN382 were inhibited by 60% and 8%, respectively. The ECF^ from both strains were analyzed on O'Farrell gels and found to be identical. Thus, i t i s unlikely that the difference i n the sensitivity of the ATPase activity to DCCD was due to differences in the ECF portion of the ECF nF n complex. 144. Table 12 Some Properties of the unc Mutants of E. c o l i used in this Thesis 3 ATPase activity Presence of ECF^ Strain -DCCD +DCCDb % Inhibition WSI 0.294 0.116 60 + NI44 0 0 - 0 CBT-302 0 0 - 0 AN382 0.199 0.183 8 + a. umol/min/mg protein. b. 63 yM DCCD c. Determined by two-dimensional gel electrophoresis of the membranes +, present; -, absent. Membrane vesicles of _E. c o l i WSI, N ,,, CBT-302 and AN382 were suspended in 50 mM Tris-HCl buffer, pH 8.0, containing 10 mM MgCI2 and 10% (v/v) glycerol at a protein concentration of 4 mg/ml. The membrane suspensions were incubated with DCCD for 30 min. at 37 C after which the ATPase activity was determined as described in MATERIALS AND METHODS. 145. SENSITIVITY OF THE MEMBRANE-BOUND ATPase ACTIVITY TO INHIBITION BY DCCD IN THE PRESENCE OF CATIONS 2+ 2+ The effect of Ca and Mg on the sensitivity of ATPase activity to DCCD in membrane vesicles of E. c o l i ML308-225, AN180, AN382 and WS1 was also examined. In a l l four strains, maximum inhibition of activity by DCCD 2+ was observed i n the presence of Mg (Fig. 18). In the wild-type strains, E. c o l i WS1, AN180 and ML308-225, the enzyme activity was inhibited half- maximally at 3-6 nmol DCCD per mg protein. By contrast, in E. c o l i AN382, the enzyme was inhibited by only 35% at 100 nmol DCCD per mg protein. Therefore, the resistance of the activity to DCCD appeared to be due to defect(s) i n the FQ portion of the ECF^FQ complex. Other methods were then used to characterize this mutant as well as the unc mutants E_. c o l i N I 4 4 and CBT-302. MEASUREMENT OF THE PROTON GRADIENT USING 9-AMINOACRIDINE Everted membrane vesicles of E. c o l i have the ECF^ portion of the ATPase complex exposed to the external medium such that during ATP hydrolysis or substrate oxidation a proton gradient i s generated in these vesicles. The proton gradient within these everted vesicles can be detected by using the fluorescent, lipophilic weak base, 9-aminoacridine. As shown i n Fig. 19 (trace 1), the fluorescent dye equilibrated across the membrane of these everted vesicles i n response to a proton gradient generated by oxi- dation of substrate (NADH). This resulted in a decrease i n fluorescence. When the system went anaerobic, the respiratory chain became non-functional and the fluorescence returned to i t s original level as a result of re-equilibration of the proton gradient across the membrane. The addition 146. F i g . 18 Sensi t iv i ty of the membrane-bound ATPase act iv i ty to Inhibit ion by DCCD i n the presence of cations. Membrane vesicles from E_. c o l i AN382, AN180, ML308-225, and WS1 were prepared i n 50 mM Tris-H2S04 buffer, pH 8.0, containing 1 mM MgCl 2 as described i n MATERIALS AND METHODS. The membrane vesicles were suspended at a protein concentration of 4 mg/ml In 50 mM Tris-H 2S04 buffer, pH 8.0, containing 10% (v/v) glycerol (*-*) and either 10 mM C a C l 2 (O-O) or 10 mM MgCl 2 (•-•) . Various concentrations of DCCD in ethanol (10 ul) were added to 1 ml of membrane vesic les . After 45 min at 37°C, the ATPase ac t iv i ty of the membrane vesicles was determined as previously described. The specif ic ac t iv i t i e s of the membrane-bound ATPase of IS. c o l i AN382, AN180, ML308-225 and WS1 i n the absence of cations were 0.09, 0.30, 0.67 2+ and 0.30 units per mg protein; i n the presence of Ca were 0.11, 0.28, 0.67 and 0.26 units per mg protein and in the presence of Mg^ + were 0.16, 0.36, 0.66 and 0.38 units per mg protein, respectively. D C C D -- nmol/mg p r o t e i n 148. NADH 111 | _ CONTROL STRIPPED USTRIPPED DCCD s _STRIPPED CONTROL ATPase UJ o ? o <n t o < t u LU CC OC O O q 2 MINUTES Fig. 19 Measurement of the proton gradient i n everted membrane vesicles using the fluorescent dye, 9-aminoacridine. Everted membrane vesicles of E_. c o l i WSI were dialyzed against a low ionic strength buffer to remove ECF^ from the membranes, as described in MATERIALS AND METHODS. The ECF^-stripped membrane vesicles were suspended i n 50 mM HEPES-KOH buffer, pH 7.5, containing 5 mM MgCl 2 at a protein concentration of 10 mg/ml. Samples (0.1 ml) were removed and assayed for the a b i l i t y to quench the fluorescence of 9-aminoacridine during substrate (NADH) oxidation, or during hydrolysis of ATP, as described i n the MATERIALS AND METHODS section. Trace 1, untreated everted membrane vesicles; Trace 2, ECF^-stripped membrane vesicles; Trace 3, ECF^-stripped everted membrane vesicles incubated with (375 uM)DCCD for 5 min at 20°C prior to assay; Trace 4, ECF^-stripped everted membrane vesicles incubated with ECF^ (90 ug protein) for 5 min at 20°C prior to assay. 149. of ATP at this point again energized the membrane and a decrease in fluorescence was observed. Quenching of fluorescence was not observed in everted vesicles which had been stripped of ECF^ (Fig. 19, trace 2). However, addition of DCCD or purified ECF^ to these stripped membranes restored i t s a b i l i t y to be energized during substrate oxidation. Stripped membranes which had been reconstituted with ECF^ could also be energized by ATP hydrolysis (Fig. 19, trace 4). These results suggested that the FQ portion of the ATPase complex i s a proton pore and that ECF^ and DCCD could block the passive proton leakage through FQ. The mutants, E. c o l i CBT-302 and AN382, as well as the wild-type strain, Ê . c o l i WS1, were investigated in a similar manner. Normal and mutant membrane vesicles were treated with 2M urea, 2M guanidine hydrochloride or 2% (w/v) silicotungstic acid to remove any subunits of ECF^ which might be retained by the membrane. In this respect, the mutant membrane vesicles were also treated with trypsin. The results shown in Fig. 20 are typical of a l l these treatments. Oxidation of ascorbate (in the presence of PMS to introduce electrons into the cytochrome region of the respiratory chain) by untreated ("washed") membranes resulted i n the quenching of fluorescence of 9-aminoacridine indicating that a proton gradient had been established across the membrane vesicle (Fig. 20, traces 1 and 3). When the system became anaerobic, addition of ATP.resulted i n the restoration of the proton gradient only in the wild-type membranes (Fig. 20, trace 1). Treatment of the membranes of the wild-type strain with urea ("stripping") resulted i n the removal of ECF^ with the loss of the abi l i t y to set-up and maintain a proton gradient (Fig. 20, trace 2). By 150. 1 MIN Fig. 20 Effect of stripping everted membrane vesicles from the parent QL* c o l i WS1) and mutant (IS. c o l i ^ 4 4 ) on the energization of the membrane. Membrane vesicles of E. c o l i WS1 and N144 were treated with 2 M urea to remove ECF^ from the membrane. The membrane vesicles were suspended at a protein concentration of 10 mg/ml i n 50 mM HEPES-KOH buffer, pH 7.5, containing 5 mM MgCl2« Samples (0.1 ml) were removed and assayed for the a b i l i t y to quench the fluorescence of 9-aminoacridine during oxidation of ascorbate (Asc) i n the presence of phenazine methosulphate (PMS) or during ATP hydrolysis, as described i n MATERIALS AND METHODS. 1 and 3 are traces of the quenching observed with the untreated everted vesicles of the parent (WS1) and the mutant ( N J 4 4 ) , respectively. 2 and 4 are traces of the quenching observed with urea-treated ("ECF^-stripped") everted vesicles of the parent (WS1) and mutant ( N J 4 4 ) , respectively. Similar results (trace 4 ) were obtained i f the membrane vesicles from the mutant ( N J 4 4 ) were treated with 2 M guanidine hydrochloride, 2% (w/v) silicotungstic acid or TPCK-trypsin. 151 ) contrast, none of the stripping procedures described earlier destroyed the capacity of the mutant membranes to generate a proton gradient during substrate oxidation (Fig-. 20, trace 4). Thus, i t i s unlikely that reten- tion of individual subunits on the membrane was responsible for the relative impermeability of these membranes to protons. Although ATPase activity was detected in the membranes of the mutant, E_. c o l i AN382 (Table 12), and the ECT^ was identical to that of the wild- type strain, ATP-dependent fluorescence quenching was not observed in this mutant. These results indicated that the lesion(s) responsible for the relative proton-impermeability of the membranes of the unc mutants, E_. col N I 4 4, CBT-302 and AN382 resided in the F Q portion of the E C F ^ complex. This was then investigated. LABELLING OF MEMBRANES OF E. c o l i WITH [ll>C]DCCD It has been shown that the inhibition of ATPase activity in E. c o l i vesicles by DCCD was associated with the labelling of a 8-9 000 dalton polypeptide (141,144). Similarly, the addition of DCCD to "stripped" everted membranes of E_. c o l i blocked proton-translocation, as shown i n Fig 19 (trace 3). Therefore, i t was postulated that the 9 000 dalton poly- peptide (DCCD-binding protein) must be involved in proton conduction (144,145). The relative impermeability of the stripped membranes of the mutant cells suggested that the lesion responsible for this might reside i this polypeptide. The presence of this polypeptide in the mutant membrane was investigated by labelling the membranes of the normal and mutant E_. c o l i with [ 1 %cpCCD followed by separation of the polypeptides by SDS- polyacrylamide gel electrophoresis. A typical result of the labelling 152. experiment of the membranes of the wild-type (E_. c o l i WS1) and the mutant (E. c o l i Nj^^) strains i s shown i n Fig. 21 (traces 1 and 2). Besides material at the top of the gel, two major peaks of radioactivity were observed both with the normal and mutant membranes. The larger of these peaks coincided with a Coomassie Blue staining polypeptide of 9 000 daltons (data not shown). The smaller peak was not associated with a Coomassie Blue staining polypeptide. A characteristic property of the DCCD-binding protein of F Q is i t s a b i l i t y to be extracted into chloroform-methanol (2:1) (144,145,150). [' "* C] -labelled membrane vesicles were extracted with chlorof orm-methanol and the proteins precipitated with ether. Gel electrophoresis of this material from both normal and mutant membranes gave mainly a single poly- peptide band of molecular weight 9 000 (data not shown) coincident with the main peak of radioactivity (Fig. 21, traces 3 and 4). Similar results were obtained with the unc mutants, E. c o l i CBT-302 and AN382 as well as with the wild-type strain E. c o l i ML308-225. These results indicated that the DCCD-binding protein occurs i n the mutants and that the aspartyl residue, which reacts with DCCD and i s involved i n proton translocation, i s present (159,160). PURIFICATION OF THE DCCD-BINDING PROTEIN Although the DCCD-binding protein was found to be present in the mutants, E_. c o l i AN382 and CBT-302 and N-j-̂ ,̂ there remained the possi- b i l i t y that the amino acid composition of the protein was altered i n these mutants. Therefore, attempts were made to purify this polypeptide. ['"CJDCCD labelled membrane vesicles of E. c o l i WS1 or ML308-225 were extracted with chloroform-methanol (2:1) and the proteins precipitated 153. Slice number Fig. 21 SDS-polyacrylamide gel electrophoresis of [ 1"C]DCCD-labelled membranes and of ether-precipitated proteins of chlorof orm-methanol extracts of the labelled membranes of E_. c o l i . Membrane vesicles of the parent (WSI) and mutant (N-j-^) strains of E. c o l i were labelled with [^CJDCCD. The DCCD-binding proteins were isolated from the labelled membrane vesicles by extraction with chloroform: methanol (2:1) followed by precipitation with ether, as described in MATERIALS AND METHODS. The labelled membrane vesicles (150-200 Mg protein) and the isolated DCCD-binding proteins (12-15 yg protein) were subjected to electrophoresis on SDS-urea (8 M) gels. Following electro- phoresis, the gels were fixed with 50% (w/v) TCA, stained with 0.25% (w/v) Coomassie Blue, cut into 2 mm slices and the radioactivity of each s l i c e determined as described previously. Traces 1 and 2 show the distribution of the radioactivity i n the [ 1" CJDCCD-labelled membranes of the parent (WSI) and the mutant ( N J 4 4 ) strains, respectively. Traces 3 and 4 show the distribution of the radioactivity of the isolated DCCD-binding protein from the parent (WSI) and mutant ( N J 4 4 ) strains, respectively. The migration positions of the molecular weight marker proteins (Mr, 94 000 - 14 400) and the bromophenol blue (BP) tracking dye are shown. 154. \ * \ • with ether. The DCCD-binding protein was further purified by thin-layer chromatography. Several bands were visible on the chromatogram (Fig. 22A). About 80% of the total radioactivity applied was associated with band 4 (Rf = 0.9) and 10% with band 1 (Rf = 0.05). The remaining radioactivity was distributed between bands 2 (Rf = 0.45) and 3 (Rf = 0.68). The material eluted from the chromatogram was analyzed by SDS-polyacrylamide gel electrophoresis. A peak of radioactivity was observed with both bands 1 and 4 (Fig. 22B), but only the radioactivity peak of band 1 coincided with a Coomassie Blue staining polypeptide of 9 000 daltons (not shown). The peak of band 4 was associated with an opaque region on the gel, which migrated with a molecular weight of less than 9 000 daltons. The amino acid composition of the material eluted from band 1 revealed the presence of the amino acids, serine, histidine and cysteine. This was diagnostic of a contaminated preparation of DCCD-binding protein since these amino acids are not present i n this polypeptide (146). Therefore, another procedure for purifying the DCCD-binding protein was needed. The [l"C]DCCD-binding protein isolated from E. c o l i WS1 by extraction of the membranes with chloroform-methanol (2:1) was chromatographed on a column of Whatman CM-32 or BioRad Cellex CM, i n the presence of organic solvents, as described i n MATERIALS AND METHODS. The elution profiles were quite different (Fig. 23). Most of the radioactive material applied to the column of BioRad Cellex CM did not bind to the resin (Fig. 23A). The unbound fraction essentially consisted of phospholipids. Protein was not detected in this fraction. By contrast, most of the radioactivity applied to the column of Whatman CM-32 was bound to the resin and eluted in the subsequent washing steps (Fig. 23B). Fractions A and B (Fig. 23B) 155. Fig. 22 Thin layer chromatography of the DCCD-binding protein. The DCCD-binding protein was isolated from [**C]DCCD-labelled membranes of IS. c o l i WSI by extraction with chlorof orm methanol (2:1) and precipitation with ether. The ether-precipitated protein(s) i n chloroform: methanol (2:1) was applied to a s i l i c a gel G plate. The chromatogram was developed by ascending chromatography in a mixture of chlorof ormmethanol: water (65:25:4) containing 20 mM HC1, as described in MATERIALS AND METHODS. The positions of the bands were visualized with iodine (PANEL A). The stained spots were scraped off the s i l i c a plate and extracted with chloro- formmethanol (2:1) containing 20 mM HC1. The extracts were dried, precipi- tated with ether and the precipitate subjected to electrophoresis on SDS- urea (8 M) gels. The gels were fixed with 50% (w/v) TCA, stained with ' 0.25% (w/v) Coomassie Blue, cut Into 2 mm slices and the radioactivity of each slice determined (PANEL B). The migration positions of the bromophenol blue (BP) tracking dye and of the molecular weight marker proteins (Mr, 94 000 - 14 400) are also shown. \ ® Solvent Front • Band 4 R f = 0.9 • \ \ \ \ \ \ \ Band 3 R, =0.68 • WWW I Band 2 R, =0.45——*• WWW Band 1 R f =0.05 • I Origin *" 157. Fig. 23 Chromatography of the DCCD-binding protein on CM-cellulose. DCCD-binding protein was extracted from [l 11 C]DCCD-labelled membranes of IS. c o l i WSI with chlorof ormmethanol (2:1) and precipitated with ether. The ether-precipitated proteins were taken up i n chloroformmethanol (2:1) and applied to a column of CM-cellulose, as described i n MATERIALS AND METHODS. PANEL A: The isolated [1"C]DCCD-binding protein (4-5 mg protein) was applied to a column of BioRad Cellex CM (0 .9 x 14 cm) e q u i l i - brated with chloroform:ethanol (2:1). The column was washed sequentially with 10 column volumes of chlorof ormmethanol (2:1) and 5 column volumes each of chlorof ormmethanol (1:1), chloroform methanol :water (10:10:1) and chlorof ormmethanol :water (5:5:1). 10 ml fractions were collected and the absorbance at 280 nm (A280) O" - -) a n d the radioactivity (O-O) determined. 1-4 refer to the fractions which were pooled and subjected to SDS-gel electrophoresis (see text). PANEL B: The experiment described for Panel A was repeated, except that the resin was Whatman CM-32 cellulose. &-280* (•-•); radioactivity, (O-O). 1-4 and A, B and C refer to the fractions which were pooled and subjected to SDS-gel electrophoresis (see Fig. 24). C, chloroform; M, methanol; H, water. 