Portal de Periódicos da CAPES

Role of the Helical Protrusion in the Conformational Change and Molecular Chaperone Activity of the Archaeal Group II Chaperonin

2004; Elsevier BV; Volume: 279; Issue: 18 Linguagem: Inglês

10.1074/jbc.m400839200

1083-351X

Ryo Iizuka, Sena So, Tomonao Inobe, Takao Yoshida, Tamotsu Zako, Kunihiro Kuwajima, Masafumi Yohda,

Enzyme Structure and Function

To elucidate the exact role of the helical protrusion of a group II chaperonin in its molecular chaperone function, three deletion mutants of the chaperonin from a hyperthermophilic archaeum (Thermococcus sp. strain KS-1) lacking one-third, two-thirds, and the whole of the helical protrusion were constructed. The helical protrusion is thought to be substituted for the co-chaperonin GroES of a group I chaperonin and to be important for binding to unfolded proteins. Protease sensitivity assays and small angle x-ray scattering experiments were performed to demonstrate the conformation change of the wild type protein and the deletion mutants by adenine nucleotides. Whereas the binding of ATP to the wild type protein induced a structural transition corresponding to the closure of the built-in lid, it did not cause significant structural changes in deletion mutants. Although the mutants effectively protected proteins from thermal aggregation, ATP-dependent protein folding ability was remarkably diminished. We conclude that the helical protrusion is not necessarily important for binding to unfolded proteins, but its ATP-dependent conformational change mediates folding of captured unfolded proteins. To elucidate the exact role of the helical protrusion of a group II chaperonin in its molecular chaperone function, three deletion mutants of the chaperonin from a hyperthermophilic archaeum (Thermococcus sp. strain KS-1) lacking one-third, two-thirds, and the whole of the helical protrusion were constructed. The helical protrusion is thought to be substituted for the co-chaperonin GroES of a group I chaperonin and to be important for binding to unfolded proteins. Protease sensitivity assays and small angle x-ray scattering experiments were performed to demonstrate the conformation change of the wild type protein and the deletion mutants by adenine nucleotides. Whereas the binding of ATP to the wild type protein induced a structural transition corresponding to the closure of the built-in lid, it did not cause significant structural changes in deletion mutants. Although the mutants effectively protected proteins from thermal aggregation, ATP-dependent protein folding ability was remarkably diminished. We conclude that the helical protrusion is not necessarily important for binding to unfolded proteins, but its ATP-dependent conformational change mediates folding of captured unfolded proteins. Protein folding is assisted by a number of molecular chaperones in vivo. The chaperonins, a ubiquitous class of molecular chaperones, form double ring complexes that mediate the folding of nascent and denatured proteins in an ATP-dependent manner (1Bukau B. Horwich A.L. Cell. 1998; 92: 351-366Google Scholar, 2Hartl F.U. Hayer-Hartl M. Science. 2002; 295: 1852-1858Google Scholar). There are two distinct groups of chaperonins: (i) group I chaperonins of eubacteria and endosymbiotic organelles (mitochondria and chloroplasts); and (ii) group II chaperonins of archaea (known as thermosome) and the eukaryotic cytoplasm (known as CCT 1The abbreviations used are: CCT, chaperonin-containing t-complex polypeptide-1; TRiC, TCP-1 ring complex; T. KS-1, hyperthermophilic archaeum Thermococcus sp. strain KS-1; αWT, T. KS-1 wild type α chaperonin; αDel I, T. KS-1 α chaperonin lacking one-third of the helical protrusion; αDel II, T. KS-1 α chaperonin lacking two-thirds of the helical protrusion; αDel III, T. KS-1 α chaperonin lacking the whole of the helical protrusion; AMP-PNP, adenosine 5′-(β, γ-imino)triphosphate; AMP-PNPhex, AMP-PNP treated with hexokinase; ADPhex, ADP treated with hexokinase; SAXS, small angle X-ray scattering; Rg, radius of gyration; Dmax, maximum particle distance; CS, citrate synthase; GFP, a heat stable mutant of green fluorescence protein in this paper. 