International Journal of Molecular Sciences
Article Development and Optimization of Expression, Purification, and ATPase Assay of KaiC for Medium-Throughput Screening of Circadian Clock Mutants in Cyanobacteria
Dongyan Ouyang 1, Yoshihiko Furuike 1,2 , Atsushi Mukaiyama 1,2 , Kumiko Ito-Miwa 3, Takao Kondo 3 and Shuji Akiyama 1,2,*
1 Research Center of Integrative Molecular Systems (CIMoS), Institute for Molecular Science, National Institute for Natural Sciences, 38 Nishigo-Naka, Myodaiji, Okazaki 444-8585, Japan; [email protected] (D.O.); [email protected] (Y.F.); [email protected] (A.M.) 2 Department of Functional Molecular Science, SOKENDAI (The Graduate University for Advanced Studies), 38 Nishigo-Naka, Myodaiji, Okazaki 444-8585, Japan 3 Division of Biological Science, Graduate School of Science and Institute for Advanced Research, Nagoya University; Furo-cho, Chikusa-ku, Nagoya 464-8602, Japan; [email protected] (K.I.-M.); [email protected] (T.K.) * Correspondence: [email protected]; Tel.: +81-564-55-7363
Received: 26 April 2019; Accepted: 3 June 2019; Published: 7 June 2019
Abstract: The slow but temperature-insensitive adenosine triphosphate (ATP) hydrolysis reaction in KaiC is considered as one of the factors determining the temperature-compensated period length of the cyanobacterial circadian clock system. Structural units responsible for this low but temperature-compensated ATPase have remained unclear. Although whole-KaiC scanning mutagenesis can be a promising experimental strategy, producing KaiC mutants and assaying those ATPase activities consume considerable time and effort. To overcome these bottlenecks for in vitro screening, we optimized protocols for expressing and purifying the KaiC mutants and then designed a high-performance liquid chromatography system equipped with a multi-channel high-precision temperature controller to assay the ATPase activity of multiple KaiC mutants simultaneously at different temperatures. Through the present protocol, the time required for one KaiC mutant is reduced by approximately 80% (six-fold throughput) relative to the conventional protocol with reasonable reproducibility. For validation purposes, we picked up three representatives from 86 alanine-scanning KaiC mutants preliminarily investigated thus far and characterized those clock functions in detail.
Keywords: circadian clock; cyanobacteria; KaiC; scanning mutagenesis; ATPase activity; temperature compensation; phosphorylation cycle
1. Introduction Circadian clocks are self-sustained oscillatory systems allowing organisms to anticipate and adapt to the daily environmental changes resulting from the Earth’s rotation [1]. This time-measuring machinery has been ubiquitously observed from eukaryotes [2] to prokaryotes [3] and is known to possess three remarkable characteristics [4]—autonomous oscillation with a period of approximately 24 h, temperature independency of the period length (temperature compensation), and entrainability by external perturbations. Principal constituents of the circadian clock systems are genes responsible for the clock functions (clock genes) and their translational products (clock proteins). Circadian rhythms arising as consequences of negative feedback regulations of clock gene transcriptions by the clock
Int. J. Mol. Sci. 2019, 20, 2789; doi:10.3390/ijms20112789 www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2019, 20, 2789 2 of 12 proteins are called transcriptional and translational oscillations (TTO) [5], while those driven solely by the clock proteins are called post-translational oscillations (PTO) [6]. The cyanobacterium Synechococcus elongatus PCC7942 (S. elongatus) is one of the organisms harboring both TTO [7] and PTO [8]. S. elongatus PTO can be reconstructed in vitro as a phosphorylation cycle of KaiC by mixing KaiC with the other two Kai proteins, KaiA and KaiB, in the presence of adenosine triphosphate (ATP) [9]. KaiC reveals rhythmic activation and inactivation of auto-kinase [9], auto-phosphatase [9], and ATPase [10] activities to affect repeated assembly and disassembly of Kai protein complexes [11,12]. KaiC ATPase is of particular interest here, as its activity is extremely low -1 (~12 ATP d )[10], temperature compensated (Q10 = 1.0) [10], and finely correlated to the frequencies of TTO as well as PTO [10,13]. These unique properties inspire development of an ATPase-based screening for KaiC clock mutants giving short, long, and/or temperature-dependent periods, as one of the effective methods to uncover origins of the temperature-compensated circadian period in the S. elongatus clock system. In this paper, we describe the current design of our in vitro ATPase-based screening system and discuss its performance using the single-alanine substitutions (alanine scanning) library of KaiC. The alanine scanning itself is an established means often used to test hypothesis on structure–function relationships in proteins [14,15]; however, its application to the ATPase-based screening is not straightforward, because several experimental steps, such as mutant expression, purification, and ATPase assay, limit its throughput. To improve the efficiency of the in vitro screening, we replaced a glutathione S-transferase (GST)-tag fused to KaiC by a hexa-histidine (His)-tag and then optimized procedures to express and purify His-tagged KaiC. Furthermore, a new high-performance liquid chromatography (HPLC) system with a four-channel temperature controller was designed to determine temperature dependencies of the ATPase activities for multiple KaiC mutants simultaneously. These improvements together with several other optimizations reduced approximately 80% of the time costs associated with the overall screening process. Thus far, 16.5% of the alanine scanning KaiC mutants (86 of 519 amino acid residues) have been preliminary expressed, purified, and analyzed with the improved throughput. For validation purposes, three representative KaiC mutants revealing modulated ATPase activity and/or its temperature dependency were subjected to in vivo TTO and in vitro PTO assays. We discuss future perspectives of the ATPase-based in vitro screening on the basis of advantages and disadvantages of the present method.