158. C / M C / M C / M / H C / M / H (2:1) (1:1) (10:10:1 ) (5:5:1) I ' 1 ' 1 ' 1 1 0 100 200 300 400 Volume - - ml 159. contained high amounts of phospholipids. Analysis of the pooled fractions (A, B and C) obtained from the column of Whatman CM-32 on SDS-polyacrylamide gels revealed that the radioactivity was distributed throughout the length of the gel in fractions A and B (Fig. 24, panels A and B). However, only the radioactivity at 9 000 daltons was coincident with a Coomassie Blue staining polypeptide (not shown). The other broad radioactivity peaks were associated with an opaque region extending throughout most of the gel. By contrast, fraction C contained essentially one peak of radioactivity (Fig. 24, panel C), which coincided with a single protein-staining band of 9 000 daltons. The fractions eluted from the column of BioRad Cellex CM were also analyzed in the same manner. Fractions 2 and 3 (Fig. 23A) contained large amounts of phospholipids, but a protein-staining band of 9 000 daltons was present only in fraction 3. The' radioactivity profile of fraction 3 on SDS-polyacrylamide gel was similar to that i n Fig. 24 (panel B). However, the fractions (fraction 4) eluted with chloroform-methanol-water (5:5:1), contained very l i t t l e phospholipid. On SDS-gels, essentially one radio- activity peak coincident with a Coomassie Blue staining polypeptide was observed, as in Fig. 24 (panel C). Therefore, BioRad Cellex CM was subsequently used for the purification of the DCCD-binding protein. Elution with chloroform-methanol-water (10:10:1) was omitted from the purification procedure (150). The amino acid composition of the material eluted with chloroform-methanol-water (5:5:1) revealed that the preparation of DCCD-binding protein was s t i l l contaminated with a polypeptide(s) containing the amino acids serine, histidine and cysteine. The contaminant(s) could be removed by adsorption chromatography on Sephadex LH-60 in the presence of chloroform-methanol Fig. 24 SDS-polyacrylamide gel electrophoresis of DCCD-binding protein obtained by chromatography on CM-cellulose. The fractions eluted from the column of Whatman CM-32 (Fig. 23, PANEL B) were pooled (A, B and C), dried and subjected to electrophoresis on SDS-urea (8 M) gels as described i n MATERIALS AND METHODS. The gels were fixed with 50% (w/v) TCA, stained with 0.25% (w/v) Coomassie Blue, cut Into 2 mm slices and the radioactivity of each sli c e determined as described previously. BP indicates the position of the bromophenol blue tracking dye. The migration positions of the molecular weight marker proteins (Mr, 94 000 - 14 400) are also shown. PANELS A, B and C correspond to the pooled fractions i n Fig. 23, PANEL B. 161. (2:1) containing 20 mM ammonium acetate (149). Two major protein peaks were obtained (Fig. 25). The DCCD-binding protein eluted as the larger of the two peaks immediately after the void volume. Judicious pooling of the fractions from this peak was required for obtaining pure DCCD-binding protein (149). AMINO ACID COMPOSITION OF THE DCCD-BINDING PROTEIN The DCCD-binding protein from normal and mutant cells were extracted with chloroform-methanol (2:1) and purified to homogeneity as described in MATERIALS AND METHODS. Their amino acid compositions were similar, suggesting that the polypeptides from the wild-type and mutant strains were identical (Table 13). Other evidence for identity was obtained by peptide- mapping of the cyanogen-bromide cleaved fragments of the DCCD-binding protein. Separation of the cleaved fragments by only one-dimensional thin-layer chromatography was not adequate to resolve a l l the fragments. Better resolution of the fragments was obtained with two-dimensional thin- layer chromatography. The pattern of migration of the peptides of the DCCD-binding protein from the normal and mutant cells were identical (Fig. 26). Therefore, the lesion responsible for the relative impermeability of the mutant membranes to protons did not appear to be present on the DCCD- binding protein. ANALYSIS OF THE MEMBRANES OF E. c o l i BY TWO-DIMENSIONAL ISOELECTRIC FOCUSING GEL ELECTROPHORESIS The previous results showed that the DCCD-binding protein from the wild-type (WSI and ML308-225) and mutant strains of E. c o l i ( N I 4 4 , AN382 CBT-302) were identical. In order to determine i f the defect responsible 162. Fig. 25 Chromatography of the CM-cellulose-purified DCCD-binding protein on Sephadex LH-60. The DCCD-binding protein (6 mg protein) purified by chromatography on CM-cellulose was applied to a column of Sephadex LH-60 (1.2 x 40 cm) which was equilibrated with chloroformmethanol ((2:1) containing 20 mM ammonium acetate. The column was eluted with this solvent and 1 ml fractions collected as described in MATERIALS AND METHODS. The fractions between the arrows were pooled and stored at -20°C. 163. Table 13 Amino Acid Composition of the DCCD-Binding Protein from Different Strains of E. c o l i Amino Acids ML 308-225 WS1 N144 AN382 CBT-302 DNA b sequence Asp 5.0 4.80 4.67 4.95 5.0 5 Thr 1.06 1.03 1.20 1.04 1.02 1 Ser 0.09 0 0.10 0.04 0.03 0 Glu 4.71 4.47 4.53 4.48 4.34 4 Pro 3.30 2.13 2.64 2.77 2.2 3 Gly 10.29 10.23 10.40 10.23 10.26 10 Ala 12.99 13.20 13.24 13.40 13.30 13 Cys 0 0 0 0 0 0 Val 5.83 5.90 5.84 5.95 5.84 6 Met 6.95(8.0)* 7.09(8.0) 6.61(8.2) 6.90(8.1 7.05(8.3) 8 Ile 7.81 7.82 7.88 7.78 7.88 8 Leu 12.41 12,39 12.50 12.21 12.37 12 Tyr 1.14 1.30 1.23 1.32 1.33 2 Phe 4.19 4.41 4.68 4.45 4.24 4 His 0 0 0 0 0 0 Lys 1.39 1.26 1.23 1.31 1.44 1 Arg 2.01 2.14 2.15 2.13 2.09 2 * values in parentheses were obtained after performic acid oxidation of the DCCD-binding protein and are relative to alanine = 13 and glycine = 10. a. Mean of four determinations; values are in mol amino acid per mol polypeptide. b. from (128, 131, 132). 164. Fig. 26 Two-dimensional thin-layer chromatography of cyanogen bromide cleaved fragments of the DCCD-binding protein of E. c o l i CBT-302. The DCCD-binding protein was treated with cyanogen bromide and the fragments (30-50 ug protein) separated by two-dimensional thin layer chromatography on cellulose. Separation of the cyanogen bromide fragments in the f i r s t dimension was by electrophoresis and i n the second dimension by ascending chromatography, as described i n MATERIALS AND METHODS. The position of migration of the peptides were visualized with ninhydrin. 165. for proton-impermeability was present on the other subunits of the FQ complex, the mutant membranes were analyzed by two-dimensional isoelectric focusing gel electrophoresis. Fig. 27 illustrates the pattern of polypeptides obtained with the membranes from the parent (WSI) and the mutant (N̂ .̂ )̂ strains. None of the subunits of ECF^ were found to be present i n the membranes of E_. c o l i N^ 4 4. The Y subunit of ECF^ i s not seen on these gels since i t migrates off because of i t s very basic nature (pi - 8.9). Careful examination of the gels shown in Fig. 27 revealed that a major polypeptide of molecular weight 18 000 (shown in the square) was absent i n the mutant, _E. c o l i N^.^. (An adjacent minor polypeptide of identical molecular weight was also absent in the mutant. This adjacent spot may have been caused by carbamylation of the major polypeptide by the cyanate formed from the urea present i n the sample. Alternatively, this doublet could be electrophoretic variants of the same subunit. Fillingame et a l . (204) have attempted to explain this occurrence as due to spontaneous deamination of the polypeptide since i t occurred most frequently i n aged samples). It i s l i k e l y that this polypeptide i s polypeptide b of FQ since i t migrates i n this position on gels of this type (126). Unfortunately, polypeptide a of FQ (24 000 daltons) does not enter these polyacrylamide gels (126). Thus, i t s presence i n the membranes of the mutants could not be determined by using this gel system. Comparison of the pattern of polypeptides obtained with the membranes of the mutants E_. c o l i CBT-302 and i t s parent, E_. c o l i CBT-1 revealed the absence of the subunits of ECF^ i n the mutant membranes. However, no differences were observed i n the 18 000 molecular weight region. By contrast, the subunits of ECF, were present i n the membranes of 166. Fig. 27 Two-dimensional Isoelectric focusing gel electrophoresis of membranes of parent (WSI) and mutant ( N J 4 4 ) strains of 15. c o l i . The first-dimension gel (horizontal direction) contained 0.4% (w/v) pH 3.5-10 and 1.6% (w/v) pH 5-7 ampholytes. The second dimension (vertical direction) was a discontinuous polyacrylamide gel (Tris-buffered system) consisting of a layer of 15% (w/v) acrylamide below a layer of 11% (w/v) acrylamide. The samples for electrophoresis were prepared by the method of Merril et a l . (175), as described in MATERIALS AND METHODS. The a, A, 6" and e subunits of the ECF^-ATPase, which are present only i n the membranes of the parent strain, are indicated. The polypeptide outlined by the rectangle i s absent i n the mutant strain. 167. the mutant E_. c o l i AN382. But again, no differences were observed in the 18 000 dalton region. However, minor changes which do not involve a change in the molecular weight or the net charge of the 18 000 dalton polypeptide of FQ could not be excluded. Furthermore, since polypeptide a (24 000 daltons) does not enter these gels, any changes in this polypeptide could not be ascertained. 168. PART III STUDIES ON. THE DCCD-BINDING PROTEIN OF THE F Q COMPLEX OF . E. c o l i r EFFECT OF ANTISERUM TO THE DCCD-BINDING PROTEIN ON THE ENERGIZATION OF THE MEMBRANE OF UREA-STRIPPED EVERTED MEMBRANE VESICLES In my study, attempts to demonstrate proton-translocating activity by reconstituting the DCCD-binding protein into phospholipids, by various methods (206,207), were not successful. But evidence from reconstitution experiments (7) involving ECF^ and ECF^-depleted membranes or by the use of mutants (25,121) suggested that the DCCD-binding protein i s directly involved i n proton translocation i n E_. c o l i . This polypeptide should be transmembranous i f i t i s the sole component of the proton translocation pathway through FQ. The orientation of the DCCD-binding protein i n the membrane of E_. c o l i was investigated by using antiserum to the DCCD-binding protein. As was shown i n the previous section, removal of ECF^ from the membrane vesicles of normal strains of E_. c o l i resulted i n the leakage of protons through FQ. Consequently, reactions such as the quenching of the fluorescence of the dye 9-aminoacridine, which required the presence of a transmembrane pH gradient, could not occur. Fluorescence quenching could be observed i f the leakage of protons through FQ was blocked by the readdition of ECF 1 or by the reaction with DCCD. Fig. 28A shows the effect on fluorescence quenching of various additions to urea-stripped everted vesicles of E_. c o l i . Bovine serum albumin at a concentration normally present i n serum (approximately 37 mg per ml) had no effect on the residual fluorescence quenching of the ECF^-depleted vesicles. Both antiserum to ECF- and preimmune serum had only a slight effect on the 169. Fig. 28 Effect of the antiserum to the DCCD-binding protein and of ECF^ on the ascorbate-oxidatipn-dependent quenching of fluorescence of 9-aminoacridine by urea-stripped everted membrane vesicles. PANEL A: 2M urea-stripped everted vesicles of E_. c o l i WS1 (10 mg protein) i n 50 mM HEPES-KOH buffer, pH 7.5, containing 5 mM MgCl 2 were Incubated for 5 h at A°C with different amounts of antiserum to the DCCD-binding protein (•-•), antiserum to ECF^ (•-•), preimmune serum (O-O) or bovine serum albumin (37 mg/ml) ( A - A ) , i n a f i n a l volume of 1 ml. Samples (0.1 ml) were assayed for the a b i l i t y to quench the fluroescence of 9-aminoacridine with ascorbate as substrate (in the presence of phenazine methosulphate) as described i n MATERIALS AND METHODS. PANEL B: Urea- stripped everted vesicles of E_. c o l i WS1 (1.0 mg protein) in 50 mM HEPES- KOH buffer, pH 7.5, containing 5 mM MgCl 2 were incubated at 20°C for 5 min with various levels of ECF^ i n a f i n a l volume of 0.1 ml. Ascorbate- oxidation-dependent quenching of fluorescence of 9-aminoacridine was measured as described i n MATERIALS AND METHODS. 1 7 0 . quenching. This small effect was l i k e l y due to non-specific interaction of serum proteins with the subunits of FQ. By contrast, antiserum to the DCCD-binding protein markedly stimulated fluorescence quenching indicating that the reaction of the antiserum with this polypeptide blocked the leakage of protons through FQ. If the data of Fig. 28A were examined as a Lineweaver-Burk plot, a value could be calculated for the maximum quenching to be expected at saturating levels of antiserum. This value (62% quenching) was i n good agreement with that observed following the addition of saturating levels of ECF^ (69% quenching, Fig. 28B). EFFECT OF ANTISERUM TO THE DCCD-BINDING PROTEIN ON THE BINDING OF ECF-̂  TO UREA-STRIPPED EVERTED MEMBRANE VESICLES The addition of antiserum to the DCCD-binding protein or ECF 1 to ECF^-depleted membrane vesicles resulted i n the stimulation of fluorescence quenching during substrate oxidation. This suggested that both ECF.^ and antiserum to the DCCD-binding protein, were reacting at the same binding site(s) to prevent the leakage of protons through FQ. The effect of antiserum to the DCCD-binding protein on the binding of ECF 1 to ECF^ depleted membranes, was examined as follows. Stripped everted vesicles of IE. c o l i WSI were incubated with different amounts of antiserum. The vesicles were sedimented by centrifugation and then washed to remove unbound antiserum. Various amounts of ECF^ were added to the suspension of the treated vesicles and the extent of binding of ECF^ measured by the , increase in the ATPase activity of the vesicles. Stripped vesicles had no ATPase activity. The extent of binding of ECF^ as a function of the amount of ECFn added to the treated and untreated vesicles i s shown as a 171. Lineweaver-Burk plot i n Fig. 29. The lines intersecting close to the ordinate suggested that antiserum to the DCCD-binding protein interferred with the binding of ECF^ in a near competitive manner. Preimmune serum at a concentration of 110 ul serum per mg membrane protein also slightly inhibited the binding of ECF^ to the stripped vesicles. However, the extent of inhibition was much lower than that seen with antiserum to the DCCD-binding protein. Similar binding studies were carried out on the unc mutants E_. c o l i N I 4 4 and CBT-302. Stripped everted vesicles of E. c o l i N l 4 4 and CBT-302 were treated with preimmune serum or antiserum to the DCCD-binding protein at a concentration of 120 ul serum per mg membrane protein. Reconstitution with ECF^ was performed as described earlier and the results shown as Lineweaver-Burk plots i n Fig. 29. In both mutants, the results were similar to those obtained with the wild-type strain, E. c o l i WS1, i n that the antiserum to the DCCD-binding protein interfered with the' binding of ECF^ in a near competitive manner. However, in contrast to the wild-type strain, preimmune serum did not inhibit the binding of ECF^ to the mutant membranes. In addition, the extent of inhibition of the binding of ECF^ by antiserum to the DCCD- binding protein was lower i n the mutants than in the wild-type membranes. This was further evidence that the FQ complex i n E_. c o l i Nj^^ and CBT-302 was different from that of the wild-type strain (E_. c o l i WS1) . EFFECT OF ANTISERUM TO THE DCCD-BINDING PROTEIN ON THE PROTON PERMEABIITY OF THE RIGHT-SIDE OUT VESICLES OF E. c o l i The experiments described earlier had detected the exposure of the DCCD-binding protein at the cytoplasmic surface of ECF1-depleted membrane 172. Fig. 29 Effect of antiserum to the DCCD-binding protein on the binding of ECF^ to urea-stripped everted membrane vesicles. Urea-stripped everted vesicles (25 mg protein) from 15. c o l i WSI, N I 4 4 and CBT-302 i n 50 mM HEPES-KOH buffer, pH 7.5, containing 10 mM MgCl 2 and 10% (v/v) glycerol were incubated at 4°C for 20 h with different amounts of antiserum i n a f i n a l volume of 4.4 - 4.5 ml ml. After a 4-fold dilution i n buffer, the mixture was centrifuged at 250 000 xg for 2.5 h. The sedimented vesicles were washed once by suspension in buffer followed by resedimentation as before. The washed vesicles were resuspended in buffer at 6.5 mg protein/ml. Different amounts of ECF^ were incubated at 4°C for 45 min with 2.5 mg of vesicles protein i n buffer. The mixtures were diluted 8 to 10-fold i n buffer and the vesicles sedimented as before. The ATPase act i v i t y of the vesicles was assayed as described i n MATERIALS AND METHODS. Activity i s expressed as units per mg protein. PANEL A: 0 y l ( A — A ) t 50 y l (•-•), 75 y l (a-n) and 110 ul (O-O) of antiserum to the DCCD-binding protein per mg of membrane protein. PANELS B and C: 0 ul ( A - A ) , and 120 ul ( A~ A) of antiserum to the DCCD-binding protein per mg membrane protein. 120 ul ( O - O ) of preimmune serum per mg membrane protein.  174. v e s i c l e s . The extent of exposure of t h i s polypeptide at the external (periplasmic) surface of the c e l l membrane was also examined by using antiserum to the DCCD-binding prot e i n . In t h i s experiment, the antiserum was p a r t i a l l y p u r i f i e d by ammonium sulphate p r e c i p i t a t i o n to reduce the buffering e f f e c t of the serum. The p a r t i a l l y p u r i f i e d antiserum was determined by crossed Immunoelectrophoresis to be s t i l l a c t i v e ( F i g . 30). The ECF-pATPase-defective mutant, E_. c o l i DL-54, r e a d i l y loses ECF^ from i t s binding s i t e s on FQ. Consequently, v e s i c l e s of DL-54 leak protons through FQ (26,205). Addition of valinomycin to K+-loaded " r i g h t - s i d e out" v e s i c l e s r e s u l t s i n an e f f l u x of K+ concomitantly with a compensatory i n f l u x of protons, provided that a pathway f o r protons i s a v a i l a b l e . This pathway i s provided i n DL-54 by the ECF^-depleted FQ proteins (26,205). The basis of t h i s assay i s i l l u s t r a t e d i n F i g . 31. The i n f l u x of protons i s detected as a r i s e i n the pH of the medium external to the v e s i c l e s . A t y p i c a l r e s u l t i s shown i n F i g . 