Amputation of a C-terminal helix of the γ subunit increases ATP-hydrolysis activity of cyanobacterial F1 ATP synthase

BBA - Bioenergetics 1859 (2018) 319–325

Contents lists available at ScienceDirect

BBA - Bioenergetics journal homepage: www.elsevier.com/locate/bbabio

Amputation of a C-terminal helix of the γ subunit increases ATP-hydrolysis activity of cyanobacterial F1 ATP synthase

T

Kumiko Kondoa,b, Yu Takeyamaa, Ei-ichiro Sunamuraa,b, Yuka Madokaa,b, Yuki Fukayaa, ⁎ Atsuko Isua,b, Toru Hisaboria,b, a b

Laboratory for Chemistry and Life Science, Tokyo Institute of Technology, Nagatsuta 4259-R1-8, Midori-Ku, Yokohama 226-8503, Japan Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Agency (JST), Tokyo 102-0075, Japan

A R T I C L E I N F O

A B S T R A C T

Keywords: FoF1 ATP synthase γ subunit ADP inhibition Cyanobacteria

F1 is a soluble part of FoF1-ATP synthase and performs a catalytic process of ATP hydrolysis and synthesis. The γ subunit, which is the rotary shaft of F1 motor, is composed of N-terminal and C-terminal helices domains, and a protruding Rossman-fold domain located between the two major helices parts. The N-terminal and C-terminal helices domains of γ assemble into an antiparallel coiled-coil structure, and are almost embedded into the stator ring composed of α3β3 hexamer of the F1 molecule. Cyanobacterial and chloroplast γ subunits harbor an inserted sequence of 30 or 39 amino acids length within the Rossman-fold domain in comparison with bacterial or mitochondrial γ. To understand the structure–function relationship of the γ subunit, we prepared a mutant F1ATP synthase of a thermophilic cyanobacterium, Thermosynechococcus elongatus BP-1, in which the γ subunit is split into N-terminal α-helix along with the inserted sequence and the remaining C-terminal part. The obtained mutant showed higher ATP-hydrolysis activities than those containing the wild-type γ. Contrary to our expectation, the complexes containing the split γ subunits were mostly devoid of the C-terminal helix. We further investigated the effect of post-assembly cleavage of the γ subunit. We demonstrate that insertion of the nick between two helices of the γ subunit imparts resistance to ADP inhibition, and the C-terminal α-helix is dispensable for ATP-hydrolysis activity and plays a crucial role in the assembly of F1-ATP synthase.

1. Introduction FoF1-ATP synthase (FoF1) synthesizes ATP from ADP and inorganic phosphate by the process of coupling with the proton translocation across the cytoplasmic membranes of bacteria, thylakoid membranes of chloroplasts, and inner membranes of mitochondria [1,2]. This enzyme can hydrolyze ATP into ADP and phosphate, when protons are transported into the opposite direction, and is therefore called FoF1 ATPase. Fo is a proton pumping unit embedded in a membrane and composed of three subunits with the stoichiometry of a1b2c8–17 [3–9]. F1 is a catalytic peripheral part, which is composed of five subunits: α3β3γ δ ε. F1 is known as a rotary motor, and the revolution of the γ subunit against the α3β3 ring is chemomechanically coupled with catalytic events at three catalytic sites [10–13]. The catalytic sites reside at the α-β interface, where the main residues are provided from the β subunit. During ATP hydrolysis, when viewed from membrane (Fo) side, the γ subunit rotates in the counterclockwise direction with 120° step, which was directly proven by using single molecule observation techniques [14]. Since

then, the molecular mechanism of the rotational catalysis has been extensively studied using this technique [15,16]. Although the basic mechanism of this rotational catalysis is highly conserved, the regulation mechanisms of this enzyme show the variety among species/organelles. Since ATP is the critical energy currency in all living cells, maintenance of the cellular ATP level, namely regulation of ATP hydrolysis and synthesis, is essential for surviving in altering environments. In bacteria and chloroplasts, two common inhibitory mechanisms for ATP hydrolysis are well studied: ADP-induced inhibition (ADP inhibition) and the inhibition caused by the intrinsic ε subunit (ε-inhibition). The former is achieved by tight binding of MgADP to the catalytic site [1,17–21], whereas the latter one is brought about by intrinsic inhibitory mechanism [22–27]. These inhibition mechanisms are regarded as important in avoiding the futile ATP-hydrolysis reaction and ensuring efficient ATP synthesis in vivo. In a previous study, we clarified that ADP inhibition is independent from ε-inhibition, and ε-inhibition can act mechanically independent from ADP inhibition using the α3β3γ

Abbreviations: FoF1, FoF1 ATP synthase; T. elongatus, Thermosynechococcus elongatus BP-1; E. coli, Escherichia coli; PCR, polymerase chain reaction; IPTG, isopropyl-β-D-thiogalactopyranoside; HEPES, N-2-hydroxyl piperazine-N′-2-ethane sulfonic acid; LDAO, lauryl dimethylamine-N-oxide; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis ⁎ Corresponding author at: Laboratory for Chemistry and Life Science, Tokyo Institute of Technology, Nagatsuta 4259-R1-8, Midori-ku, Yokohama 226-8503, Japan. E-mail address: [email protected] (T. Hisabori). https://doi.org/10.1016/j.bbabio.2018.02.004 Received 6 October 2017; Received in revised form 9 February 2018; Accepted 14 February 2018 Available online 19 February 2018 0005-2728/ © 2018 Elsevier B.V. All rights reserved.

