N-terminal deletion of the c subunit affects the stabilization and activity of chloroplast ATP synthase Zhang-Lin Ni, Hui Dong and Jia-Mian Wei Shanghai Institute of Plant Physiology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China ATP synthase occurs ubiquitously on energy-transduc- ing membranes such as chloroplast thylakoid mem- branes, mitochondrial inner membranes, and bacterial plasma membranes. This enzyme catalyzes ATP syn- thesis by a proton motive force across the membrane formed by the respiratory chain or photosynthetic elec- tron transport (ATPase in Escherichia coli recently reviewed in [1,2], and ATP synthase in chloroplasts in [3,4]). The general structural features of the enzyme are highly conserved among different organisms. The enzyme in chloroplasts consists of two parts: CF 0 and CF 1 .CF 0 , a membrane-spanning complex, conducts proton flux through the thylakoid membrane and pro- vides affinity sites for the CF 1 complex. CF 1 , extrinsic to the membrane, contains the nucleotide-binding and catalytic sites, and can hydrolyze ATP at high rates after appropriate treatment [4,5]. The CF 1 complex consists of five types of subunit with the stoichiometry a 3 b 3 cde. The first high-resolution X-ray structure of ATP synthase was of bovine mitochondrial F 1 in 1994 [6]. The structure is essentially unchanged in X-ray studies of bovine F 1 inhibited by N,N¢-dicyclohexylcarbodi- imide (Fig. 1) [7]. Nucleotide bound to all three cata- lytic sites in the aluminum fluoride-inhibited form of bovine F 1 [8]. The X-ray structures show that the a and b subunits alternate with each other to form a hexamer surrounding a central cavity, where a coiled- coil structure formed by the N-terminal and C-ter- minal helices of the c subunit penetrates. The three catalytic sites of F 1 are located on the b subunits, where the sites interface subunit a in three different conformational states. The importance of the c subunit in the catalytic cycle has been demonstrated previ- ously, showing that it is probably related to the sequential conformational changes in the ab pairs in addition to being responsible for the generation of a high-affinity nucleotide-binding site on the b subunits Keywords ATP synthase; chloroplast; glutathione S-transferase pull-down assay; abc assembly; c subunit Correspondence J M. Wei, Shanghai Institute of Plant Physiology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 300 Fenglin Road, Shanghai 200032, China Fax: +86 21 54924015 Tel: +86 21 54924230 E-mail: wjm@iris.sipp.ac.cn (Received 25 November 2004, revised 22 December 2004, accepted 17 January 2005) doi:10.1111/j.1742-4658.2005.04570.x Five truncation mutants of chloroplast ATP synthase c subunit from spin- ach (Spinacia oleracea) lacking 8, 12, 16, 20 or 60 N-terminal amino acids were generated by PCR by a mutagenesis method. The recombinant c genes were overexpressed in Escherichia coli and assembled with ab sub- units into a native complex. The wild-type (WT) abc assembly i.e. abcWT exhibited high Mg 2+ -dependent and Ca 2+ -dependent ATP hydrolytic activity. Deletions of eight residues of the c subunit N-terminus caused a decrease in rates of ATP hydrolysis to 30% of that of the abWT assembly. Furthermore, only  6% of ATP hydrolytic activity was retained with the sequential deletions of c subunit up to 20 residues compared with the activ- ity of the abWT assembly. The inhibitory effect of the e subunit on ATP hydrolysis of these abc assemblies varied to a large extent. These observa- tions indicate that the N-terminus of the c subunit is very important, together with other regions of the c subunit, in stabilization of the enzyme complex or during cooperative catalysis. In addition, the in vitro binding assay showed that the c subunit N-terminus is not a crucial region in bind- ing of the e subunit. Abbreviations CF 0 , the hydrophobic