2. Results
2.1. Optimization of the Expression and Purification of KaiC In an expression vector (pGEX-6P-1 kaiC) originally constructed by Iwasaki et al. [16], the C-terminus of GST is fused to the N-terminus of S. elongatus KaiC consisting of 519 amino acid residues. KaiC is thus expressed as an N-terminally GST-tagged form and has to be treated with proteinases for a long time to cleave the GST-tag off. To skip this time-consuming treatment, we constructed an expression vector encoding KaiC in a C-terminally His-tagged form (pET-3a kaiC). In this study, Escherichia coli(E. coli)BL21(DE3)pLysE was used as a host for His-tagged KaiC over-expression (Step I in Figure1a), as it allowed more expression in a culture volume of 400 mL within a shorter culture time than BL21(DE3) strain (see details in Materials and Methods). In the purification process (Step II in Figure1a), His-tagged KaiC was first extracted by suspending cell pellets in 25 mL of a buffer containing 20 mM Tris/HCl, 0.15 M NaCl, 0.5 mM ethylenediaminetetraacetic acid (EDTA), 1 mM ATP, 5 mM MgCl2, 1 mM dithiothreitol (DTT), and 40 U/mL benzonase at pH 8.0. After 1 h stirring, the cells were then crushed by sonication. The extraction of His-tagged KaiC was further promoted by adding tritonX-100 solution to a final concentration of 1% and by stirring for another 1 h. A supernatant fraction separated by centrifugation (lane 3 in Figure1b) was applied to a free-packed Ni-Sepharose affinity column (bead volume of 2 mL) equilibrated with a binding buffer Int. J. Mol. Sci. 2019, 20, 2789 3 of 12
containing 20 mM Tris/HCl, 500 mM NaCl, 5 mM MgCl2, 1 mM ATP, and 130 mM imidazole at pH 7.4 (Step IIIInt. in J. Mol. Figure Sci. 20191a)., 20 The, x FOR column PEER REVIEW after loading the sample was washed with 25 mL of the 3 of binding 12 buffer. His-tagged KaiC was eluted with an elution buffer containing 20 mM Tris/HCl, 500 mM NaCl, containing 20 mM Tris/HCl, 500 mM NaCl, 5 mM MgCl2, 1 mM ATP, and 130 mM imidazole at pH 5 mM MgCl , 1 mM ATP, and 480 mM imidazole at pH 7.4. Immediately after the elution, the fractions 7.4 (Step2 III in Figure 1a). The column after loading the sample was washed with 25 mL of the binding containingbuffer. imidazole His-tagged were KaiC subjectedwas eluted towith a PD-10an elution desalting buffer containing column (GE20 mM Healthcare) Tris/HCl, 500 equilibrated mM NaCl, with 20 mM5 TrismM/ MgClHCl,2, 0.1 1 mM M ATP, NaCl, and 0.5 480 mM mM EDTA, imidazole 1 mM at pH ATP, 7.4. Immediately 5 mM MgCl after2, and the elution, 1 mM the DTT fractions at pH 8.0 to removecontaining imidazole imidazole because were of subjected a poor compatibility to a PD-10 desalting between column imidazole (GE Healthcare) and His-tagged equilibrated KaiC. with All of the washing,20 mM Tris/HCl, eluting, 0.1 and M NaCl, buffer 0.5 exchanging mM EDTA, 1 steps mM ATP, were 5 thusmM MgCl conducted2, and 1 mM quickly DTT at to pH minimize 8.0 to the exposureremove of His-tagged imidazole because KaiC to of imidazole a poor compatibility to prevent between the sample imidazole from and ending His-tagged with aKaiC. cloudy All solution,of or, eventhe worse, washing, precipitation. eluting, and buffer exchanging steps were thus conducted quickly to minimize the exposure of His-tagged KaiC to imidazole to prevent the sample from ending with a cloudy solution, Impurities remaining even after Ni-Sepharose affinity chromatography (lane 5 in Figure1b) were or, even worse, precipitation. separated throughImpurities two-step remaining chromatographic even after Ni-Sepharose purification affinity procedurechromatography (Steps (lane IV 5 and in Figure V in Figure 1b) 1a). The desaltedwere separated sample through was first two-step loaded chromatographic to an anion-exchange purification column procedure (Resource (Steps IV Q)and equilibrated V in Figure with 20 mM1a). Tris The/HCl, desalted 0.1 M sample NaCl, 0.5was mM first EDTA, loaded 1to mM an anion-exchange ATP, 5 mM MgCl column2, and (Resource 1 mM DTT Q) equilibrated at pH 8.0, and the resultantwith main 20 mM fractions Tris/HCl, (lane 0.1 6M in NaCl, Figure 0.51 mMb) were EDTA, subjected 1 mM ATP, to a 5 size-exclusion mM MgCl2, and column 1 mM DTT (Superdex at pH 200) 8.0, and the resultant main fractions (lane 6 in Figure 1b) were subjected to a size-exclusion column equilibrated with 20 mM Tris/HCl, 0.15 M NaCl, 0.5 mM EDTA, 1 mM ATP, 5 mM MgCl2, and 1 mM (Superdex 200) equilibrated with 20 mM Tris/HCl, 0.15 M NaCl, 0.5 mM EDTA, 1 mM ATP, 5 mM DTT at pH 8.0 to achieve high purities of more than 95% (lane 7 in Figure1b). According to the results MgCl2, and 1 mM DTT at pH 8.0 to achieve high purities of more than 95% (lane 7 in Figure 1b). of moreAccording than 86 to mutants the results tried of more thus than far, 86 the mutants final yieldtried thus was far, maintained the final yield at was least maintained 0.6 mg per at least 400 mL of terrific0.6 broth mg per medium 400 mL forof terrific each sample.broth medium This for was each nearly sample. the This amount was nearly (~0.7 the mg) amount of His-tagged (~0.7 mg) KaiC sufficientof His-tagged to measure KaiC its sufficient ATPase activityto measure at fourits ATPase different activity temperatures at four different (Step temperatures VI in Figure (Step1a) VI [10 ]. in Figure 1a) [10].
(a)
(c) (b)
FigureFigure 1. Optimized 1. Optimized schemes schemes for for expression, expression, purification, purification, and ATPase-assay and ATPase-assay of hexa-histidine of hexa-histidine (His)- (His)-taggedtagged KaiC. KaiC. (a)) Overview of of the the protocol. protocol. (b) ( b Sodium) Sodium dodecyl dodecyl sulfate-polyacrylamide sulfate-polyacrylamide gel gel electrophoresiselectrophoresis (SDS-PAGE) (SDS-PAGE) analysis analysis of His-tagged of His-tagged KaiC KaiC during during its its purification. purification. LaneLane 1, molecular mass mass makers; lane 2, the pre-purified glutathione S-transferase (GST)-tagged KaiC; lane 3, the makers; lane 2, the pre-purified glutathione S-transferase (GST)-tagged KaiC; lane 3, the supernatant supernatant fraction after sonication; lane 4, the pellet fraction after sonication; lane 5, the proteins fractionenriched after sonication; on Ni-Separose lane affinity 4, the pellet column fraction and then after eluted; sonication; lane 6; the lane main 5, the fraction proteins eluted enriched from on Ni-Separose affinity column and then eluted; lane 6; the main fraction eluted from Resource Q column;
lane 7, the main fraction separated by Superdex 200 column. (c) Pseudo-parallel treatment of three sets of two KaiC mutants to achieve six KaiC mutants per week. Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 4 of 12
Resource Q column; lane 7, the main fraction separated by Superdex 200 column. (c) Pseudo-parallel Int. J.treatment Mol. Sci. 2019 of, three20, 2789 sets of two KaiC mutants to achieve six KaiC mutants per week. 4 of 12 2.2. Development of the HPLC System Equipped with a Multi-Channel High-Precision Temperature Controller 2.2. Development of the HPLC System Equipped with a Multi-Channel High-Precision Temperature Controller ATPases are enzymes catalyzing the hydrolysis of ATP into adenosine diphosphate (ADP) and phosphateATPases (Pi). are Time enzymes courses catalyzing of ADP theproduction hydrolysis have of ATPbeen intomeasured adenosine thus diphosphate far using conventional (ADP) and columnsphosphate and (Pi). HPLC Time to coursesevaluate of the ADP steady-state production ATPase have beenactivity measured of KaiC thus[10,17]. far Because using conventional most of the commercialcolumns and HPLCs HPLC toare evaluate equipped the with steady-state just one ATPase sample activity plate of thermostated KaiC [10,17 ]. at Because one particular most of temperature,the commercial multiple HPLCs rounds are equipped of additional with experiments just one sample at different plate thermostatedtemperatures atwere one required particular to estimatetemperature, Q10 values. multiple rounds of additional experiments at different temperatures were required to estimateTo saveQ10 values. this time cost, we developed a set of four sample tables, each thermostated independentlyTo save this at timedifferent cost, wetemperatures, developed aand set ofinstalled four sample it into tables, a commercial each thermostated HPLC (Figure independently 2a). Nine sampleat different holders temperatures, were fabricated and installed on each it sample into a commercial table. A Pt100 HPLC sensor (Figure was2a). directly Nine sampleinserted holders into a water-filledwere fabricated reference on each vial sample placed table. at A one Pt100 of sensor the nine was sample directly holders inserted so into that a water-filled the samples reference were maintainedvial placed at at the one target of the temperature nine sample with holders a precision so that of the ± 0.01 samples °C (Figure were maintained2b). This standard at the targetsetup allowedtemperature the simultaneous with a precision analysis of 0.01of (maximally)C (Figure2 eightb). This different standard samples/references setup allowed the at simultaneousfour different ± ◦ temperatures,analysis of (maximally) saving the eighttime required different for samples reasonable/references estimations at four of di thefferent Q10 values temperatures, while maintaining saving the experimentaltime required quality. for reasonable estimations of the Q10 values while maintaining experimental quality.