BBA - Bioenergetics 1859 (2018) 319–325

K. Kondo et al.

Fig. 1. The structural position and the genetic location of the mutations in this study. A, a structure of the γ subunit in bovine F1 structure (Protein Data Bank code 1E79). Dashed lines and a scissor indicate the position of the mutation. The N-terminal part, γN, and the C-terminal part, γC, are shown in blue and sky-blue, respectively. An apparent position of 30-amino acid insertion specific to cyanobacteria is colored in red. The figure was generated by PyMOL. B, the domain architecture of the γ subunit. C, construction of plasmids for expressing α3β3γnick, α3β3γstop, α3β3γlinker and α3β3γthrom. All the mutants were originally constructed from α3β3γwild, which was described in Konno et al. [30]. A stop codon, followed by Shine–Dalgarno sequence and a start codon, the nucleotide sequence encoding a flexible linker (GGSGG), or the sequence encoding a thrombin-recognition sequence, LVPRGS, was inserted between 666th and 667th bases of the gene encoding the γ subunit for α3β3γnick, α3β3γlinker, and α3β3γthrom, respectively. For α3β3γstop, C-terminal part was totally deleted. Black boxes indicate His10-tag fused to the N-terminus of the β subunit.

inhibition and ε-inhibition extents [28]. However, the molecular mechanisms underlying those regulations have remained elusive in the light of structure–function relationship of the γ subunit. In this study, we prepared a mutant F1-ATP synthase of T. elongatus. The mutant expresses the nicked-γ subunits, consisting of N-terminal αhelix along with the Rossman-fold domain including the inserted sequence and the remaining C-terminal part (Fig. 1A, B). This mutation resulted in an increase of ATP-hydrolysis activity. We also explored the effect of post-assembly cleavage of the γ subunit. Altogether, our results uncover an unexpected role of the γ subunit and its C-terminus.

sub-complex and the ε subunit obtained from a thermophilic cyanobacterium, Thermosynechococcus elongatus BP-1 (T. elongatus) [27]. The γ subunit, which is the rotary shaft of F1 motor, comprises Nterminal and C-terminal helices domains, and a protruding Rossmanfold domain is located between these two major helices parts (Fig. 1). The N-terminal and C-terminal helices domains of the γ assemble into an antiparallel coiled-coil structure and are almost embedded into the stator ring composed of α3β3 in the F1 molecule. In addition, cyanobacterial and chloroplast γ subunits harbor an inserted sequence of 30 or 39 amino acids length within the Rossman-fold in comparison with bacterial or mitochondrial ATP synthases [28]. Although the cyanobacterial inserted sequence lacks nine amino acid sequence containing two cysteine residues compared with the chloroplast γ, whose redox status is the key for the redox regulation in the case of chloroplast ATP synthase, the overall sequences of the γ subunits from photosynthetic organisms are highly conserved [29]. We previously showed that the inserted sequence in the cyanobacterial γ provides the ability to frequently shift into an ADP-inhibition state, and the deletion of this sequence made the enzyme less sensitive to ε-inhibition [30,31]. In addition, we found that ATP-hydrolysis activity is significantly affected by the restriction of the relative position of two central α-helices to each other by way of a disulfide bond formation using the introduced cysteines. This structural restriction affected both ADP inhibition and εinhibition, suggesting relative slippage of central α-helices in the γ subunit that can regulate the enzyme activity by changing both ADP

2. Materials and methods 2.1. Materials ATP, Biotin-PEAC5-maleimide, and thrombin were obtained from ORYENTAL YEAST (Tokyo, Japan), Dojindo (Kumamoto, Japan), and Wako Pure Chemical Industries (Osaka, Japan), respectively. Pyruvate kinase, lactate dehydrogenase, and NADH were purchased from Roche Diagnostics (Basel, Switzerland). Other chemicals were of the highest commercially available grade. 2.2. Strains An ATP-synthase mutant strain derived from BL21 (DE3) of 320

BBA - Bioenergetics 1859 (2018) 319–325

K. Kondo et al.

2.5. Measurement of ATP-hydrolysis activity

Escherichia coli (E. coli), [Tcr, BL21 (DE3) uncΔ702, asnA::Tn10], which is deficient in the expression of endogenous ATP-synthase proteins, was used for expression of the wild-type and mutants of α3β3γ [32,33].

ATP-hydrolysis activities were measured as described in a previous study [31]. The measurements were performed at 25 °C in the presence of an ATP-regenerating system (50 mM HEPES–NaOH, 100 mM potassium chloride, 2 mM magnesium dichloride, 2 mM phosphoenolpyruvate, 50 μg ml−1 pyruvate kinase/lactate dehydrogenase, 0.2 mM NADH, and 2 mM ATP). ATP hydrolysis was started by adding 2–10 μg of α3β3γ complexes to 1.2 ml of assay buffer with or without lauryl dimethylamine-N-oxide (LDAO) at a final concentration of 0.1%. ATP-hydrolysis activity was measured by monitoring the decrease in NADH absorption at 340 nm. The activities were determined from the slope in the steady state (at least 60 s after the addition of α3β3γ).

