- Email: [email protected]
Facilitated release of substrate protein from prefoldin by chaperonin
FEBS Letters 579 (2005) 3718–3724
FEBS 29686
Facilitated release of substrate protein from prefoldin by chaperonin Tamotsu Zakoa,b,c,*,1, Ryo Iizukab, Mina Okochib, Tomoko Nomurab, Taro Uenoa, Hisashi Tadakumaa, Masafumi Yohdab, Takashi Funatsua,c,d,* a
Department of Physics, School of Science and Engineering, Waseda University, 3-4-1 Okubo, Shinjyuku-ku, Tokyo 169-8555, Japan b Department of Biotechnology and Life Science, Tokyo University of Agriculture and Technology, 2-24-16 Naka-cho, Koganei-shi, Tokyo 184-8588, Japan c Laboratory of Bio-Analytical Chemistry, Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan d Core Research for Evolutional Science and Technology, JST, Kawaguchi-shi, Saitama 332-0012, Japan Received 29 March 2005; revised 13 May 2005; accepted 26 May 2005 Available online 13 June 2005 Edited by Jesus Avila
Abstract Prefoldin is a chaperone that captures a proteinfolding intermediate and transfers it to the group II chaperonin for correct folding. However, kinetics of interactions between prefoldin and substrate proteins have not been investigated. In this study, dissociation constants and dissociation rate constants of unfolded proteins with prefoldin were firstly measured using fluorescence microscopy. Our results suggest that binding and release of prefoldin from hyperthermophilic archaea with substrate proteins were in a dynamic equilibrium. Interestingly, the release of substrate proteins from prefoldin was facilitated when chaperonin was present, supporting a handoff mechanism of substrate proteins from prefoldin to the chaperonin. 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Prefoldin; Chaperonin; Molecular chaperone; Hyperthermophilic archaea
1. Introduction Prefoldin is a heterohexameric chaperone that cooperates with the group II chaperonin [1]. Prefoldin captures actins and tubulins in the unfolded state and transfers them to a cytosolic chaperonin, chaperonin containing TCP-1 (CCT) for functional folding [2–4]. Prefoldin exists in archaea as well as * Corresponding authors. Fax: +81 48 462 4658 (T. Zako); +81 3 5802 3339 (T. Funatsu). E-mail addresses: [email protected] (T. Zako), [email protected] (T. Funatsu).
group II chaperonins [4], irrespective of their lack of actins or tubulins. To date, archaeal prefoldins from Methanobacterium thermoautotrophicum (MtGimC) [5,6] and Pyrococcus horikoshii OT3 (PhPFD) [7,8] have been functionally characterized. They are capable of stabilizing non-native proteins and releasing them for subsequent chaperonin-dependent folding in vitro. Crystal structure of MtGimC [9] and electron microscopy of eukaryotic (bovine and human) prefoldin [10] have shown that prefoldin has the appearance of a jellyfish consisting of a double beta barrel assembly with six long coiled coils protruding from it. Biochemical and structural studies indicate that these tentacles bind to substrate proteins in a concerted manner [6,8–10] and can move outward for efficient interaction with substrate proteins [6]. Observation of prefoldin–CCT complexes by electron microscopy has revealed that the outer regions of the prefoldin tentacles interact with the inner regions of the apical domains of the chaperonin [10]. Although these data support a handoff mechanism of substrate proteins from prefoldin to the chaperonin, the detailed mechanism remains to be solved. In this study, we examined the dissociation kinetics of PhPFD and substrate proteins [porcine heart citrate synthase (CS) and green fluorescent protein (GFP)]. We found that binding and release of PhPFD with substrate proteins in the folding or unfolding intermediate state were in a dynamic equilibrium. The release of substrate proteins from PhPFD was facilitated in the presence of chaperonin, supporting a handoff mechanism of substrate proteins from prefoldin to the chaperonin.
2. Materials and methods
1
Present address: Bioengineering Laboratory, RIKEN (The Institute of Physical and Chemical Research), 2-1, Hirosawa, Wako, Saitama 351-0198, Japan. Abbreviations: PhPFD, prefoldin from Pyrococcus horikoshii; CS, citrate synthase; GFP, green fluorescent protein; IPMDH, 3-isopropylmalate dehydrogenase; tc8atc5b, PhPFD mutant with 8- and 5-amino acid truncations from C-terminus of a and b subunit; aN472C, N472C point mutant of Thermococcus strain KS-1 chaperonin a-subunit; 488bio-CPN, aN472C modified with Alexa Fluor 488 and biotin; BSA, bovine serum albumin; TIRFM, total internal reflection fluorescence microscopy; koff, dissociation rate constant; KD, dissociation constant; buffer A, a solution containing 25 mM HEPES, pH 7.4, 100 mM KCl, 5 mM MgCl2
2.1. Proteins CS was purchased from Sigma (St. Louis, MO). GFP was purified as described previously [11,12]. 3-Isopropylmalate dehydrogenase (IPMDH) from Thermus thermophilus strain HB8 was purified as described [13]. PhPFD and its mutant tc8atc5b which has 8- and 5-amino acid truncations from C-terminus of a and b subunit [8] were expressed in E. coli BL21 (DE3) and purified as described [7,8]. In order to immobilize and visualize chaperonin, Asn472, which is located on the outer surface of the equatorial domain of Thermococcus strain KS-1 (T.KS-1) chaperonin a-subunit [14], was replaced by Cys (termed aN472C) using a QuikChange Site-directed Mutagenesis Kit (Stratagene, CA). The aN472C was purified as described for the wild-type [12,15].