(a) (b)
Figure 2. High-performanceHigh-performance liquid liquid chromatography chromatography (HPLC) (HPLC) system system equipped with four-channel temperaturetemperature controller. controller. ( (aa)) Photograph Photograph of of an an HPLC HPLC equipped equipped with with four four sample tables controlled independentlyindependently at at different different temperatures. ( (bb)) Schematic Schematic top top and and side side views views of of the the four four sample sample tables, tables, eacheach carrying nine sample holders thermostated by a Peltier device. 2.3. Throughput, Performance, and Reproducibility of the Protocol 2.3. Throughput, Performance, and Reproducibility of the Protocol Our experiences working along with the present protocol suggest that a better throughput is Our experiences working along with the present protocol suggest that a better throughput is expected by expressing, purifying, and analyzing three sets of two KaiC mutants in a pseudo-parallel expected by expressing, purifying, and analyzing three sets of two KaiC mutants in a pseudo-parallel manner (Figure1c). One factor limiting further parallel processing by one worker is the requirement manner (Figure 1c). One factor limiting further parallel processing by one worker is the requirement for His-tagged KaiC to be separated immediately from imidazole using the open column. The other for His-tagged KaiC to be separated immediately from imidazole using the open column. The other is a preventive measure against unexpected actions of mistaking one sample for another. Even with is a preventive measure against unexpected actions of mistaking one sample for another. Even with these practical limitations, our method permits a six-fold throughput (six KaiC mutants per week) these practical limitations, our method permits a six-fold throughput (six KaiC mutants per week) for analyzing the ATPase activities at four different temperatures as compared to the conventional for analyzing the ATPase activities at four different temperatures as compared to the conventional analysis of GST-tagged KaiC with ordinary HPLC (just under one KaiC mutant per week) [10,17]. analysis of GST-tagged KaiC with ordinary HPLC (just under one KaiC mutant per week) [10,17]. To assess the impact of single-alanine substitutions, we estimated a distribution of relative ATPase To assess the impact of single-alanine substitutions, we estimated a distribution of relative activities (Figure3a) using a preliminary dataset for 86 KaiC mutants prepared /analyzed along with ATPase activities (Figure 3a) using a preliminary dataset for 86 KaiC mutants prepared/analyzed the present protocol. As expected, the distribution included the KaiC mutants carrying activated and along with the present protocol. As expected, the distribution included the KaiC mutants carrying inactivated ATPases. activated and inactivated ATPases. Reproducibility is another important factor required for screening processes. For validation Reproducibility is another important factor required for screening processes. For validation purposes, we chose three KaiC mutants (KaiC-P37A, KaiC-T228A, and KaiC-C306A) from the purposes, we chose three KaiC mutants (KaiC-P37A, KaiC-T228A, and KaiC-C306A) from the preliminary dataset and estimated those ATPase activities using three independent preparations. preliminary dataset and estimated those ATPase activities using three independent preparations. Total errors (SD) arising during the expression, the purification, and the ATPase assay were 7~14% Total errors (SD) arising during the expression, the purification, and the ATPase assay were 7~14% (Table1), well below the di fference in the ATPase activities between the hyper-activated and the hyper-inactivated mutants (Figure3a). Furthermore, most of the 86 KaiC mutants could be Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 5 of 12 Int. J. Mol. Sci. 2019, 20, 2789 5 of 12 (Table 1), well below the difference in the ATPase activities between the hyper-activated and the expressedhyper-inactivated and purified mutants successfully (Figure 3a). through Furthermore, the optimized most of protocolthe 86 KaiC except mutants for eight could mutants be expressed with poorand expression purified successfully (Figure1a), yielding, through on the average, optimized 2.4 protocol1.1 mg (per except 400 mL for of eight terrific mutants broth medium) with poor of ± His-taggedexpression KaiC(Figure mutants 1a), yielding, sufficient on to assayaverage, both 2.4 ATPase ± 1.1 activitymg (per and 400 phosphorylation mL of terrific broth cycle medium) (Figure3b). of TheseHis-tagged observations KaiC mutants indicate sufficient that the combined to assay both usage ATPase of the activity present protocoland phosphorylation and the developed cycle (Figure HPLC system3b). These provides observations the opportunity indicate of that screening the combined the ATPase usage mutants of the ofpresent KaiC withprotocol reasonable and the throughput, developed performance,HPLC system and provides reproducibility. the opportunity of screening the ATPase mutants of KaiC with reasonable throughput, performance, and reproducibility.
(a) (b)
FigureFigure 3. 3.Preliminary Preliminary datasetdataset forfor 8686 KaiCKaiC mutants.mutants. ((aa)) DistributionDistribution ofof thethe relativerelative ATPaseATPase activities.activities. (b(b)) Distribution Distribution of of the the final final yields yields (per (per 400 400 mL ofmL terrific of terrific broth broth medium) medium) of His-tagged of His-tagged KaiC. Mean KaiC. values Mean forvalues KaiC-WT, for KaiC-WT, KaiC-P37A, KaiC-P37A, KaiC-T228A, KaiC-T228A, and KaiC-C306A and KaiC-C306A are represented are represented as blue, green, as blue, magenta, green, andmagenta, red bars, and respectively, red bars, respectively, with standard with deviation standard from deviation three independentfrom three independent experiments. experiments.