2.3. Expression plasmids and protein preparation A plasmid for expression of α3β3γ from T. elongatus, pTR19FR, in which a deca-histidine tag was fused to the N-terminus of β subunit, was originally constructed for the single molecule experiments [30]. Using this plasmid as a background (the wild-type plasmid, which expresses α3β3γ, corresponding to α3β3γwild in this study), all the mutants were constructed by the way of megaprimer polymerase chain reaction (PCR) method [34]. For the first PCR, two primers were used: BP_1_gamma_Fw for the forward primer and gamma_222_nick_Rv, gamma_222_stop_Rv, gamma_222_linker_Rv, or gamma_222_throm_Rv for the reverse primers. For the second PCR, BP_1_gamma_Rv and the DNA fragment obtained from the first PCR were used as primers. Both the resultant DNA fragments and pTR19FR were treated by NheI and EcoRI and then ligated. The desired plasmids were selected by DNA sequencing. For a plasmid for expression of C-terminal hexa-histidinetagged γc, a DNA fragment encoding C-terminal 93-amino-acid peptide of the γ subunit was amplified using gamma_C_Fw and gamma_C_Rv, and was fused to a NdeI/XhoI site of pET23a (Novagen, Madison, USA) using hot-fusion reactions [35]. The primers used for the preparation of various mutants are listed in Table S1.

2.6. Immunoblot analysis Proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a PVDF membrane (Immun-Blot, Bio-Rad). Antibodies against γc were prepared using Histagged recombinant proteins as antigen. Chemiluminescence was detected using horseradish peroxidase-conjugated secondary antibodies and Supersignal West Femto Maximum Sensitivity Substrate (Life Technologies, Carlsbad, USA) and visualized by LAS 3000mini (GE Healthcare). Images were digitized using ImageJ software. 2.7. Post-assembly cleavage

2.4. Protein preparation Thrombin (0.01–100 Unit) was added to 10 μg of α3β3γ complexes. An equal volume of thrombin in a buffer containing 20 mM sodium phosphate, pH 8.0, and 100 mM sodium chloride was added to the α3β3γ at 25 °C, followed by measurements of ATP-hydrolysis activities. Alternatively, the reaction was stopped by adding 5% (w/v, final concentration) trichloroacetic acid. After centrifugation, the supernatant was removed, and the remaining oxidant was washed away with 80 μl of ice-cold acetone. Furthermore, the mixture was centrifuged and airdried after removing the supernatant. The pellet was dissolved in 63 mM Tris/HCl (pH 6.8), 5% (v/v) 2-mercaptoethanol, and 2% (w/v) SDS followed by electrophoresis on 16% polyacrylamide gel, and staining with Coomassie-Blue G-250.

Expression and purification of α3β3γ complexes were performed as described [30] with slight modifications. E. coli strain BL21 (DE3) uncΔ702, transformed with the plasmids, were cultured in 2 × YT medium containing 100 μg ml−1 ampicillin and 0.2 mM isopropyl-β-Dthiogalactopyranoside (IPTG) at 37 °C for 20 h. The complexes were purified by Ni-affinity chromatography and applied to size-exclusion column chromatography on a Superdex 200 Increase column 10/300 (GE Healthcare, Little Chalfont, England), which was equilibrated with 100 mM potassium phosphate, pH 7.0, 100 mM KCl, 0.1 mM MgCl2, and 0.1 mM ATP. The elution peak positioned at around 11 ml at a flow rate of 0.5 ml min−1 was collected, followed by labeling with Biotin-PEAC5maleimide at 20 °C for overnight. The excess reagents were removed using NAP-5 column (GE Healthcare), which was equilibrated with 50 mM N-2-hydroxyl piperazine-N′-2-ethane sulfonic acid (HEPES), pH 8.0, and 100 mM KCl (HK buffer). For immunoblot analysis and post-assembly cleavage, after Ni-affinity chromatography, the complexes were purified by size-exclusion chromatography with HK buffer supplemented with 0.1 mM MgCl2 and 0.1 mM ATP. The complexes obtained were then flash frozen in liquid nitrogen and stored at −80 °C until further use. Protein concentrations were determined using the Bradford method (Bio-Rad Protein Assay, Bio-Rad Laboratories, Inc., Hercules, USA). For the expression of γc, transformed cells of E. coli strain BL21 (DE3) pLysS were cultured at 37 °C, followed by induction with 1 mM IPTG at 18 °C for overnight. Cells were then collected by centrifugation and broken by two passages through a French pressure cell press. The supernatant was applied to a nickel-NTA Superflow column (Qiagen, Hilden, Germany) equilibrated with 20 mM sodium phosphate, pH 8.0, 100 mM sodium chloride, and 10 mM imidazole. The column was washed with the same buffer containing 20 mM imidazole; the proteins were eluted with the same buffer containing 200 mM imidazole, followed by size-exclusion column chromatography on a Superdex 75 column 10/300 (GE Healthcare). The γc proteins eluted as a stable complex with DnaK were used for generating antibodies and applied as a control in immunoblot analyses, of which the purity was 80–90% for γc. Expression and purification of the ε subunit were performed as described previously [30].