0014-5793/$30.00 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2005.05.061
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
3719
Substrate proteins were labeled with fluorescent dyes for detection by fluorescence microscopy. 18 lM CS or 72 lM GFP was incubated with 35 lM Cy3 or 150 lM Cy5 Mono-reactive Dye (Amersham Biosciences) for 2 h at room temperature in 25 mM HEPES, pH 8.5, 20 mM CaCO3, 100 mM KCl, 5 mM MgCl2. The labeled substrate proteins were separated from the unreacted dyes by gel filtration. The molar ratios of Cy3 to CS, Cy5 to CS and Cy5 to GFP were 0.8, 1.0 and 1.1, respectively. aN472C modified with Alexa Fluor 488-C5-maleimide (Alexa 488; Molecular Probes, OR) and biotinPEAC5-maleimide (Dojindo Laboratory, Kumamoto, Japan) (termed 488bio-CPN) was obtained as described [16]. The molar ratio of Alexa 488 to aN472C was 1.1 throughout this study. Streptavidin was purchased from Sigma. Biotinylated bovine serum albumin (BSA) was prepared as described [16]. 2.2. Microscopy Fluorescently labeled proteins were observed by total internal reflection fluorescence microscopy (TIRFM) [16,17]. Cy5-, Cy3-, and Alexa Fluor 488-labeled proteins were illuminated with a He–Ne laser (1.0 mW, 632.8 nm, model GLG5350; NEC Corporation, Tokyo, Japan), a green solid-state laser (2.8 mW, 532 nm, l-Green model 4601; Uniphase, San Jose, CA), and a blue solid-state laser (2.0 mW, 473 nm, ML0250A, Nippon Avionics Co., Japan), respectively. The fluorescence emission from the specimen was collected with an oilimmersion microscope objective (1.40 NA, 100·, PlanApo; Olympus, Tokyo, Japan). Images were taken with a SIT camera (C2400-08; Hamamatsu Photonics, Shizuoka, Japan) coupled to an image intensifier (VS4-1845; Video Scope International, VA) and recorded onto videotapes for subsequent analysis. Images were digitally captured by PC and the average intensities of 4 video frames were measured using the Scion Image software (Scion Corporation, ML). 2.3. Measurement of dissociation rate constants of substrate proteins from PhPFD complex Dissociation rate constants (koff) of substrate proteins from PhPFD were obtained by quantifying the amount of substrate proteins bound to PhPFD by TIRFM (Fig. 1A). 20 nM of Cy3-CS or Cy5-GFP (denatured at pH 1.2 for 10 min) was mixed with 20 nM PhPFD in buffer A (25 mM HEPES, pH 7.4, 100 mM KCl, 5 mM MgCl2) and incubated for 10 min at 50 C where the archaeal proteins used in this study are active. Then, 200 nM unlabeled substrate proteins were added into the solution and further incubated at 23 or 50 C. 20 ll of the sample solution were then taken at various incubation times and infused into a flow cell (20 ll in volume) consisting of a glass slide overlaid with a coverslip and two spacers (50 lm in thickness) at room temperature (23 C). The flow cell was then rinsed with 100 ll of buffer A to wash out unbound substrate proteins. Subsequently, the flow cell was filled with 20 ll of buffer A containing an oxygen scavenger system (25 mM glucose, 2.5 lM glucose oxidase, 10 nM catalase, 10 mM dithiothreitol) to suppress photobleaching [18]. The fluorescence intensity of the labeled substrate proteins bound to PhPFD was quantified by TIRFM. The value of koff was calculated by exponential fitting to obtained results using KaleidaGraph 3.5 (Synergy Software, PA). 2.4. Measurement of equilibrium constants of PhPFD and substrate proteins Schematic drawing of the experiment was shown in Fig. 2A. Various concentrations of Cy3-CS or Cy5-GFP (denatured at pH 1.2 for 10 min) were mixed with fixed concentrations of PhPFD (20, 50 or 100 nM) in buffer A and incubated at 50 C for 10 min. Then, 20 ll of the sample solution were infused into a flow cell. Because PhPFD strongly bound to a glass surface, but not for substrate proteins, PhPFD remained on the glass surface after washing of the flow cell with 100 ll of buffer A. The fluorescence intensity was measured by TIRFM to determine the amount of substrate proteins bound to PhPFD. KD were determined by fitting to the data using Eq. (1) [19]
Fluorescence intensity ¼
ðK D þ ½PFD0 þ ½Cy3-CS0 Þ