Table 1. Temperature dependency of ATPase activity of KaiC, period length of in vitro post-translational Table 1. Temperature dependency of ATPase activity of KaiC, period length of in vitro post- oscillations (PTO), and period length of in vivo transcriptional and translational oscillations (TTO). translational oscillations (PTO), and period length of in vivo transcriptional and translational oscillations (TTO). In Vitro Phosphorylation Cycle In Vivo Bioluminescence KaiC In Vitro ATPase Activity 1,2 1,3 Rhythm 1 In vitro Phosphorylation In vivo Bioluminescence KaiC In vitro(d 1) 4ATPase ActivityQ 1,2 Period (h) 4 Q Period (h) 5 Q 6 − 10 Cycle 1,3 10 Rhythm 1 10 1.05 0.01 WT 13.4 (d−1)0.9 4 1.01 Q0.0410 Period 23.1 0.3 (h) 4 1.09Q100.00 Period 24.8 0.2(h) 5 Q±10 6 ± ± ± ± ± (0.99 0.01) ± 1.051.01 ± 0.010.01 *** P37AWT 13.4 12.9 ± 0.91.6 * 1.03 1.01 ±0.02 0.04 * 24.6 23.1 ±0.3 0.3 *** 1.08 1.09 ±0.00 0.00 *** 28.1 24.8 ±0.1 0.2 **** ± ± ± ± ± ± (0.93 0.01 ***) (0.99 ±± 0.01) 1.17 0.03 ** T228A 9.8 0.9 *** 1.18 0.06 ** 30.2 0.4 **** 1.15 0.01 *** 30.2 0.2 **** 1.01 ±± 0.01*** P37A 12.9± ± 1.6* 1.03± ± 0.02* 24.6± ± 0.3*** 1.08± ± 0.00*** 28.1± ± 0.1**** (0.94 0.03 *) ± (0.931.09 ±0.01 0.01 ******) C306A 15.9 2.2 * 1.14 0.05 ** 19.6 0.1 **** 1.14 0.02 ** 20.8 0.0 **** ± ± ± ± ± ± (1.051.17 ±0.01 0.03 ****)** T228A 9.8 ± 0.9*** 1.18 ± 0.06** 30.2 ± 0.4**** 1.15 ± 0.01*** 30.2 ± 0.2**** ± 1 Results are shown as the mean standard deviation from three independent preparations and measurements.(0.94 ± 0.03*) 2 ± 3 Steady-state ATPase activity of KaiC alone in the absence of both KaiA and KaiB. Phosphorylation rhythms of *** 4 5 1.09 ± 0.01 C306AKaiC and15.9 its mutants± 2.2* in the 1.14 presence ± 0.05 of** both 19.6 KaiA ± and0.1**** KaiB. 1.14Measurements ± 0.02** at 30 20.8◦C. ± Measurements0.0**** at 29 ◦C. 6 **** Estimates using data taken at 25 and 29 ◦C. Values in parenthesis correspond to estimates using data(1.05 taken ± at 0.01 25, ) 1 Results29, and are 35 shown◦C. T-tests as the against mean KaiC-WT: ± standard **** significantdeviation ( pfrom< 0.001), three *** independent significant (p
Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 6 of 12
Int. J. Mol. Sci. 2019, 20, 2789 6 of 12
40 ◦C (Figure4d). KaiC-T228A revealed similar or more pronounced shortening of the period length from 33.2 h at 25◦C to 26.8 h at 35◦C (Figure4c). These observations support the e ffectiveness of in vitro ATPase-based screening in identifying KaiC mutations affecting the period length and/or its temperature compensation in PTO.
(a) (b)
(c) (d)
FigureFigure 4. 4.Phosphorylation Phosphorylation/dephosph/dephosphorylationorylationcycles cycles ofof ((aa)) KaiC-WT,KaiC-WT, ( (bb)) KaiC-P37A, KaiC-P37A, ((cc)) KaiC-T228A,KaiC-T228A, andand ( d(d)) KaiC-C306A KaiC-C306A at at di differentfferent temperatures. temperatures. Blue,Blue, green,green, orange,orange, andand redred markers markers correspond correspond to to thethe fractions fractions of of phosphorylated phosphorylated KaiC KaiC (f (phosfphos)) atat 25,25, 30,30, 35,35, andand 4040 ◦°C,C, respectively.respectively. Dark-coloredDark-coloredcircles circles correspondcorrespond toto thethe meanmean fromfrom independent independent preparationspreparations andand measurementsmeasurements (pale-colored(pale-colored squares,squares, triangles,triangles, andand diamonds).diamonds). ThinThin lineslines representrepresent thethe fittingfitting ofof thethe meanmean toto EquationEquation (1)(1) describeddescribed inin MaterialsMaterials and and Methods. Methods. Int. J. Mol. Sci. 2019, 20, 2789 7 of 12 Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 7 of 12 Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 7 of 12
(a) (b) (a) (b)
Figure 5. Cross-correlationalCross-correlational plots ofof ((aa)) inin vivovivo frequencyfrequency ofof bioluminescencebioluminescence rhythms rhythms ( f(fTTOTTO), in vitro Figure 5. Cross-correlational plots of (a) in vivo frequency of bioluminescence rhythms (fTTO), in vitro frequency of KaiC phosphorylation cycles ((ffPTO),), and and inin vitro vitro ATPaseATPase activity activity of of KaiC KaiC alone alone at at 30 °C,◦C, frequency of KaiC phosphorylation cycles (fPTO), and in vitro ATPase activity of KaiC alone at 30 °C, and of (b) the Q1010 values for bioluminescence rhythmsrhythms( (QQ10,TTO10,TTO,, 25~29 25~29 °C),◦C), KaiC KaiC phosphorylation phosphorylation cycles and of (b) the Q10 values for bioluminescence rhythms (Q10,TTO, 25~29 °C), KaiC phosphorylation cycles (Q10,PTO, ,25~35 25~35 °C),◦C), and and inin vitro vitro ATPase activityactivity ofof KaiCKaiC alonealone (Q(Q10,ATPase10,ATPase, ,25~35 25~35 °C).◦C). Filled Filled circles, (Q10,PTO, 25~35 °C), and in vitro ATPase activity of KaiC alone (Q10,ATPase, 25~35 °C). Filled circles, triangles, squares,squares, and and diamonds diamonds correspond correspond to KaiC-WT,to KaiC-WT, KaiC-P37A, KaiC-P37A, KaiC-T228A, KaiC-T228A, and KaiC-C306A, and KaiC- triangles, squares, and diamonds correspond to KaiC-WT, KaiC-P37A, KaiC-T228A, and KaiC- C306A,respectively. respectively. Red and Red blue and markers blue correspondmarkers correspon to in vitrod to(PTO) in vitro and (PTO)in vivo and(TTO) in data,vivo respectively.