32. Addition of DCCD to the v e s i c l e s blocked the movement of protons through FQ ( F i g . 33B) . A maxi- mum of about 60-65% of the t o t a l protons entering the v e s i c l e s could be i n h i b i t e d by DCCD (880 P M ) . The r e s i d u a l uptake of protons was l i k e l y due to passive movement of the protons across the membrane. In contrast to the e f f e c t of DCCD, the antiserum to the DCCD-binding pr o t e i n had no e f f e c t on t h e - t o t a l amount of protons taken up ( F i g . 33A). However, i f the data were analyzed with respect to the rate, not extent, of proton uptake, an e f f e c t was observed with antiserum to the DCCD-binding p r o t e i n . These r e s u l t s are shown i n F i g . 33 (panels C and D). The rate of proton uptake was i n h i b i t e d to a maximum of 65% at a DCCD concentration of 440-880 P M . Antiserum to the DCCD-binding p r o t e i n also 175. Fig. 30 Crossed Immunoelectrophoresis of antiserum to the DCCD-binding protein. Crossed Immunoelectrophoresis of the DCCD-binding protein was carried out as described i n MATERIALS AND METHODS. UPPER SLIDE: Antigen in the first-dimension was 4 vg DCCD-binding protein in Bjerrum buffer, pH 8.8, containing 100 mM glycine, 38 mM Tris base and 1% (w/v) Triton X-100. 150 ul of antiserum to the DCCD-binding protein was present in the second- dimension gel. LOWER SLIDE: Antigen i n the first-dimension was 8 Ug DCCD-binding protein in Bjerrum buffer, pH 8.8. 175 ul of ammonium sulphate-purified antiserum to the DCCD-binding protein was present in the second-dimension gel. 176. Fig. 31 Schematic representation of the proton-pathway provided by the "right-side out" vesicles of _E. c o l i DL-54. The figure on the l e f t shows that the addition of valinomycin to K +-loaded "right-side out" vesicles of IS. c o l i results in the efflux of K + with a concomitant influx of protons through FQ. The influx of proton i s blocked by addition of antibody to polypeptide(s) of FQ (figure on the right). Details of this type of experiment are described i n the legend to Fig. 32. Val, valinomycin; Ab, antibody; RSO, "right-side out" vesicles. (Drawings courtesy of Dr. P.D. Bragg.) 177. r" Valinomycin Fig. 32 Effect of DCCD and of antiserum to the DCCD-binding protein on the proton permeability of "right-side out" membrane vesicles. E_. c o l i DL-54 was grown to the late exponential phase and converted to spheroplasts by treatment with lysozyme in the presence of EDTA. The spheroplasts were then loaded with K as described i n MATERIALS AND METHODS. The pH changes occurring following the addition of 3 ul of valinomycin ( 5 mg/ml) to 0.85 ml K +-loaded vesicles ( 1 . 5 mg protein) i n 0 . 4 M sucrose - 1 0 mM MgCl 2 were measured with a glass pH electrode connected to a Fisher Accumet Model 325 expanded scale pH meter as described in the MATERIALS AND METHODS section. In some experiments, the vesicles were pre-incubated with DCCD (15 u l i n ethanol) or with ammonium sulphate-purified antiserum to the DCCD-binding protein for 4 5 min at 2 0 ° C . Each assay was internally calibrated by addition of a known con- centration of acid ( H C 1 or H 2 S O 4 ) . 1 , no addition; 2 , 50 ul antiserum to the DCCD-binding protein; 3, 880 uM DCCD. The amounts and rates of proton influx are summarized i n Fig. 33. 178. 240 Q. 3 C 0) (0 + o E c 120 - ® • 1 I 1 1 120 c E a a 3 C a> ro O E c 5 0 S e r u m - u l 100 4 0 0 D C C D - u M 8 0 0 Fig. 33 Effect of DCCD and of antiserum to the DCCD-binding protein on the proton-permeability of "right-side out" membrane vesicles of l i . c o l i DL-54. The experiment was carried out as described in the legend to Fig. 32. PANELS A and B show the amounts of protons taken up by the "right-side out" vesicles, whereas PANELS C and D show the rates of proton-uptake, in the presence of DCCD or antiserum to the DCCD-binding protein. 179. decreased the rate of proton uptake, but only to a maximum of 30%. Thus, there appeared to be some correlation between the effect of DCCD oh the rate of proton uptake and on the total amount of protons taken up by these vesicles. This relationship was not shown when the antiserum to the DCCD-binding protein was used, suggesting that DCCD and antiserum to the DCCD-binding protein affected proton movement through FQ by different mechanisms. EFFECT OF ANTISERUM TO THE DCCD-BINDING PROTEIN ON THE ENERGIZATION OF THE MEMBRANE OF NATIVE EVERTED MEMBRANE VESICLES The effect of antiserum on the quenching of fluorescence during substrate oxidation and ATP hydrolysis by native everted vesicles was also investigated. Everted vesicles of E_. c o l i WS1 were incubated i n the presence of different amounts of antiserum and then assayed for fluorescence quenching as previously described. Bovine serum albumin and preimmune serum did not affect fluorescence quenching during oxidation of ascorbate (in the presence of PMS) (Fig. 34, panel A). By contrast, antiserum to the DCCD-binding protein increased the level of quenching from 50% to 60%. Addition of saturating levels of ECF^ to these vesicles stimulated the quenching to 70%. Similar results were obtained with quenching during ATP hydrolysis (Fig. 34, panel B). Bovine serum albumin and preimmune serum had, at the most, only a slight effect on the fluorescence quenching of the everted vesicles. Stimulation of fluorescence quenching was observed when anti- serum to the DCCD-binding protein was present. These results suggested that there was loss of some ECF1 from the 180. Volume -- ul Fig. 34 Effect of antiserum to the DCCD-binding protein on the energization of untreated everted membrane vesicles. Membrane vesicles of E. c o l i WSI (10 mg protein) i n 50 mM HEPES-KOH buffer, pH 7.5, containing 5 mM MgCl 2 were incubated for 5 h at 4°C with various amounts of antiserum to the DCCD-binding protein (•-•), antiserum to ECFj (•-•), preimmune serum ( A - A ) t or bovine serum albumin (37 mg/ml) (•-•), i n a f i n a l volume of 1 ml. Samples (0.1 ml) were assayed for the a b i l i t y to quench the fluorescence of a 9-arainoacridine during: PANEL A, oxidation of ascorbate i n the presence of phenazine methosulphate; PANEL B, ATP hydrolysis. 181. membrane during preparation. The most striking effect was observed when native everted vesicles were treated with antiserum to ECF^. During substrate oxidation, anti- serum to ECF^ caused a decrease in the quenching of fluorescence (Fig. 34, panel A) indicating that the membranes became leaky to protons. Fluorescence quenching during ATP hydrolysis was completely abolished i n vesicles treated with antiserum to ECF^ (Fig. 34, panel B). The cause(s) of the decrease i n fluorescence quenching during substrate oxidation i n the treated vesicles i s not known. It has been suggested (208,209) that the membrane and/or membrane proteins undergo conformational changes during membrane energization. Therefore, i t i s possible that antiserum to ECF^ prevented the conformational change of ECF^ during substrate oxidation such that the FQ was not completely blocked. Alternatively, the antiserum to ECF^ could have caused the detachment of ECF^ from the membrane (65). Quenching during ATP hydrolysis was abolished i n vesicles treated with antiserum to ECF^ because of inhibition of the ATPase activity rather than because of membranes becoming leaky to protons. Addition of antiserum to ECF1 to purified ECF 1 inhibited the ATPase activity. ENERGIZATION OF THE MEMBRANE OF TRYPSIN-TREATED UREA-STRIPPED EVERTED VESICLES ATP and respiration-dependent fluorescence quenching could be restored to urea-stripped everted vesicles reconstituted with ECF^. By contrast, ECF^ could not restore these ac t i v i t i e s i n stripped vesicles which had been treated with trypsin (Table 14). Since the energized membrane was detected indirectly by the quenching of the fluorescence of 9-aminoacridine, Table 14 Energization of the membrane of trypsin-treated everted membrane vesicles of E. c o l i 182. Percent Quenching System NADH Ascorbate/PMS ATP Urea-treated vesicles 7.5 5.9 0-2 + DCCD 35.1 7.3 0-2 + ECF1 40.5 28.3 37.5 Trypsin -treated stripped vesicles 5.0 5.2 3.5 + DCCD 15.0 6.4 0-2 + ECF1 7.1 4.8 4.3 Urea-treated everted vesicles of E. c o l i WSI were treated with TPCK- trypsin. The treated vesicles were suspended in 50mM HEPES-KOH buffer, pH 7.5, containing 5mM MgCI^ and 10% (v/v) glycerol at a protein concentration of 10 mg/ml. Samples (0.1 ml.) were removed and assayed for the a b i l i t y to quench the fluorescence of 9-aminoacridine during sub- strate (NADH; ascorbate, in the presence of PMS) oxidation or during hydrolysis of ATP, as described in MATERIALS AND METHODS. In experiments in which DCCD or ECF, was used, each was incubated with the membrane 1 o suspension for 5 min. at 20 C, prior to assay. The f i n a l concentration of DCCD in the assay mixture was 375 uM and the amount of ECF used was 80 - 100 yg protein. 1 183. i t was not known i f the absence of quenching i n trypsin-treated vesicles was due to the complete destruction of the respiratory chain or to the vesicles becoming leaky to protons as a result of protease treatment. It was unlikely that the absence of fluorescence quenching activity was due to the destruction of the respiratory chain since NADH-oxidation- dependent fluorescence quenching i n the presence of DCCD was observed (Table 14). Quenching was not observed during ascorbate oxidation because DCCD inhibited i t s activity. This was supported by the observation that trypsin-treatment of the membranes from the unc mutants, E_. c o l i AN382 and CBT-302 did not destroy NADH- or ascorbate-oxidation-dependent fluorescence quenching (Fig. 20). Since ECF^ did not inhibit the proton-translocating properties of FQ i n trypsin-treated vesicles, i t was of interest to determine i f antiserum to DCCD-binding protein could restore respiration-dependent fluorescence quenching. Urea-stripped everted vesicles of E. c o l i WS1 were treated with trypsin and then incubated with different amounts of antiserum. Fig. 35 shows the effect of various additions to trypsin-treated vesicles on the fluorescence quenching. These results appeared to be similar to those i n Fig. 28, panel A, for urea-treated vesicles. Bovine serum albumin had no effect on the residual fluorescence quenching. Preimmune serum and antiserum to ECF^ only slightly stimulated the quenching of fluorescence. By contrast, antiserum to the DCCD-binding protein markedly stimulated fluorescence quenching indicating that reaction of the antiserum with this polypeptide blocked the leakage of protons through F Q . These results were different from those obtained with urea-treated everted vesicles (Fig. 28A) i n two ways. F i r s t l y , the maximum quenching 184. V o l u m e -- ul Fig. 35 Effect of antiserum to the DCCD-binding protein on the ascorbate-oxidation^dependent quenching of fluorescence of 9-aminoacridine by trypsin-treated everted membrane vesicles, Urea-stripped everted vesicles of _E. c o l i WS1 were treated with TPCK- trypsin at 20°C for 30 min as described i n MATERIALS AND METHODS. The trypsin-treated vesicles (10 mg) in 50 mM HEPES-KOH buffer, pH 7.5, containing 5 mM MgCl 2 were incubated for 5 h at 4°C with various levels of antiserum to the DCCD-binding protein (•-•), antiserum to E C F 2 ( • - • ) , preimmune serum (O-O) or bovine serum albumin (37 mg/ml) ( A - A ) , in a f i n a l volume of 1 ml. Samples (0.1 ml) were assayed for the a b i l i t y to quench the fluorescence of 9-aminoacridine with ascorbate as substrate (in the presence of phenazine methosulphate) as described i n the MATERIALS AND METHODS section. 185. expected at saturating levels of antiserum was lower (42% quenching) than with urea-treated everted vesicles (62% quenching). Secondly, the addition of ECF^ to trypsin-treated vesicles (at levels which caused maximum fluorescence quenching in urea-treated vesicles), did not stimulate fluorescence quenching. The lower extent of quenching may be due to damage of the membrane by trypsin, causing an increase in proton leakage other than through FQ. BINDING OF ECF 1 TO PROTEASE-TREATED MEMBRANE VESICLES Treatment of stripped everted vesicles with trypsin did not completely destroy the respiratory chain. However, the absence of fluorescence quenching in trypsin-treated vesicles reconstituted with ECF^ suggested that the coupling- and/or ECF^-binding sites on the membrane were affected by protease treatment. This was examined as follows. Urea-stripped everted vesicles of E. c o l i WSI were treated with trypsin and reconstituted with ECF^ as described earlier. vThe extent of binding of ECF^ was measured by the increase in the ATPase activity of the vesicles since the trypsin- or urea-treated vesicles had no ATPase activity. ATP- and respiration-dependent fluorescence quenching was not observed in these trypsin-treated vesicles which had been reconstituted with ECF^ The results i n Fig. 36 (panel A) shows that the digestion of the stripped vesicles with trypsin for up to 4 h at 37°C did not affect the a b i l i t y of the membranes to bind ECF^. Trypsin-treatment of stripped membranes of the unc mutants, E. c o l i N I 4 4 and CBT-302 also did not alter the capacity of these membranes to bind ECF (Fig. 36, panels B and C). 186. F i g . 36 B i n d i n g of ECF^ to t r y p s i n - t r e a t e d everted membrane v e s i c l e s . U r e a - s t r i p p e d everted v e s i c l e s of 15. c o l i WSI, N 1 4 4 and CBT-302 were t r e a t e d w i t h TPCK-trypsin and then r e c o n s t i t u t e d w i t h v a r i o u s l e v e l s of ECF! as d e s c r i b e d i n MATERIALS AND METHODS. Treatment w i t h TPCK-trypsin was f o r 4 h at 37°C. The amount of membrane v e s i c l e p r o t e i n r e c o n s t i t u t e d w i t h ECF;L was 2 mg. The extent of b i n d i n g of ECF^ to u r e a - s t r i p p e d or t r y p s i n - t r e a t e d ( A - A ) v e s i c l e s was measured by the in c r e a s e I n ATPase a c t i v i t y . A c t i v i t y i s expressed i n u n i t s per mg p r o t e i n . 187. E C F 1 - - H g 188. Double-reciprocal plots of the data i n Fig. 36 (not shown) revealed that the maximum ATPase activity to be expected in urea-treated membrane vesicles of E_. c o l i WS1, N I 4 4 and CBT-302, at saturating levels of ECF^, was similar (8.33 units per mg protein). The half-saturation values for E. c o l i WS1, N T / .. and CBT-302 were determined to be 0.55, 1.11 — 144 and 1.11 mg ECF^, respectively. This suggested that the urea-treated vesicles of E_. c o l i WS1 (wild-type) had a higher a f f i n i t y for ECF1, as would be expected. Similar analysis of the data on the binding of ECF to "trypsin-treated vesicles resulted in more complex kinetics (not shown). Attempts to abolish the a b i l i t y of urea-treated vesicles to bind ECF^ were made also by treating the stripped vesicles with Staphylococcus aureus Vg protease. Stripped vesicles which were treated with Vg protease were s t i l l able to bind ECF1 (Fig. 37). Analysis of the data on Lineweaver-Burk plots, revealed that the maximum activity expected at saturating levels of ECF^ was 8.33 units per mg protein for E_. c o l i WS1 only. Similar analysis of the data on the binding of ECF^ to Vg protease-treated vesicles of E_. c o l i and CBT-302 were also very complex (not shown). EFFECT OF DCCD ON THE ATPase ACTIVITY OF THE ECF̂ ^ BOUND TO TRYPSIN-TREATED VESICLES The activity of the intact membrane-bound ATPase enzyme can be inhibited by DCCD. Since trypsin-treated vesicles were s t i l l capable of binding ECF^, i t was of interest to determine i f the activity of the rebound ATPase enzyme was s t i l l sensitive to DCCD. Urea-stripped everted vesicles of E_. c o l i WS1 were treated with trypsin and then reconstituted with ECF.. as previously described. The F i g . 37 B i n d i n g o f E C F ^ t o S t a p h y l o c o c c u s a u r e u s VR p r o t e a s e - t r e a t e d e v e r t e d membrane v e s i c l e s . U r e a - s t r i p p e d v e s i c l e s o f _E. c o l i WSI , ^ 4 4 and CBT-302 were t r e a t e d w i t h S t a p h y l o c o c c u s a u r e u s VR p r o t e a s e and t h e n r e c o n s t i t u t e d w i t h v a r i o u s l e v e l s o f ECF-L a s d e s c r i b e d i n MATERIALS AND METHODS. T r e a t m e n t w i t h Vg p r o t e a s e was f o r 4 h a t 3 7 ° C . 2 mg o f membrane v e s i c l e p r o t e i n was r e c o n s t i t u t e d w i t h E C F ^ . The e x t e n t o f b i n d i n g o f E C F ^ t o u r e a - s t r i p p e d ( • - • ) o r Vg p r o t e a s e - t r e a t e d ( • - • ) v e s i c l e s was m e a s u r e d by t h e i n c r e a s e i n ATPase a c t i v i t y . A c t i v i t y i s e x p r e s s e d i n u n i t s p e r mg p r o t e i n . E C F 1 H g 191. reconstituted vesicles were incubated with different amounts of DCCD and the ATPase activity measured. The results are illustrated in Fig. 38. After a short period of incubation (45 min at 37°C), the enzyme activities present in the native membrane vesicles and in the trypsin-treated vesicles were inhibited by about 40% at a concentration of 120 nmol DCCD per mg protein. With a longer period of incubation (i2 h at 4°C), the inhibition of the ATPase acitivty i n both cases increased to 60%. Under identical conditions, the ATPase activity of the purified ECF^ was inhibited by 5-10%. Therefore, DCCD inhibited the ATPase activity of the rebound ECF^ to an extent similar to that found in native everted membrane vesicles. IMMUNOPRECIPITATION OF THE ECF-^Q COMPLEX WITH ANTISERUM Since ECF^ could s t i l l bind to trypsin-treated vesicles, i t was conceivable that the reconstituted vesicles could be solubilized with detergents and the ECF^FQ complex immunoprecipitated with antiserum to determine which polypeptide of FQ was l i k e l y to be affected by trypsin- treatment. I n i t i a l l y , attempts were made to determine which polypeptides of the ECFJFQ complex could be immunoprecipitated. " Native everted vesicles of c°li WS1 were washed with a low ionic strength buffer in the presence of protease inhibitors and then solubilized with Aminoxid WS-35 as described in the legend to Fig. 39. The solubilized fraction was treated with the test antiserum. The immunoprecipitate obtained with antiserum to the DCCD-binding protein contained many bands when analyzed by SDS-polyacrylamide gel electrophoresis (Fig. 39). The major protein-staining polypeptides had 192. Fig. 38 Effect of DCCD on the ATPase activity of the ECF± bound to trypsin-treated everted membrane vesicles. Urea-stripped everted vesicles of E_. c o l i WSI were treated with TPCK-trypsin for 45 min at 20°C as described i n MATERIALS AND METHODS. Reconstituti on of the vesicles with ECF-i was carried out as follows! 