portion of chloroplast ATP synthase; CF 1 , coupling factor one; GST, glutathione S-transferase; WT, wild-type. FEBS Journal 272 (2005) 1379–1385 ª 2005 FEBS 1379 by its rotation within the a 3 b 3 core [1–4,6–8]. In F 1 , rotation of the c subunit coupled with ATP hydrolysis was confirmed by its direct observation in the move- ment of a fluorescence-labeled actin filament, which was attached to the c subunit of the thermophilic bac- terial F 1 subcomplex, a 3 b 3 c [9], e subunit F 1 [10,11] and CF 1 [12]. The catalytic core of the enzyme is a 3 b 3 c, despite ab exhibiting lower rates of ATP hydrolysis [13]. It is generally accepted that ATP synthase generates coup- ling between cooperative catalysis and proton translo- cation during hydrolysis ⁄ synthesis processes. However, the precise catalytic mechanism of F 1 -ATPase is still unknown [1,14]. With respect to CF 1 , it is also pro- posed that isolated CF 1 operates through a full 360° rotation, like other F 1 -ATPases [12] . CF 1 is unique in that thiol modulation, the structural basis of which is an insert of about 20 amino acids including a regula- tory disulfide bond, is reversibly oxidized and reduced. The N-terminus and C-terminus of c subunits from dif- ferent organisms are highly conserved [15]. Deletion of the 20 amino acids in the C-terminus of the c subunit resulted in an active chloroplast enzyme [16]. Crystal structures (Fig. 1) reveal that the N-terminal domain of the c subunit makes contact with the b E sub- unit C-terminal domain containing the conserved DELSEED motif, which is thought to be important for energy-coupling rotation of the c subunit by steric interaction. This indicates the possible importance in the c N-terminal domain during catalytic cooperativity [4,7,8]. cSer8 substitution with a Cys residue resulted in it being cross-linked with a different b region in the presence of Mg 2+ -ADP or Mg 2+ -ATP [17]. cMet23 substitution caused ATPase uncoupling [18], which was suppressed by amino-acid replacements between 269 and 280 in the C-terminal domain [19]. It has been demonstrated by fluorescence mapping [20] and cross- linking [21] that the e subunit is in close proximity to the c subunit. The e subunit interacts directly with the c subunit [22–24], but does so with a higher affinity when the c subunit is assembled with the a 3 b 3 core [25, 26]. The hybrid enzyme from the a,b subunits of a thermophilic bacterium and the mutant CF 1 c subunit (D194–230) was insensitive to added e subunit [25]. Studies by Gao et al. [26], who developed an in vitro reconstitution system by assembling the ab complex with an isolated c subunit, showed that this complex was able to obtain the reconstituted core enzyme com- plex as effectively as the native a 3 b 3 c. Recently, a hybrid F 1 -ATPase from Rhodospirillum rubrum or chloroplast subunits was used to study the mechanism of photosyn- thetic F 1 -ATPase [27,28]. In the present study, we exam- ined the importance of the CF 1 N-terminus of the c subunit during hydrolytic turnover using this reconsti- tution system and the binding of e to c through a gluta- thione S-transferase (GST) pull-down assay. To do this, we selectively deleted 8, 12, 16, 20 or 60 residues from the N-terminus of the c subunit. The reconstituted abc assemblies were tested for ATP hydrolytic activity. The results show that the c subunit’s N-terminus is very important for stabilization of the enzyme complex. Results Overexpression and assembly of the c truncated mutants All the plasmids listed in Table 1 were transformed into the expression strain E. coli BL21 (DE3) ⁄ pLysS. The spinach chloroplast atpC gene constructed in the pET11b expression vector had a high expression level in E. coli. More than 100 mg recombinant c protein was obtained per litre of culture medium. Overexpres- sion of the cloned polypeptides in E. coli resulted in the