2.5. Measurement of dissociation rate constants of substrate proteins from PhPFD in the presence of chaperonin Dissociation of substrate proteins from PhPFD in the presence of aN472C was quantified by TIRFM (Fig. 3A). First, complex of substrate protein (Cy3-CS or Cy5-GFP) and PhPFD was formed as described in Section 2.3 and incubated at 50 C for 10 min. Then, 200 nM unlabeled substrate protein and 400 nM aN472C (or 2.6 lM IPMDH as a negative control) were added to the sample and further incubated at 50 C. After the various incubation times, 20 ll of the sample solution were taken and infused into a flow cell. koff of substrate proteins from PhPFD were obtained as described above. A PhPFD mutant tc8atc5b which had reduced affinity with chaperonin was also used as a control. 2.6. Single molecule imaging of chaperonin and substrate proteins by TIRFM A flow cell of which glass surface was coated with streptavidin was prepared beforehand to immobilize 488bio-CPN as follows. Buffer A containing 43 lM biotinylated BSA was infused into the flow cell, being washed, then 17 lM streptavidin was infused, and washed again. On one hand, the solution containing 100 nM PhPFD and 50 nM Cy5CS in buffer A was incubated at 50 C for 10 min, followed by the addition of 100 nM 488bio-CPN and additional 10 min incubation. The solution was diluted approximately 1000-fold with buffer A, and immediately infused into the flow cell coated with streptavidin to immobilize 488bio-CPN at 23 C. After washing free protein molecules with 100 ll buffer A, the flow cell was filled with 20 ll buffer A containing the oxygen scavenger system (Fig. 4A). Images were taken as described in Microscopy. At least 5 fields of images were recorded for each assay, and statistical analysis was performed for two independent experiments. The positions of 488bio-CPN and Cy5-CS in the same field were marked individually. Complexes of 488bio-CPN and Cy5-CS were revealed as positions where the two fluorescence signals were superimposed coincidently. The agreement ratio (%) was calculated by dividing the number of Cy5-CS molecules bound to 488bio-CPN by the total number of 488bio-CPN molecules. Similar experiments were performed by changing the order of the addition of PhPFD, CS and 488bio-CPN.
3. Results and discussion 3.1. Dissociation of substrate proteins from PhPFD complexes Previously we have performed kinetic studies of the interaction between PhPFD and group II chaperonin using SPR sensor [8,17]. However, we failed to measure binding and dissociation constants between prefoldin and unfolded substrate proteins by such conventional methods because nonspecific binding of unfolded substrate proteins to the supports or instruments was significant and unavoidable. In this study, we have developed a method based on fluorescence microscopy taking advantages of the following characteristics of PhPFD. First, PhPFD binds to a glass slide by electrostatic interaction so strong that amount of immobilized prefoldin did not decrease after washing. Secondly, PhPFD retained its binding activity to substrate protein after immobilization judging from the fact that the amount of substrate protein bound to PhPFD did not decrease after immobilization (data not shown). In addition, importantly, non-specific binding of the substrate proteins on the glass slide was negligible (5%). The dissociation of substrate proteins from PhPFD complexes was measured as shown in Fig. 1A. After mixing with
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðK D þ ½PFD0 þ ½Cy3-CS0 Þ2 4 ½PFD0 ½Cy3-CS0 2
.
ð1Þ
3720
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
Fig. 1. Measurement of koff of substrate proteins from PhPFD (A) Experimental procedure for assaying the dissociation of substrate proteins from PhPFD using fluorescence microscopy. First, fluorescent substrate proteins and PhPFD were mixed and incubated in a test tube for complex formation. Subsequently, excess amount of unlabeled substrate proteins were added, and aliquots of the solution taken at various incubation times were introduced into a flow cell. After washing out free proteins, the amount of substrate proteins bound to PhPFD attached to a glass surface was quantified by TIRFM. (B,C) Dissociation of substrate proteins (B, Cy3-CS; C, Cy5-GFP) from PhPFD at 23 C (triangles) and 50 C (circles). Broken lines and solid lines indicate the results of exponential fitting to the data at 23 and 50 C, respectively.
fluorescently labeled substrate proteins (Cy3-CS or Cy5-GFP) and subsequent incubation with excess unlabeled substrates, PhPFD was immobilized on the slide glass surface. Then the unbound substrate proteins were washed out and the amount of remaining fluorescently labeled substrate proteins bound to the glass via PhPFD was quantified. Fig. 1B and C show the dissociation of Cy3-CS and Cy5-GFP from PhPFD, respectively. The values of koff determined by single exponential fitting to the data were 1.0 · 103 s1 (23 C) and 5.9 · 103 s1 (50 C)
for Cy3-CS, and 5.5 · 104 s1 (23 C) and 3.0 · 103 s1 (50 C) for Cy5-GFP, respectively. Dissociation of substrate proteins from PhPFD at room temperature (23 C) was much slower than that at 50 C, suggesting that dissociation of substrate proteins from PhPFD during the preparation and observation of fluorescence microscopy at 23 C was negligible. Therefore, for further experiments, proteins were incubated at 50 C where the archaeal proteins are active, and microscopic observation was performed at room temperature (23 C).