40 mg of membrane vesicle protein i n 50 mM HEPES-KOH buffer, pH 7.5, contain- ing 10 mM MgCl 2 and 10% (v/v) glycerol was reconstituted with 2.4 mg of ECF^ in a f i n a l volume of 4 ml. The suspension was incubated at 4°C for 45 min, diluted 8-10 fold with buffer, and the vesicles sedimented by centrifugation at 250 000 xg for 2.5 h. The sedimented vesicles were resuspended i n buffer and samples containing 4 mg membrane protein incubated with various levels of DCCD i n a f i n a l volume of 1 ml either for 45 min at 37°C or for 45 min at 37°C followed by 12 h at 4°C before the ATPase activity was determined. Reconstituted urea-treated (•-•) or trypsin-treated ( D - D ) vesicles incubated for 45 min at 37°C; reconstituted urea-treated (•-•) or trypsin-treated (O-O) vesicles incubated for 45 min at 37°C followed by 12 h at 4°C. ECFj^ ( A - A ) treated with DCCD for 45 min at 37°C followed by 12 h at 4°C. The specific a c t i v i t i e s of ECF X and of the reconstituted urea-stripped and trypsin-treated vesicles were 18.5, 1.34 and 1.08 units/mg protein, respectively. 193. Fig. 39 SDS-polyacrylamide gel electrophoresis of the ECF^FQ complex immunoprecipitated with antiserum. Everted membrane vesicles of E_. c o l i ML 308-225 were suspended in 50 mM Tris-HCl buffer, pH 7.4 containing 100 mM KC1, 40 mM e-aminocaproic acid, 24 mM p-aminobenzamidine, 0.8 mM PMSF, 0.4% (w/v) Merthiolate and 2% (v/v) methanol at a protein concentration of 10 mg/ml. The membrane vesicles were solubilized with Aminoxid WS-35 at a detergent to protein ratio of 6, as described in MATERIALS AND METHODS. To 10 ml of solubilized material was added 0.5 ml of antiserum to the DCCD-binding protein or antiserum to ECF^ and the mixture incubated at 4°C for 24-30 h. The precipitate was collected by centrifugation at 12 000 xg for 30 min and washed once with 0.9% (w/v) NaCl containing 1% (w/v) Aminoxid WS-35, and twice with 0.9% (w/v) NaCl. The immunoprecipitate was subjected to SDS-electrophoresis, as described by Laemmli (171). The separating gel consisted of 13% (w/v) acrylamide and the stacking gel was 4% (w/v) acrylamide. The gel was stained with 0.1% (w/v) Coomassie Blue as described previously. Lanes a and c, purified ECF^; lane b, immuno- precipitate obtained with antiserum to the DCCD-binding protein; lanes d and e, immunoprecipitate obtained with antiserum to ECF-̂  with lane e containing twice the amount of material as that i n lane d. The migration positions of the subunits of ECF-̂ , of the DCCD-binding protein (DBP), and of the 18 000 dalton polypeptide of FQ are Indicated. 194. molecular weights of 60 000, 55 000, 32 000, 30 000, 28 000, 22-25 000 1 (stained diffusely), 18 000, 11 000 and 9 000. The polypeptide of 9 000 daltons migrated as a diffuse Coomassie Blue staining region and was identified as the DCCD-binding protein since i t co-migrated with purified DCCD-binding protein. However, the 56 000 (a), 52 000 (B), 22 000 (6) and 13 000 (e) dalton subunits of ECF^ were absent. Similarly, the subunits of ECF^ could not be detected i n the immunoprecipitate when the ATPase complex, purified by Phenyl-Sepharose CL-4B and centrifugation at 250 000 xg for 16-17 h, was treated with antiserum to the DCCD-binding protein. It i s possible that the interaction of the antibody to the DCCD- binding protein with the ECF^FQ complex resulted i n the displacement of ECF^ since i t has been shown that the antibody binds close to the ECF^- binding site on F Q (Fig. 29A). By contrast, treatment of the solubilized fraction with antiserum to ECF^ precipitated polypeptides with molecular weights of 94 000, 90 000, 56 000 (a), 52 000 (fi), 48 000, 35 000, 32 000, 30 000, 28 000, 22-25 000 (stained diffusely), 18 000, 13 000 (e), 11 000 and 9 000 (stained diffusely). The presence of the a, B and e subunits suggested that the 32 000 dalton polypeptide was lik e l y to be the y subunit of ECF^. The 9 000 dalton polypeptide was identified as the DCCD-binding protein as before. These results indicated that the immuno- precipitate obtained with antiserum to ECF^ contained a more "intact" ECF-JFQ complex. Urea-stripped vesicles of E_. c o l l WS1 also were treated with trypsin and reconstituted with ECF^. The reconstituted vesicles were solubilized with Aminoxid WS-35 and the solubilized fraction immunoprecipitated with antiserum to ECF^. The immunoprecipitates were analyzed on the two- dimensional gel electrophoresis system of O'Farrell (176). This technique 195. was used to analyze the contents of the immunoprecipitates because of the characteristic migration patterns of the a, J3, <S and e subunits of ECF^ and the 18 000 dalton subunit (polypeptide b) of F Q (Fig. 27). These preliminary results are shown in Fig. 40. Many polypeptides were present i n the precipitate from the urea-stripped vesicles which had been reconsti- tuted with ECF 1 (Fig. 40A). The ECF^FQ complex in this preparation was presumably intact. The <*, J3, e subunits of ECF^ were easily distinguishable. The 6 subunit of ECF^ was missing i n this preparation. This was probably due to the susceptibility of this subunit to proteases (Dr. Helga Stan-Lotter, personal communication) and the problem was further enhanced by the prolonged incubation period (30-36 h) involved i n the experiment. In addition to these three subunits of ECF^, a polypeptide of 18 000 daltons was present (shown in square) and was l i k e l y to be polypeptide b of F Q . Similarly, the <*, J3 and e subunits of ECF^ were present in the precipitate obtained from the trypsin-treated vesicles which had been reconstituted with ECF^ (Fig. 40 B). In contrast to the urea-treated vesicles, trypsin-treatment resulted i n the cleavage of the 18 000 dalton subunit. This polypeptide was absent, even though the gel was loaded with more protein than i n Fig. 40A. (Note the relative staining intensity of the B and e subunits of ECF^) Analysis of the precipitates from the urea- and trypsin-treated vesicles on one dimensional SDS-polyacrylamide gels revealed that the 9 000 dalton polypeptide (DCCD-binding protein) did not appear to be affected by trypsin-treatment. However, the possibility that trypsin digested a fraction of the DCCD-binding protein as well as the 24 000 dalton subunit (polypeptide a) of F n could not be excluded. 196. Fig. 40 Two-dimensional gel electrophoresis of the ECF^FQ complex obtained by immunoprecipitation with antiserum to ECF-̂ . Urea-stripped everted vesicles of E_. c o l i ML308-225 were treated with TPCK-trypsin and suspended in 50 mM HEPES-KOH buffer, pH 7.5, containing 10 mM MgCl 2 and 10% (v/v) glycerol at a protein concentration of 10 mg/ml. Twenty ml of the treated vesicles were reconstituted with 0.3 ml of ECF^ (9.5 mg/ml) as described in MATERIALS AND METHODS. The reconstituted vesicles were suspended in 50 mM Tris-HCl buffer, pH 7.4 containing 100 mM KC1, 40 mM e-aminocaproic acid, 24 mM p-aminobenzamidine, 0.8 mM PMSF, 0.4% (w/v) Merthiolate and 2%(v/v) methanol at a protein concentration of 10 mg/ml and solubilized with Aminoxid WS-35 at a detergent to protein ratio of 6, as described in the legend to Fig. 39. To 25 ml of the solubilized fraction was added 1.5 ml of antiserum to ECF]_ and the ( mixture incubated at 4°C for 30-36 h. The immunoprecipitate was collected by centrifugation at 12 000 xg for 30 min and washed three times with 50 mM Tris-H2S04 buffer, pH 8.0, containing 10 mM MgCl2- The immunoprecipitate was subjected to two-dimensional gel electrophoresis. The first-dimension gel (horizontal direction) consisted of 0.8% (w/v) pH 3.5-10 and 1.6% (w/v) pH 5-7 ampholytes. The second-dimension gel (vertical direction) was a linear gradient of 7.5-16.5% (w/v) acrylamide in the Tris-buffered system. The gels were stained with 0.1% (w/v) Coomassie Blue, as described previously. PANEL A: Immunoprecipitate obtained from urea-stripped everted vesicles reconstituted with ECF^. PANEL B: Immunoprecipitate obtained from trypsin-treated everted vesicles reconstituted with ECF^. The a, Q and e subunits of ECF^ are indicated. The polypeptide outlined by the rectangle i s absent in the immunoprecipitate from trypsin- treated everted vesicles. ( A ) Urea - treated 198. DETECTION BY SOLID PHASE RADIOIMMUNE ASSAY OF THE REACTION OF ANTIBODY WITH MEMBRANE VESICLES The experiments described earlier, in which the decrease i n proton leakage through FQ i n the presence of antiserum to the DCCD-binding protein was measured, were attempts to determine i f the DCCD-binding protein was transmembranous. The a b i l i t y of the antibody to the DCCD-binding protein to react with this polypeptide in "right-side out" and everted membrane vesicles was also examined by the competitive inhibition assay (195). In this assay, the DCCD-binding protein of the membrane vesicles (urea-stripped everted vesicles or right-side out vesicles) competed for the antibody with the DCCD-binding protein immobilized in microtitre wells. Poly-L-lysine was required to immobilize the antigen. The t i t r a t i o n curve i n Fig. 41 could not be reproduced If the antigen was passively dried (55-60°C) onto the microtitre wells. The results of the competitive inhibition assay are shown in Fig. 42 (panel A). The right-side out vesicles of E_. c o l i WS1 were approximtely one-third as effective i n binding the antibody compared with urea-stripped everted vesicles. Almost identical results were obtained with the vesicles from E_. c o l i ML380-225. In both of these strains, 50% inhibition of binding was obtained with a concentration of 10 ug everted vesicle protein per ml, while that by the right-side out vesicles of E_. c o l i WS1 and ML308-225 was 26.6 yg protein per ml. The a b i l i t y of both types of vesicles to bind the antibody, but with different effectiveness, could be due to the presence of a mixed population of antibodies (i.e. antibodies to determinants exposed on the inside (cytoplasmic side) and outside of the c e l l ) . Alternatively, the binding of antibody to the right-side out vesicles was likely due to reaction with 199. Fig. 41 Titration of the DCCD-binding protein with antiserum to this polypeptide. DCCD-binding protein (3.3 yg) i n 100 mM sodium borate buffer, pH 8.8, containing 2% (w/v) Triton X-100 was immobilized onto polylysine- coated microtitre plate wells. Non-specific binding sites were quenched with RIA buffer and the immobilized DCCD-binding protein was titrated with seri a l dilutions of antiserum to the DCCD-binding protein (•-•) or preimmune serum (O-O). The extent of binding of the rabbit antiserum was measured with 1 2 5 I - l a b e l l e d goat anti-rabbit immunoglobulin as described in MATERIALS AND METHODS. 200. Fig. 42 Inhibition of antibody binding to immobilized DCCD-binding protein by membrane vesicles of 15. c o l i , PS3 and rat l i v e r mitochondria, and by phospholipid vesicles. Urea-stripped everted vesicles were suspended i n 50 mM HEPES-KOH buffer, pH 7.5, containing 10 mM MgCl 2 and 10% (v/v) glycerol. "Right- side out" vesicles were suspended i n 0.4 M sucrose containing 10 mM MgCl 2. The phospholipids were taken up in 100 mM sodium borate buffer, pH 8.8 containing 2% (w/v) Triton X-100. DCCD-binding protein (3.3 yg protein) was immobilized to polylysine-coated microtitre plate wells. Inhibition of antibody binding to the immobilized polypeptide by various amounts of membrane vesicles or phospholipid vesicles was measured by the "competitive inhibition assay" as described i n MATERIALS AND METHODS. PANEL A: (•-•), urea-stripped everted vesicles of E. c o l i WSI; (O-O), right-side out vesicles of E_. c o l i WSI. PANEL B: (•-•), urea-stripped everted vesicles of 15. c o l i WSI; (O-O), urea-stripped everted vesicles of PS3; (•-•), phosphate-washed mitochondrial inner membranes; ( A - A ) , sonicated phosphate-washed mitochondrial inner membranes; (•-•), urea- stripped phosphate-washed mitochondrial inner membanes. PANEL C: (•-•) urea-stripped everted vesicles of E. c o l i WSI; ( A - A ) , soybean phospha- tidylcholine vesicles; (O-O), egg yolk phosphatidylcholine; (•-•), synthetic phosphatidylcholine (l-palmityl-2-oleoyl phosphatidylcholine). Concentration of the everted membrane vesicles i s expressed i n yg protein/ml whereas the phospholipid concentration i s expressed as yg/ml (w/v).  202. re-oriented ATPase complexes (210-212) or to a contamination of the right- side out vesicles with a population of inside-out vesicles. In contrast to the results obtained with E. c o l i WSI and ML308-225, 50% inhibition of binding by the urea-stripped vesicles of the unc mutants E_. c o l i DL-54, N l 4 4 and AN382 occurred at a concentration of 19 yg everted vesicle protein per ml. In addition, 50% inhibition of binding by right-side out vesicles of E_. c o l i DL-54 and were at 24 and 19 yg membrane protein per ml, respectively. The cause(s) of this difference i n the value of the 50% inhibition of binding by everted and right-side out vesicles of the wild-type and mutant strains of E. c o l i i s not known. However, Cox et a l . (140) have proposed, on the basis of genetic studies, that the a and /or B subunits of ECF.. were needed for the proper / 1 assembly of a functional FQ. In i t s (their) absence, the (FQ polypeptide) DCCD-binding protein was suggested to insert randomly into the membrane. Therefore, i f the DCCD-binding protein does exist as a "loop" i n the membrane, then i n a random insertion, one would expect that 50% of the antigenic determinant normally exposed on the cytoplasmic surface to be also exposed on the external surface of the c e l l . In this event, differen- tiation between "right-side" out and "inside-out" vesicles would not be possible and the 50% inhibition of antibody-binding by both types of vesicles should be identical. The ATPase complex of the thermophilic bacterium PS3 has been well characterized (4). In spite of considerable homology between the DCCD- binding proteins of IS. c o l i and PS3 (146), the antibody reacted only weakly with everted membrane vesicles of PS3 (Fig. 42, panel B). Everted inner mitochondrial membrane vesicles (submitochondrial particles) from rat l i v e r , or purified mitochondrial membranes from which MF, had been stripped, did 203. not react with the antibody to the E_. c o l i DCCD-binding protein, at least at the concentration of antigen used (Fig. 42, panel B). It would appear that a very high concentration of membrane vesicles (E_. c o l i , PS3 or mito- chondrial membranes) could result in the trapping or non-specific inter- action of antibody with the vesicles. As shown in Fig. 42 (panel C), at extremely high concentrations (greater than 1.8 mg per ml) of phospholipids, a significant amount of antibody was bound by the phospholipid vesicles. REACTION SITE(S) FOR THE ANTIBODY ON THE DCCD-BINDING PROTEIN The antiserum to the DCCD-binding protein could bind to the purified DCCD-binding protein. Therefore, attempts were made to determine the reaction site(s) for the antibody on this polypeptide. The DCCD-binding protein was modified with several group-specific reagents (213). The modified polypeptide was then used i n the competitive inhibition assay. The immobilized antigen i n the microtitre wells was unmodified DCCD-binding protein. Cleavage of the polypeptide at methionyl residues with cyanogen bromide, or oxidation of methionyl residues with performic acid to greater than 95% (Table 13) (184,213), resulted in a reduction in the af f i n i t y of the modified polypeptide for the antibody by almost two orders of magnitude (Fig. 43). The polypeptides modified by these two reagents were soluble in borate-Triton buffer only up to a level of 90-100 ug protein per ml. Although the extent of cleavage of the DCCD-binding protein by TPCK-trypsin or Staphylococcus aureus Vg protease was not known, the af f i n i t y of the protease-treated polypeptide was also reduced 3-fold. Modifiction of the arginyl residue(s) of the DCCD-binding protein with phenylglyoxal or 2,3-butanedione had no effect on i t s reaction with the antibody. Similarly, 20A. 6] log Protein Fig. 43 Inhibition of antibody binding to immobilized DCCD-binding protein by protease-treated or chemically-modified DCCD-binding protein. The protease-treated or chemically-modified DCCD-binding protein was suspended i n 100 mM sodium borate buffer, pH 8.8, containing 2% (w/v) Triton X-100. DCCD-binding protein (3.3 yg) was immobilized to polylysine- coated microtitre plate wells. Inhibition of antibody binding to the im- mobilized polypeptide by various amounts of protease-treated or chemically- modified DCCD-binding protein was measured by the "competitive inhibition assay", as described i n MATERIALS AND METHODS. PANEL A: ( X - X ) , untreated DCCD-binding protein; (•-•), 2,3-butanedipne-treated DCCD-binding protein; (O-O), phenylglyoxal-treated DCCD-binding protein; (•-•), performic acid-treated DCCD-binding protein; ( A _ A ) , antibody replaced in (control) experiments by preimmune serum. PANEL B: (•-•), untreated DCCD-binding protein; (O-O), TPCK-trypsin-treated DCCD-binding protein; (*-*), VQ- protease-treated DCCD-binding protein; (•-•), cyanogen bromide-treated DCCD-binding protein. 