accumulation of insoluble inclusion bodies. The inclusion bodies were solubilized in 4 m urea and recovered by slow dialysis as described previously [16]. Like the wild-type and native c subunits [16], all the c mutants tended to aggregate during dialysis when pro- tein concentration was high. On SDS ⁄ PAGE, wild- type c protein and the truncated polypeptides migrated for distances consistent with the extent of truncation Fig. 1. Schematic representation of the bovine heart mitochondria F 1 produced from pdb file 1e79 using DEEPVIEW [37]. Subunit a is on the left, and subunit b is on the right. The N-terminal 20 residues of subunit c are depicted in gray. N-Terminal deletion of the c subunit Z L. Ni et al. 1380 FEBS Journal 272 (2005) 1379–1385 ª 2005 FEBS (Fig. 2A). Each of the c constructs reacted with c anti- serum on immunoblots (data not shown). The trun- cated polypeptides were designated cDN8 to cDN60 according to the number of amino-acid residues dele- ted from the c subunit N-terminus. The N-terminus amino-acid sequence of c subunit from chloroplasts is shown in Table 1. ATP hydrolytic activity of the mutant assemblies We tested ATP hydrolytic activity of the reconstituted assemblies. Incubation of the native c protein from chloroplasts with isolated ab subunits (Fig. 2B) resul- ted in their assembly into a stable, highly active abc complex under optimal conditions. The cloned c poly- peptide was identical with the native c subunit in its ability to form a fully active core enzyme complex [16,26]. Mg 2+ -ATPase activity was measured in the presence of sodium sulfite, a strong stimulator of ATP hydrolysis [16]. The relative rates of ATP hydrolysis of these reconstituted assemblies in the presence of either Ca 2+ or Mg 2+ as the bivalent cation substrate were compared (Fig. 3). The wild-type assembly exhibited the maximum activity [14.4 lmol P i Æ(mg pro- tein) )1 Æmin )1 ), consistent with previous results showing that assembling the cloned c polypeptide with the iso- lated ab subunits resulted in a fully active core enzyme complex [26], although there were some differences among the hydrolytic rates. Deletion of eight residues from the c subunit N-terminus impaired the ATP hydrolytic ability, despite differences between Mg 2+ - ATPase and Ca 2+ -ATPase. The deletion of 12 residues resulted in a greater decrease in Ca 2+ -ATPase activity. About 6% of ATP hydrolytic activity retained on dele- tion of 20 residues. When 60 residues were deleted, the rate of ATP hydrolysis of the reconstituted assembly was similar to the ab complex containing no c subunit. Interaction of subunits c and e in vitro GST pull-down assays were used to detect c–e interac- tion. The full-length cDNA encoding the e subunit was fused to the C-terminus of the GST gene in the expres- sion plasmid, and the GST-fusion protein was over- expressed in E. coli. As shown in Fig. 4, GST alone did not bind to the wild-type c subunit ( cWT); in con- trast, GST–e was able to bind directly to all the c con- structs. It was also able to bind to the cDN8, cDN12 Table 1. Amino-acid and primer sequences of truncated mutants. Listed below are the amino-acid sequences and PCR primers of truncation mutants (D)ofthec subunit of spinach chloroplast ATP synthase. The numbers in the plasmids indicate the numbers of residues deleted fol- lowed by N designating N-terminus truncation, respectively. The truncations begin with deletion of eight residues with successive deletion of four residues up to 20 residues from the N-terminus, and N-terminal mutants with deletion of 60 residues (cDN60). Plasmid Amino-acid sequences Forward primers (5¢-3¢) Reverse primers (5¢-3¢) pET11-cWT ANLRELRDRIGSVKNTQKITEAMKLVAAAK-31 TTTGTCATATGGCAAACCTCCGTGAGC pET11-cDN8 …8RIGSVKNTQKITEAMKLVAAAK-31 GGCCATATGCGGATCGGATCAGTCAAA ATTCCGGAC pET11-cDN12 …12VKNTQKITEAMKLVAAAK-31 TCGGCCATATGGTCAAAAACACGCAGAAG