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
3721
Fig. 2. Equilibrium binding of Cy3-CS to PhPFD. (A) Experimental procedure for measurement of KD of a substrate protein to PhPFD using fluorescence microscopy. Substrate proteins were incubated with PhPFD at 50 C. Then an aliquot of the solution was introduced into a flow cell. After washing out free proteins, the amount of complexes of substrate proteins and PhPFD attached to a glass surface were quantified by TIRFM. (B) The amount of Cy3-CS molecules bound to PhPFD at 50 C was measured with various concentrations of Cy3-CS ([Cy3-CS]0) and fixed concentrations of PhPFD ([PFD]0) (20 nM: squares; 50 nM: triangles; 100 nM: circles). The molar ratio of Cy3-CS to PhPFD varied from 0.5 to 6. KD values were obtained by fitting the data using Eq. (1) and fitted results are shown in lines.
3.2. KD and the second-order association rate constants between substrate proteins and PhPFD Dissociation constants (KD) of substrate proteins to PhPFD was measured using TIRFM (Fig. 2A). Denatured Cy3-CS was incubated with fixed concentrations of PhPFD (20, 50 or 100 nM) in buffer A at 50 C for 10 min, and the solution was infused into a flow cell. After washing out free substrate proteins with buffer A, the amount of substrate proteins bound to PhPFD was measured by TIRFM. KD was determined by fitting the data with Eq. (1) (Fig. 2B). KD values obtained at PhPFD concentrations of 20, 50 and 100 nM were 15, 20 and 20 nM, respectively. The average value of KD was 19 ± 3 nM. KD of Cy5-GFP to PhPFD was also determined and was 5 nM (data not shown). Using the obtained values of KD and koff, the second-order association rate constants (kon) were calculated, and these were 3.0 · 105 M1 s1 for Cy3-CS and 6.0 · 105 M1 s1 for Cy5GFP. Taking these results into account, we conclude that the substrate protein–PhPFD complexes are not stable, in other words, the complex is in a dynamic equilibrium of association and dissociation.
3.3. Chaperonin enhances dissociation of substrate proteins from PhPFD complexes Dissociation of substrate proteins from PhPFD complex in the presence of chaperonin (aN472C) was measured (Fig. 3). aN472C facilitated the release of Cy3-CS from PhPFD, which was demonstrated by 5-fold increase in koff in the presence of aN472C (Fig. 3B and G, koff = 2.8 · 102 s1). aN472C also accelerated the release of Cy5-GFP from PhPFD about 5-fold (Fig. 3C, koff = 1.1 · 102 s1). This effect was absent when IPMDH was used, indicating that the accelerated release of substrate protein was aN472C specific (Fig. 3D and G, koff = 5.4 · 103 s1). A PhPFD mutant, tc8atc5b, which was found to have similar affinity with substrate proteins as that of the wild-type, but had impaired affinity with chaperonin [8], is used to further confirm the effect of prefoldin–chaperonin interaction (Fig. 3E). koff of Cy3-CS from tc8atc5b without chaperonin is 9.0 · 103 s1 (triangles and a broken line of Fig. 3E), which is similar to that of wild-type. As expected, chaperonin did not promote the release of substrate protein from tc8atc5b (circles and a solid line of Fig. 3E, koff = 4.7 · 103 s1). Taken all together, we conclude that facilitated release of substrate
3722
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
Fig. 3. Dissociation of substrate proteins from PhPFD in the presence (circles and fitted solid lines) or absence (triangles and fitted broken lines) of chaperonin. koff values were calculated by exponentially fitting to data points and fitted results are shown in lines. Error bars indicate standard deviations. (A) Experimental procedure for measurement of koff in the presence of chaperonin. See also Fig. 1A for the procedure in the absence of chaperonin. (B) Dissociation of Cy3-CS from PhPFD in the presence or absence of aN472C. (C) Dissociation of Cy5-GFP from PhPFD in the presence or absence of aN472C. (D) Dissociation of Cy3-CS from PhPFD in the presence or absence of IPMDH (negative control for aN472C). (E) Dissociation of Cy3-CS from PhPFD mutant (tc8atc5b) in the presence or absence of aN472C. (F) Dissociation of Cy3-CS from PhPFD in a solution containing 5 mM ATP in the presence or absence of aN472C. (G) Relative koff values of Cy3-CS from the complex with PhPFD.
protein from PhPFD is caused by a specific interaction between prefoldin and the chaperonin. The acceleration effect of aN472C on Cy3-CS dissociation from PhPFD was reduced when 5 mM ATP was added (Fig. 3F and G, koff = 1.8 · 102 s1), suggesting that the promoted dissociation of substrate proteins from prefoldin is related to the chaperonin conformational state since we have recently shown that the conformation of chaperonin changes to the closed conformation in the presence of ATP [15].
This facilitated release of substrate protein would be caused by the formation of ternary complex of prefoldin, substrate protein and chaperonin, as suggested by MartinBenito et al. [10] for eukaryotic prefoldin, actin and CCT complex. In other words, our observation supports the idea that the ternary complex is also formed for archaeal prefoldin system, which is reasonable since the structural similarities of prefoldins and type II chaperonins from archaea and eukaryote are high.