(TTO) data, C306A, respectively. Red and blue markers correspond to in vitro (PTO) and in vivo (TTO) data, Thickrespectively. lines represent Thick lines linear represent fits to the linear data fits with to the correlation data with coe correlatifficientson (r). coefficients (r). respectively. Thick lines represent linear fits to the data with correlation coefficients (r). To examine the correspondence to in vivovivo clockclock properties,properties, wewe constructedconstructed a bioluminescentbioluminescent To examine the correspondence to in vivo clock properties, we constructed a bioluminescent cyanobacterial reporter reporter strain strain [7,9] [7,9 carrying] carrying each each of the of theKaiC-P37A, KaiC-P37A, the KaiC-T228A, the KaiC-T228A, and the and KaiC- the cyanobacterial reporter strain [7,9] carrying each of the KaiC-P37A, the KaiC-T228A, and the KaiC- C306AKaiC-C306A substitutions substitutions (Figure (Figure 6). As 6shown). As shown in Table in 1, Table the period1, the period length lengthof the bioluminescence of the bioluminescence rhythm C306A substitutions (Figure 6). As shown in Table 1, the period length of the bioluminescence rhythm wasrhythm nearly was matched nearly matched to that toof thatin vitro of in PTO vitro asPTO was as observed was observed previously previously [9]. Interestingly, [9]. Interestingly, the was nearly matched to that of in vitro PTO as was observed previously [9]. Interestingly, the bioluminescencethe bioluminescence rhythm rhythm of the of the strain strain carrying carrying the the KaiC-C306A KaiC-C306A mutation mutation was was gradually gradually accelerated accelerated bioluminescence rhythm of the strain carrying the KaiC-C306A mutation was gradually accelerated in aa temperature-dependent temperature-dependen mannert manner (Figure (Figure6d). The6d). QThe10 value Q10 ofvalue the KaiC-C306Aof the KaiC-C306A strain was strain estimated was in a temperature-dependent manner (Figure 6d). The Q10 value of the KaiC-C306A strain was estimatedusing the profilesusing the taken profiles at 25 taken and 29 at◦ C25 to and be 1.0929 °C to0.01 be and1.09was ± 0.01 slightly and was smaller slightly than smaller that (1.14 than 0.02)that estimated using the profiles taken at 25 and 29 °C± to be 1.09 ± 0.01 and was slightly smaller than± that (1.14of in ± vitro 0.02)PTO of in (t-test: vitro PTOp = (t-test:0.027). p It = should0.027). It be should noted, be however, noted, however, that the thatQ10 thevalue Q10 of value KaiC-C306A of KaiC- (1.14 ± 0.02) of in vitro PTO (t-test: p = 0.027). It should be noted, however, that the Q10 value of KaiC- wasC306A always was always larger thanlarger that than of that KaiC-WT of KaiC-WT in both in casesboth cases of PTO of (t-test:PTO (t-test:p = 0.031)p = 0.031) and and TTO TTO (t-test: (t- C306A was always larger than that of KaiC-WT in both cases of PTO (t-test: p = 0.031) and TTO (t- test:p = 0.006) p = 0.006) (Table (Table1). Although 1). Although the period the lengthperiod for length the KaiC-T228A for the KaiC-T228A strain was strain unexpectedly was unexpectedly prolonged test: p = 0.006) (Table 1). Although the period length for the KaiC-T228A strain was unexpectedly prolongedfrom 30.2 from0.2 to30.2 32.9 ± 0.2 0.1to 32.9 by raising± 0.1 by the raising temperature the temperature from 29 from to 35 29◦C to (Figure 35 °C (Figure6c), its 6c),Q10 itsvalue Q10 prolonged± from 30.2 ± 0.2± to 32.9 ± 0.1 by raising the temperature from 29 to 35 °C (Figure 6c), its Q10 (1.17value (1.170.03) ± calculated0.03) calculated using using the period the period lengths lengths at 25 and at 25 29 and◦C was29 °C consistent was consistent with the within the vitro inresult vitro value± (1.17 ± 0.03) calculated using the period lengths at 25 and 29 °C was consistent with the in vitro (resultQ10 = (1.15Q10 = 1.150.01) ± (t-test:0.01) (t-test:p = 0.321) p = 0.321) (Table (Table1). A similar1). A similar tendency tendency was also was observed also observed for KaiC-P37A for KaiC- result (Q10 ±= 1.15 ± 0.01) (t-test: p = 0.321) (Table 1). A similar tendency was also observed for KaiC- P37Afor some for unknownsome unknown reasons reasons (Figure (Figure6b). While 6b). While the consistency the consistency of the ofQ the10 values Q10 values between between PTO PTO and P37A for some unknown reasons (Figure 6b). While the consistency of the Q10 values between PTO TTOand TTO can becan influenced be influenced by the by temperature the temperature range range used used for evaluation for evaluation (Table (Table1), in 1), vitro in ATPase-basedvitro ATPase- and TTO can be influenced by the temperature range used for evaluation (Table 1), in vitro ATPase- basedscreening screening is, in principle, is, in principle, beneficial bene forficial studying for studying both PTO both and PTO TTO. and TTO. based screening is, in principle, beneficial for studying both PTO and TTO.
(a) (b) (a) (b)
Figure 6. Cont.
Int. J. Mol. Sci. 2019, 20, 2789 8 of 12 Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 8 of 12
(c) (d)
FigureFigure 6.6. InIn vivovivobioluminescence bioluminescence rhythmsrhythms ofof ((aa)) KaiC-WT,KaiC-WT, ((bb)) KaiC-P37A,KaiC-P37A, ((cc)) KaiC-T228A,KaiC-T228A, andand ((dd)) KaiC-C306AKaiC-C306A atat didifferentfferent temperatures. temperatures. Blue,Blue, green,green, andand redred circlescircles correspondcorrespond toto profilesprofiles recordedrecorded atat 25,25, 29,29, and and 35 35◦ C,°C, respectively. respectively. For For clarity clarity of of comparison, comparison, each each profile profile is shifted is shifted horizontally horizontally so that so that the peakthe peak matches matches 0 h. 0 h.