205. treatment of the DCCD-binding p r o t e i n w i t h chloramine T or hydrogen peroxide a l s o d i d not a f f e c t i t s a b i l i t y to bind the antibody (not shown). But the extent of m o d i f i c a t i o n of the DCCD-binding p r o t e i n by these reagents was not known. In c o n t r a s t to the e f f e c t of t r y p s i n and Vg protease on the a b i l i t y v v of the p u r i f i e d DCCD-binding p r o t e i n to react w i t h the antibody, urea- s t r i p p e d v e s i c l e s t r e a t e d w i t h t r y p s i n or Vg protease d i d not cause a re d u c t i o n i n the a f f i n i t y f o r the antibody ( F i g . .44). S i m i l a r r e s u l t s were obtained w i t h u r e a - s t r i p p e d v e s i c l e s which had been t r e a t e d w i t h phenylglyoxal or Chloramine T (not shown). BINDING OF ECFj^ BY PURIFIED DCCD-BINDING PROTEIN The a b i l i t y of ur e a - s t r i p p e d v e s i c l e s of the unc mutant E_. c o l i ^ I 4 4 » w n * c h presumably contains only the DCCD-binding p r o t e i n i n the FQ complex, suggested that t h i s polypeptide may be in v o l v e d i n the binding of ECF^ Indeed, ECF.^ could be bound by the p u r i f i e d DCCD-binding p r o t e i n . This was shown i n two ways. In the f i r s t method, DCCD-binding p r o t e i n immobilized i n m i c r o t i t r e p l a t e w e l l s ("fixed antigen") was reacted w i t h ECF^ ("free a n t i g e n " ) . Bound ECF^ was t i t r a t e d w i t h various d i l u t i o n s of anti-ECF^ serum. The extent of- binding of the antibody was then measured w i t h 1 2 s I - l a b e l l e d goat a n t i - r a b b i t immunoglobulin. (This assay i s s c h e m a t i c a l l y represented i n F i g . 45.) Nonspecific binding of antiserum to the f i x e d antigen (and to the w e l l s ) and n o n s p e c i f i c binding of ECF^ to the w e l l s were correc t e d f o r by o m i t t i n g the ECF^ and the DCCD-binding p r o t e i n , r e s p e c t i v e l y . As shown i n F i g . 46, panel A, n o n s p e c i f i c binding of anti-ECF^ serum was n e g l i g i b l e . S i g n i f i c a n t n o n s p e c i f i c binding of ECF could be detected but t h i s was much l e s s than the binding of ECF.. 206. log P r o t e i n Fig. 44 Inhibition of antibody binding to immobilized DCCD-binding protein by protease-treated or chemically-modified everted membrane vesicles. Urea-stripped everted membrane vesicles of E. c o l i WS1 were treated with phenylglyoxal, Staphylococcus aureus Vs protease, or TPCK-trypsin as described i n MATERIALS AND METHODS. The treated vesicles were then suspended i n 50 mM HEPES-KOH buffer, pH 7.5, containing 10 mM MgCl 2 and 10% (v/v) glycerol. DCCD-binding protein (3.3 ug) was immobilized to polylysine-coated microtitre plate wells. Inhibition of antibody binding to the immobilized polypeptide by various amounts of protease-treated or chemically-modified vesicles was measured by the "competitive inhibition assay" as described i n the MATERIALS AND METHODS section. (•-•), urea- stripped everted vesicles; ( A _ A ) , phenylglyoxal-treated everted vesicles; ( A - A ) , V 3 protease-treated everted vesicles; (O-O), TPCK-trypsin-treated everted vesicles. 207. • DCCD-BP a Rabbit anti-Rj 1 2 ^ 1 ? 5 i > Goat anti- rabbit lg F i g . 45 Schematic representation of the radioimmune binding assay. This figure shows the binding of the free antigen (F^) to the fixed antigen (DCCD-binding protein). The amount of F^ bound was measured by treating the contents of the wells with antiserum to F^ and f i n a l l y with 1 2 5 I - l a b e l l e d goat anti-rabbit immunoglobulin. See MATERIALS AND METHODS for deta i l s . (Drawings courtesy of Dr. P.D. Bragg.) 208. Fig. 46 Binding of ECF^ to the DCCD-binding protein. See MATERIALS AND METHODS for details of the binding assay (Solid Phase Radioimmune Assay) PANEL A; Immobilized DCCD-binding protein (3.3 yg protein) i n the presence (•-•) or absence ( A _ A) of ECF^ (19 yg protein) was titrated with various dilutions of anti-ECF^ serum. (O-O), ECF^ (19 yg protein) added but the immobilized DCCD-binding protein was omitted. PANEL B: Immobilized ECF^ (19 yg protein) i n the presence (•-•) or absence (O-O) of DCCD-binding protein (3.3 yg protein) was titrated with various dilutions of anti-DCCD-binding protein serum. ( A - A ) , DCCD-binding protein was present but immobilized ECFj_-was omitted. 209. found in the presence of the DCCD-binding protein. The extent of nonspecific binding could be decreased by lowering the concentration of ECF^ used i n the experiment. In the second method (Fig. 46, panel B), the fixed antigen was purified ECF1 and the binding of DCCD-binding protein to this was measured using antiserum to the polypeptide. Again, significantly more DCCD-binding protein was bound to ECF^ than i n the control. Prior reaction of immobilized DCCD-binding protein with ECF^ did not affect i t s subsequent reaction with antibody (Fig. 47, panel A). (A similar result was obtained for the reaction of immobilized ECF^ with i t s antibody (Fig. 47, panel B).) There are two possible explanations of this. Either the reaction of the DCCD-binding protein with i t s antibody i s sufficiently strong to displace pre-bound ECF-ĵ  or the ECF^-binding site on the polypeptide i s separate from the binding site for the antibody. As described below, our results favour the latter explanation. REACTION SITE(S) ON ECFj^ FOR THE DCCD-BINDING PROTEIN The subunits of ECF 1 responsible for binding to the DCCD-binding protein were explored. Treatment of ECF^ with TPCK-trypsin followed by reisolation of the enzyme on a sucrose gradient resulted i n the complete removal of the 6 and e subunits (Fig. 48) and cleavage of the amino- terminal fifteen residues of the <* subunits (68,214). The y subunit also appeared to be cleaved i n our preparation. The trypsin-treated ECF^ was bound by the DCCD-binding protein almost as effectively as the native enzyme (Fig. 49, panel A). Pronase treatment of ECF^ reduced the extent of binding dramatically due to extensive cleavage of a l l subunits of the ECF . Treatment of ECF.. with other proteases also reduced the extent 210. E a Reciprocal D i l u t i o n x 10 Fig. 47 Effect of ECF^ on the binding of anti-DCCD-binding protein serum to the DCCD-binding protein and the effect of DCCD-binding protein on the binding of anti-ECF! serum to ECF^. See MATERIALS AND METHODS for details of the radioimmune binding assay. PANEL A: Immobilized DCCD-binding protein (3.3 Pg protein) was titrated with s e r i a l dilutions of anti-DCCD-binding protein serum in the absence (•-•) and presence (n-o) of ECF^ (19 Pg protein). (*-*), ECF]^ (19 Pg protein) was added but the immobilized DCCD-binding protein was omitted. PANEL B: Immobilized ECF^ (19 Pg protein) was titrated with s e r i a l dilutions of anti-ECF-^ serum in the absence (•-•) and presence (O-O) of DCCD-binding protein (3.3 pg). ( A - A ) , DCCD-binding protein (3.3 pg) was added but the immobilized ECF^ was omitted. 211. Fig. 48 SDS-polyacrylamide gel electrophoresis of subunits of ECF^. The subunits of ECF^ were prepared from E. c o l i ML308-225 as described i n MATERIALS AND METHODS. The SDS-gel consisted of 13% (w/v) acrylamide with a 4% (w/v) stacking gel, i n the Tris-buffered system. The gel was stained with 0.1% Coomassie Blue as previously described. Lane a, purified ECFj; lane b, TPCK-trypsin treated ECFj; lane c, purified a subunit of ECF-̂ ; lane d, purified J3 subunit of ECFj^. The migration positions of the subunits of ECF^ (a - e) are indicated. Fig. 49 Binding of ECF^ to the DCCD-binding protein: Effect of protease treatment. ECF^ and DCCD-binding protein were treated with various proteases and the binding assay carried out as described i n MATERIALS AND METHODS. PANEL A: Binding of DCCD-binding protein to TPCK-trypsin treated ECF;L. Immobilized ECF X ( • - • ) (19 yg protein) and TPCK-trypsin treated ECF-L (•-•) (14 yg protein) were titrated with DCCD-binding protein. PANEL B: Binding of DCCD-binding protein to protease-treated ECF^. ECFj^ (19 yg protein) was treated with Vg protease (•-•), a-chymotrypsin (A~ A) or pronase (•-•), or not treated (•-•), and immobilized on polylysine- coated microtitre plate wells and then titrated with DCCD-binding protein. PANEL C: Binding of ECF^ to protease-treated DCCD-binding protein. Titration by ECF-ĵ  of immobilized DCCD-binding protein (3.3 yg) which had been treated with Vg protease ( D - D ) , a-chymotrypsin ( A - A ) , TPCK- trypsin (O-O), or pronase (•-•), or not treated (•-•).  214. of i t s binding of the DCCD-binding protein but the effect was not as large as that seen with pronase (Fig. 49, panel B). Similarly, the binding of ECF^ to the immobilized DCCD-binding protein was almost completely abolished when the polypeptide was treated with pronase or TPCK-trypsin prior to immobilization (Fig. 49, panel C). The role of the major subunits of ECF^ in binding the DCCD-binding protein was investigated following the purification of the a and B subunits from the salt-dissociated enzyme by the procedure of Dunn and Futai (43). The purity of these subunits i s shown i n Fig. 48. While the isolated a subunit was about 100% pure, the isolated Q subunit was s t i l l contaminated (5-10%) with some low molecular weight polypeptides. These isolated subunits were immobilized i n polylysine-treated microtitre plate wells and the extent of binding of the DCCD-binding protein measured using i t s antibody. Both subunits bound the DCCD-binding protein effectively (Fig., 50). EFFECT OF CHEMICAL MODIFICATION OF THE DCCD-BINDING PROTEIN ON ITS REACTION WITH ECFj^ By contrast with i t s inhibitory effect on the binding of antibody, performic oxidation of methionyl residues of the DCCD-binding protein had l i t t l e effect on the binding of ECF^ (Fig. 51, panel A). However, modifi- cation of i t s arginyl residues with phenylglyoxal and 2,3-butanedione reduced the a f f i n i t y of the DCCD-binding protein for ECF1 with a relatively small effect on the total ECF^-binding capacity of the polypeptide (Fig. 52, panel A). Double-reciprocal plots of binding versus concentration gave half-saturation values (under the conditions of this experiment) of 0.26, 0.59 and 0.95 wg for untreated, phenylglyoxal-, and \ 215. Fig. 50 Binding of DCCD-binding protein to subunits of ECF^. See MATERIALS AND METHODS for details of the radioimmune binding assay. Titration of the immobilized ECF^ (•-•) (19 ug protein), a subunit ( A — A ) (6 ug protein) and J3 subunit (O-O) (9 ug protein) by DCCD-binding protein. 216. Fig. 51 Effect of chemical modification of the DCCD-binding protein on the binding of ECF^. The experimental details are described i n the MATERIALS AND METHODS section. The following modification was made. The chemically-modified and the mock-treated DCCD-binding protein (controls) were suspended in 90% formic acid at a protein concentration of 0.13 mg/ml. 25 ul of this mixture (3.3 Ug DCCD-binding protein) was immobilized on polylysine- coated microtitre plate wells and the binding experiment carried out as described previously. PANEL A: Titration of the immobilized mock-treated (•-•) or performic acid-treated ( A~ A) DCCD-binding protein (3.3. ug protein) with ECF^. PANEL B: Titration of the immobilized mock-treated (•-•) or cyanogen bromide-treated (O-O) DCCD-binding protein (3.3 ug protein) with ECFj. F i g . 52 Effect of modification of the arginyl residue(s) of the DCCD-binding protein on the binding of ECFj. Details of the experiment are described i n MATERIALS AND METHODS. PANEL A: T i t ra t ion of immobilized ECFj (19 yg protein) with mock- treated (•-•), phenylglyoxal-treated (O-O) and 2,3-butanedione-treated ( A - A ) DCCD-binding protein. PANEL B: The above data analyzed as a Lineweaver-Burk plot . 218. 219. 2,3-butanedione-treated DCCD-binding protein, respectively (Fig. 52, panel B). This suggests that ECF1 binds to the polypeptide i n the region of i t s arginyl residues and i s supported by the finding that the binding of ECF^ to trypsin-treated DCCD-binding protein was significantly reduced (Fig. 49C). These arginyl residues are located i n the central polar region (residues 41 to 50) (Fig. 55) of the DCCD-binding polypeptide molecule. This region should remain intact when the polypeptide i s cleaved by cyanogen bromide at methionyl residues 17 and 57. As shown in Fig. 51 (panel B), the DCCD-binding protein had increased a f f i n i t y for ECF1 following cleavage by cyanogen bromide. Half-saturation of the binding sites on the untreated and cleaved polypeptide under the conditions of the experiments were given by 0.55 and 0.2 ug ECF.̂ , respectively. The ECF^binding capacities of the polypeptides were similar. EFFECT OF PHENYLGLYOXAL ON THE BINDING OF ECF^ TO UREA-STRIPPED EVERTED VESICLES Since i n vitro binding of ECF^ to the DCCD-binding protein was reduced by treatment of the polypeptide with phenylglyoxal, i t was of interest to determine the effect on intact vesicles. The effect of phenylglyoxal on the binding of ECF^ to urea-stripped vesicles was examined as follows. Stripped everted vesicles of E_. c o l i WS1 and N-j-^ were treated with phenylglyoxal and then washed to remove any unreacted phenylglyoxal. Various amounts of ECF^ were added to the treated vesicles and the extent of binding of ECF^ was measured by the increase i n ATPase activity of the vesicles. Phenylglyoxal inhibited the binding of ECF.̂  to the stripped membranes of the wild-type (E_. c o l i WS1) and the mutant (E_. c o l i N ) strains (Fig. 53). 220. ECF., - - ug F i g . 53 Bindi n g of ECF^ t o p h e n y l g l y o x a l - t r e a t e d v e s i c l e s . U r e a - s t r i p p e d v e s i c l e s of E_. c o l i WSI and were t r e a t e d w i t h p h e n y l g l y o x a l as des c r i b e d i n MATERIALS AND METHODS. R e c o n s t i t u t i o n w i t h ECF-L was c a r r i e d out as f o l l o w s : 2 mg samples of the t r e a t e d membrane v e s i c l e s a t 10 mg p r o t e i n per ml i n 50 mM HEPES-KOH b u f f e r , pH 7.5, c o n t a i n i n g 10 mM M g C l 2 and 10% (v/v) g l y c e r o l were incubated w i t h v a r i o u s l e v e l s of ECF-L at 4°C f o r 45 min i n a f i n a l volume of 0.5 ml. The mixture was then d i l u t e d 8-10 f o l d w i t h b u f f e r and the membranes sedimented at 250 000 xg f o r 2.5 h. The membranes were resuspended i n b u f f e r and the ATPase a c t i v i t y i n the r e c o n s t i t u t e d u r e a - s t r i p p e d (•-•) and p h e n y l g l y o x a l - t r e a t e d (O-O) v e s i c l e s determined as described p r e v i o u s l y . Enzyme a c t i v i t y i s expressed as u n i t s per mg p r o t e i n . 221. It was possible that the effect of phenylglyoxal on the binding of ECF^ was not due to the modification of the arginyl residue(s) of the DCCD-binding protein. In order to determine whether the arginyl residues of the DCCD-binding protein were modified, the urea-stripped vesicles of E_. c o l i WSI were treated with [7- 1 1 1 C]phenylglyoxal. The DCCD-binding protein was extracted from the treated vesicles with chloroform-methanol (2:1) and precipitated with ether. Gel electrophoresis of this material gave three polypeptide bands of molecular weights 96 000, 19 000 and 9 000. The 19 000 dalton polypeptide was stained very intensely with Coomassie Blue (Fig. 54, panel A). However, more than 95% of the total radioactivity recovered from the gel was associated with the protein-staining peak of 9 000 daltons (DCCD-binding protein). The remaining radioactivity was coincident with the peak of 96 000 daltons (Fig. 54, panel B). The product(s) of the reaction,of phenylglyoxal with the arginyl residues has been reported to be quite unstable at alkaline pHs (213). This may l i k e l y be the cause of the decreased amount of radioactivity recovered from the gel following staining with Coomassie Blue (Fig. 54, panel A). These results suggested that the arginyl residue(s) of the DCCD-binding protein were modified in the stripped vesicles. Treatment of the untreated vesicles with chloroform-methanol (2:1) does not normally result i n the extraction of the 19 000 dalton polypeptide. Only the DCCD-binding protein i s extracted from the membrane. Therefore, the extraction of a 19. 000 dalton polypeptide from the phenylglyoxal-treated vesicles suggested that the reaction of the arginyl residue(s) of this polypeptide with phenylglyoxal caused the polypeptide to become quite non-polar. This resulted i n i t s extraction from the membrane with chloroform-methanol (2:1) (Fig. 54B). 222. Fig. 54 SDS-polyacrylamide gel electrophoresis of the DCCD-binding protein of E. c o l i labelled with [7- l"C]phenylglyoxal. 