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
3723
Fig. 4. Single molecule imaging of Cy5-CS and 488bio-CPN complexes (A) The schematic drawing of the experiment. Complex of PhPFD and Cy5CS was mixed with 488bio-CPN and incubated at 50 C for 10 min in a test tube. The solution was infused into a flow cell and 488bio-CPN was attached to a glass slide via biotinylated BSA and streptavidin. The fluorescence of individual 488bio-CPN and Cy5-CS molecules were observed by TIRFM. (B) Single molecule imaging of 488bio-CPN immobilized on the glass slide (top) and Cy5-CS bound to 488bio-CPN (bottom). Individual complexes of 488bio-CPN bound to Cy5-CS were revealed as spots where the two fluorescence signals are superimposed (arrow heads). Scale bar, 5 lM.
One possible reason for the facilitated substrate release is conformational change of prefoldin. Outward conformational change of archaeal prefoldin upon substrate protein binding was observed by Lundin et al. [6]. So, it is possible to assume that the similar outward conformational change of the prefoldin tentacles upon chaperonin binding results in a faster release of the substrate protein. Another possible reason is a loss of cooperativity of prefoldin toward substrate protein. Since the six tentacles of prefoldin have been shown to bind to substrate proteins in a concerted manner [6,8,9], it is likely that the cooperative binding ability of prefoldin to substrate proteins decrease upon chaperonin binding to one(s) of prefoldin tentacles, which would facilitate a release of substrate proteins from prefoldin. 3.4. Single molecule imaging of chaperonin–substrate protein complexes To confirm the delivery of the substrate protein arrested by PhPFD to the chaperonin, single molecule observation was carried out by TIRFM (Fig. 4A). The solution containing complex of PhPFD and Cy5-CS was mixed with aN472C modified with Alexa Fluor 488 and biotin (488bio-CPN) and incubated at 50 C for 10 min. After capturing 488bio-CPN on the glass slide via streptavidin in a flow cell, the fluorescence of individual 488bio-CPN and Cy5-CS molecules were observed by TIRFM. The positions of 488bio-CPN and Cy5-CS were identified independently using illumination by blue and red laser, respectively. As shown in Fig. 4B, fluorescent spots of Cy5-CS were observed at the same positions of the fluorescence spots of 488bio-CPN. These superpositions revealed the binding of Cy5-CS to 488bioCPN. The agreement ratio of Cy5-CS molecules bound to 488bio-CPN was about 40% (142/358). Co-incubation with another protein, IPMDH, did not affect the ratio (42%, 286/666), indicating the specific binding of protein substrates to 488bioCPN. Addition of PhPFD after the formation of the substrate protein and chaperonin complexes did not decrease the ratio of Cy5-CS bound to 488bio-CPN (43%, 353/813). This result indicates that PhPFD cannot plunder the substrate protein bound to ‘‘open form’’ of chaperonin without ATP [15,20]. It is also possible that substrate protein trapped by chaperonin is unreachable from PhPFD by steric hindrance of helical protrusion, since the substrate protein binding site of chaperonin is
now thought to be located inside face of the apical domain, just below the helical protrusion [14,20]. Further experiments to examine the effect of ATP, would be necessary to understand the mechanisms for binding of prefoldin to chaperonin and for substrate hand-off in vivo.
4. Conclusion Although it has been postulated that unfolded proteins are transferred from prefoldin to the chaperonin, the detailed hand-off mechanism has remained unsolved to date. Our kinetic study could partly explain the detail of the mechanism. Prefoldin and substrate proteins are in a dynamic equilibrium of association and dissociation. Upon addition of chaperonin, release of the substrate from prefoldin is facilitated by about 5fold. This observation indirectly supports the presence of ternary complex of prefoldin, substrate protein and chaperonin. After the facilitated release of substrate proteins from prefoldin, they should enter the cavity of the chaperonin since prefoldin binds to the chaperonin as ‘‘the lid of a pot’’ [10]. Our results support the idea that substrate proteins are efficiently folded by the type II chaperonin in cooperation with prefoldin. Acknowledgements: We thank T. Yoshida and K. Watanabe for helpful suggestions and critical reading of the manuscript. The IPMDH expression plasmid was a kind gift from A.Yamagishi (Tokyo University of Pharmacy and Life Science). The work reported here is a part of the 21st Century COE (Center of Excellence) program for the ‘‘Physics of Self-organized Systems’’ (T.F.) and the ‘‘Future Nano-Materials’’ (M.Y.) research and education projects, which are supported financially by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT). This study was also supported by grants-in-aid for Specially Promoted Research (T.F.) and for scientific research on priority areas (14037266 and 15032254 to H.T., and 13033008, 14037216 and 15032212 to M.Y.), and a grant from the National Project on Protein Structural and Functional Analyses (M.Y) from MEXT.
References [1] Geissler, S., Siegers, K. and Schiebel, E. (1998) A novel protein complex promoting formation of functional alpha- and gammatubulin. EMBO J. 17, 952–966.