1The abbreviations used are: CCT, chaperonin-containing t-complex polypeptide-1; TRiC, TCP-1 ring complex; T. KS-1, hyperthermophilic archaeum Thermococcus sp. strain KS-1; αWT, T. KS-1 wild type α chaperonin; αDel I, T. KS-1 α chaperonin lacking one-third of the helical protrusion; αDel II, T. KS-1 α chaperonin lacking two-thirds of the helical protrusion; αDel III, T. KS-1 α chaperonin lacking the whole of the helical protrusion; AMP-PNP, adenosine 5′-(β, γ-imino)triphosphate; AMP-PNPhex, AMP-PNP treated with hexokinase; ADPhex, ADP treated with hexokinase; SAXS, small angle X-ray scattering; Rg, radius of gyration; Dmax, maximum particle distance; CS, citrate synthase; GFP, a heat stable mutant of green fluorescence protein in this paper. or TRiC) (3Kubota H. Hynes G. Willison K. Eur. J. Biochem. 1995; 230: 3-16Google Scholar, 4Gutsche I. Essen L.O. Baumeister W. J. Mol. Biol. 1999; 293: 295-312Google Scholar). Despite the relatively low sequence homology between the two groups, their domain architectures are very similar. Both of them are folded into three domains. The equatorial domain contains the ATP binding site and is involved in intra- and inter-ring contacts. The apical domain is involved in the binding to substrate proteins. The intermediate domain connects the equatorial and apical domains of each subunit and transfers the ATP-induced conformational changes from the equatorial to the apical domain (5Braig K. Ortwinowski Z. Hedge R. Boisvert D.C. Joachimiak A. Horwich A. Sigler P.B. Nature. 1994; 371: 578-586Google Scholar, 6Ditzel L. Lowe J. Stock D. Stetter K.O. Huber H. Huber R. Steinbacher S. Cell. 1998; 93: 125-138Google Scholar). The most striking structural difference between them is the lid of the central cavity of the chaperonin complex. The co-chaperonin GroES/cpn-10 serves as a lid for group I chaperonins in a heptameric dome-like structure. GroES/cpn-10 interacts with one or both GroEL rings in an ATP-regulated fashion, thereby sealing the cavity from the outside. It is demonstrated that a polypeptide of up to ∼57 kDa is able to fold within the GroES-sealed cavity (7Sakikawa C. Taguchi H. Makino Y. Yoshida M. J. Biol. Chem. 1999; 274: 21251-21256Google Scholar). On the other hand, there is no GroES-like co-chaperonin for group II chaperonins. Instead, group II chaperonins have a built-in lid, which is composed of an extension of the apical domain called the helical protrusion. The helical protrusion is thought to play the equivalent role of GroES, sealing off the central cavity of the chaperonin complex (6Ditzel L. Lowe J. Stock D. Stetter K.O. Huber H. Huber R. Steinbacher S. Cell. 1998; 93: 125-138Google Scholar, 8Gutsche I. Holzinger J. Rauh N. Baumeister W. May R.P. J. Struct. Biol. 2001; 135: 139-146Google Scholar, 9Llorca O. Martin-Benito J. Grantham J. Ritco-Vonsovici M. Willison K.R. Carrascosa J.L. Valpuesta J.M. EMBO J. 2001; 20: 4065-4075Google Scholar, 10Yoshida T. Kawaguchi R. Taguchi H. Yoshida M. Yasunaga T. Wakabayashi T. Yohda M. Maruyama T. J. Mol. Biol. 2002; 315: 73-85Google Scholar, 11Meyer A.S. Gillespie J.R. Walther D. Millet I.S. Doniach S. Frydman J. Cell. 2003; 113: 369-381Google Scholar, 12Iizuka R. Yoshida T. Shomura Y. Miki K. Maruyama T. Odaka M. Yohda M. J. Biol. Chem. 2003; 278: 44959-44965Google Scholar). Furthermore, this region is assumed to be involved in substrate binding (6Ditzel L. Lowe J. Stock D. Stetter K.O. Huber H. Huber R. Steinbacher S. Cell. 1998; 93: 125-138Google Scholar, 13Klumpp M. Baumeister W. Essen L.O. Cell. 1997; 91: 263-270Google Scholar). However, the exact role of the helical protrusion during the chaperonin functional cycle remains unclear. We have been studying the protein folding mechanism of group II chaperonins using the chaperonin from a hyperthermophilic archaeum, Thermococcus sp. strain KS-1. The natural chaperonin derived from T. KS-1 is a hetero-oligomer of α and β subunits, and the composition varies with the growth temperature (14Yoshida T. Ideno A. Hiyamuta S. Yohda M. Maruyama T. Mol. Microbiol. 