2.8. Rotation assay The rotation assay was performed as described [31], with some modifications. In total, 15 μl of biotinylated complexes were infused into a flow chamber, followed by incubation for 2 min at room temperature. The flow chamber was washed with 50 μl of 20 mM potassium phosphate (pH 8.0), 100 mM potassium chloride, and 0.5% BSA. An amount of 15 μl of streptavidin-coated beads (340 nm) in the same buffer was then infused into the flow chamber. After incubation for 15 min, followed by washing with 50 mM HEPES–KOH (pH 8.0) and 100 mM KCl, rotation was initiated by addition of 50 μl of an assay buffer (50 mM HEPES–KOH, pH 8.0, 100 mM KCl, 0.5 mM MgCl2, 20 μM ATP, 100 μg ml−1 pyruvate kinase, and 2 mM phosphoenolpyruvate). Rotation of the beads was observed using phase-contrast microscopy and video-recorded for further analysis. The pause/rotation duration was analyzed from three to nine molecules. Pauses of more than 1 s or continuous unidirectional revolutions of more than 10 rotations were used for generating each histogram. The rotation speeds were determined from continuous unidirectional rotations lasting more than 30 s. 3. Result 3.1. Nick insertion up-regulates ATP hydrolysis activity Using a recombinant expression system of α3β3γ from T. elongatus in 321

BBA - Bioenergetics 1859 (2018) 319–325

K. Kondo et al.

Fig. 2. ATP-hydrolysis activities (A), SDS-PAGE analysis (B), and immunoblot analysis (C) of α3β3γwild, α3β3γnick, or α3β3γlinker. A, ATP-hydrolysis activities were measured using an ATP-regenerating system. Gray and white bars indicate the activities in the presence and absence of LDAO (final concentration of 0.1%), respectively. The assay was conducted at 25 °C. The activities were determined from the slope in the steady state. The results of six or seven technical replicates from two independent experiments were averaged (mean ± SD). B, purified fractions of the complexes (2.5 μg per lane) were electrophoresed on 16% polyacrylamide gel, followed by Coomassie-Blue G-250 staining. Theoretical mass of α, β, γ, γN and γC are 54.3, 51.8, 35.0, 25.0 and 9.96 kDa, respectively. From α3β3γnick, γN was detected, but γC was not. C, estimation of γC content in α3β3γnick. A total of 5 μg of α3β3γnick from three independent experiments were loaded, as well as 26 pg–1.664 ng of γC-His6 as control. Electrophoresed proteins on 16% polyacrylamide gel were transferred to PVDF membrane, followed by immunoblotting with antibodies against γC-His6; chemiluminescence was detected. The theoretical mass of γCHis6 is 11.1 kDa. Signal intensities were then quantified using ImageJ software, followed by calculation of γC content as shown in Fig. S2.

Table 1 ATP-hydrolysis activities of α3β3γwild, α3β3γnick, and α3β3γlinker. α3β3γwild ATP-hydrolysis activity (μmol Pi min

−1

−1

mg

)

–LDAO +LDAO

Activation ratio (+LDAO/−LDAO) a

1.29 ± 0.08 25.4 ± 1.48 19.6

α3β3γnick

α3β3γlinker a

9.38 ± 0.64 (7.25) 21.9 ± 1.2 (0.864)a 2.34

5.67 ± 1.02 (4.38)a 32.0 ± 3.01 (1.26)a 5.65

The ratio of the mutant activities against that of α3β3γwild were calculated.

terminal His-tagged γC (Fig. 2C), the amounts of γN was calculated by the subtraction of γC from the total γ. Then, the resulting stoichiometry was estimated as α3β3(γN + γC): α3β3γN = 1: 120 (Fig. S2A, B and C, details of the analyses are explained in the figure legend). In contrast, α3β3γN was not obtained by size-exclusion chromatography, when the proteins were expressed in E. coli with a plasmid harboring genes encoding α, β, and γN, but not γC (Fig. 1C and S1C, α3β3γstop). In this sample, ATP-hydrolysis activities were not detected at all. These results suggest that γC is indispensable for the assembly of α3β3γ complex, whereas it is dispensable for ATP-hydrolysis process. In other words, it is indicated that α3β3γN itself exhibits 7-fold higher ATP-hydrolysis activity. Accordingly, we found that disassembled α3β3γnick was not reconstituted by incubation at 37 °C, whereas α3β3γwild was partially reconstituted into the complex (Fig. S3).

E.coli, we introduced a nick into the γ subunit, namely a stop codon followed by Shine–Dalgarno sequence and a start codon, between 666th and 667th bases of the gene encoding the γ subunit (Fig. 1C), yielding the N-terminal 222 amino acids length protein (γN) with a theoretical mass of 25 kDa and the remaining C-terminal 94 amino acids length protein (γC) of 10 kDa. The recombinant mutant complexes (α3β3γnick) as well as the wild type (α3β3γwild) were purified using nickel-affinity chromatography followed by size-exclusion column chromatography. Although the retention time of α3β3γnick was quite similar to that of α3β3γwild, the final protein yield was much lower, instead of the relatively higher peaks corresponding to α:β hetero-dimers and β monomers (Fig. S1A and B). Their ATP-hydrolysis activities were then examined using ATP-regenerating system [30] in the absence and presence of LDAO (Fig. 2A). LDAO is a detergent, which is known to recover F1 ATPase from ADP-inhibition state [36]. Without LDAO, α3β3γnick showed 7.3-fold higher ATP-hydrolysis activities than that of α3β3γwild (Table 1). In contrast, the addition of LDAO enhanced the activities of α3β3γwild up to 20-fold, whereas only 2-fold in α3β3γnick. These results suggest that ADP inhibition is partially cancelled in α3β3γnick, implying that nick insertion assists the release of MgADP from the catalytic sites.