ACGGATCCA pET11-cDN16 …16QKITEAMKLVAAAK-31 TTGGCCATATGCAGAAGATCACCGAAGCA ATTAATCTC pET11-cDN20 … 20 EAMKLVAAAK-31 TCCGGCATATGGAAGCAATGAAGCTCGTC pET11-cDN60 …60 TE-62 TCGCGCATATGACTGAGGATGTTGATGTT AB Fig. 2. Gel electrophoresis profiles of prepa- rations of inclusion bodies and the purified ab complex. Preparations were analyzed by SDS ⁄ PAGE on 15% polyacrylamide gels, and the proteins were stained with Coomas- sie Brilliant Blue R250. Each lane contained 6 lg protein. (A) Lane 1, CF 1 ; lanes 2–7, inclusion body preparations of cWT, DcN8, DcN12, DcN16, DcN20 and DcN60, respect- ively. (B) Lane 1, CF 1 (–de); lane 2, isolated ab complex. Z L. Ni et al. N-Terminal deletion of the c subunit FEBS Journal 272 (2005) 1379–1385 ª 2005 FEBS 1381 and cDN16 molecules with similar affinity to cWT. Deletion of 20 residues resulted in slight impairment of the binding activity. Inhibitory effects of the e subunit Given that the N-terminal deletions of the c subunit barely inhibited the interaction between c and e, fur- ther studies were performed to confirm the responses of the mutant assemblies to the inhibitory e subunit. The inhibitory responses of the Ca 2+ -ATPase activity of the different mutant assemblies were examined after the addition of the e subunit (Fig. 5). The abcWT assembly that exhibited the highest Ca 2+ -ATPase activity was inhibited by 84%. The inhibitory effects of the e subunits on hydrolytic rates in the mutant (cDN8, cDN12, cDN16 and cDN20) assemblies varied greatly, ranging from 35% to 63% reduction in hydro- lysis. Discussion The studies presented here focused on the importance of the N-terminus of the c subunit during ATP hydro- lysis in addition to examining the role of binding and inhibition of the e subunit. A schematic representation of bovine heart mitochondria F 1 is shown in Fig. 1. The conserved N-terminal region of the c subunit forms an antiparallel left-handed coiled coil with the C-terminal part, which penetrates into a cavity formed by the a 3 b 3 hexamer . The c subunit’s N-terminus makes contact with the C-terminal domain of the b E subunit, which contains the conserved DELSEED motif [7,8]. It is generally accepted that the c subunit confers the asymmetric properties of the catalytic sites by interacting with the a and b subunits, resulting in catalytic cooperativity. The permanent asymmetry of Fig. 3. Relative ATPase activity of reconstituted abc assemblies. Mg 2+ -ATPase (black columns) and Ca 2+ -ATPase (gray columns) activities were analyzed as described in Experimental procedures. The ATPase activities [lmol P i Æ(mg protein) )1 Æmin )1 ]ofthecWT assembly (100%) were: Mg 2+ -ATPase, 13.8; Ca 2+ -ATPase, 14.4. The activities of both the ab complex and the abDcN60 assemble were about 0.3. Fig. 4. Interactions of wild-type c and the engineered c with e in vitro. GST pull-down assays were performed as described in Experimental procedures. GST or GST–e fusion proteins were bound to glutathione–Sepharose 4B beads. 6 lg of each of the c construct proteins were incubated with 20 lL of the beads bound to GST–e. After extensive washing, proteins eluted from the beads were subjected to Western-blot analysis with c antiserum. GST alone was used as a negative control by incubating 20 lL of the beads bound to GST with 6 lg of the wild-type c recombinant. Fig. 5. Inhibition of the Ca 2+ -ATPase of the reconstituted abc assemblies by addition of e subunit. Ca 2+ -ATPase activities were analyzed as described in Experimental procedures. Assemblies were incubated with the e subunit at the molar ratio 10e:1ab.The inset shows the Ca 2+ -ATPase activities [lmol P i Æ(mg pro- tein) )1 Æmin )1 ] in the absence of e subunit, which are set as 100%. N-Terminal deletion of the c subunit Z L. Ni et al. 1382 FEBS Journal 272 (2005) 1379–1385 ª 2005 FEBS isolated CF 1 was found in labeling experiments with Lucifer