2001; 39: 1406-1413Google Scholar). In addition, each subunit forms a double ring homo-oligomer capable of ATP-dependent protein folding in vitro (10Yoshida T. Kawaguchi R. Taguchi H. Yoshida M. Yasunaga T. Wakabayashi T. Yohda M. Maruyama T. J. Mol. Biol. 2002; 315: 73-85Google Scholar). In this study, we constructed helical protrusion deletion mutants of T. KS-1 α chaperonin (termed αDel I, αDel II, and αDel III; see Fig. 1) and characterized them. Protease sensitivity assays and small angle x-ray scattering (SAXS) studies indicated that the presence of ATP did not significantly change the conformation of the deletion mutants. They were able to prevent protein aggregation but unable to stimulate the refolding of denatured proteins. We have concluded that the helical protrusion is essential for the conformational change to mediate ATP-dependent protein folding rather than for substrate binding. Bacterial Strains, Plasmids, and Reagents—Escherichia coli strains used in this study were DH5α for plasmid preparation and BL21(DE3) for protein expression. The plasmid pK1Eα2 was used as a template for PCR and for the expression of T. KS-1 wild type α chaperonin (αWT) (15Yoshida T. Yohda M. Iida T. Maruyama T. Taguchi H. Yazaki K. Ohta T. Odaka M. Endo I. Kagawa Y. J. Mol. Biol. 1997; 273 (Correction (2000) J. Mol. Biol.299, 1399–1400): 635-645Google Scholar). The expression plasmid for GFP, pET21c-GFPUV(His), was a gift from Dr. H. Taguchi. Ex Taq™ DNA polymerase and restriction endonucleases were the products of Takara Bio Inc. (Shiga, Japan). ATP, ADP, and thermolysin were purchased from Wako Pure Chemicals (Osaka, Japan). AMP-PNP and citrate synthase from porcine heart and Thermoplasma acidophilum were obtained from Sigma. The AMP-PNP and ADP were purified by anion exchange chromatography (16Horst M. Oppliger W. Feifel B. Schatz G. Glick B.S. Protein Sci. 1996; 5: 759-767Google Scholar) or treated with hexokinase (Roche Diagnostics) plus glucose as described (17Motojima F. Yoshida M. J. Biol. Chem. 2003; 278: 26648-26654Google Scholar) before use to remove contaminating ATP. Protein concentrations were determined with a Bradford assay kit (Bio-Rad) using bovine serum albumin as the standard (18Bradford M.M. Anal. Biochem. 1976; 72: 248-254Google Scholar). Construction of Helical Protrusion Deletion Mutants and Chaperonin Purification—Helical protrusion deletion mutants of the α subunit, αDel I (Δ256–265), αDel II (Δ251–270), and αDel III (Δ245–276), were each generated by two separate polymerase chain reactions using the pK1Eα2 plasmid as a template. For construction of the αDel I mutant, DNAs encoding the N-terminal moiety (residue 1–255) and the C-terminal moiety (residue 266–548) were amplified using primer pairs containing NdeI/SalI and SalI/BamHI sites, respectively. The amplified fragments were subcloned into the pT7Blue T-vector (Novagen). The inserted DNA fragments were excised with NdeI/SalI and SalI/BamHI and co-ligated into pET23a (Novagen) at the sites of NdeI and BamHI. The linker sequence connecting the N- and C-terminal moieties was Val-Asp (GTC-GAC; SalI recognition sequence). The other two deletion mutants were constructed using the same procedure. All constructs were verified by DNA sequencing. The chaperonins were expressed and purified as described previously (12Iizuka R. Yoshida T. Shomura Y. Miki K. Maruyama T. Odaka M. Yohda M. J. Biol. Chem. 2003; 278: 44959-44965Google Scholar). Thermal Aggregation Measurement—The thermal aggregation of citrate synthase (CS) from porcine heart was monitored by measuring light scattering at 500 nm with a spectrofluorometer (RF-5300PC, Shimadzu, Kyoto, Japan) at 50 °C. Native CS (120 nm, as a monomer) was incubated in the assay buffer (50 mm Tris-HCl pH8, 25 mm MgCl2, 100 mm KCl, and 5 mm dithiothreitol) in the presence or absence of chaperonins (120 nm). The assay buffer was preincubated at 50 °C and continuously stirred throughout the measurement. Refolding Assay—The GFP used in this paper is a heat-stable mutant with alanine inserted between Met-1 and Ser-2, a His tag in the C-terminal region, and amino acid substitutions of F99S, M153T, V163A, and L165F (7Sakikawa C. Taguchi H. Makino Y. Yoshida M. J. Biol. Chem. 1999; 274: 21251-21256Google Scholar, 10Yoshida T. Kawaguchi R. Taguchi H. Yoshida M. Yasunaga T. Wakabayashi T. Yohda M. Maruyama T. J. Mol. Biol. 2002; 315: 73-85Google Scholar). It was purified as described (19Iizuka R. Yoshida T. Maruyama T. Shomura Y. Miki K. Yohda M. Biochem. Biophys. Res. Commun. 2001; 289: 1118-1124Google Scholar). The refolding assay was carried out at 60 °C. GFP (5 μm) was denatured in the folding buffer (50 mm Tris-HCl pH7.5, 25 mm MgCl2, 100 mm KCl, and 5 mm dithiothreitol) containing 100 mm HCl at room temperature and diluted 100-fold in the folding buffer with or without chaperonins (100 nm). 