1.25 ml of urea-stripped everted vesicles of 15. c o l i WS1 (13.5 mg protein per ml) i n 50 mM HEPES-KOH buffer, pH 7.5, containing 10 mM MgCl 2 and 10% (v/v) glycerol was mixed with 15 ul of 16.45 mM [7- 1"C]phenyl- glyoxal (specific activity, 15.1 mCi/mmol) (f i n a l concentration, 195 uM). The reaction mixture was incubated at 20°C for 2 h after which the phenyl- glyoxal concentration was increased to 34.1 mM with the non-radioactive reagent. The mixture was incubated for another 60 min. The reaction was stopped by addition of 1 volume of 100 mM arginine hydrochloride in buffer. After incubation for another 30 min at 20°C, the reaction mixture was diluted 10-15 fold with buffer and then centrifuged at 250 000 xg for 2.5 h. The sedimented vesicles were resuspended i n d i s t i l l e d water (10 ml) and the DCCD-binding protein extracted with chloroform-methanol (2:1) as described in MATERIALS AND METHODS. The ether precipitate containing the DCCD-binding protein was subjected to electrophoresis on an SDS-urea (8M) gel. The gel was fixed with 50% (w/v) TCA for 4-6 h at 20°C, stained with 0.1% (w/v) Coomassie Blue, sliced into 1 mm segments and the radioactivity of each sli c e determined. PANEL A: Coomassie Blue-stained gel of the [7- 1 HC]phenylglyoxal labelled proteins. The migration positions of the molecular weight marker proteins (Mr, 94 000 - 14 400), the 19 000 dalton polypeptide, the DCCD-binding protein (DBP) and the bromophenol blue (BP) tracking dye are indicated. PANEL B: Distribution of the radioactivity of identical gels which had been cut into 1 mm slices immediately after fixing in 50% (w/v) TCA ( A - A ) or after staining with Coomassie Blue (•-•). BP indicates the position of the bromophenol blue tracking dye. The migration positions of the molecular weight marker proteins (Mr, 94 000 - 14 400) are also shown. 223. 67 K—• 43 K—• 30K—• 20.1 K—• ll «-19 K 14.4K—• - « - D B P « • BP 224. Furthermore, the absence of radioactivity associated with the 19 000 dalton polypeptide indicated the reactive arginyl residue(s) on this poly- peptide had a low af f i n i t y for phenylglyoxal such that reaction occurred only at a high concentration of phenylglyoxal (34 mM cold phenylglyoxal). Labelling with phenylglyoxal was also suggested by the brownish-yellow polypeptide band (19 000 dalton) observed when the gel was fixed with 50% TCA prior to staining with Coomassie Blue. It has been reported (215) that the reaction of the arginyl residue(s) of a polypeptide with phenylglyoxal results i n an orange-brown product. Therefore, the reactive arginyl residue(s) of the DCCD-binding protein must have a higher a f f i n i t y for phenylglyoxal. 225. DISCUSSION . PURIFICATION OF THE ECFJFQCOMPLEX Attempts to purify the ECF^FQ complex by gel f i l t r a t i o n chromato- graphy only were not successful. The majority of the proteins coeluted with the ATPase activity. In addition, the activity was eluted as a broad peak suggesting that the ATPase complexes were of different molecular weights (Table 11). The enzyme eluted as larger aggregates (M , 680 000 - 890 000) i n the presence of non-ionic detergents than in the presence of ionic detergents (M r > 450 000 - 580 000). This difference i n the molecular weight of the solubilized enzyme was li k e l y due to the asso- ciation of the detergent molecules with the membrane proteins to form lipoprotein-detergent complexes as suggested by Tanford and Reynolds (216). Therefore, the size of the eluted enzyme and the broadness of the activity peak w i l l depend on the extent of association of the detergent molecules with the ECF^FQ complex and on the CMC of the detergent. In the present study, the best purification of the ECF^FQ complex was achieved by chromatography on Phenyl-Sepharose CL-4B followed by sedimentation of the enzyme at 250 000 xg for 16-17 h. The purified enzyme was judged to be an intact ECF^FQ complex. This was determined by the sensitivity of i t s activity to inhibition by DCCD (Fig. 14) and through labelling studies involving [ x"C]-iodoacetic acid and 5-iodoacetamido- fluorescein. The purified ECF^FQ complex consisted of eleven major polypeptides of molecular weight 56 000, 52 000, 32 000, 30 000, 28 000, 22 000, 18 000, 14 000, 12 000, 9 000 and 7 500. Minor bands of 85 000, 71 000 and 24 000 were also present. The polypeptides with molecular weights of 56 000 («), 52 000 (B), 32 000 (Y), 22 000 (6), and 12 000 (e) 226. were subunits of ECF^ Labelling studies with ['"CJDCCD revealed that the polypeptide of 9 000 daltons was the DCCD-binding protein. The polypeptides of molecular weights 28 000, 18 000 and 14 000 may also be subunits of FQ . The identification of some of these polypeptides as subunits of FQ was made by comparison with the purification of others. The subunit composition of FQ obtained in the present study and those obtained by other workers i s summarized in Table 15. The DCCD-binding protein (8-9 000 daltons) and the 18-19 000 dalton polypeptides are clearly subunits of FQ . However, i t was not clear whether the 28 000, 24 000 and 14 000 dalton subunits were polypeptides of F Q . Genetic and DNA sequencing evidence (42,128-133) indicates that the ECF^FQ complex consists of eight subunits of molecular weights 55 264 (a), 50 157 (J3), 34 100 (y), 30 258 (a), 19 310 ( 6 ) , 17 233 (b), 14 194 (e) and 8 365 (c), and are coded for by the unc genes listed i n Table 4. The polypeptides of molecular weight 30 258 (a), 17 233 (b) and 8 365 (c) are the subunits of FQ, although polypeptide a migrates with an apparent molecular weight of 24 000 on SDS-polyacrylamide gels (138). Foster and Fillingame (94) reported the identical subunit composition to the above in a preparation of the ECF^FQ complex. This subunit composition was subsequently confirmed by Friedl et a l . ( 1 0 8 ) . Schneider and Altendorf (107) have also purified the ECF.F. complex J 1 0 and showed that the FQ also consisted of three subunits. But, i n contrast to that found by Foster and Fillingme ( 9 4 ) , the FQ subunits had molecular weights of 28 000, 19 000 and 8 300. The 24 000 dalton polypeptide was present in only a minor amount. In this respect, the ECF^FQ complex appears to be similar to that purified i n this thesis. However, from the DNA sequence of the unc operon and the known anomalous migration of poly- 227. Table 15 Subunit Composition of F Q from Various ECF^F and FQ Purifications. Reference Source Subunit composition of FQ (Mr x 1 0 " 3 ) Foster and Fillingame (94) F i F o 24; 19; 8.4 a Negrin et a l (9) F o 24; 19; 8.4 a F r i e d l and Schairer (108) F i F o 24; 19; 8.3 F o 24; 19; 8.3 Schneiders. Altendorf (107) F i F o 28; 19; 8 . 3 a F o 19; 14; 8.3 This Thesis F i F o 28; 24; 18; 14; 9 a a. This subunit identified as the DCCD - binding protein using [ 1 4C] DCCD 228. peptide a on SDS gels, i t i s unlikely that the 28 000 dalton polypeptide i s a subunit of FQ. Therefore, the question remains as to the identity of the 28 000 dalton polypeptide. Foster and Fillingame (9T4) also reported that a 14 000 dalton polypep- tide copurified with the preparation of ECF^FQ complex only from cells grown in a medium containing succinate, acetate and malate. The 14 000 dalton polypeptide could be the Gene 1 product of the unc operon (128). The presence of a 28 000 and a 14 000 dalton subunit in the E C F ^ complex purified in the present study suggest that the former might be a dimer of the latter. This possibility was f i r s t postulated by Schneider and Altendorf (107). They found that urea-treatment of their purified ECF^FQ complex consisting of the 28 000, 19 000 and 8 300 dalton poly- peptides as the major subunits of FQ resulted in the disappearance of the 28 000 dalton polypeptide, with a concomitant appearance of a 14 000 dalton polypeptide. The resulting FQ preparation consisted of the 19 000, 14 000 and 8 300 dalton polypeptides. The function,of the 14 000 dalton subunit i s not known, although i t has been postulated (128) that i t might function in regulating the assembly of the FQ polypeptides. Schneider and Altendorf (107) also reported that the absence of the 14 000 dalton polypeptide in FQ preparations resulted in a non-functional FQ complex. However, recent genetic studies have shown that the 14 000 dalton subunit (Gene 1 product) i s not essential for the biosynthesis or the activity of a functional ECF^FQ complex (217). C r i t i c a l examination of the densitometric scans of the gels of the ECF^FQ complex prepared by other workers (105-108) clearly revealed the presence of variable amounts of 28 000 and 14 000 dalton polypeptides. This contrasts with the apparent stoichiometric amounts of these polypep- - 229. t i d e s found i n the pr e p a r a t i o n of ECF^FQ complex described i n t h i s t h e s i s . The causes of t h i s i n c o n s i s t e n c y i n the r e l a t i v e amounts and type(s) of subunits of FQ i n the various preparations are not known. I t i s p o s s i b l e t h a t these d i f f e r e n c e s are a f u n c t i o n of the d e t a i l s of the p u r i f i c a t i o n procedures (Tables 2 and 3 ) , of the c e l l s t r a i n , of the growth c o n d i t i o n s (94), of the oli g o m e r i c nature of the FQ polypeptides (149,155) and of p r o t e o l y t i c d i g e s t i o n (95). A l t e r n a t i v e l y , these d i f f e r e n c e s could a l s o be due to the p o s s i b i l i t y that there are polypeptides which are c l o s e l y associated w i t h the ECF^FQ complex i n v i v o . For example, the polypeptides required by the l o c a l i z e d proton hypothesis (218). According to t h i s hypothesis, protons are not extruded i n t o the e x t e r n a l medium during membrane e n e r g i z a t i o n . Instead, the protons are l o c a l i z e d w i t h i n the b i l a y e r and are transported through s p e c i f i c channels c o n s i s t i n g of a s e r i e s of connecting polypeptides between the r e s p i r a t o r y chain and the ATPase complex. Such polypeptides could c o p u r i f y w i t h the ECF^FQ complex. In c o n c l u s i o n , i t i s l i k e l y that the polypeptides of molecular weight 28 000, 24 000, 18 000, 14 000 and 9 000 are subunits of F Q or are c l o s e l y a s s ociated w i t h i t i n the membrane. SOME MUTANTS OF E. c o l l DEFECTIVE IN PROTON TRANSLOCATION Removal of ECF^ from membranes of normal s t r a i n s of 15. c o l i r e s u l t s i n the leakage of protons through FQ. As a r e s u l t , r e a c t i o n s such as the quenching of fluorescence of the dye 9-aminoacridine ( F i g . 19), or the energy-dependent transhydrogenation of NADP + by NADH (23), which r e q u i r e the presence of a transmembrane proton gradient cannot occur. The unc 230. mutants of E_. c o l i N-j.^ and CBT-302 were isolated as ATPase-deficient mutants (Table 7). Although the membranes of these mutants were found to be lacking ECF^ (Table 12), they were s t i l l capable of maintaining a transmembrane proton gradient (Fig. 20). Several explanations for this are possible. (i) Although intact ECF^ i s absent i n these mutants, a subunit of the ECF^ could s t i l l be present and prevent the leakage of protons through FQ. Mutants i n which only the J3 or 6 subunits of ECF^ are retained by the membrane (following removal of ECF^) without i t becoming leaky have been described (219-221). ( i i ) F Q or the DCCD-binding protein could be absent from the mutant membranes, ( i i i ) The aspartyl residue of the DCCD-binding protein with which DCCD reacts specifically to block proton translocation through FQ has been replaced (159-161). Mutants i n which this aspartyl residue has been replaced by a glycyl or glutamyl residue are defective i n proton translocation (Table 6). (iv) A polypeptide of FQ other than the DCCD-binding protein and which i s involved i n the proton channel could be absent from or modified i n the mutant membranes. These possibilities were investigated in the unc mutants, E_. c o l i N^.^, CBT-302 and AN 382. Although the membranes of E_. c o l i AN 382 contain an active ECF^, the membrane cannot be energized through the hydrolysis of ATP, suggesting that there i s a defect i n the FQ portion of the ATPase complex. The extensive stripping procedures used to remove ECF^ and subunits of ECF^ resulted i n the loss of the a b i l i t y of the normal membranes (E_. c o l i WS1) to set up and maintain a proton gradient. By contrast, none of these treatments destroyed the capacity of the mutant membranes to generate and maintain a proton gradient. Therefore, i t was unlikely that retention of individual subunits of ECF^ on the membranes of these mutants was responsible for the relative impermeability of these membranes to protons. / 231. The absence of the a, B, 6 and e subunits of ECF^ on the membranes of E_. c o l i N j 4 4 and CBT-302 was confirmed by comparing their polypeptide composition with that of the parent strains, 15. c o l i WSI and CBT-1. The presence of a [1"C]DCCD-binding protein of 8 000 daltons i n the membranes of these mutants, and the similarity of the amino acid composi- tions to the parent strains (Table 13), indicated that the DCCD-binding protein occurred in these three mutants and that the DCCD-reacting aspartyl residue was present. Therefore, the defect(s) i n FQ responsible for the relative impermeability of these three mutant membranes to protons was li k e l y to be on the other subunits(s) of FQ (polypeptides a and b). Analysis of the membranes of these mutants by two-dimensional isoelectric focusing gel electrophoresis revealed that the 18 000 dalton polypeptide b / of FQ was missing in the mutant, E. c o l i N.^ (Fig. 27). This d i f f e r - ence was not observed in the membranes of the unc mutants, E_. c o l i , AN382 and CBT-302. However, changes in this polypeptide which did not affect i t s molecular weight or i t s net charge could not be excluded. Recently, Friedl et a l . (222) have reported the isolation of an unc mutant of E_. c o l i which contained only the 24 000 dalton (polypeptide a) and the DCCD-binding protein (polypeptide c) of the ECF^FQ complex. The absence of the 18 000 dalton subunit (polyeptide b) of FQ i n this mutant drastically reduced the permeabiity of the membranes to protons. Thus, the mutant, 15. c o l i N^^^ appears to be similar to that isolated by Friedl et a l . (222). From thetarrangement of the genes of the unc operon, i t would appear either that E_. c o l i N-j-^ i s a deletion mutant with only the proximal unc B and E genes (coding for polypeptide a and c of FQ) of the whole operon being retained or there i s a polar effect of the mutation on the expression of the distal genes of the operon. 232. Therefore, these results suggest that the DCCD-binding protein and polypeptide a are insufficient to form a functional proton channel through the membrane. However, i t i s not known i f polypeptide b forms part of the proton channel or acts indirectly by influencing the conformation of the DCCD-binding protein and/or of polypeptide a of F Q . Fillingame et a l . (204) have recently characterized the defect in the unc mutant E_. c o l i AN382 by genetic techniques. It was shown that the lesion responsible for relative impermeabiity of the membranes to protons i n this mutant was due to an amber-suppressible, chain-terminating mutation in the unc B gene that resulted in the loss of the 24 000 dalton subunit (polypeptide a) from the FQ complex. Although the membranes of E_. c o l i CBT-302 did not contain any subunits of ECFp no change(s) i n the composition of FQ could be detected". Thus, the nature of the lesion responsible for proton impermeability in E. c o l i CBT-302 could not be characterized. Therefore, the picture which emerges from the studies on the unc mutants, E_. c o l i N^.^ and AN.382 is that a l l three subunits of FQ (polypep- tides a, b and c) are required for a functional proton channel and this has recently been confirmed by the results of Friedl et a l . (232). This i s also supported by the finding that an unc mutant containing only the 24 000 dalton subunit (polypeptide a) of FQ was relatively impermeable to protons (140). This picture i s complicated by the observation that another unc mutant of E_. c o l i containing a l l three subunits of FQ (a, b and c) was also relatively impermeable to protons (140). Apparently, the presence of a l l three subunits of FQ i s not adequate to form a functional proton channel through the membrane. Cox et a l . (140) have suggested that only the proper assembly of a l l three subunits i n the membrane w i l l result i n 233. functional proton channel and that proper assembly of FQ requires the involvement of the a and/or Q subunit of ECF^. But the presence of the a and 13 subunits i n the membranes of E_. c o l i AN382 suggests that mutations in the-polypeptide(s) of FQ can also affect the assembly of a functional proton channel. The a b i l i t y of the F Q of _E. c o l i AN382 to bind ECT^, although more weakly than in the wild-type strain, in the absence of the 24 000 dalton subunit of FQ suggests that this polypeptide i s not necessary for the binding of ECF^ to the membrane. This i s supported by the finding that the membranes of another unc mutant containing only the 24 000 dalton subunit of F Q was incapable of binding ECF^ (140). Therefore, the ab i l i t y of the isolated membranes of E_. c o l i N̂ .̂ ,̂ containing only the 24 000 and 9 000 dalton subunits of F Q, to bind purified ECF 1 suggests that the DCCD-binding polypeptide i s responsible for binding ECF^ to the membrane. The absence of the 24 000 or the 18 000 dalton polypeptide i n the mutant affected only the af f i n i t y , but not the capacity of the membrane for ECF^. The decreased a f f i n i t y of the mutant membranes (half saturation of binding for E. c o l i WSI, 0.55 mg ECF^; E. c o l i 1.10 mg ECF^) could be explained on the basis that the other FQ subunits most likely influence the conformation of the DCCD-binding protein. Alternatively, the DCCD- binding protein could affect the conformation of the missing polypeptide, thus causing i t to interact with ECF^. Therefore, the membranes of the unc mutants, E_. c o l i N^.