3. Discussion 3. Discussion In this paper, the effectiveness of in vitro ATPase-based screening was evaluated in terms of its In this paper, the effectiveness of in vitro ATPase-based screening was evaluated in terms of its throughput, performance, and reproducibility. The present throughput of six KaiC mutants per week throughput, performance, and reproducibility. The present throughput of six KaiC mutants per week allows us to estimate the time required for completing the whole-KaiC scanning mutagenesis to be allows us to estimate the time required for completing the whole-KaiC scanning mutagenesis to be approximately 1.6 years. The in vivo bioluminescence screening system [7,9] may provide an equivalent approximately 1.6 years. The in vivo bioluminescence screening system [7,9] may provide an or even higher throughput given the availability of the multiple apparatuses placed in independently equivalent or even higher throughput given the availability of the multiple apparatuses placed in temperature-controlledindependently temperature-controlled rooms. The developed rooms. The set ofdeveloped four sample set of tables four issample compact tables enough is compact to be installedenough to to be the installed sample to compartment the sample compartment of ordinary HPLC of ordinary systems, HPLC providing systems, a moderate-throughputproviding a moderate- screeningthroughput of thescreening KaiC clock of the mutants KaiC clock with mutants a reasonable with setupa reasonable cost and setup space. cost and space. TheThe resultsresults ofof thethe threethree representativerepresentative mutantsmutants supportsupport thethe performanceperformance ofof ourour ATPase-basedATPase-based screeningscreening system.system. AsAs shownshown inin TableTable1 1and and Figure Figure5, 5, the the KaiC KaiC mutations mutations showing showing lower lower and and higher higher ATPaseATPase activities activities were were demonstrated demonstrated to to give give longer long ander and shorter shorter periods, periods, respectively, respectively, in both in both PTO PTO and TTO.and TTO. This correspondenceThis correspondence means means that the that ATPase the ATPase activity activity is the excellent is the excellent measure measure of the period of the lengthperiod inlength the cyanobacterial in the cyanobacterial circadian circadian clock. clock. It deserves to be noted that there was an effect of KaiA and/or KaiB on the Q10 values in PTO. It deserves to be noted that there was an effect of KaiA and/or KaiB on the Q10 values in PTO. As As shown in Table1, the Q10 values of PTO (1.09 0.00) for KaiC-WT were somewhat larger than that shown in Table 1, the Q10 values of PTO (1.09 ± 0.00)± for KaiC-WT were somewhat larger than that of of ATPase (1.01 0.04) (t-test: p = 0.062). The similar tendency was also confirmed in KaiC-P37A (t-test: ATPase (1.01 ± ±0.04) (t-test: p = 0.062). The similar tendency was also confirmed in KaiC-P37A (t-test: p = 0.067). These observations may be explained by the fact that the Q10 value of ATPase becomes p = 0.067). These observations may be explained by the fact that the Q10 value of ATPase becomes slightlyslightly a ffaffectedected by by temperature temperature when when both KaiAboth andKaiA KaiB and coexist KaiB [10coexist]. The less[10]. significant The less correlationsignificant of Q10,ATPase with Q10,TTO than with Q10,PTO (Figure5b) may suggest a pronounced KaiA /KaiB-related correlation of Q10,ATPase with Q10,TTO than with Q10,PTO (Figure 5b) may suggest a pronounced eKaiA/KaiB-relatedffect in vivo. It is surprising effect in vivo. that, givenIt is surprising this potential that, KaiA given/KaiB-related this potential effect, KaiA/KaiB-related our system is sensitive effect, enoughour system to isolate is sensitive the mutations enough (KaiC-T228Ato isolate the andmutations KaiC-C306A), (KaiC-T228A resulting and in KaiC-C306A), slight changes resulting of the Q 10in value by ~0.1 unit. slight changes of the Q10 value by ~ 0.1 unit. TheThe temperaturetemperature dependencedependence ofof amplitudesamplitudes oftenoften providesprovides cluesclues toto elucidateelucidate oscillatoryoscillatory mechanismsmechanisms behind and and to to construct construct models. models. Intere Interestingly,stingly, the theamplitude amplitude of PTO of PTOwas reduced was reduced as the astemperature the temperature decreased decreased (Figure (Figure4), while4), such while temp sucherature temperature dependence dependence was not wasobvious not for obvious that of forTTO that (Figure of TTO 6). In (Figure the case6). of In cyanobacterial the case of cyanobacterial PTO, the decreased PTO, the amplitude decreased at lower amplitude temperatures at lower is temperaturesexplained by isa explainedtransition by throug a transitionh the Hopf through bifurcation the Hopf [18]. bifurcation The recent [18]. Theobservation recent observation of TTO in ofmammalian TTO in mammalian cultured cells cultured suggests cells a suggests mechanism a mechanism of temperature-amplitude of temperature-amplitude coupling to coupling explain toits explaintemperature-compensated its temperature-compensated period but period temperatur but temperature-sensitivee-sensitive amplitudes amplitudes [19]. [19Although]. Although the thereduced/enhanced reduced/enhanced amplitudes amplitudes cannot cannot be be targeted targeted di directlyrectly by by our our ATPase-based ATPase-based inin vitro screening,screening, thethe temperature-dependenttemperature-dependent KaiC KaiC mutants mutants screened screened by by our our protocol protocol may may constitute constitute another another library library toto searchsearch forfor mutantsmutants exhibiting exhibiting an an irregular irregular temperature temperature dependence dependence of of the the PTO PTO/TTO/TTO amplitudes. amplitudes. While two (KaiC-P37A and KaiC-T228A) of the three mutations characterized here reside in the N-terminal domain of KaiC, one (KaiC-C306A) locates in its C-terminal domain. The N-terminal domain of KaiC is the domain mostly responsible for ATPase [10,13]. According to the crystal
Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 9 of 12
structure of KaiC [20], Pro37 and Thr228 are close to an ATP molecule bound in the N-terminal domain with 12 and 4 Å distances, respectively (Figure 7). We infer from these observations that both KaiC-P37A and KaiC-T228A mutations directly affect the structure of the active site of ATPase in the Int.N-terminal J. Mol. Sci. 2019domain., 20, 2789 On the other hand, Cys306 is distant from the active site by over 34 Å.9 ofThe 12 present screening is robust enough to identify a KaiC mutation affecting the ATPase activity indirectly through allosteric interactions, providing a practical means to uncover the regulatory While two (KaiC-P37A and KaiC-T228A) of the three mutations characterized here reside in the mechanisms of ATPase in KaiC. N-terminal domain of KaiC, one (KaiC-C306A) locates in its C-terminal domain. The N-terminal While the whole-KaiC alanine scanning is under way, there is every reason to be optimistic in domain of KaiC is the domain mostly responsible for ATPase [10,13]. According to the crystal structure screening the clock mutants of KaiC using the present method. This is because roughly half of the of KaiC [20], Pro37 and Thr228 are close to an ATP molecule bound in the N-terminal domain with preliminary dataset thus far examined revealed the changes in the period length and/or its 12 and 4 Å distances, respectively (Figure7). We infer from these observations that both KaiC-P37A temperature-dependency (Figure 3a). Further modulations of the period length and/or its and KaiC-T228A mutations directly affect the structure of the active site of ATPase in the N-terminal temperature-dependency can be expected by mutating a critical residue identified through the domain. On the other hand, Cys306 is distant from the active site by over 34 Å. The present screening present protocol into residues other than alanine. The developed ATPase-based screening is a is robust enough to identify a KaiC mutation affecting the ATPase activity indirectly through allosteric powerful tool for comprehensive mapping of the structural units regulating slow but temperature- interactions, providing a practical means to uncover the regulatory mechanisms of ATPase in KaiC. compensated ATPase of KaiC.