1 mm ATP was added to the mixture 8 min after the dilution. The fluorescence of GFP at 510 nm with excitation at 396 nm was continuously monitored with a JASCO FP-6500 spectrofluorometer (JASCO, Tokyo, Japan). The reaction mixtures were continuously stirred at 60 °C throughout the experiments. As a control, native GFP was diluted in the folding buffer without chaperonins. The fluorescence intensity of native GFP was taken as 100%. CS from T. acidophilum was subjected to a refolding assay at 50 °C. CS (19.8 μm, as a monomer) was denatured in 50 mm HEPES-KOH pH 7.5 containing 6 m guanidine hydrochloride and 5 mm dithiothreitol for 30 min at 50 °C and then diluted 60-fold in the dilution buffer (50 mm HEPES-KOH pH 7.5, 50 mm MgCl2, and 300 mm KCl) in the absence or presence of chaperonins (0.5 μm). The refolding reactions were conducted for 60 min at 50 °C. 1 mm ATP was added to the mixture 10 min after the dilution. At the indicated time points, aliquots were removed from the mixture, and the recovered enzyme activity was assayed as described by Furutani et al. (20Furutani M. Iida T. Yoshida T. Maruyama T. J. Biol. Chem. 1998; 273: 28399-28407Google Scholar). The activity of the native enzyme at the same concentration is taken as 100%. Protease Sensitivity Assay—Chaperonins (50 nm) were incubated with or without nucleotide (1 mm) for 10 min at 65 °C while being continuously mixed. The assay buffer was TNM buffer (50 mm Tris-HCl, pH 7.5, 100 mm NaCl, and 25 mm MgCl2) or TKM buffer (50 mm Tris-HCl, pH 7.5, 100 mm KCl, and 25 mm MgCl2). Digestion with thermolysin (1 ng/μl) was carried out for 10 min at 65 °C. Aliquots of the reaction mixture were precipitated using 30% (w/v) trichloroacetic acid and then analyzed on 15% polyacrylamide gels containing SDS. Gels were stained with Coomassie Brilliant Blue. SAXS Measurement—The SAXS experiments were performed at beamline 15A of the Photon Factory in the High Energy Accelerator Research Organization, Tsukuba, Japan. The measurements were done at the protein concentrations of 3–7 mg/ml in TNM buffer at 58.5 °C. Samples were incubated with or without adenine nucleotides (αWT, 1 mm; deletion mutants, 3 mm) for 5 min at 58.5 °C before data collection. Scattering patterns were recorded by a CCD-based x-ray detector that consisted of a beryllium-windowed x-ray image intensifier (Be-XRII) (Hamamatsu, V5445P-MOD), an optical lens, a CCD image sensor, and a data acquisition system (Hamamatsu C7300), as described (21Amemiya Y. Ito K. Yagi N. Asano Y. Wakabayashi K. Ueki T. Endo T. Rev. Sci. Instrum. 1995; 66: 2290-2294Google Scholar, 22Arai M. Ito K. Inobe T. Nakao M. Maki K. Kamagata K. Kihara H. Amemiya Y. Kuwajima K. J. Mol. Biol. 2002; 321: 121-132Google Scholar). The experimental details and the analyses of the scattering data were essentially the same as described (22Arai M. Ito K. Inobe T. Nakao M. Maki K. Kamagata K. Kihara H. Amemiya Y. Kuwajima K. J. Mol. Biol. 2002; 321: 121-132Google Scholar). Pair distribution (P(r)) functions were calculated by using the GNOM package (23Semenyuk A.V. Svergun D.I. J. Appl. Crystallogr. 1991; 24: 537-540Google Scholar). The Q range used for the calculation was from 0.0155 to 0.2 Å-1. The values of the radius of gyration (Rg) and the maximum particle dimension (Dmax) were estimated from the P(r) function. Construction of Helical Protrusion Deletion Mutants—The helical protrusion region of the α subunit of T. KS-1 chaperonin is composed of 32 amino acid residues (residue 245–276) (Fig. 1). To examine its role in the conformational change and molecular chaperone functions, we constructed three helical protrusion deletion mutants (termed αDel I, αDel II, and αDel III), which lack one-third, two-thirds, and the whole of the helical protrusion, respectively (Fig. 1). Each construct was expressed in E. coli and purified in the same manner as the wild type. The deletion of the helical protrusion had no effect on the oligomer formation and ATP hydrolysis (data not shown). Protease Sensitivity Assays of Adenine Nucleotide-induced Conformational Changes of α Chaperonins—Conformational changes of αWT and deletion mutants induced by adenine nucleotides were compared using protease sensitivity assays (12Iizuka R. Yoshida T. Shomura Y. Miki K. Maruyama T. Odaka M. Yohda M. J. Biol. Chem. 2003; 278: 44959-44965Google Scholar) (Fig. 2). The chaperonins were subjected to proteolysis by thermolysin in the absence and presence of adenine nucleotides. To exclude the effects of contaminating ATP in AMP-PNP (a non-hydrolysable analogue of ATP) and ADP, the experiments were conducted using nucleotides purified by treatment with hexokinase plus glucose (17Motojima F. Yoshida M. J. Biol. Chem. 2003; 278: 26648-26654Google Scholar). Those treated with hexokinase are referred to as AMP-PNPhex and ADPhex, respectively. αWT was sensitive to digestion by thermolysin in the absence of nucleotide or the presence of ADPhex. In contrast, incubation with ATP rendered αWT resistant to thermolysin. AMP-PNPhex also led to protease resistance in the absence but not the presence of potassium ion (Fig. 2A). This result confirmed that our previous result is not caused by the contamination of ATP in ADP or AMP-PNP. The same experiments were performed using αDel I, αDel II, and αDel III (Fig. 2, B–D). Almost the same pattern was observed in the digestion of αDel I, irrespective of the presence of adenine nucleotides and potassium ion (Fig. 2B). However, neither αDel II nor αDel III was subjected to proteolysis by thermolysin (Fig. 2, C and D). These results are well consistent with our previous observation (12Iizuka R. Yoshida T. Shomura Y. Miki K. Maruyama T. Odaka M. Yohda M. J. Biol. Chem. 2003; 278: 44959-44965Google Scholar) that the proteases mainly attack the helical protrusion in a nucleotide-free or ADP-bound state. SAXS Analyses of Adenine Nucleotide-induced Conformational Changes of α Chaperonins—To study the conformation change of αWT and deletion mutants further, small angle x-ray scattering (SAXS) measurements were carried out. The measurements were conducted at 58.5 °C under potassium-free conditions where chaperonin-catalyzed ATP hydrolysis was restricted (12Iizuka R. Yoshida T. Shomura Y. Miki K. Maruyama T. Odaka M. Yohda M. J. Biol. Chem. 2003; 278: 44959-44965Google Scholar). AMP-PNP and ADP purified by anion exchange chromatography (16Horst M. Oppliger W. Feifel B. Schatz G. Glick B.S. Protein Sci. 1996; 5: 759-767Google Scholar) or treatment with hexokinase plus glucose (17Motojima F. Yoshida M. J. Biol. Chem. 2003; 278: 26648-26654Google Scholar) were also used in the measurements. Fig. 3, A–C show the pair distribution (P(r)) functions of αWT (A), αDelI(B), and αDel II (C) in the absence or presence of adenine nucleotides, which were calculated from a Fourier transformation of the intensity functions. Two different peaks were observed in the P(r) curve of αWT (Fig. 3A), suggesting the presence of two different conformations. In contrast, the P(r) function of the nucleotide-free form has a peak at an r value of ∼105 Å (Fig. 3A; black dotted line). The P(r) function was not affected by the addition of ADPhex (Fig. 3A, gray line). The peak shifted to a lower r value of ∼100 Å in the presence of ATP (Fig. 3A, black line). This indicates that the binding of ATP leads to a conformation distinct from the other two forms. The same P(r) curve was obtained in the presence of AMP-PNPhex (Fig. 3A, gray dotted line). The P(r) curves of αDel I in the presence of ATP and AMP-PNPhex differed from those under nucleotide-free conditions and in the presence of ADPhex, but the changes were marginal (Fig. 3B). αDel II exhibited almost no change in its P(r) curve upon the addition of ATP or AMP-PNP (Fig. 3C). The structural parameters obtained in this study and calculated from the crystal structure are summarized in Table I. The values for Rg and the Dmax were computed from the P(r) function using the program GNOM (23Semenyuk A.V. Svergun D.I. J. Appl. Crystallogr. 