Yellow [29], which possibly indicated that the N-terminal part of subunit c remains in contact with the a E and b E subunits during the complete catalytic turnover without a full 360° rotation [4]. However, direct observation of the movement of subunit c in iso- lated CF 1 was also determined recently to resemble that of E. coli F 1 or thermophilic bacterial F 1 subcom- plex, thereby favoring multisite catalysis with a full 360° rotation. The data presented here reveal that deletion of 60 residues eliminated almost all hydrolytic activity of the reconstituted assembly; moreover, the removal of eight residues abolished most of the activity. When up to 20 residues were deleted, very low ATP hydro- lytic activity was retained (Fig. 3). There are two possible explanations for the above results. (a) The deletions impaired the stability of the reconstituted assemblies and the efficient assembly of the ab com- plex with recombinant c constructs; or (b) the struc- tural change in the truncated c construct altered the asymmetry conformation of a 3 b 3 c, thereby affecting transmission of conformational signals between cata- lytic sites, which resulted in impaired normal catalytic cooperativity. These observations indicate that the N-terminal part of subunit c is indispensable and functions with other regions of c during stabilization of the abc complex and rotational catalysis. Consis- tent with previous studies, we also found that Ca 2+ - dependent and Mg 2+ -dependent ATP hydrolytic activities were different, indicating different catalytic mechanisms [28]. It is well established that the e subunit, a regulatory protein of ATPase, binds to c directly and rotates with the c and c (homologous to CF 0 -III in chloroplasts) subunits as a part of a rotor (cec 10 ) [1,2]. The inter- action of c and e increases when c penetrates into a a 3 b 3 hexamer [26]. The e subunit binds to CF 1 with an apparent dissociation constant of < 10 )10 m [20]. In the e subunit, the sites of c interaction with e were mapped to between R49 and R70, and the C-terminal part beyond K199 [24]. The regulatory c regions of CF 1 seem to be very important for e subunit binding [30]. Our results also show that the e subunit stably binds to c, which is consistent with earlier studies [23–25]. Deletion of 20 residues from the N-terminal region did not markedly decrease the binding affinity between the c and e subunits in vitro (Fig. 4). The engineered c subunit bound to the e subunit with almost identical affinity, although the inhibitory effects of e subunit varied with the number of resi- dues removed. The high binding affinity of a 3 b 3 c for the e subunit is essential for inhibition of ATPase catalysis, which is much higher than that of binding e to individual c [26]. The truncation seemed not to change the binding affinity between c and e (Fig. 5), but the altered asymmetrical conformation of a 3 b 3 c resulting from deletions in c is enough to decrease the binding of e to a 3 b 3 c. Meanwhile, impaired cata- lytic cooperativity may not be efficiently inhibited by the e subunit. Both of the above may partially explain the varied inhibitory effect of e on ATP hydrolysis. Taken together, we have shown that the N-terminal region of the c subunit is important for stabilization of a 3 b 3 c or cooperative catalysis of isolated CF 1. The N-terminal region of subunit c in CF 1 is not crucial for binding of subunit e in vitro. Further studies are needed to determine candidate residues participating in transmission of conformation signals and efficient energy coupling of the c subunit N-terminus. Experimental procedures Materials Restriction endonucleases, T4 DNA ligase, Klenow frag- ment, and Pfu and Taq DNA polymerase were purchased from Takara (Dalian, China) and Promega (Shanghai, China). Sephadex G-50 was purchased from Pharmacia (Uppsala, Sweden). DEAE-cellulose was