Figure 7. Three representative KaiC mutations (Pro37, Thr228, and Cys306) mapped on the crystal Figure 7. Three representative KaiC mutations (Pro37, Thr228, and Cys306) mapped on the crystal structure of wild-type KaiC carrying no alanine substitutions (accession code: 2GBL) [20]. Shown are structure of wild-type KaiC carrying no alanine substitutions (accession code: 2GBL) [20]. Shown are distances to the nearest ATP in the N-terminal domain. distances to the nearest ATP in the N-terminal domain. While the whole-KaiC alanine scanning is under way, there is every reason to be optimistic in4. Materials screening and the Methods clock mutants of KaiC using the present method. This is because roughly half of the preliminary dataset thus far examined revealed the changes in the period length and/or 4.1. Expressions and Purifications its temperature-dependency (Figure3a). Further modulations of the period length and /or its temperature-dependencyTransformed E. coli cancells bewere expected pre-cultured by mutating in 10 mL a critical of Luria-Bertani residue identified (LB) broth through medium the presentcontaining protocol 0.04 intomg/mL residues ampicillin other thanand alanine.0.034 mg/mL The developed chloramphenicol ATPase-based at 37 screening°C for 16 is h a powerfuland then tooltransferred for comprehensive to 400 mL of mapping terrific ofbroth the structuralmedium co unitsntaining regulating 0.04 mg/mL slow but ampicillin. temperature-compensated Bacterial growth ATPasewas monitored of KaiC. with OD600 every 30 min prior to isopropyl β-D-thiogalactopyranoside (IPTG) induction. After the culture reached OD600 of 1.5 (approximately 3 h), over-expression of His-tagged 4.KaiC Materials was induced and Methods by adding IPTG to a final concentration of 0.1 mM. The cells were further harvested for approximately 5 h until OD600 reached ~6. The OD600 was measured after 10-fold 4.1. Expressions and Purifications dilution of the culture medium. The harvested cells were collected by centrifugation and then kept at −20 °CTransformed until use. E. coli cells were pre-cultured in 10 mL of Luria-Bertani (LB) broth medium containing 0.04 mgMultiple/mL ampicillin open columns and 0.034 filled mg with/mL Ni-Sepharose chloramphenicol resin atand 37 PD-10◦C for (GE 16 h Healthcare, and then transferred Chicago, IL, to 400US) mLwere of used terrific to purify broth medium multiple containing KaiC mutants 0.04 simultaneously. mg/mL ampicillin. Both Bacterial KaiA and growth KaiB were was monitored expressed withas GST-tagged OD600 every forms 30 minand priorpurified to isopropylafter the cleavageβ-d-thiogalactopyranoside of the GST-tag, as described (IPTG) induction. previously After [21]. the culture reached OD600 of 1.5 (approximately 3 h), over-expression of His-tagged KaiC was induced by adding4.2. Development IPTG to a finalof Temperature-Controlled concentration of 0.1 mM. Sample The cellsTables were further harvested for approximately 5 h until OD reached ~6. The OD was measured after 10-fold dilution of the culture medium. An aluminum600 sample-holder produced600 to achieve the best fit to sample vials (Hitachi, Tokyo, The harvested cells were collected by centrifugation and then kept at 20 C until use. Japan) was placed on an assembly of a Peltier device, a cooling fin, and− a fan◦ (Alpha, Shizuoka, Japan). Multiple open columns filled with Ni-Sepharose resin and PD-10 (GE Healthcare, Chicago, IL, The sample temperature in each sample table was monitored directly with a Pt100 sensor and US) were used to purify multiple KaiC mutants simultaneously. Both KaiA and KaiB were expressed controlled by a high-precision temperature controller (CELL System, Kanagawa, Japan). as GST-tagged forms and purified after the cleavage of the GST-tag, as described previously [21].
Int. J. Mol. Sci. 2019, 20, 2789 10 of 12
4.2. Development of Temperature-Controlled Sample Tables An aluminum sample-holder produced to achieve the best fit to sample vials (Hitachi, Tokyo, Japan) was placed on an assembly of a Peltier device, a cooling fin, and a fan (Alpha, Shizuoka, Japan). The sample temperature in each sample table was monitored directly with a Pt100 sensor and controlled by a high-precision temperature controller (CELL System, Kanagawa, Japan).
4.3. ATPase Measurements ATPase measurements were conducted routinely by using an HPLC system (EXTREMA, JASCO, Tokyo, Japan) equipped with the four sample tables each thermostated independently at 24, 27, 30, and 35 C. ADP was separated from ATP on an Inert Sustain C18 column (2 µm, 3.0 50 mm) ◦ × (GL Sciences, Tokyo, Japan) at a flow rate of 0.4 mL min-1. After the measurement, the ADP concentrations were calculated from their peak areas as reported previously [10,17]. ATPase activity was determined from a slope of a linear fit to time-evolution of the produced ADP. The activity was defined as the number of ADP produced by one KaiC molecule in a monomer unit per day.
4.4. KaiC Phosphorylation/Dephosphorylation Cycle Phosphorylation/dephosphorylation cycles were reconstructed in vitro by mixing KaiC or its mutants with both KaiA and KaiB in the presence of ATP at 25, 30, and 35 ◦C[9]. A fraction of the phosphorylated KaiC (f phos) at each time point was analyzed by using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) [21] and LOUPE [22]. The period lengths were estimated by fitting f phos values to the following equation:
t + Φ f = B + A cos 2π (1) phos P where A is amplitude, B is baseline, t is incubation time, Φ is phase, and P is period. Q10 values were determined from the slope of an Arrhenius plot of cycle frequency as described previously [23].
4.5. In Vivo Bioluminsescence Assay The bioluminescence assays were conducted as previously reported [7,9]. The KaiC open reading frame in pCkaiABC [7] was mutagenized using QuickChange II Site-Directed Mutagenesis Kits (Agilent Technologies, Santa Clara, CA, US) to change P37, T228, and C306. Cyanobacterial cells with the kaiBC-reporter cassette were pre-cultured for 3 days at 30 ◦C on BG-11 solid medium under constant light (LL) conditions [45 µE m-2 S-1 from light emitting diode daylight lamp]. After a dark treatment for 12 h, cells were then transferred to LL conditions at 25, 29, or 35 ◦C. The bioluminescence from the cells was recorded using a photomultiplier tube detector at 25, 29, and 35 ◦C.
Author Contributions: D.O., T.K., and S.A. conceived and designed the study; D.O. conducted expressions, purifications, and ATPase assays. D.O., Y.F., and K.I.-M. conducted in vivo experiments; D.O., A.M., Y.F., and S.A. analyzed the data; D.O. and S.A. prepared the manuscript with inputs from all authors. Funding: This research was supported by JSPS KAKENHI Grant Numbers JP17H06165, JP18K06171, and JP16K14685. Acknowledgments: We thank Tomoe Nishikawa for her technical support for the bioluminescence assays. Conflicts of Interest: The authors declare no conflict of interest.