1991; 24: 537-540Google Scholar). In αWT, the conformation in the presence of ATP is the most compact, with an Rg value of 67.6 ± 0.4 Å and a Dmax value of 168 ± 3 Å. The calculated Rg and Dmax are close to values predicted from the crystal structure of Thermoplasma chaperonin (6Ditzel L. Lowe J. Stock D. Stetter K.O. Huber H. Huber R. Steinbacher S. Cell. 1998; 93: 125-138Google Scholar) or the hypothetical asymmetric structure, i.e. one open and one closed ring. The nucleotide-free form and ADP-bound form have a stretched conformation, with Rg of 74.1 ± 0.2 and 72.5 ± 0.6 Å, and a Dmax of 194 ± 2 and 184 ± 4 Å, respectively. Thus, the presence of ATP reduces the value of Rg by ∼6.5 Å and the value of Dmax by ∼26 Å. The conformation in the AMP-PNP-bound state is very similar to that in the ATP-bound state (Table I). The Rg and Dmax values of the conformations without the addition of nucleotides and with ADP are in agreement with those of the open conformation of the Thermoplasma α-thermosome observed by cryo-electron microscopy (25Nitsch M. Walz J. Typke D. Klumpp M. Essen L.O. Baumeister W. Nat. Struct. Biol. 1998; 5: 855-857Google Scholar). On the other hand, the addition of ATP and purified AMP-PNP to αDel I and αDel II did not induce a significant conformational change. These results clearly showed that the archaeal chaperonin takes an open conformation in the nucleotide-free or ADP-bound state and changes to the closed conformation on binding to ATP. The result is consistent with previous observations (8Gutsche I. Holzinger J. Rauh N. Baumeister W. May R.P. J. Struct. Biol. 2001; 135: 139-146Google Scholar, 12Iizuka R. Yoshida T. Shomura Y. Miki K. Maruyama T. Odaka M. Yohda M. J. Biol. Chem. 2003; 278: 44959-44965Google Scholar). In the case of CCT, the closure of the built-in lid was induced not by the binding but by the hydrolysis of ATP (11Meyer A.S. Gillespie J.R. Walther D. Millet I.S. Doniach S. Frydman J. Cell. 2003; 113: 369-381Google Scholar). There seems to be a difference in the mechanism for the closure of the built-in lid between archaeal chaperonins and CCT.Table IStructural parameters of group II chaperoninsProteinNucleotideRgDmaxÅÅαWTaStructural parameter obtained by SAXS, this studyFree74.1 ± 0.2194 ± 2+ ATP67.6 ± 0.4168 ± 3+ AMP-PNPhex68.0 ± 0.3171 ± 2+ ADPhex72.5 ± 0.6184 ± 4αDel IaStructural parameter obtained by SAXS, this studyFree75.6 ± 0.4192 ± 2+ ATP73.6 ± 0.4189 ± 3+ AMP-PNPhex73.8 ± 0.1195 ± 3+ ADPhex75.5 ± 0.2193 ± 2αDel IIaStructural parameter obtained by SAXS, this studyFree73.2 ± 0.1194 ± 4+ ATP71.0 ± 0.3189 ± 5+ AMP-PNPpurbAMP-PNP purified by anion exchange chromatography (16)73.0 ± 0.4188 ± 2+ ADP72.0 ± 0.5186 ± 3CCTcStructural parameter obtained by SAXS with values taken from Meyer et al. (11)Free70.2 ± 0.5195 ± 5+ ATP64.7 ± 0.6171 ± 5+ AMP-PNP70.6 ± 0.4195 ± 3+ ADP69.4 ± 0.8193 ± 3+ ATPγS70.5 ± 0.9195 ± 4+ ADP, AlFx67.8 ± 0.9182 ± 5Open conformationdStructural parameter obtained by calculation with values taken from Meyer et. al. (11)71.3 ± 0.6203 ± 4Asymmetric conformationdStructural parameter obtained by calculation with values taken from Meyer et. al. (11)67.6 ± 0.7179 ± 4Closed conformationdStructural parameter obtained by calculation with values taken from Meyer et. al. (11)64.8 ± 0.7164 ± 5a Structural parameter obtained by SAXS, this studyb AMP-PNP purified by anion exchange chromatography (16Horst M. Oppliger W. Feifel B. Schatz G. Glick B.S. Protein Sci. 1996; 5: 759-767Google Scholar)c Structural parameter obtained by SAXS with values taken from Meyer et al. (11Meyer A.S. Gillespie J.R. Walther D. Millet I.S. Doniach S. Frydman J. Cell. 2003; 113: 369-381Google Scholar)d Structural parameter obtained by calculation with values taken from Meyer et. al. (11Meyer A.S. Gillespie J.R. Walther D. Millet I.S. Doniach S. Frydman J. Cell. 2003; 113: 369-381Google Scholar) Open table in a new tab Helical Protrusion Is Not Necessarily Important for Substrate Binding—The effect of deletion mutants on the aggregation of unfolded protein was examined using CS from porcine heart, a homo-dimer of 48 kDa subunits. The time course of CS aggregation was monitored by light scattering. When native CS was diluted in the assay buffer incubated at 50 °C, it aggregated immediately to increase the light scattering (Fig. 4, filled squares). The decrease of light scattering after 10 min of incubation is probably due to the precipitation of the aggregates. At a 1:1 molar ratio of αWT oligomer to CS monomer, the thermal aggregation of CS was almost completely inhibited (Fig. 4, filled circles). All deletion mutants also inhibited the thermal aggregation of CS at an equimolar ratio, despite a slight decrease of efficiency (Fig. 4, filled triangles, open squares, and open circles). Also, the effect of each deletion mutant changes with the concentration (data not shown). These