3.3. Post-assembly cleavage of γ subunit into γN and γC promotes ATP hydrolysis We further analyzed the effect of the nick insertion by post-assembly cleavage of the γ subunit. For this purpose, another mutant complex, α3β3γthrom, was constructed, which harbors an insertion of a thrombinrecognition sequence between γN and γC (Fig. 1C). We obtained a similar amount of the complex to α3β3γwild (Fig. S1D), and after addition of thrombin, the two bands corresponding to γN and γC were detected in this mutant complex by SDS-PAGE, followed by Coomassie-Blue G-250 staining (Fig. S4A), whereas no effect of thrombin was observed in α3β3γwild (Fig. S4B). When thrombin was added to α3β3γ in a ratio of

3.2. γC is dispensable for ATP hydrolysis SDS-PAGE analysis revealed that γC was markedly decreased in the purified fraction of α3β3γnick compared to other subunits, α, β, and γN (Fig. 2B). By immunoblotting with specific antibodies against the C322

BBA - Bioenergetics 1859 (2018) 319–325

K. Kondo et al.

Fig. 3. Post-assembly cleavage of the γ subunit. A, thrombin treatment for cleavage of the γ subunit into γN and γC. Thrombin was added to α3β3γthrom at a ratio of 1 Unit/100 μg of complexes at 25 °C, and the cleavage was stopped by adding 5% (w/v, final concentration) trichloroacetic acid. 10 μg of proteins were loaded per lane. B, The effect of the thrombin treatment on ATP-hydrolysis activities (gray bars) and the percentage of γN (black bars), which was detected by the SDS-PAGE analysis shown in A. ATP-hydrolysis activities were measured, as described in Fig. 2. C, ATP-hydrolysis activities before and after 30-min treatment with thrombin at 25 °C (white bars and gray bars, respectively). Thrombin was added to α3β3γthrom or α3β3γwild, at a ratio of 1 Unit/100 μg of complexes. After the treatment, ATP-hydrolysis activities were measured in the presence/absence of 0.1% LDAO.

Table 2 Activities of ATP hydrolysis before or after 30-min treatment with thrombin. α3β3γwild Thrombin treatment ATP-hydrolysis activity (μmol Pi min−1 mg−1) Activation ratio (+LDAO/−LDAO)

–LDAO +LDAO

− 1.82 ± 0.06 31.7 ± 1.78 17.4

1 U: 100 μg, the cleavage reaction was supposed to be completed within 30–60 min (Fig. 3A). Upon this treatment, ATP-hydrolysis activities were upregulated, and the time-course of change in the activity was correlated with that of the cleavage (Fig. 3B). Furthermore, we found that 30-min treatment with thrombin elevated the level of ATP-hydrolysis activities as high as 3-folds in α3β3γthrom (Fig. 3C and Table 2). Activation ratios upon addition of LDAO for non-treated and thrombintreated α3β3γthrom were 14 and 5.1, respectively (Fig. 3C and Table 2). These results demonstrate that cleavage of the γ certainly assists restoration from ADP inhibition.

α3β3γthrom + 1.63 ± 0.07 29.4 ± 0.65 18.0

− 2.26 ± 0.50 32.5 ± 5.24 14.4

+ 6.63 ± 0.23 34.0 ± 1.40 5.12

observation techniques (Fig. 4C). The results indicated that the average rotation rates at the rotating period were not significantly altered by the addition of thrombin: 3.3 ± 1.2 revolutions per second (rps) for untreated α3β3γthrom (Fig. 4A and S5) and 4.4 ± 1.0 rps for those treated with thrombin (Fig. 4B and S5), while that of the wildtype was reported as 3.8 ± 1.1 rps [31]. These results indicate that the rates of ATP hydrolysis are not affected by the nicking, which supports our hypothesis that the high activity is attributed to its lower ADP-inhibition level. Yet, no significant differences were observed in both pause durations and rotation durations (Fig. S6, [29]). We also analyzed the rotation behaviors of α3β3γwild and α3β3γnick, although no significant differences were observed between them (Fig. S5 and S7). The reason for these results will be discussed in the next section.

3.4. Rotation properties of α3β3γthrom and α3β3γnick To confirm our hypothesis, we analyzed rotational behaviors of α3β3γthrom with or without thrombin treatment, using single molecule 323

BBA - Bioenergetics 1859 (2018) 319–325

K. Kondo et al.

Fig. 4. Rotation of the γ subunit in α3β3γthrom. The typical time courses of α3β3γthrom before (A, n = 9) and after (B, n = 7) the addition of thrombin (1 Unit/100 μg complexes) are shown. Rotations were analyzed with custom software [14]. C, The scheme of the rotation experiments, which were performed as previously described [31]. Biotinylated α3β3γthrom complexes were infused into the flow chamber on a glass plate, followed by binding with streptavidin-coated beads. Rotation of the duplex beads was monitored with a conventional optical microscope type IX 71 (Olympus, Tokyo, Japan) with a 100× objective lens.