3724 [2] Hansen, W.J., Cowan, N.J. and Welch, W.J. (1999) Prefoldinnascent chain complexes in the folding of cytoskeletal proteins. J. Cell Biol. 145, 265–277. [3] Siegers, K., Waldmann, T., Leroux, M.R., Grein, K., Shevchenko, A., Schiebel, E. and Hartl, F.U. (1999) Compartmentation of protein folding in vivo: sequestration of non-native polypeptide by the chaperonin–GimC system. EMBO J. 18, 75– 84. [4] Vainberg, I.E., Lewis, S.A., Rommelaere, H., Ampe, C., Vandekerckhove, J., Klein, H.L. and Cowan, N.J. (1998) Prefoldin, a chaperone that delivers unfolded proteins to cytosolic chaperonin. Cell 93, 863–873. [5] Leroux, M.R., Fandrich, M., Klunker, D., Siegers, K., Lupas, A.N., Brown, J.R., Schiebel, E., Dobson, C.M. and Hartl, F.U. (1999) MtGimC, a novel archaeal chaperone related to the eukaryotic chaperonin cofactor GimC/prefoldin. EMBO J. 18, 6730–6743. [6] Lundin, V.F., Stirling, P.C., Gomez-Reino, J., Mwenifumbo, J.C., Obst, J.M., Valpuesta, J.M. and Leroux, M.R. (2004) Molecular clamp mechanism of substrate binding by hydrophobic coiled-coil residues of the archaeal chaperone prefoldin. Proc. Natl. Acad. Sci. USA 101, 4367–4372. [7] Okochi, M., Yoshida, T., Maruyama, T., Kawarabayasi, Y., Kikuchi, H. and Yohda, M. (2002) Pyrococcus prefoldin stabilizes protein-folding intermediates and transfers them to chaperonins for correct folding. Biochem. Biophys. Res. Commun. 291, 769–774. [8] Okochi, M., Nomura, T., Zako, T., Arakawa, T., Iizuka, R., Ueda, H., Funatsu, T., Leroux, M. and Yohda, M. (2004) Kinetics and binding sites for interaction of prefoldin with group II chaperonin: contiguous non-native substrate and chaperonin binding sites in archaeal prefoldin. J. Biol. Chem. 279, 31788– 31795. [9] Siegert, R., Leroux, M.R., Scheufler, C., Hartl, F.U. and Moarefi, I. (2000) Structure of the molecular chaperone prefoldin: unique interaction of multiple coiled coil tentacles with unfolded proteins. Cell 103, 621–632. [10] Martin-Benito, J., Boskovic, J., Gomez-Puertas, P., Carrascosa, J.L., Simons, C.T., Lewis, S.A., Bartolini, F., Cowan, N.J. and Valpuesta, J.M. (2002) Structure of eukaryotic prefoldin and of its
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
[11] [12]
[13]
[14]
[15]
[16] [17] [18]
[19] [20]
complexes with unfolded actin and the cytosolic chaperonin CCT. EMBO J. 21, 6377–6386. Sakikawa, C., Taguchi, H., Makino, Y. and Yoshida, M. (1999) On the maximum size of proteins to stay and fold in the cavity of GroEL underneath GroES. J. Biol. Chem. 274, 21251–21256. Iizuka, R., Yoshida, T., Maruyama, T., Shomura, Y., Miki, K. and Yohda, M. (2001) Glycine at the 65th position plays an essential role in ATP-dependent protein folding by Archael group II chaperonin. Biochem. Biophys. Res. Commun. 289, 1118–1124. Numata, K., Muro, M., Akutsu, N., Nosoh, Y., Yamagishi, A. and Oshima, T. (1995) Thermal stability of chimeric isopropylmalate dehydrogenase genes constructed from a thermophile and a mesophile. Protein Eng. 8, 39–43. Shomura, Y., Yoshida, T., Iizuka, R., Maruyama, T., Yohda, M. and Miki, K. (2004) Crystal structures of the group II chaperonin from thermococcus strain KS-1: steric hindrance by the substituted amino acid, and inter-subunit rearrangement between two crystal forms. J. Mol. Biol. 335, 1265–1278. Iizuka, R., Yoshida, T., Shomura, Y., Miki, K., Maruyama, T., Odaka, M. and Yohda, M. (2003) ATP binding is critical for the conformational change from an open to closed state in archaeal group II chaperonin. J. Biol. Chem. 278, 44959–44965. Taguchi, H., Ueno, T., Tadakuma, H., Yoshida, M. and Funatsu, T. (2001) Single-molecule observation of protein-protein interactions in the chaperonin system. Nat. Biotechnol. 19, 861–865. Zako, T., Funatsu, T. and Yohda, M. (2004) Kinetic analysis of interactions between archaeal prefoldin and chaperonin. Recent Res. Dev. Biophys. 3, 475–483. Harada, Y., Sakurada, K., Aoki, T., Thomas, D.D. and Yanagida, T. (1990) Mechanochemical coupling in actomyosin energy transduction studied by in vitro movement assay. J. Mol. Biol. 216, 49–68. Burston, S.G., Ranson, N.A. and Clarke, A.R. (1995) The origins and consequences of asymmetry in the chaperonin reaction cycle. J. Mol. Biol. 249, 138–152. Iizuka, R., So, S., Inobe, T., Yoshida, T., Zako, T., Kuwajima, K. and Yohda, M. (2004) Role of the helical protrusion in the conformational change and molecular chaperone activity of the archaeal group II chaperonin. J. Biol. Chem. 279, 18834–18839.