findings show that the helical protrusion region is not indispensable for the recognition and binding of the substrate protein. Helical Protrusion Is Essential for Chaperonin-mediated Protein Refolding—To investigate whether deletion mutants can promote the refolding of denatured proteins, we first used GFP, a monomeric 27-kDa protein, as the substrate protein (Fig. 5A). We employed a heat stable mutant GFP to perform assays at 60 °C (7Sakikawa C. Taguchi H. Makino Y. Yoshida M. J. Biol. Chem. 1999; 274: 21251-21256Google Scholar, 10Yoshida T. Kawaguchi R. Taguchi H. Yoshida M. Yasunaga T. Wakabayashi T. Yohda M. Maruyama T. J. Mol. Biol. 2002; 315: 73-85Google Scholar). When acid-denatured GFP without fluorescence was diluted in the folding buffer at a neutral pH, it refolded spontaneously and the level of fluorescence recovered. Because of the high temperature for GFP refolding, the recovery speed and yield were not high. The yields of spontaneous folding at 15 min after the dilution was estimated to be ∼20% (Fig. 5A, filled squares). In the presence of chaperonins in the folding mixture (at a 2:1 molar ratio of chaperonin to GFP), spontaneous refolding of GFP was inhibited in each case. These results show that deletion mutants are able to capture unfolded GFP in concordance with the above results (Fig. 4). When ATP was added to the mixture 8 min after the dilution of GFP, αWT enhanced the productive refolding of GFP (Fig. 5A, filled circles), whereas deletion mutants failed to initiate GFP folding at a measurable rate (Fig. 5A, filled triangles, open squares, and open circles). Another substrate protein, CS from an acidothermophilic archaeum, T. acidophilum, a homo-dimer of 43 kDa subunits, was tested (Fig. 5B). Only 4% of the activity of CS was recovered without chaperonin (Fig. 5B, filled squares). All of the chaperonin variants used were able to bind denatured CS and inhibit the spontaneous folding of CS. When ATP was added, αWT promoted a 25% reactivation of CS activity (Fig. 5B, filled circles). By contrast, deletion mutants had almost no effect on the proper refolding and enzymatic activity of CS (Fig. 5B, filled triangles, open squares, and open circles). Thus, these findings suggest that the helical protrusion region is indispensable for ATP-dependent protein folding rather than for substrate binding. The helical protrusion is characteristic of all group II chaperonins (13Klumpp M. Baumeister W. Essen L.O. Cell. 1997; 91: 263-270Google Scholar). Although it is thought to be important for the function of the group II chaperonins, its role remains unclear. To clarify this issue, we have constructed helical protrusion deletion mutants of T. KS-1 α chaperonin (αDel I, αDel II, and αDel III; see Fig. 1) and characterized them. It has been suggested that the helical protrusion acts as a built-in lid for the central cavity (6Ditzel L. Lowe J. Stock D. Stetter K.O. Huber H. Huber R. Steinbacher S. Cell. 1998; 93: 125-138Google Scholar, 8Gutsche I. Holzinger J. Rauh N. Baumeister W. May R.P. J. Struct. Biol. 2001; 135: 139-146Google Scholar, 9Llorca O. Martin-Benito J. Grantham J. Ritco-Vonsovici M. Willison K.R. Carrascosa J.L. Valpuesta J.M. EMBO J. 2001; 20: 4065-4075Google Scholar, 10Yoshida T. Kawaguchi R. Taguchi H. Yoshida M. Yasunaga T. Wakabayashi T. Yohda M. Maruyama T. J. Mol. Biol. 2002; 315: 73-85Google Scholar, 11Meyer A.S. Gillespie J.R. Walther D. Millet I.S. Doniach S. Frydman J. Cell. 2003; 113: 369-381Google Scholar, 12Iizuka R. Yoshida T. Shomura Y. Miki K. Maruyama T. Odaka M. Yohda M. J. Biol. Chem. 2003; 278: 44959-44965Google Scholar). Protease sensitivity assays and the SAXS experiments revealed that ATP induces a drastic conformational change in αWT under potassium-free conditions (Figs. 2A and 3A). The observed difference between the absence and presence of ATP is ∼26 Å in terms of the value of Dmax (Table I). AMP-PNP induced a similar conformational change in the absence of potassium ion (Figs. 2 and 3 and Table I). Thus, it is concluded that the hydrolysis of ATP is dispensable for this change. ATP and AMP-PNP also induced the conformational change of αDel I, but the extent of the change was much less than that of αWT (Fig. 3B and Table I). In addition, significant structural changes of αDel II were not observed in the presence of ATP or