Abbreviations
HPLC high-performance liquid chromatography ATP adenosine triphosphate ADP adenosine diphosphate PTO post-translational oscillation TTO transcriptional and translational oscillation Int. J. Mol. Sci. 2019, 20, 2789 11 of 12
References
1. Pittendrigh, C.S. Temporal organization: Reflections of a Darwinian clock-watcher. Annu. Rev. Physiol. 1993, 55, 16–54. [CrossRef][PubMed] 2. Reppert, S.M.; Weaver, D.R. Molecular analysis of mammalian circadian rhythms. Annu. Rev. Physiol. 2001, 63, 647–676. [CrossRef][PubMed] 3. Johnson, C.H.; Golden, S.S.; Ishiura, M.; Kondo, T. Circadian clocks in prokaryotes. Mol. Microbiol. 1996, 21, 5–11. [CrossRef][PubMed] 4. Johnson, C.H.; Elliott, J.; Foster, R.; Honma, K.; Kronauer, R. Fundamental Properties of Circadian Rhythms. In Chronobiology: Biological Timekeeping; Dunlap, J.C., Loros, J.J., DeCoursey, P.J., Eds.; Sinauer Associates, Inc. Publishers: Sunderland, MA, USA, 2004; Volume 3, pp. 67–105. ISBN -13. 5. Young, M.W.; Kay, S.A. Time zones: A comparative genetics of circadian clocks. Nat. Rev. Genet. 2001, 2, 702–715. [CrossRef] 6. O’Neill, J.S.; van Ooijen, G.; Dixon, L.E.; Troein, C.; Corellou, F.; Bouget, F.Y.; Reddy, A.B.; Millar, A.J. Circadian rhythms persist without transcription in a eukaryote. Nature 2011, 469, 554–558. [CrossRef] [PubMed] 7. Ishiura, M.; Kutsuna, S.; Aoki, S.; Iwasaki, H.; Andersson, C.R.; Tanabe, A.; Golden, S.S.; Johnson, C.H.; Kondo, T. Expression of a gene cluster kaiABC as a circadian feedback process in cyanobacteria. Science 1998, 281, 1519–1523. [CrossRef] 8. Tomita, J.; Nakajima, M.; Kondo, T.; Iwasaki, H. No transcription-translation feedback in circadian rhythm of KaiC phosphorylation. Science 2005, 307, 251–254. [CrossRef] 9. Nakajima, M.; Imai, K.; Ito, H.; Nishiwaki, T.; Murayama, Y.; Iwasaki, H.; Oyama, T.; Kondo, T. Reconstitution of circadian oscillation of cyanobacterial KaiC phosphorylation in vitro. Science 2005, 308, 414–415. [CrossRef] 10. Terauchi, K.; Kitayama, Y.; Nishiwaki, T.; Miwa, K.; Murayama, Y.; Oyama, T.; Kondo, T. ATPase activity of KaiC determines the basic timing for circadian clock of cyanobacteria. Proc. Natl. Acad. Sci USA 2007, 104, 16377–16381. [CrossRef] 11. Kageyama, H.; Nishiwaki, T.; Nakajima, M.; Iwasaki, H.; Oyama, T.; Kondo, T. Cyanobacterial circadian pacemaker: Kai protein complex dynamics in the KaiC phosphorylation cycle in vitro. Mol. Cell 2006, 23, 161–171. [CrossRef] 12. Akiyama, S.; Nohara, A.; Ito, K.; Maeda, Y. Assembly and disassembly dynamics of the cyanobacterial periodosome. Mol. Cell 2008, 29, 703–716. [CrossRef] 13. Abe, J.; Hiyama, T.B.; Mukaiyama, A.; Son, S.; Mori, T.; Saito, S.; Osako, M.; Wolanin, J.; Yamashita, E.; Kondo, T.; et al. Circadian rhythms. Atomic-scale origins of slowness in the cyanobacterial circadian clock. Science 2015, 349, 312–316. [CrossRef] 14. Petegem, V.F.; Duderstadt, K.E.; Clark, K.A.; Wang, M.; Minor, D.L., Jr. Alanine-scanning mutagenesis defines a conserved energetic hotspot in the CaValpha1 AID-CaVbeta interaction site that is critical for channel modulation. Structure 2008, 16, 280–294. [CrossRef] 15. Samai, P.; Shuman, S. Structure-function analysis of the OB and latch domains of chlorella virus DNA ligase. J. Biol. Chem. 2011, 286, 22642–22652. [CrossRef] 16. Iwasaki, H.; Taniguchi, Y.; Ishiura, M.; Kondo, T. Physical interactions among circadian clock proteins KaiA, KaiB and KaiC in cyanobacteria. EMBO J. 1999, 18, 1137–1145. [CrossRef] 17. Murayama, Y.; Mukaiyama, A.; Imai, K.; Onoue, Y.; Tsunoda, A.; Nohara, A.; Ishida, T.; Maeda, Y.; Terauchi, K.; Kondo, T.; et al. Tracking and visualizing the circadian ticking of the cyanobacterial clock protein KaiC in solution. EMBO J. 2011, 30, 68–78. [CrossRef][PubMed] 18. Murayama, Y.; Kori, H.; Oshima, C.; Kondo, T.; Iwasaki, H.; Ito, H. Low temperature nullifies the circadian clock in cyanobacteria through Hopf bifurcation. Proc. Natl. Acad. Sci. USA 2017, 114, 5641–5646. [CrossRef] 19. Kurosawa, G.; Fujioka, A.; Koinuma, S.; Mochizuki, A.; Shigeyoshi, Y. Temperature–amplitude coupling for stable biological rhythms at different temperatures. PLoS Comput. Biol. 2017, 13, e1005501. [CrossRef] 20. Pattanayek, R.; Wang, J.; Mori, T.; Xu, Y.; Johnson, C.H.; Egli, M. Visualizing a circadian clock protein: Crystal structure of KaiC and functional insights. Mol. Cell 2004, 15, 375–388. [CrossRef] 21. Nishiwaki, T.; Satomi, Y.; Nakajima, M.; Lee, C.; Kiyohara, R.; Kageyama, H.; Kitayama, Y.; Temamoto, M.; Yamaguchi, A.; Hijikata, A.; et al. Role of KaiC phosphorylation in the circadian clock system of Synechococcus elongatus PCC 7942. Proc. Natl. Acad. Sci. USA 2004, 101, 13927–13932. [CrossRef] Int. J. Mol. Sci. 2019, 20, 2789 12 of 12
22. Furuike, Y.; Abe, J.; Mukaiyama, A.; Akiyama, S. Accelerating in vitro studies on circadian clock systems using an automated sampling device. Biophys. Phys. 2016, 13, 235–241. [CrossRef][PubMed] 23. Akiyama, S. Structural and dynamic aspects of protein clocks: How can they be so slow and stable? Cell Mol. Life Sci. 2012, 69, 2147–2160. [CrossRef][PubMed]
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).