3.5. Insertion of a flexible linker

disulfide bond formation increased their ATP-hydrolysis activities [28]. Thus, it was postulated that the conformational change of the γ subunit is due to the relative slippage of the two helices, which is the main cause of altered enzyme activity. This hypothesis is supported by our results from α3β3γlinker, which exhibited the elevated activity. We tried to reconstitute α3β3γnick complex using the γC protein. However, fusing N-terminal tag, which does not appear to interfere with its interaction with α3β3, and introducing protease recognition sequence into γC-His6 greatly affected its solubility. Nevertheless, we found that coexpression of γC is required to obtain the complex (Fig. 1 and S1). Furthermore, we obtained the significant difference in the reconstitution efficiency between α3β3γwild and α3β3γnick, as shown in Fig. S3. Similar results were obtained from the thrombin-treated α3β3γthrom and the untreated α3β3γthrom. As shown in Fig. S9, thrombin treatment reduced the reconstitution efficiencies of α3β3γthrom. All these results point toward the unexpected role of γC in assembly of α3β3γ. Previously, using a reconstitution system of spinach CF1, Samra et al. reported that neither N-terminal (1–196) nor C-terminal (207–323) γ subunit alone could assemble with the α and β subunits to form an active complex [38]. Though, in their experiments, the insertional region is included in the C-terminal part, these data are in good accordance with our finding that γC is required for α3β3γ assembly. It was previously shown that cyanobacteria-specific insertion sequence plays a crucial role in ADP inhibition and indirectly regulates the ε-inhibition extent [30,31]. In this report, γN harbors this part in its C-terminus. We explored the effect for ε-inhibition, and no significant differences were detected between α3β3γnick and α3β3γwild (Fig. S8) [30]. Although little is known about how the change of the γ subunit affects ε-inhibition, our findings provide important insights into the mechanism; ε-inhibition is conferred by the insertion part of the γ itself. Buchert et al. have recently proposed that restrictive movements of the γ-termini lower the extent of ADP inhibition, from the results of complexes containing chimeric γ subunit derived from cyanobacteria and chloroplast, which harbors a redox-regulated cysteine pair [39]. This is consistent with our findings; the rigidity of the central helices of the γ subunit modulates the extent of ADP inhibition, albeit by an unknown mechanism. As the thrombin-cleavage site appears to be positioned far from the catalytic site in α3β3γ [40,41], it is reasonable that rotation speed was not affected by the thrombin treatment. However, we could not detect any significant differences in pause/rotation durations with or without thrombin treatment of α3β3γthrom, probably due to very slight change in the rotation behaviors. Further analyses including estimation of torque of the rotation are required to address whether the upregulation of ATP-hydrolyzing activities is solely due to

To elucidate the mechanisms underlying the elevation of ATP-hydrolysis activities in those mutants, we further introduced a 5-aminoacid linker, GGSGG, at the position of the nick that allowed flexibility in the two domains: γN and γC (Fig. 1C, α3β3γlinker). We found that the ATP-hydrolysis activity was increased 4.4-fold than that of α3β3γwild, whereas the activation ratio by the addition of LDAO was only 5.7 (Fig. 2A and Table 1). This result indicates that a flexibility in two helices upregulate the ATP-hydrolysis activities, at least in part, by way of cancellation of ADP inhibition. 3.6. The effect of the nick insertion on the epsilon inhibition ε-inhibition is another major mechanism regulating activities of F1ATP synthase. We explored the effect for this, and no significant differences were detected between α3β3γnick and α3β3γwild. The apparent dissociation constant of the ε subunit obtained for α3β3γnick was 0.80 nM, and a maximum inhibition of 92.5% was achieved (Fig. S8), while those for the wildtype were previously reported as 2.1 ± 0.3 nM and 100%, respectively [30]. 4. Discussion In this study, we demonstrated that the nick-inserted mutant, α3β3γnick, showed higher ATP-hydrolysis activity as compared with α3β3γwild. As opposed to the expectation, the data demonstrated that the purified fraction was almost devoid of γC, and the content of γC was less than 1/120. This result implies that α3β3γN, and not α3β3(γN + γC), retains the enhanced activity as it is unlikely that the α3β3(γN + γC) would exhibit the elevated activity as high as 120-fold. Both post-assembly cleavage and insertion of a flexible linker to the γ subunit resulted in the acceleration of ATP-hydrolysis activities. All these results suggest that conformational rigidity of the central shaft, conferred by γC, regulates ATP-hydrolysis activity through ADP inhibition. A number of previous studies have explored how the shaft plays a crucial role in the rotation and ATP hydrolysis. Sokolov et al. reported that a deletion of 20 amino acids at the C-termini of the γ subunit of CF1 decreased ATP-hydrolysis activity to 19%, whereas α3β3γ complex was assembled as effectively as the wild type [37]. In such a mutant, it is conceivable that the interaction of the N-terminal α-helix and the truncated Cterminal α-helix in the complex somehow inhibit the activity. Sunamura et al. constructed the mutants, where the two central helices of the γ are cross-linked by disulfide bonds, and demonstrated that the 324

BBA - Bioenergetics 1859 (2018) 319–325

K. Kondo et al.

cancellation of ADP inhibition. [13]

5. Conclusion

[14]

This study demonstrates that cleavage of the γ subunit at the cyanobacteria-specific insertional region results in the upregulation of ATP-hydrolysis activity of α3β3γ complexes. LDAO treatment indicated that complexes containing the cleaved γ are less prone to ADP inhibition. Results from immunoblot analysis indicate that the complexes containing the split γ subunits were mostly devoid of γC. On the contrary, when the proteins were expressed in E. coli with a plasmid harboring genes encoding α, β, and γN, but not γC, α3β3γN was not obtained by size-exclusion chromatography, suggesting that γC is indispensable for the assembly of α3β3γ complex.

[15]

[16] [17]

[18] [19]

Transparency document The http://dx.doi.org/10.1016/j.bbabio.2018.02.004 with this article can be found, in online version.