AMP-PNP (Fig. 3C and Table I). These findings clearly demonstrate that the conformational change of αWT induced by ATP binding is a movement of the helical protrusion. From the structural parameters of the model structures (Table I), it could be concluded that the closing of the chaperonin cavity is induced by the binding of ATP. Importantly, we observed that the deletion mutants have no significant ATP-dependent protein-folding activity, even though they are capable of binding unfolded proteins (Fig. 5). Taken together, we conclude that productive protein folding requires a complete closure of the ring cavity. It is likely that even a partial deletion of the region makes it impossible to seal off the central cavity and encapsulate the bound substrate. As was expected, it is thought that the helical protrusion of group II chaperonins plays an equivalent role to GroES in group I chaperonins. The helical protrusion was also assumed to be involved in substrate binding because of the presence of a large hydrophobic patch in the region (6Ditzel L. Lowe J. Stock D. Stetter K.O. Huber H. Huber R. Steinbacher S. Cell. 1998; 93: 125-138Google Scholar, 13Klumpp M. Baumeister W. Essen L.O. Cell. 1997; 91: 263-270Google Scholar). In GroEL, the substrate binding sites involve a number of hydrophobic residues that can generally bind hydrophobic regions in unfolded proteins (1Bukau B. Horwich A.L. Cell. 1998; 92: 351-366Google Scholar, 2Hartl F.U. Hayer-Hartl M. Science. 2002; 295: 1852-1858Google Scholar). In contrast, the removal of the helical protrusion did not impair the activity for inhibition of the thermal aggregation of CS (Fig. 4). Also, the deletion mutants inhibited the spontaneous refolding of GFP and CS (Fig. 5). These results show that the helical protrusion region is not necessarily important for the recognition and binding of the substrate protein. Similar findings are reported in CCT. A recent study has shown that cleavage of the protrusion does not affect substrate binding ability (11Meyer A.S. Gillespie J.R. Walther D. Millet I.S. Doniach S. Frydman J. Cell. 2003; 113: 369-381Google Scholar). The crystal structure of the CCTγ apical domain suggests that the substrate binding sites are located within the central cavity (26Pappenberger G. Wilsher J.A. Roe S.M. Counsell D.J. Willison K.R. Pearl L.H. J. Mol. Biol. 2002; 318: 1367-1379Google Scholar). Also, electron microscopic studies of CCT-substrate protein complexes show that actin appears to be bound well below the helical protrusion and that tubulin seems to interact with a much broader region of the apical domains, including parts of the protrusion (9Llorca O. Martin-Benito J. Grantham J. Ritco-Vonsovici M. Willison K.R. Carrascosa J.L. Valpuesta J.M. EMBO J. 2001; 20: 4065-4075Google Scholar, 27Llorca O. Martin-Benito J. Ritco-Vonsovici M. Grantham J. Hynes G.M. Willison K.R. Carrascosa J.L. Valpuesta J.M. EMBO J. 2000; 19: 5971-5979Google Scholar). It is speculated that there is a "true" substrate binding site in the apical domain, although the helical protrusion contributes slightly to the binding to unfolded proteins. Several studies indicated that group II chaperonin subunits have a cluster of relatively conserved residues on the inside face of the apical domain, just below the helical protrusion (24Shomura Y. Yoshida T. Iizuka R. Maruyama T. Yohda M. Miki K. J. Mol. Biol. 2004; 335: 1265-1278Google Scholar, 28Archibald J.M. Blouin C. Doolittle W.F. J. Struct. Biol. 2001; 135: 157-169Google Scholar, 29Archibald J.M. Roger A.J. J. Mol. Biol. 2002; 316: 1041-1050Google Scholar). Archibald et al. pointed out that this region appears to be in direct contact with actin and tublin in CCT-substrate complexes (28Archibald J.M. Blouin C. Doolittle W.F. J. Struct. Biol. 2001; 135: 157-169Google Scholar). It is thought that the region is a new candidate for the substrate binding site. The helical protrusion might have a role in accessing substrate protein in the cavity, as expected from crystallographic studies (6Ditzel L. Lowe J. Stock D. Stetter K.O. Huber H. Huber R. Steinbacher S. Cell. 1998; 93: 125-138Google Scholar, 13Klumpp M. Baumeister W. Essen L.O. Cell. 1997; 91: 263-270Google Scholar). We thank M. Arai, K. Maki, K. Kamagata S. Enoki, A. Kadooka, and T. Oroguchi for generous efforts on SAXS measurements.