^, AN382 and CBT-302 could s t i l l bind ECF^ but not with as high an aff i n i t y as in the wild-type strain (E. c o l i WSI). Studies on the unc mutants, c o l i AN382 and N^ 4 4 suggest that a l l three subunits of FQ may be needed for optimal ECF^-binding a f f i n i t y and are absolutely required for a functional proton 234. channel. ORIENTATION OF THE DCCD-BINDING PROTEIN IN THE MEMBRANE An understanding of the arrangement of the polypeptides of FQ in the membrane i s obviously an important prerequisite for, determining the mechanism of proton translocation through FQ . The orientation of the DCCD-binding protein in the membrane of E_. c o l i was studied by using antiserum against this polypeptide. The antiserum to the DCCD-binding protein blocked the leakage of protons through FQ in urea-stripped everted vesicles suggesting that the DCCD-binding protein was exposed on the cytoplasmic surface of the c e l l membrane (Fig. 28A). Competitive inhibition assays confirmed that the antibodies to the DCCD-binding protein reacted preferentially with the cytoplasmic surface of the c e l l membrane. Although the antibody was raised against the purified DCCD-binding protein, i t could s t i l l recognize and bind to this polypeptide when i t was assembled as an oligomer i n the membrane. The antibody i s presumably recognizing a limited amino acid sequence i n the molecule since i t i s unlikely that the isolated polypeptide would retain i t s native oligomeric structure following extraction by chloroformmethanol (2:1). The amino acid sequence of the DCCD-binding polypeptide exposed on the cytoplasmic , surface of the c e l l membrane and which i s recognized by the^antibody i s not known. However, rn vitro studies with purified DCCD-binding protein have shown that the oxidation of methionyl residues in the isolated polypeptide almost completely abolished i t s a b i l i t y to react with the antibody. Cleavage of the polypeptide at the methionyl residues with cyanogen bromide 235. had a similar effect. These results are consistent with the antibody- binding site being close to one or more methionyl residues. If the binding- site on the isolated polypeptide is the same as that of the membrane-bound polypeptide, then the most l i k e l y methionyl residue i s that at position 57 (Fig. 55). This assignment depends on the validity of the evidence that tyrosyl residues 10 and 73 are on the periplasmic surface of the membrane (223). Modification of arginyl residues of the DCCD-binding protein (in the isolated or membrane-bound form) with phenylglyoxal and 2,3-butanedione had l i t t l e effect on the binding of the antibody (Fig. 43). Thus i t i s unlikely that the antibody reacts with residues 41 to 50 of the polar segment of the polypeptide molecule. On this basis, the DCCD-binding protein molecule must have a looped arrangment in the membrane and the other methionyl residues at position 6, 11, 16, 17, 65 and 75 could not be close to the cytoplasmic surface of the membrane. _ The exposure of the DCCD-binding protein^on the external (periplasmic) surface of the c e l l membrane could not be unambiguously determined. Addition of valinomycin to K+-loaded "right-side out" vesicles of E_. c o l i DL-54 resulted in an efflux of K + concomitantly with a compensatory influx of protons primarily through the ECF^-depleted FQ proteins. This influx of protons was inhibited by DCCD. The total amount of protons as well as the rate of proton uptake was inhibited by DCCD to a maximum of 65%. By contrast, antiserum to the DCCD-binding protein did not reduce the total amount of protons taken up by these vesicles, but did reduce the rate of proton uptake to a maximum of 30% (Fig. 33). The absence of an effect on the movement of the total amount of protons could be due to several causes. (i) The DCCD-binding protein may not be exposed on the external surface of the c e l l membrane. However, the 236. results of Schneider et a l . (223) indicate that the amino- and carboxy- terminals of this polypeptide are exposed on the external surface, ( i i ) The determinants of the DCCD-binding protein on the external surface are inaccessible to the antibodies, ( i i i ) The portion(s) of the DCCD-binding protein exposed on the external surface may be so weakly antigenic that the antiserum does not,contain antibodies to them. (iv) The antibodies did bind to the portion of the molecule exposed on the external surface without affecting proton translocation. The effect of the antiserum to the DCCD-binding protein on the rate of proton movement could also be due to several causes. (i) Antibodies to the determinants on the DCCD-binding protein exposed on the external surface of the membrane are present. ( i i ) There may be a re-orientation of the ATPase complex to the external surface during preparation of the membranes. ( i i i ) The "right-side out" vesicles may be contaminated with a population of "inside-out" vesicles. Energetically, this appears to be more feasible than ( i i ) . However, Wickner (210), Adler and Rosen (211) and Owen and Kaback (212), have suggested that the re-orientation event occurs readily. The extent to which this occurs seems to vary between workers and perhaps i s a function of the c e l l strain and of the details of the procedures used to prepare the vesicles. Thus one explanation of the above results i s that the antibody i s reacting preferentialy with the DCCD-binding protein exposed on the cytoplasmic surface of the c e l l membrane of E_. c o l i and that the binding to right-side out vesicles i s due to the reaction with re-orientated ATPase complexes. Recently, van der Plas et a l . (224) reported that 31% of the ATPase enzyme re-orientated to the external surface in right-side out vesicles. This would account for the results that "right-side out" vesicles of E. c o l i WS1 and ML308-225 (Fig. 42 A) were only one-third as effective 237. in competing for the antibody to the DCCD-binding protein compared with everted vesicles. It i s l i k e l y that the problems associated with the determination of the antigenic sites of the DCCD-binding protein which are exposed and the possible re-orientation of the ATPase enzyme across the membrane may only be solved by using monoclonal antibodies. INTERACTION OF THE DCCD-BINDING PROTEIN WITH ECFj^ The amino acid sequence of the DCCD-binding protein i s suggestive of structure. Four regions of the molecule can be recognized (Fig. 55) (146,150). Seven amino-terminal, predominantly polar amino acids are followed by a non-polar region (residues 8 to 32). Amino acids 34 to 52 are predominantly polar. The fourth region (residues 53 to 79) is non- polar. It has been suggested by Altendorf et a l . (150) that the two non-polar regions of the molecule are transmembranous, forming a looped structure i n which the middle polar region i s exposed at the cytoplasmic surface of the membrane. The looped arrangement i s supported by the findings that (i) replacement of amino acyl residue 28 (with isoleucine or valine, Table 6) i n the f i r s t non-polar segment influenced the reaction of DCCD with aspartyl-61 i n the other non-polar segment (162), ( i i ) tyrosyl residues 10 and 73 were accessible on the periplasmic (outer) surface of the c e l l membrane (223), and ( i i i ) antiserum against the DCCD-binding protein blocked the leakage of protons through FQ i n ECF^-stripped everted membrane vesicles (Fig. 28A). The accessibility of the DCCD-binding protein at the cytoplasmic surface suggested that i t could react with ECF^. In contrast to the effect of modification of amino acyl residues of the DCCD-binding protein on i t s reaction with antibodies, the effect on the 238. Fig. 55 Amino acid sequence of the DCCD-binding protein of E_. c o l i (146) The amino acid sequence of the DCCD-binding protein i s drawn as a "hairpin" loop. The two nonpolar regions of the molecule are shown. DCCD i s bound to the circled aspartyl residue. The methionyl (Q) and arginyl (-¥") residues of the polypeptide are also Indicated. 239. binding of ECF^ was different. Modification of the arginyl residues of the DCCD-binding protein with phenylglyoxal 2,3-butanedione (Fig. 52), or treatment of the DCCD-binding protein with TPCK-trypsin (Fig. 49C), reduced the binding of ECF^ to the polypeptide. However, oxidation of the methionyl residues of the"DCCD-binding protein, or cleavage of the chain at these residues with cyanogen bromide, did not affect the reaction of ECF^ with the polypeptide (Fig. 51). Therefore, the ECF^binding site must be near residues 41 to 50 i n the polar segment of the polypeptide. This site i s consistent with the "looped model" in which this region i s proposed to be at the cytoplasmic surface of the c e l l membrane (223). The independence of the ECF^-binding and the antibody-binding sites was confirmed directly. Binding of ECF^ to the DCCD-binding protein did not prevent the binding of the antibody. The reverse was also true (Fig. 47). However, i t was shown (Fig. 29) with everted membrane vesicles that the binding of antibody interfered with the rebinding of ECF^ to ECF^-stripped membranes. The most likely reason i s that the antibody-binding site i s closer to the ECF 1-binding site i n the oligomeric form of the DCCD-binding protein present in the membrane (5,146). Such conformational differences of the DCCD-binding protein in vivo and in vitro i s also suggested by the lower level of inhibition by antiserum to the DCCD-binding protein in the binding of ECF^ to the membranes of the unc mutants, E_. coli_ N-j-^ and CBT-302 (Fig. 29). As discussed earlier, the polypeptides of FQ may not be correctly assembled in these two mutants. The significance of the binding of ECF1 by the isolated DCCD-binding- protein i s unclear; Evidence for the in vivo involvement of the DCCD- binding protein i n binding F^-ATPase to the membrane comes from the purification of the F1 F n complex from Clostridium pasteurianum (16) and \ 240. from the chloroplast (225). In both cases, the DCCD-binding protein was the only subunit of the FQ complex. It was mentioned previously that arginyl residues are involved in the interaction of ECF^ with the isolated DCCD-binding protein, and they also appear to be involved in the interaction of ECF^ with FQ in everted vesicles. Treatment of ECF^-stripped everted vesicles with phenylglyoxal, an arginyl-modifying reagent, almost completely abolished the rebinding of ECF^ in both the wild-type (E_. c o l i WSI) and the mutant strain (E. c o l i N ^ ) (Fig. 53). Labelling of stripped everted vesicles with [7- 1*C]phenylglyoxal modified the arginyl residue(s) of the DCCD-binding protein (Fig. 54). This suggests that the binding of ECF^ to the isolated DCCD-binding protein may be physiologically ' significant. Binding to the polypeptide involved the a and/or fl subunits of ECF^ since both of these subunits could bind independently. This i s not surprising i n view of their sequence homology (226). However, the 6 and e subunits of ECF^ have also been implicated in the binding of ECF^ to the membranes of E_. c o l i (6,63). These subunits were not required for the Interaction with the isolated DCCD-binding protein. Mutants have been isolated i n which the & (unc D) or 6" (unc H) subunits are retained by the membranes in the absence of other subunits (219-221). The polypeptide(s) with which the B or 6 subunits interact has not been identified, but clearly ECF^ i s able to form linkages with FQ not involving the 6 and e subunits. Recently, Andreo et a l . (227) have concluded also that the <5 subunit of chloroplast F^ is not absolutely required for binding to the membrane, but is required to block the leakage of protons through F Q . Therefore, i t is becoming clearer that ECF^ may not be attached to the membrane solely by a "stalk" consisting of the 6 and e subunits (19,20). It has been suggested instead (231), that a large portion of the ECF, 241. molecule i s embedded in the membrane and that the extent to which i t is embedded i s regulated by the energy state of the c e l l . Walker et a l . (228) have proposed that ECF 1 i s linked via i t s <S and e subunits to polypeptide b of FQ. The amino acid sequence of polypeptide i b suggests that the molecule is composed of two a-helical regions which are anchored by the amino-terminal sequence of non-polar amino acids. The helices are suggested to provide a pathway for conduction of protons to ECF^ (132,228). Walker et a l . (228) also quoted unpublished work of Hoppe et a l . (229) which suggests that this polypeptide of FQ is particularly sensitive to degradation by proteases. Similar results (Fig. 40) have been obtained in my studies. The immunoprecipitate obtained from the ECF^-reconstituted trypsin-treated vesicles did not contain a 18 000 dalton polypeptide indicating that i t had been removed by proteolysis. Hoppe et a l . (233) have recently confirmed that trypsin-treatment of stripped-everted vesicles of E_. c o l i results i n the proteolytic degradation of the 18 000 dalton (polypeptide b) of F Q. In the present study, trypsin-treatment of ECF^-stripped everted vesicles of E. c o l i did not prevent the rebinding of ECF^. The rebound ECF^ was s t i l l sensitive to DCCD. However, even following rebinding of ECF^, FQ was s t i l l leaky to protons. (A somewhat similar behaviour was observed with trypsin-treated submitochondrial particles (230).) One possible explanation of these data i s that in the absence of polypeptide b, destroyed by protease treatment, the proton pathway through FQ and the 6 and e subunits of ECF^ i s disrupted. In this case, binding of ECF^ to F Q could be explained by the interaction of the a and/or J3 subunits with the DCCD-binding protein. The involvement of the DCCD-binding protein in the proton pathway i s supported by the fact that the antibody to this polypep- 242. tide w i l l also block proton leakage even in the trypsin-treated vesicles (Fig. 35). , In summary, the results presented in this thesis are consistent with the lopped arrangement of the DCCD-binding protein in the membrane, as proposed by Altendorf et a l . (150), ih which the polar central region of this molecule i s at the cytoplasmic surface of the c e l l membrane. This region may interact with the a and/or Q subunits of ECF^. Additional linkage(s) to the FQ may involve the interaction of the <S and e subunits of ECF with polypeptide b of F n as proposed by Walker et a l . (228). 243. REFERENCES 1. Simoni, R.D. and Postma, P.W. (1975) Ann. Rev. Biochem. 43, 523-554. 2. Harold, F.M. (1977) Curr. Top. Bioenerg. 6_, 83-149. 3. Haddock, B.A. and Jones, C.W. (1977) Bacteriol. Rev. 4l_, 47-99. 4. Kagawa, Y., Sone, N., Hirata, H. and Yoshida, M. (1979) J. Bioenerg. Biomemb. 11_, 39-78. 5. Fillingame, R.H. (1980) Ann. Rev. Biochem. _49, 1079-1113. 6. Dunn, S.D. and Heppel, L.A. (1981) Arch. Biochem. Biophys. 210, 421-436. / 7. Bragg, P.D. (1979) in Microbiology (Schlessinger, D., ed.), pp. 54-57, Amer. Soc. Microbiol., Washington, D.C. 8. Futai, M. and Kanazawa, H. (1980) Curr. Top. Bioenerg. l£, 181-215. 9. Negrin, R.S., Foster, D.L., and Fillingame, R.H. (1980) J. Biol. Chem. 255, 5643-5648. 10. Friedl, P. and Schairer, H.U. (1981) FEBS Lett. 128, 261-264. 11. Mitchell, P. (1961) Nature (London) 191, 144-148. ( i 12. Boyer, P.D., Chance, B., Ernster, L., Mitchell, P., Racker, E. and Slater, E.C. (1977) Ann. Rev. Biochem. 46, 955-1026. 13. Mitchell, P. (1979) Eur. J. Biochem. 95, 1-20. 14. Mitchell, P. (1976) Biochem. Soc. Trans. 4_, 399-430. 15. Kashket, E.R. (1982) Biochemistry 21, 5534-5538. 16. Clarke, D.J., Fuller, F.M. and Morris, J.G. (1979) Eur. J. Biochem. ^8, 597-612. 17. Abrams, A., McNamara, P. and Johnson, F.B. (1960) J. Biol. Chem. 253, 3659-3662. 18. Voelz, H. (1964) J. Bacteriol. J56, 1196-1198. 19. Openheim, J.D. and Salton, M.J.R. (1973) Biochim. Biophys. Acta 298, 297-322. 20. Hinkle, P.C. and McCarty, R.E. (1978) Sci. Amer. 238, 104-123. 244. 21. Kanner, B.I., Nelson; N. and Gutnick, D.L. (1975) Biochim. Biophys. Acta 396, 347-359. 22. Kobayashi, H. and Anraku, Y. (1972) J. Biochem. 71. 387-399. 23. Bragg, P.D. and Hou, C. (1972) FEBS Lett.^28, 309-312. 24. Bragg, P.D. (1979) i n Membrane Proteins in Energy Transduction (Capaldi, R.A., ed.) pp. 341-449, Marcel Decker, Inc., New York, N.Y. 25. Downie, J.A., Gibson, F. and Cox, G.B. (1979) Ann. Rev. Biochem. 4£, 103-131. 26. Bragg, P.D. and Hou, C. (1973) Biochem. Biophys. Res. Commun. 50, 729-736. 27. Rosen, B.P. (1973) J. Bacteriol. 116, 1124-1129. 28. Altendorf, K., Harold, F.M. and Simoni, R.D. (1974) J. Biol. Chem. 249, 4587-4593. 29. Boonstra, J., Gutnick, D.L. and Kaback, H.R. (1975) J. Bacteriol. 124, 1248-1255. 30. Hasan, S.M. and Rosen, B.P. (1977) Biochim. Biophys. Acta 459, 225-240. 31. Kanazawa, H. and Futai, M. (1979) FEBS Lett. 105, 275-277. 32. Fillingame, R.H. and Wopat, A.F. (1978) J. Bacteriol. 134, 687-689. 33. Davies, P.L. and Bragg, P.D. (1972) Biochim. Biophys. Acta 266, 273-284. 34. Laget, P.P. (1978) Arch. Biochem. Biophys. 189, 122-131. 35. Paradies, H.H., Mertens, G., Schmid, R., Schneider, E. and Altendorf, K. (.1982) Biophys. J. 37, 195-197. 36. Paradies, H.H. and Schmidt, V.D. (1979) J. Biol. Chem. 254, 5257-5263. , 37. Satre, M. and Zaccal, G. (1979) FEBS Lett. 102, 244-248. 38. Yoshida, M., Sone, N., Hirata, H., Kagawa, Y. and Ui, N. (1979) J. Biol. Chem. 254, 9525-9533. 39. Biketov, S.F., Kasho, V.N., Kozlov, I.A., Mileykovskaya, Y.I., Ostrovsky, D.N., Skulachev, V.P., Tikhonora, G.V. and Tsuprun, V.L. (1982) Eur. J. Biochem. 129, 241-250. 40. Kagawa, Y. (1982) i n Membranes and Transport (Martonosi, A.N., ed.) Vol. 1, pp. 439-446, Plenum Press, New York, N.Y. 245. 41. Bragg, P.D. and Hou, C. (1975) Arch. Biochem. Biophys. 167, 311-321. 42. Saraste, M., Gay, N.J., Eberle, A., Runswick, M.J. and Walker, J.E. (1981) Nucleic Acids Res. 9, 5287-5296. 43. Bragg, P.D. (1975) J. Supramol. Struct. _3, 297-303. 44. Yoshida, M., Sone, N., Hirata, H. and Kagawa, Y. (1978) Biochem. Biophys. Res. Commun. j?4, 117-122. 