[20]

associated

[21]

Acknowledgements

[22]

We acknowledge Biomaterials Analysis Division of Tokyo Institute of Technology for DNA sequencing, and Materials Analysis Division for supporting mass spectrometry analysis. We are grateful to Dr. Keisuke Yoshida (Tokyo Institute of Technology) for N-terminal amino acid sequence analysis. We thank Drs. Kazunori Sugiura and Satoshi Hara (Tokyo Institute of Technology) for their helpful suggestions. This work was supported in part by the Core Research for Evolutional Science and Technology program (CREST) from the Japan Science and Technology Agency (JST) and Grants-in-Aid for Scientific Research (Grant 16H06556 to T.H.) from the Japan Society for the Promotion of Science (JSPS), and in part by Dynamic Alliance for Open Innovation Bridging Human, Environment and Materials.

[23]

[24]

[25]

[26]

[27]

[28]

Appendix A. Supplementary data [29]

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bbabio.2018.02.004.

[30]

References

[31]

[1] P.D. Boyer, The ATP synthase–a splendid molecular machine, Annu. Rev. Biochem. 66 (1997) 717–749. [2] M. Yoshida, E. Muneyuki, T. Hisabori, ATP synthase–a marvellous rotary engine of the cell, Nat. Rev. Mol. Cell Biol. 2 (2001) 669–677. [3] D. Stock, A.G. Leslie, J.E. Walker, Molecular architecture of the rotary motor in ATP synthase, Science 286 (1999) 1700–1705. [4] N. Mitome, T. Suzuki, S. Hayashi, M. Yoshida, Thermophilic ATP synthase has a decamer c-ring: indication of noninteger 10:3 H+/ATP ratio and permissive elastic coupling, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 12159–12164. [5] D. Pogoryelov, J. Yu, T. Meier, J. Vonck, P. Dimroth, D.J. Muller, The c15 ring of the Spirulina platensis F-ATP synthase: F1/F0 symmetry mismatch is not obligatory, EMBO Rep. 6 (2005) 1040–1044. [6] D. Pogoryelov, C. Reichen, A.L. Klyszejko, R. Brunisholz, D.J. Muller, P. Dimroth, T. Meier, The oligomeric state of c rings from cyanobacterial F-ATP synthases varies from 13 to 15, J. Bacteriol. 189 (2007) 5895–5902. [7] B. Ballhausen, K. Altendorf, G. Deckers-Hebestreit, Constant c10 ring stoichiometry in the Escherichia coli ATP synthase analyzed by cross-linking, J. Bacteriol. 191 (2009) 2400–2404. [8] I.N. Watt, M.G. Montgomery, M.J. Runswick, A.G.W. Leslie, J.E. Walker, Bioenergetic cost of making an adenosine triphosphate molecule in animal mitochondria, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 16823–16827. [9] S. Schulz, M. Wilkes, D.J. Mills, W. Kühlbrandt, T. Meier, Molecular architecture of the N-type ATPase rotor ring from Burkholderia pseudomallei, EMBO Rep. 18 (2017) 526–535. [10] H. Noji, R. Yasuda, M. Yoshida, K. Kinosita Jr., Direct observation of the rotation of F1-ATPase, Nature 386 (1997) 299–302. [11] R. Yasuda, H. Noji, M. Yoshida, K. Kinosita Jr., H. Itoh, Resolution of distinct rotational substeps by submillisecond kinetic analysis of F1-ATPase, Nature 410 (2001) 898–904. [12] T. Nishizaka, K. Oiwa, H. Noji, S. Kimura, E. Muneyuki, M. Yoshida, K. Kinosita Jr.,

[32]

[33]

[34] [35]

[36]

[37]

[38]

[39]

[40] [41]