obtained from Whatmann (Uppsala, Sweden) and hydroxyapatite from Bio-Rad (Hercules, CA, USA). Other reagents were all standard AR grade. Generation of c truncation mutants and GST-fusion protein Plasmid pJLA503-pchlc, a gift from S. Engelbrecht (Uni- versity of Osnabruck, Germany), contains the atpC genes encoding the ATP synthase c subunit [31]. The wild-type c fragment was PCR amplified from pJLA503-pchlc, and five truncation mutants were PCR generated with the mutagen- esis primers (Table 1). The PCR products were digested with NdeI and BamHI, and subsequently subcloned into the pET11b expression vector. The resulting plasmids were confirmed by DNA sequencing. For construction of the GST–e fusion protein, full-length cDNA encoding the e subunit was amplified by PCR using pJLA-eWT [23] as a template with the following primers: 5 ¢-GACGGATCC CCATGACCTTAAATCTTTGT-3¢ as the 5¢ primer and 5¢-ATAGTCGACCTGGTTACGAAGAAATCG-3¢ as the 3¢ primer. The PCR products were cleaved with BamHI and EcoRI, and cloned into plasmid pGEX-5X-1 (Amer- sham-Pharmacia Biotech, Shanghai, China). The resulting plasmid pGEX-5X-e was confirmed to be inframe with the GST cassette by DNA sequencing. Z L. Ni et al. N-Terminal deletion of the c subunit FEBS Journal 272 (2005) 1379–1385 ª 2005 FEBS 1383 Solubilization and folding of overexpressed c mutants and GST-fusion protein The resulting pET11b plasmids containing the atpC gene were transformed into the expression strain E. coli BL21(DE3) ⁄ pLysS. The resulting E. coli cells were grown at 37 °C in Luria–Bertani medium containing l-ampicillin. Cells were induced with 0.4 mm isopropyl thio-b-d-galacto- side in mid-exponential phase, incubated for 7 h, and har- vested as described previously [32]. Solubilization and folding of the insoluble c polypeptide were performed according as described previously [16]. Overexpression and collection of GST or GST–e fusion protein in E. coli were carried out as previously described [23]. Preparation and reconstitution of an ab complex and c mutants An ab complex was isolated from CF 1 (– de) as described previously [26]. Reconstitution of the complex and c mutants was carried out as described previously [26]. The incubated mixture was assayed directly for ATP hydrolytic activity. In vitro binding assays of e with c subunit and immunoblotting analysis Equal amounts of GST or GST–e fusion proteins were bound to glutathione–Sepharose 4B beads in binding buffer [50 mm Tris ⁄ HCl, 100 mm NaCl, 1 mm EDTA, 1% (v ⁄ v) Triton X-100, 1 mm phenylmethanesulfonyl fluoride and 10% (v ⁄ v) glycerol, pH 8.0] at 4 °C for 2 h. The bound glutathione–Sepharose 4B beads were washed three times with binding buffer to remove unbound fusion proteins. These beads were incubated with equal amounts of the c constructs for 4 h at 4 °C in binding buffer and washed four times with binding buffer to remove unbound proteins. Subsequently, the beads were suspended in 2· SDS loading buffer and boiled for 3 min. Proteins released from the beads were analyzed by SDS ⁄ PAGE (15% polyacrylamide gel) [23], transferred to nitrocellulose membrane, and detec- ted by western immunoblot analysis using an ECL Western Blotting Detection System (Amersham) and c antiserum. CF 1 and CF 1 (–de) preparation and measurement of ATPase activity CF 1 and CF 1 (–d) were prepared from fresh market spinach as described previously [33,34]. Before use, the proteins were desalted on Sephadex G-50 centrifuge columns [35]. ATPase activities were determined by measuring phosphate release for 5–10 min at 37 °C. The assay was performed in 1 mL volumes of assay mixture containing 50 mm Tricine ⁄ NaOH, pH 8.0, and 5 mm ATP. Ca 2+ -ATPase was carried out in the presence of 5 mm CaCl 2 , and Mg 2+ - ATPase in the presence of 2 mm MgCl 2 and 20 mm Na 2 SO 3. The reaction was stopped by adding 200 lL 20% trichloroacetic acid. c Antiserum was