45. Pedersen, P.L. (1975) J. Bioenerg. 6_, 243-275. 46. Shavit, N. (1980) Ann. Rev. Biochem. 49̂ , 111-138. 47. Baird, B.A. and Hammes, G.G. (1979) Biochim. Biophys. Acta 549, 31-53. 48. Amzel, L.M. and Pedersen, P.L. (1978) J. Biol. Chem. 252, 4743-4748. 49. Merchant, S., Shaner, S.L. and Selman, B.R. (1983). J. Bio l . Chem. 258, 1026-1031. 50. Harris, D.A. (1978) Biochim. Biophys. Acta 463, 245-273. 51. Maeda, M., Kobayashi, H., Futai, M. and Anraku, Y. (1976) Biochem. Biophys. Res. Commun. _70, 228-234. 52. Maeda, M., Kobayashi, H., Futai, M. and Anraku, Y. (1977) J. Biochem. 82, 311-314. 53. Nelson, N., Kanner, B.I. and Gutnick, D.L. (1974) Proc. Natl. Acad. Sci., U.S.A. 71_, 2720-2724. 54. Futai, M., Sternweiss, P.C. and Heppel, L.A. (1974) Proc. Natl. Acad. Sci., U.S.A. 71_, 2725-2729. 55. Hanson, R.L. and Kennedy, E.P. (1973) J. Bacteriol. 114, 772-781. 56. Bragg, P.D., Davies, P.L. and Hou, C. (1973) Arch. Biochem. Biophys. 159, 664-670. 57. Smith, J.B. and Sternweiss, P.C. (1975) Biochem. Biophys. Res. Commun. 62, 764-771. 58. Smith, J.B. and Sternweiss, P.C. (1977) Biochemistry 16_, 306-311. 59. Smith, J.B. and Sternweiss, P.C. (1977). Biochemistry 16_, 4020-4025. 60. Younis, H.M., Winget, G.D. and Racker, E. (1977) J. Biol. Chem. 252, 1814-1818. 61. Yoshida, M., Okamoto, H., Sone, N., Hirata, H. and Kagawa, Y. (1977) Proc. Natl. Acad. Sci., U.S.A. 74, 936-940.' 246. 62. Abrams, A., Jensen, C. and M o r r i s , D.H. (1976) Biochem. Biophys. Res. Commun. j>9, 804-811. 63. Sternweiss, P.C. (1978). J . B i o l . Chem. 253, 3123-3128. 64. Laget, P.P. and Smith, J.B. (1979) Arch. Biochem. Biophys. 197, 83-89. - i 65. Smith, J.B. and Sternweiss, P.C. (1982) Arch. Biochem. Biophys. 217, 376-387. 66. Van de Stadt, R.J., de Boer, B.L. and Van Dam, K. (1973) Biochim. Biophys. Acta. 292, 338-349. 67. Asami, K., J u n t t i , K. and E r n s t e r , L. (1970) Biochim. Biophys. Acta 205, 307-311. 68. Dunn, S.D., Heppel, L.A. and Full m e r , C.S. (1980) J . B i o l . Chem. 255, 6891-6896. 69. Larson, R. and Smith, J.B. (1977) Biochemistry .16, 4266-4270. 70. Dunn, S.D. (1982) J . B i o l . Chem. 257, 7354-7359. 71. Vogel, G. and S t e i n h a r t , R. (1976) Biochemistry 15_, 208-216. 72. Dunn, S.D. and F u t a i , M. (1980) J . B i o l . Chem. 255, 113-118. 73. Bragg, P.D. and Hou, C. (1977) Arch. Biochem. Biophys. 178, 486-494. 74. Vogel, G., S c h a i r e r , H.U. and S t e i n h a r t , R. (1978) Eur. J . Biochem. JB7, 155-160. 75. Dunn, S.D. (1978) Biochem. Biophys. Res. Commun. J32, 596-602. 76. Kanazawa, H., S a i t o , S. and F u t a i , M. (1978) J . Biochem. 84, 1513-1517. 77. S a t r e , M., Lun a r d i , J . , Pougeois, R. and V i g n a i s , P.V. (1979) Biochemistry 18_, 3134-3140. 78. Pougeois, R., Satre, M. and V i g n a i s , P.V. (1980) FEBS L e t t . 117, 344-348. 79. V e r h e i j e n , J.H., Postma, P.W. and Van Dam, K. (1978) Biochim. Biophys. Acta 502, 345-353. 80. L u n a r d i , J . , S a t r e , M. and V i g n a i s , P.V. (1981) Biochemistry 20, 473-480. 81. Bragg, P.D., St a n - L o t t e r , H. and Hou, C. (1981) Arch. Biochem. Biophys. 207, 290-299. 247. 82. Bragg, P.D., Stan-Lotter, H. and Hou, C. (1982) Arch. Biochem. Biophys. 213, 669-679. 83. Schatz, G., Penefsky, H.S. and Racker, E. (1967) J. Biol. Chem. 242, 2552-2560. 84. Liu, S.S., Tsai, H.L., Tsou, T.M. and Gao, F.H. (1976) Acta Biochem. Biophys. Sin. 8_, 307-317. 85. Liu, S.S., Gao, F.H., Tsai, H.L. and Ding, Y.Z. (1982) ' Biophys. J. 37, 88-91. 86. Futai, M., Kanazawa, H. and Takeda, K. (1980) Biochem. Biophys. Res. Commun. 96, 227-234. 87. Takeda, K., Hirano, M., Kanazawa, H., Nukiwa, N., Kagawa, Y. and Futai, M. (1982) J. Biochem. 91, 695-701. 88. Bragg, P.D. and Hou, C. (1976) Biochem. Biophys. Res. Commun. 72, 1042-1048. 89. Bragg, P.D. and Hou, C. (1980) Eur. J. Biochem. 106, 495-503. 90. Helenius, A. and Simons, K. (1975) Biochim. Biophys. Acta 415, 29-79. 91. Helenius, A., McCaslin, D.R., Fries, E. and Tanford, C. (1979) in Methods i n Enzymology (Fleischer, S. and Packer, L., eds.) Vol. 56, pp. 735-749, Academic Press, Inc., New York, N.Y. 92. Gonenne, A. and Ernst, R. (1978) Anal. Biochem. 87, 28-38. 93. Ryrie, I.J. (1977) Arch. Biochem. Biophys. 184, 464-475. 94. Foster, D.L. and Fillingame, R.H. (1979) J. Biol. Chem. 254, 8230-8236. 95. Ryrie, I.J. and Gallagher, A. (1979) Biochim. Biophys. Acta 545, 1-14. 96. Ludwig, B., Prochaska, L. and Capaldi, R.A. (1980) Biochemistry 19, 1516-1523. 97. Soper, J.W., Decker, G.L. and Pedersen, P.L. (1979) J. Biol. Chem. ' 254, 11170-11176. 98. Sebald, W. (1977) Biochim. Biophys. Acta 463, 1-27. 99*. Pick, U. and Racker, E. (1979) J. Biol. Chem. 254,' 2793-2799. 100. Nelson, N., Nelson, H. and Schatz, G. (1980) Proc. Natl. Acad. Sci., U.S.A. 77_, 1361-1364. 101. Zakharov, S.D. and Mal'yan, A.N. (1981) Biokhimiya 46, 1236-1243. 248. 102. Sone, N., Yoshida, M., Hirata, H. and Kagawa, Y. (1975) J. Bio l . Chem. 250, 7917-7923. 10a. Sone, N., Yoshida, M., Hirata, H. and Kagawa, Y. (1978) Proc. Natl. Acad. Sci., U.S.A. 75_, 4219-4223. 104. Hare, J.F. (1975) Biochem. Biophys. Res. Commun. 66, 1329-1337. 105. Rosen, B.P. and Hasan, S.M. (1979) FEBS Lett. 104, 339-342. 106. Friedl, P., Friedl, C. and Schairer, H.U. (1979) Eur. J. Biochem. 100, 175-180. 107. Schneider, E. and Altendorf, K. (1980) FEBS Lett. 116, 173-176. 108. Friedl, P. and Schairer, U. (1981) FEBS Lett. 128, 261-264. 109. Okamoto, H., Sone, N., Hirata, H., Yoshida, M. and Kagawa, Y. (1977) J. Biol. Chem. 252, 6125-6131. 110. Cohen, N.S., Lee, S.-H. and Brodie, A.F. (1978) J. Supramol. Struct. 8_, 111-117. 111. Leimgruber, R.M., Jensen, C. and Abrams, A. (1981) J. Bacteriol. 147, 363-372. 112. Sone, N., Yoshida, M., Hirata,- H. and Kagawa, Y. (1977) J. Biol. Chem. 252, 2956-2960. 113. Sone, N., Takeuchi, Y., Yoshida, M. and Ohno, K. (1977) J. Biochem. 82_, 1751-1758. 114. Foster, D.L., Mosher, M.E., Futai, M. and Fillingame, R.H. (1980) J. Biol. Chem. 255, 12037-12041. 115. Kanazawa, H., Miki, T., Tamura, F., Yura, T. and Futai, M. (1979) Proc. Natl. Acad. Sci., U.S.A. 76_, 1126-1130. 116. Butlin, J.D., Cox, G.B. and Gibson, F. (1971) Biochem. J. 124, 75-81. 117. Schairer, H.U. and Haddock, B.A. (1972) Biochem. Biophys. Res. Commun. _48, 544-551. 118. Butlin, J.D., Cox, G.B. and Gibson, F. (1973) Biochim. Biophys. Acta 292, 366-375. 119. Gutnick, D.L., Kanner, B.I. and Postma, P.W. (1972) Biochim. Biophys. Acta 283, 217-222. 120. Cox, G.B.', Gibson, F. and McCann, L. (1973) Biochem. J. 134, 1015-1021. 121. Gibson, F. (1982) Proc. R. Soc. (London) 215, 1-18. 249. 122. Ames, B.N. (1966) i n Methods in Enzymology (Estabrook, R.W. and Pullman, M.E., ed.) Vol. 8̂, pp. 115-118, Academic Press, Inc., New York, N.Y. 123. Gibson, F., Cox, G.B., Downie, J.A. and Radik, J. (1977) Biochem. J. 164, 193-198. 124. Gibson, F., Cox, G.B., Downie, J.A. and Radik, J. (1977) Biochem. J. 162, 665-670. 125. Gibson, F., Downie, J.A., Cox, G.B. and Radik, J. (1978) J. Bacteriol. 134, 728-736. 126. Downie, J.A., Langman, L., Cox, G.B., Yanofsky, C. and Gibson, F. (1980) J. Bacteriol. 143, 8-17. 127. Downie', J.A., Cox, G.B., Langman, L., Ash, G., Becker, M. and Gibson, F. (1981) J. Bacteriol. 145, 200-210. 128. Gay, N.J. and Walker, J.E. (1981) Nucleic Acids Res. 9_, 3919-3926. 129. Mabuchi, K., Kanazawa, H., Kayano, T. and Futai, M. (1981) Biochem. Biophys. Res. Commn. 102, 172-179. 130. Kanazawa, H., Kayano, T., Mabuchi, K. and Futai, M. (1981) Biochem. Biophys. Res. Commun. 103, 604-612. 131. Kanazawa, H., Mabuchi, K., Kayano, T., Noumi, T., Sekiya, T. and Futai, M. (1981) , Biochem. Biophys. Res. Commun. 103, 613-620. 132. Nielsen, J., Hansen, F.G., Hoppe, J., Friedl, P. and von Meyenburg, K. (1981) Mol. Gen. Genet. 184, 33-39. 133. Kanazawa, H., Kayano, T., Kiyasu, T. and Futai, M. (1982) Biochem. Biophys. Res. Commun. 105, 1257-1264. 134. Brusilow, W.S.A., Klionsky, D.J. and Simoni, R.D. (1982) J . Bacteriol. 151, 1363-1371. 135. Brusilow, W.S.A., Gunsalus, R.P., Hardeman, E.C, Decker, K.P. and Simoni, R.D. (1981) J. Biol. Chem. 256, 3141-3144. 136. Bridgen, J. and Walker, I.D. (1976) Biochemistry 15, 792-797. 137. Anderson, S., Bankier, A., Barrel, B.G., de Bruijn, M.H.L., Coulson, A.R., Drouin, J., Eperori^ I.C., Nierlich, D.P., Roe, B.A., Sanger, F., Schreier, P.H., Smith, A.J., Staden, R. and Young, I.G. (1981) Nature (London) 290, 457-465. 138. Steffens, K., K i l t z , H.H., Schneider, E., Schmid, R. and Altendorf, - K. (1982) FEBS Lett. 142, 151-154." / 250. 139. Decker, K.P., Brusilow, W.S.A., Gunsalus, R.P. and Simoni, R.D. (1982) J. Bacteriol. 152, 815-821. 140. Cox, G.B., Downie, J.A., Langman, L., Senior, A.E., Ash, G., Fayle, D.R.H. and Gibson, F. (1981) J. Bacteriol. 148, 30-42. 141. Altendorf, K. and Zitzman, W. (1975) FEBS Lett. 59, 268-272. 142. Sebald, W., Hoppe, J. and Wachter, E. (1979) in Function and Molecular Aspects of Biomembrane Transport (Quagliariello, E., ed.) pp. 63-74, Elsevier/North-Holland, New York, N.Y. 143. Beechey, R.B., Roberton, A.M., Holloway, C.T. and Knight, I.G. (1967) Biochemistry 3867-3879. 144. Fillingame, R.H. (1975) J. Bacteriol. 124, 870-883. 145. Cattell, K.J., Lindop, C.R., Knight, I.G. and Beechey, R.B. (1971) Biochem. J. 125, 169-177. 146. Sebald, W. and Hoppe, J. (1981) Curr. Top. Bioenerg. JL2, 1-64. 147. Sierra, M.F. and Tzagoloff, A. (1973) Proc. Natl. Acad. Sci., U.S.A. 70, 3155-3159. 148. Blondin, G.A. (1979) Biochem. Biophys. Res. Commun. 87_, 1087-1094. 149. - Fillingame, R.H. (1976) J. Biol. Chem. 251, 6630-6637. 150. Altendorf, K., Hammel, U., Deckers, G., K i l t z , H.-H. and Schmid, R. (1979) i n Function and Molecular Aspects of Biomembrane Transport (Quagliariello, E., Ed.) pp. 53-61, Elsevier/North-Holland, New York, N.Y. v 151. Nelson, N., Eytan, E., Notsani, B.-E., Sigrist, H., Sigrist-Nelson, K. and Gitler, C. (1977) Proc. Natl. Acad. Sci., U.S.A. 74, 2375-2378. 152. Celis, H. (1980) Biochem. Biophys. Res. Commun. 92., 26-31. 153. Criddle, R.S., Packer, L. and Shieh, P. (1977) Proc. Natl. Acad. Sci., U.S.A. _74, 4306-4310. 154. Criddle, R.S., Johnston, R., Packer, L., Shieh, P. and Konishi, T. (1979) in Cation Flux Across Biomembranes (Mukohata, Y. and Packer, L., eds.) pp. 399-407,., Academic Press, Inc., New York, N.Y. 155. Foster, D.L. and Fillingame, R.H. (1982) J. Biol. Chem. 257, 2009-2015. 156. Kagawa, Y., Sone, Ni, Yoshida, M., Hirata, H. and Okamoto, Hv (1976) J. Biochem. 80, 141-151. 251. 157. Turner, G., Imam, G. and Kuntzel, H. (1979) Eur. J. Biochem. 97, 565-571. ,158. Schmid, R., K i l t z , H.H., Schneider, E. and Altendorf, K. (1981) FEBS Lett. 125, 97-100. 159. Hoppe, J., Schairer, H.U. and Sebald, W. (1980) FEBS Lett. 109, 107-111. 160. Wachter, E., Schmid, R., Deckers, G. and Altendorf, K. (1980) FEBS Lett. 113, 265-270. 161. Hoppe, J., Schairer, H.U., Friedl, P. and Sebald, W. (1982) FEBS Lett. 145, 21-24. 162. Hoppe, J., Schairer, H.U. and Sebald, W. (1980) Eur. J. Biochem. 112, 17-24. 163. Engelman, D.M. and Zaccai, G. (1980) Proc. Natl. Acad. Sci., U.S.A. 77_, 5894-5898. 164. Bragg, P.D., Davies, P.L. and Hou, C. (1972) Biochem. Biophys. Res. Commun. 47_, 1248-1255. 165. Yoshida, M., Sone, N., Hirata, H. and Kagawa, Y. (1975) J. Biol. Chem. 250, 7910-7916. 166. Singh, A.P. and Bragg, P.D. (1976) Biochim. Biophys. Acta 423, 450-461. 167. Hirata, H., Altendorf, K. and Harold, F.M. (1974) J. Biol. Chem. 249, 2939-2945. 168. Soper, J.W. and Pedersen, P.L. (1979) in Methods in Enzymology (Fleischer, S. and Packer, L., eds.) Vol 55, pp. 328-333, Academic Press, Inc., New York, N.Y. 169. Rouser, G., Kritchevsky, G., Yamamoto, A., Simon, G., G a l l i , C. and Bauman, A.J. (1969) i n Methods in Enzymology (Lowenstein, J.M., ed.) Vol 1_4, pp. 272-317, Academic Press, Inc., New York, N.Y. 170. Penefsky, H.S. (1974) J. Biol. Chem. 249, 3579-3585. 171. Laemmli, U.K. (1970) Nature (London) 227, 680-685. 172. Ames, G.F. (1974) J. Biol. Chem. 249, 634-644. 173. Cox, G.B., Downie, J.A., Fayle, D.R.H., Gibson, F. and Radik, J. (1978) J. Bacteriol. 133, 287-292. 174. Weber, K. and Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412. 252. 175. Merrll, C.R., Switzer, R.C. and Van Keuren, M.L. (1979) Proc. Natl. Acad. Sci., U.S.A. 76., 4335-4339. 176. O'Farrell, P;H. (1975) J. Bio l . Chem. 250, 4007-4021. 177. Nowotny, A. (1979) Basic Exercises in Immunochemistry, pp. 1-3, Springer-Verlag, New York, N.Y. 178. Bjerrum, O.J. and Lundhal, P. (1974) Biochim. Biophys. Acta 342, 69-80. 179. Mayer, R.J. and Walker, J.H. (1980) Immunochemical Methods in Biological Sciences: Enzymes and Proteins, pp. 131-132, Academic Press, Inc., New York, N.Y. 180. Fairbanks, G., Steck, T.L. and Wallach, D.F.H. (1971) Biochemistry 10, 2606-2617. 181. Cuatrecasas, P.J. and Anfinsen, C.B. (1971) i n Methods i n Enzymology (Jakoby, W.B., ed.) Vol. 22, pp. 345-378, Acadmic Press, Inc., New York, N.Y. 182. Moore, S. and Stein, W.H. (1963) i n Methods in Enzymology (Colowick, S.P. and Kaplan, N.O., eds.) Vol. 6, pp. 819-831, Academic Press, Inc., New York, N.Y. 183. Sebald, W., Machleidt, W. and Wachter, E. (1980) Proc. Natl. Acad. Sci., U.S.A. 77, 785-789. 184. Hirs, C.H.W. (1967) i n Methods i n Enzymology (Hirs, C.H.W., ed.) Vol 11, pp. 59-62, Academic Pres, Inc., New York, N.Y. 185. Guide to the Care and Use of Experimental Animals (1980) (Canadian Council on Animal Care, ed.) Vol. 1_, Ottawa, Ont. 186. Herbert, W.J. (19.78) i n Handbook of Experimental Immunology, (Weir, H., ed.), 3rd ed. , Vol. _3_, Appen. 4.16-4.17, Blackwell Scientific Publications, Oxford, Great Britain. 187. Garvey, J.S., Cremer, N.E. and Sussdorf, D.H. (1977) Methods i n Immunology, 3rd ed., pp. 36-38, W.A. Benjamin, Inc., Reading, M.A. 188. Weeke, B. (1973) Scand. J. Immunol. Vol. 2_, Suppl. 1, 15-35. 189. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 190. Bradford, M.M. (1976) Anal. Biochem. 72., 248-254. 191. Singh, A.P. and Bragg, P.D. (1976) Eur. J. Biochem. 67, 177-186. 192. Rottenberg, H. and Lee, CP. (1975) Biochemistry 14, 2675-2680. 253. 193. Pudek, M.R. (1976) Ph.D. Thesis, University of British Columbia. 194. Beers, R.F. and Sizer, I.W. (1952) J. Biol. Chem. 195, 133-150. 195. MacKenzie, D. and Molday, R.S. (1982) J. Biol. Chem. 257, 7100-7105. 196. Roisin, M-P. and Kepes, A. (1973) Biochim. Biophys. Acta 305, 249-259. 197. Carreira, J. and Munoz, E. (1975) Mol. Cell . Biochem. 9_, 85-95. 198. Bragg, P.D. and Hou, C. (1976) Arch. Biochem. Biophys. 174, 553-561. 199. Ritz-Gold, C.J. and Brodie, A.F. (1979) Biochim. Biophys. Acta 547, 18-26. 200. Huberman, M. and Salton, M.R.J. (1979) Biochim. Biophys. Acta 547, 230-240. - 201. Kita, K., Yamato, I. and Anraku, Y. (1978) J. Biol. Chem. 253, 8910-8915^. 202. Yon, R.J. (1978) Int. J. Biochem. j), 373-379. 203. Dickie, P. and Weiner, J.H. (1979) Can. J. Biochem. 57, 813-821. 204. Fillingame, R.H., Mosher, M.E., Negrin, R.S. and Peters, L.K. (1983) J. Biol. Chem. 258, 604-609. 205. Altendorf, K., Harold, F.M., and Simoni, R.D. (1974) J." Biol. Chem. 249, 4587-4593. 206. Kagawa, Y. and Racker, E. (1971) J. Biol. Chem. 246, 5477-5487. 207. Racker, E. (1973) Biochem. Biophys. Res. Commun. 55_, 224-230. 208. Ohta, S., Tsuboi, M., Yoshida, M., and Kagawa, Y. (1980) Biochemistry 19_, 2160-2165. 209. Dunn, S.D. (1980) J. Biol. Chem. 255, 11857-11860. 210. Wickner, W. (1976) J. Bacteriol. 127, 16,2-167. 211. Adler, L.W. and Rosen, B.P. (1977) J. Bacteriol. 129, 959-966. 212. Owen, P. and Kaback, H.R. (1979) Biochemistry 18, 1422-1426. 213. Means, G.E. and Feeney, R.E. (1971) Chemical Modification of Proteins, pp. 195-197, Holden-Day, San Francisco, C.A. 214. Bragg, P.D. and Hou, C. (1978) Can. J. Biochem. _56, 559-564. 215. Takahashi, K. (1968) J. Biol. Chem. 243, 6171-6179. 254. 216. Tanford, C. and Reynolds, J.A. (1976) Biochim. Biophys. Acta 457, 133-170. 217. von Meyenburg, K., Jorgensen, B.B., Nielsen, J. and Hansen, F.G. (1982) Mol. Gen. Genet. 188, 240-248. 218. Srere, P.A. (1982) Trends Biochem. Sci. _7, 375-378. 219. Senior, A.E., Fayle, D.R.H., Downie, J.A., Gibson, F. and Cox, G.B. (1979) Biochem. J. 180, 111-118. 220. Humbert, R., Brusilow, W..A., Gunsalus, R.P., Klionsky, D.J. and Simoni, R.D. (1983) J. Bacteriol. 153, 416-422. 221. Noumi, T. and Kanazawa,, H. (1983) Biochem. Biophys. Res. Commun. I l l , 143-149. 222. Friedl, P., Bienhaus, G., Hoppe, J. and Schairer, U. (1981) Proc. Natl. Acad. Sci., U.S.A. 78, 6643-6646. 223. Schneider, E., Schmid, R., Deckers, G., Steffens, K., K i l t z , H.H. and Altendorf, K. (1981) i n Vectorial Reactions in Electron and Ion Transport in Mitochondria and Bacteria (Palmieri, F., ed.) pp. 231-234, Elsevier/North-Holland, New York, N.Y. 224. van der Plas, J., Hellingwerf, K.J., Seijen, H.G., Guest, J.R.,' Weiner, J.H. and Konings, W.N. (1983) J. Bacteriol. 153, 1027-1037. 225. Suss, K.-H. and Manteuffel, R. (1983) FEBS Lett. 153, 134-140. 226. Walker, J.E., Eberle, A., Gay, N.J., Runswick, M.J. and Saraste, M. (1982) Biochem. Soc. Trans. jLO, 203-206. 227. Andreo, C.S., Patrie, W.J. and McCarty, R.E. (1982) J. Biol. Chem. 257, 9968-9975. 228. Walker, J.E., Saraste, M. and Gay, N.J. (1982) Nature 298, 867-869. 229. Hoppe, J. et a l . Eur. Bioenerget. Conf. Rep. (1982) 2_, 85-86. 230. Pedersen, P.L., Hullihen, J. and Wehrle, J.P. (1981) J. Biol. Chem. 256, 1362-1369. 231. Kozlov, I.A. and Skulachev, V.P. (1977) Biochim. Biophys. Acta 463, >29-89. 232. Friedl, P., Hoppe, J., Gunsalus, R.P., Michelsen, 0., von Meyenburg, K. and Schairer, H.U. (1983) EMBO. J. 2, 99-103. 233. Hoppe, J., Friedl, P., Schairer, H.U., Sebald, W., von Meyenburg, K. and Jorgensen, B.B. (1983) EMBO. J. 2_, 105-110.

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 10 24
Japan 4 0
United States 3 0
City Views Downloads
Beijing 7 0
Tokyo 4 0
Lanzhou 3 0
Mountain View 2 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items