FEBS 29686
Facilitated release of substrate protein from prefoldin by chaperonin Tamotsu Zakoa,b,c,*,1, Ryo Iizukab, Mina Okochib, Tomoko Nomurab, Taro Uenoa, Hisashi Tadakumaa, Masafumi Yohdab, Takashi Funatsua,c,d,* a
Department of Physics, School of Science and Engineering, Waseda University, 3-4-1 Okubo, Shinjyuku-ku, Tokyo 169-8555, Japan b Department of Biotechnology and Life Science, Tokyo University of Agriculture and Technology, 2-24-16 Naka-cho, Koganei-shi, Tokyo 184-8588, Japan c Laboratory of Bio-Analytical Chemistry, Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan d Core Research for Evolutional Science and Technology, JST, Kawaguchi-shi, Saitama 332-0012, Japan Received 29 March 2005; revised 13 May 2005; accepted 26 May 2005 Available online 13 June 2005 Edited by Jesus Avila
Abstract Prefoldin is a chaperone that captures a proteinfolding intermediate and transfers it to the group II chaperonin for correct folding. However, kinetics of interactions between prefoldin and substrate proteins have not been investigated. In this study, dissociation constants and dissociation rate constants of unfolded proteins with prefoldin were firstly measured using fluorescence microscopy. Our results suggest that binding and release of prefoldin from hyperthermophilic archaea with substrate proteins were in a dynamic equilibrium. Interestingly, the release of substrate proteins from prefoldin was facilitated when chaperonin was present, supporting a handoff mechanism of substrate proteins from prefoldin to the chaperonin. 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Prefoldin; Chaperonin; Molecular chaperone; Hyperthermophilic archaea
1. Introduction Prefoldin is a heterohexameric chaperone that cooperates with the group II chaperonin [1]. Prefoldin captures actins and tubulins in the unfolded state and transfers them to a cytosolic chaperonin, chaperonin containing TCP-1 (CCT) for functional folding [2–4]. Prefoldin exists in archaea as well as * Corresponding authors. Fax: +81 48 462 4658 (T. Zako); +81 3 5802 3339 (T. Funatsu). E-mail addresses: [email protected] (T. Zako), [email protected] (T. Funatsu).
group II chaperonins [4], irrespective of their lack of actins or tubulins. To date, archaeal prefoldins from Methanobacterium thermoautotrophicum (MtGimC) [5,6] and Pyrococcus horikoshii OT3 (PhPFD) [7,8] have been functionally characterized. They are capable of stabilizing non-native proteins and releasing them for subsequent chaperonin-dependent folding in vitro. Crystal structure of MtGimC [9] and electron microscopy of eukaryotic (bovine and human) prefoldin [10] have shown that prefoldin has the appearance of a jellyfish consisting of a double beta barrel assembly with six long coiled coils protruding from it. Biochemical and structural studies indicate that these tentacles bind to substrate proteins in a concerted manner [6,8–10] and can move outward for efficient interaction with substrate proteins [6]. Observation of prefoldin–CCT complexes by electron microscopy has revealed that the outer regions of the prefoldin tentacles interact with the inner regions of the apical domains of the chaperonin [10]. Although these data support a handoff mechanism of substrate proteins from prefoldin to the chaperonin, the detailed mechanism remains to be solved. In this study, we examined the dissociation kinetics of PhPFD and substrate proteins [porcine heart citrate synthase (CS) and green fluorescent protein (GFP)]. We found that binding and release of PhPFD with substrate proteins in the folding or unfolding intermediate state were in a dynamic equilibrium. The release of substrate proteins from PhPFD was facilitated in the presence of chaperonin, supporting a handoff mechanism of substrate proteins from prefoldin to the chaperonin.
2. Materials and methods
1
Present address: Bioengineering Laboratory, RIKEN (The Institute of Physical and Chemical Research), 2-1, Hirosawa, Wako, Saitama 351-0198, Japan. Abbreviations: PhPFD, prefoldin from Pyrococcus horikoshii; CS, citrate synthase; GFP, green fluorescent protein; IPMDH, 3-isopropylmalate dehydrogenase; tc8atc5b, PhPFD mutant with 8- and 5-amino acid truncations from C-terminus of a and b subunit; aN472C, N472C point mutant of Thermococcus strain KS-1 chaperonin a-subunit; 488bio-CPN, aN472C modified with Alexa Fluor 488 and biotin; BSA, bovine serum albumin; TIRFM, total internal reflection fluorescence microscopy; koff, dissociation rate constant; KD, dissociation constant; buffer A, a solution containing 25 mM HEPES, pH 7.4, 100 mM KCl, 5 mM MgCl2
2.1. Proteins CS was purchased from Sigma (St. Louis, MO). GFP was purified as described previously [11,12]. 3-Isopropylmalate dehydrogenase (IPMDH) from Thermus thermophilus strain HB8 was purified as described [13]. PhPFD and its mutant tc8atc5b which has 8- and 5-amino acid truncations from C-terminus of a and b subunit [8] were expressed in E. coli BL21 (DE3) and purified as described [7,8]. In order to immobilize and visualize chaperonin, Asn472, which is located on the outer surface of the equatorial domain of Thermococcus strain KS-1 (T.KS-1) chaperonin a-subunit [14], was replaced by Cys (termed aN472C) using a QuikChange Site-directed Mutagenesis Kit (Stratagene, CA). The aN472C was purified as described for the wild-type [12,15].