raised by subcuta- neous injections into rabbits. Inclusion bodies containing recombinant c polypeptide were subjected to SDS ⁄ PAGE. Proteins were recovered from Coomassie Brilliant Blue R250-stained gel bands and used for the immunization of rabbits. Protein concentration was measured by the method of Bradford [36]. Acknowledgements This work was supported by the National Natural Sci- ence Foundation of China (30170078) and State Key Basic Research and Development Plan (G1998010100). References 1 Senior AE, Nadanaciva S & Weber J (2002) The mole- cular mechanism of ATP synthesis by F 1 F 0 -ATP synthase. Biochim Biophys Acta 1553, 188–211. 2 Weber J & Senior AE (2003) ATP synthesis driven by proton transport in F 1 F 0 -ATP synthase. FEBS Lett 545, 61–70. 3 McCarty RE, Evron Y & Johnson EA (2000) The chloro- plast ATP synthase: a rotary enzyme? Annu Rev Plant Physiol Plant Mol Biol 51, 83–109. 4 Richter ML, Hein R & Huchzermeyer B (2000) Impor- tant subunit interactions in the chloroplast ATP synthase. Biochim Biophys Acta 1458, 326–342. 5 Richter ML, Patrie WJ & McCarty RE (1984) Prepara- tion of the e subunit and e subunit-deficient chloroplast coupling factor 1 in reconstitutively active forms. J Biol Chem 259, 7371–7373. 6 Abrahams JP, Leslie AGW, Lutter R & Walker J (1994) Structure at 2.8 A ˚ resolution of F 1 -ATPase from bovine heart mitochondria. Nature 370, 621–628. 7 Gibbons C, Montgomery MG, Leslie AG & Walker JE (2000) The structure of the central stalk in bovine F 1 -ATPase at 2.4 A ˚ resolution. Nat Struct Biol 7 , 1055– 1061. 8 Menz RI, Walker JE & Leslie AG (2001) Structure of bovine mitochondrial F 1 -ATPase with nucleotide bound to all three catalytic sites: implications for the mechan- ism of rotary catalysis. Cell 106, 331–341. 9 Noji H, Yasuda R, Yoshida M & Kinosita K Jr (1997) Direct observation of the rotation of F 1 -ATPase. Nature 386, 299–302. 10 Noji H, Hasler K, Junge W, Kinosita K Jr, Yoshida M & Engelbrecht S (1999) Rotation of Escherichia coli F 1 - ATPase. Biochem Biophys Res Commun 260, 597–599. 11 Omote H, Sambonmatsu N, Saito K, Sambongi Y, Iwamoto-Kihara A, Yanagida T, Wada Y & Futai M N-Terminal deletion of the c subunit Z L. Ni et al. 1384 FEBS Journal 272 (2005) 1379–1385 ª 2005 FEBS (1999) The c subunit rotation and torque generation in F 1 -ATPase from wild-type or uncoupled mutant Escheri- chia coli. Proc Natl Acad Sci USA 96, 7780–7784. 12 Hisabori T, Kondoh A & Yoshida M (1999) The c sub- unit in chloroplast F 1 -ATPase can rotate in a unidirec- tional and counter-clockwise manner. FEBS Lett 463 , 35–38. 13 Matsui T & Yoshida M (1995) Expression of the wild- type and the Cys- ⁄ Trp-less a 3 b 3 c complex of thermophi- lic F 1 -ATPase in Escherichia coli. Biochim Biophys Acta 1231, 139–146. 14 Ren H & Allison WS (2000) On what makes the c sub- unit spin during ATP hydrolysis by F 1 . Biochim Biophys Acta 1458, 221–233. 15 Miki J, Maeda M, Mukohata Y & Futai M (1988) The c-subunit of ATP synthase from spinach chloroplasts. Primary structure deduced from the cloned cDNA sequence. FEBS Lett 232, 221–226. 16 Sokolov M, Lu L, Tucker W, Gao F, Gegenheimer PA & Richter ML (1999) The 20 C-terminal amino acid residues of the chloroplast ATP synthase c subunit are not essential for activity. J Biol Chem 274, 13824–13829. 17 Aggeler R, Cai SX, Keana JF, Koike T & Capaldi RA (1993) The c subunit of the Escherichia coli F 1 -ATPase can be cross-linked near the glycine-rich loop region of a b subunit when ADP + Mg 2+ occupies catalytic sites but not when ATP + Mg 2+ is bound. J Biol Chem 268, 20831–20837. 18 Nakamoto RK, Maeda M & Futai M (1993) The c sub- unit of the Escherichia coli ATP synthase. Mutations in the carboxyl-terminal region restore energy coupling to the amino-terminal mutant c Met-23 fi Lys. J Biol Chem 268, 867–872. 