325

Chemomechanical coupling in F1-ATPase revealed by simultaneous observation of nucleotide kinetics and rotation, Nat. Struct. Mol. Biol. 11 (2004) 142–148. R. Watanabe, H. Noji, Chemomechanical coupling mechanism of F1-ATPase: catalysis and torque generation, FEBS Lett. 587 (2013) 1030–1035. R. Yasuda, H. Noji, K. Kinosita Jr., M. Yoshida, F1-ATPase is a highly efficient molecular motor that rotates with discrete 120 degree steps, Cell 93 (1998) 1117–1124. K. Adachi, K. Oiwa, T. Nishizaka, S. Furuike, H. Noji, H. Itoh, M. Yoshida, K. Kinosita Jr., Coupling of rotation and catalysis in F1-ATPase revealed by singlemolecule imaging and manipulation, Cell 130 (2007) 309–321. R. Watanabe, H. Noji, Timing of inorganic phosphate release modulates the catalytic activity of ATP-driven rotary motor protein, Nat. Commun. 5 (2014) 3486. I.B. Minkov, A.F. Fitin, E.A. Vasilyeva, A.D. Vinogradov, Mg2+-induced ADP- dependent inhibition of the ATPase activity of beef heart mitochondrial coupling factor F1, Biochem. Biophys. Res. Commun. 89 (1979) 1300–1306. D. Bar-Zvi, N. Shavit, Modulation of the chloroplast ATPase by tight ADP binding. Effect of uncouplers and ATP, J. Bioenerg. Biomembr. 14 (1982) 467–478. E.A. Vasilyeva, I.B. Minkov, A.F. Fitin, A.D. Vinogradov, Kinetic mechanism of mitochondrial adenosine triphosphatase. ADP-specific inhibition as revealed by the steady-state kinetics, Biochem. J. 202 (1982) 9–14. J.M. Zhou, Z.X. Xue, Z.Y. Du, T. Melese, P.D. Boyer, Relationship of tightly bound ADP and ATP to control and catalysis by chloroplast ATP synthase, Biochemistry 27 (1988) 5129–5135. J.G. Digel, A. Kishinevsky, A.M. Ong, R.E. McCarty, Differences between two tight ADP binding sites of the chloroplast coupling factor 1 and their effects on ATPase activity, J. Biol. Chem. 271 (1996) 19976–19982. N. Nelson, H. Nelson, E. Racker, Partial resolution of the enzymes catalyzing photophosphorylation. XII. Purification and properties of an inhibitor isolated from chloroplast coupling factor 1, J. Biol. Chem. 247 (1972) 7657–7662. M.L. Richter, W.J. Patrie, R.E. McCarty, Preparation of the ε subunit and ε subunitdeficient chloroplast coupling factor 1 in reconstitutively active forms, J. Biol. Chem. 259 (1984) 7371–7373. R. Aggeler, R.A. Capaldi, Nucleotide-dependent movement of the ε subunit between α and β subunits in the Escherichia coli F1F0-type ATPase, J. Biol. Chem. 271 (1996) 13888–13891. Y. Kato, T. Matsui, N. Tanaka, E. Muneyuki, T. Hisabori, M. Yoshida, Thermophilic F1-ATPase is activated without dissociation of an endogenous inhibitor, ε subunit, J. Biol. Chem. 272 (1997) 24906–24912. K.F. Nowak, V. Tabidze, R.E. McCarty, The C-terminal domain of the ε subunit of the chloroplast ATP synthase is not required for ATP synthesis, Biochemistry 41 (2002) 15130–15134. H. Konno, A. Isu, Y. Kim, T. Murakami-Fuse, Y. Sugano, T. Hisabori, Characterization of the relationship between ADP- and ε-induced inhibition in cyanobacterial F1-ATPase, J. Biol. Chem. 286 (2011) 13423–13429. E.I. Sunamura, H. Konno, M. Imashimizu, M. Mochimaru, T. Hisabori, A conformational change of the γ subunit indirectly regulates the activity of cyanobacterial F1-ATPase, J. Biol. Chem. 287 (2012) 38695–38704. Y. Kim, H. Konno, Y. Sugano, T. Hisabori, Redox regulation of rotation of the cyanobacterial F1-ATPase containing thiol regulation switch, J. Biol. Chem. 286 (2011) 9071–9078. H. Konno, T. Murakami-Fuse, F. Fujii, F. Koyama, H. Ueoka-Nakanishi, C.G. Pack, M. Kinjo, T. Hisabori, The regulator of the F1 motor: inhibition of rotation of cyanobacterial F1-ATPase by the ε subunit, EMBO J. 25 (2006) 4596–4604. E.I. Sunamura, H. Konno, M. Imashimizu-Kobayashi, Y. Sugano, T. Hisabori, Physiological impact of intrinsic ADP inhibition of cyanobacterial FoF1 conferred by the inherent sequence inserted into the γ subunit, Plant Cell Physiol. 51 (2010) 855–865. A.K. Joshi, S. Ahmed, G. Ferro-Luzzi Ames, Energy coupling in bacterial periplasmic transport systems. Studies in intact Escherichia coli cells, J. Biol. Chem. 264 (1989) 2126–2133. N.N. Nichols, C.S. Harwood, PcaK, a high-affinity permease for the aromatic compounds 4-hydroxybenzoate and protocatechuate from Pseudomonas putida, J. Bacteriol. 179 (1997) 5056–5061. O. Landt, H.P. Grunert, U. Hahn, A general method for rapid site-directed mutagenesis using the polymerase chain reaction, Gene 96 (1990) 125–128. C. Fu, W.P. Donovan, O. Shikapwashya-Hasser, X. Ye, R.H. Cole, Hot Fusion: an efficient method to clone multiple DNA fragments as well as inverted repeats without ligase, PLoS One 9 (2014) e115318. S.D. Dunn, R.G. Tozer, V.D. Zadorozny, Activation of Escherichia coli F1-ATPase by lauryldimethylamine oxide and ethylene glycol: relationship of ATPase activity to the interaction of the ε and β subunits, Biochemistry 29 (1990) 4335–4340. M. Sokolov, L. Lu, W. Tucker, F. Gao, P.A. Gegenheimer, M.L. Richter, The 20 Cterminal amino acid residues of the chloroplast ATP synthase γ subunit are not essential for activity, J. Biol. Chem. 274 (1999) 13824–13829. H.S. Samra, F. Gao, F. He, E. Hoang, Z. Chen, P.A. Gegenheimer, C.L. Berrie, M.L. Richter, Structural analysis of the regulatory dithiol-containing domain of the chloroplast ATP synthase γ subunit, J. Biol. Chem. 281 (2006) 31041–31049. F. Buchert, H. Konno, T. Hisabori, Redox regulation of CF1-ATPase involves interplay between the γ-subunit neck region and the turn region of the βDELSEED-loop, Biochim. Biophys. Acta 1847 (2015) 441–450. C. Gibbons, M.G. Montgomery, A.G. Leslie, J.E. Walker, The structure of the central stalk in bovine F1-ATPase at 2.4 Å resolution, Nat. Struct. Biol. 7 (2000) 1055–1061. G. Cingolani, T.M. Duncan, Structure of the ATP synthase catalytic complex (F1) from Escherichia coli in an auto-inhibited conformation, Nat. Struct. Mol. Biol. 18 (2011) 701–707.