19 Nakamoto RK, al-Shawi MK & Futai M (1995) The ATP synthase c subunit. Suppressor mutagenesis reveals three helical regions involved in energy coupling. J Biol Chem 270, 14042–14046. 20 Snyder B & Hammes GG (1985) Structural organization of chloroplast coupling factor. Biochemistry 24, 2324– 2331. 21 Schulenberg B, Wellmer F, Lill H, Junge W & Engelbrecht S (1997) Cross-linking of chloroplast F 0 F 1 -ATPase subunit e to c without effect on activity. e and c are parts of the rotor. Eur J Biochem 249, 134– 141. 22 Dunn SD (1982) The isolated c subunit of Escherichia coli F 1 ATPase binds the e subunit. J Biol Chem 257, 7354–7359. 23 Shi XB, Wei JM & Shen YG (2001) Effects of sequen- tial deletions of residues from the N- or C-terminus on the functions of e subunit of the chloroplast ATP syn- thase. Biochemistry 40, 10825–10831. 24 Dunn SD (1997) e-binding regions of the c subunit of Escherichia coli ATP synthase. Biochim Biophys Acta 1319, 177–184. 25 Tang C & Capaldi RA (1996) Characterization of the interface between a and e subunits of Escherichia coli F 1 -ATPase. J Biol Chem 271, 3018–3024. 26 Gao F, Lipscomb B, Wu I & Richter ML (1995) In vitro assembly of the core catalytic complex of the chloroplast ATP synthase. J Biol Chem 270, 9763–9769. 27 Tucker WC, Du Z, Hein R, Richter ML & Gromet-Elha- nan Z (2000) Hybrid Rhodospirillum rubrum F 0 F 1 ATP synthases containing spinach chloroplast F 1 b or a and b subunits reveal the essential role of the a subunit in ATP synthesis and tentoxin sensitivity. J Biol Chem 275, 906– 912. 28 Du Z, Tucker WC, Richter ML & Gromet-Elhanan Z (2001) Assembled F 1 -(ab) and Hybrid F 1 -a 3 b 3 c-ATPases from Rhodospirillum rubrum a, wild type or mutant b, and chloroplast c subunits. Demonstration of Mg 2+ versus Ca 2+ -induced differences in catalytic site struc- ture and function. J Biol Chem 276, 11517–11523. 29 Nalin CM, Snyder B & McCarty RE (1985) Selective modification of an a subunit of chloroplast coupling factor 1. Biochemistry 24, 2318–2324. 30 Hisabori T, Motohashi K, Kroth P, Strotmann H & Amano T (1998) The formation or the reduction of a disulfide bridge on the c subunit of chloroplast ATP synthase affects the inhibitory effect of the e subunit. J Biol Chem 273, 15901–15905. 31 Lill H, Burkovski A, Altendorf K, Junge W & Engelbrecht S (1993) Complementation of Escherichia coli unc mutant strains by chloroplast and cyanobacter- ial F 1 -ATPase subunits. Biochim Biophys Acta 1144, 278–284. 32 Ni ZL, Wang DF & Wei JM (2002) Substitutions of the conserved Thr42 increased the roles of the e subunit of maize CF 1 as CF 1 inhibitor and proton gate. Photosyn- thetica 40, 517–522. 33 Jagendorf AT (1982) Isolation of chloroplast coupling factor (CF 1 ) and of its subunits. In Methods in Chloro- plast Molecular Biology (Edelman M, Hallick RB & Chua N-H, eds), pp. 881–897. Elsevier Biomedical Press, Amsterdam. 34 Richter ML, Snyder B, McCarty RE & Hammes GG (1985) Binding stoichiometry and structural mapping of the e polypeptide of chloroplast coupling factor 1. Biochemistry 24, 5755–5763. 35 Penefsky HS (1977) Reversible binding of P i by beef heart mitochondrial adenosine triphosphatase. J Biol Chem 252, 2891–2899. 36 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein util- izing the principle of protein-dye binding. Anal Biochem 72, 248–254. 37 Schwede T, Kopp J, Guex N & Peitsch MC (2003) SWISS-MODEL: an automated protein homology- modeling server. Nucleic Acids Res 31, 3381–3385. Z L. Ni et al. N-Terminal deletion of the c subunit FEBS Journal 272 (2005) 1379–1385 ª 2005 FEBS 1385 . TCGGCCATATGGTCAAAAACACGCAGAAG ACGGATCCA pET11-cDN16 …16QKITEAMKLVAAAK-31 TTGGCCATATGCAGAAGATCACCGAAGCA ATTAATCTC pET11-cDN20 … 20 EAMKLVAAAK-31 TCCGGCATATGGAAGCAATGAAGCTCGTC pET11-cDN60. TTTGTCATATGGCAAACCTCCGTGAGC pET11-cDN8 …8RIGSVKNTQKITEAMKLVAAAK-31 GGCCATATGCGGATCGGATCAGTCAAA ATTCCGGAC pET11-cDN12 …12VKNTQKITEAMKLVAAAK-31 TCGGCCATATGGTCAAAAACACGCAGAAG