0014-5793/$30.00 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2005.05.061
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
3719
Substrate proteins were labeled with fluorescent dyes for detection by fluorescence microscopy. 18 lM CS or 72 lM GFP was incubated with 35 lM Cy3 or 150 lM Cy5 Mono-reactive Dye (Amersham Biosciences) for 2 h at room temperature in 25 mM HEPES, pH 8.5, 20 mM CaCO3, 100 mM KCl, 5 mM MgCl2. The labeled substrate proteins were separated from the unreacted dyes by gel filtration. The molar ratios of Cy3 to CS, Cy5 to CS and Cy5 to GFP were 0.8, 1.0 and 1.1, respectively. aN472C modified with Alexa Fluor 488-C5-maleimide (Alexa 488; Molecular Probes, OR) and biotinPEAC5-maleimide (Dojindo Laboratory, Kumamoto, Japan) (termed 488bio-CPN) was obtained as described [16]. The molar ratio of Alexa 488 to aN472C was 1.1 throughout this study. Streptavidin was purchased from Sigma. Biotinylated bovine serum albumin (BSA) was prepared as described [16]. 2.2. Microscopy Fluorescently labeled proteins were observed by total internal reflection fluorescence microscopy (TIRFM) [16,17]. Cy5-, Cy3-, and Alexa Fluor 488-labeled proteins were illuminated with a He–Ne laser (1.0 mW, 632.8 nm, model GLG5350; NEC Corporation, Tokyo, Japan), a green solid-state laser (2.8 mW, 532 nm, l-Green model 4601; Uniphase, San Jose, CA), and a blue solid-state laser (2.0 mW, 473 nm, ML0250A, Nippon Avionics Co., Japan), respectively. The fluorescence emission from the specimen was collected with an oilimmersion microscope objective (1.40 NA, 100·, PlanApo; Olympus, Tokyo, Japan). Images were taken with a SIT camera (C2400-08; Hamamatsu Photonics, Shizuoka, Japan) coupled to an image intensifier (VS4-1845; Video Scope International, VA) and recorded onto videotapes for subsequent analysis. Images were digitally captured by PC and the average intensities of 4 video frames were measured using the Scion Image software (Scion Corporation, ML). 2.3. Measurement of dissociation rate constants of substrate proteins from PhPFD complex Dissociation rate constants (koff) of substrate proteins from PhPFD were obtained by quantifying the amount of substrate proteins bound to PhPFD by TIRFM (Fig. 1A). 20 nM of Cy3-CS or Cy5-GFP (denatured at pH 1.2 for 10 min) was mixed with 20 nM PhPFD in buffer A (25 mM HEPES, pH 7.4, 100 mM KCl, 5 mM MgCl2) and incubated for 10 min at 50 C where the archaeal proteins used in this study are active. Then, 200 nM unlabeled substrate proteins were added into the solution and further incubated at 23 or 50 C. 20 ll of the sample solution were then taken at various incubation times and infused into a flow cell (20 ll in volume) consisting of a glass slide overlaid with a coverslip and two spacers (50 lm in thickness) at room temperature (23 C). The flow cell was then rinsed with 100 ll of buffer A to wash out unbound substrate proteins. Subsequently, the flow cell was filled with 20 ll of buffer A containing an oxygen scavenger system (25 mM glucose, 2.5 lM glucose oxidase, 10 nM catalase, 10 mM dithiothreitol) to suppress photobleaching [18]. The fluorescence intensity of the labeled substrate proteins bound to PhPFD was quantified by TIRFM. The value of koff was calculated by exponential fitting to obtained results using KaleidaGraph 3.5 (Synergy Software, PA). 2.4. Measurement of equilibrium constants of PhPFD and substrate proteins Schematic drawing of the experiment was shown in Fig. 2A. Various concentrations of Cy3-CS or Cy5-GFP (denatured at pH 1.2 for 10 min) were mixed with fixed concentrations of PhPFD (20, 50 or 100 nM) in buffer A and incubated at 50 C for 10 min. Then, 20 ll of the sample solution were infused into a flow cell. Because PhPFD strongly bound to a glass surface, but not for substrate proteins, PhPFD remained on the glass surface after washing of the flow cell with 100 ll of buffer A. The fluorescence intensity was measured by TIRFM to determine the amount of substrate proteins bound to PhPFD. KD were determined by fitting to the data using Eq. (1) [19]
Fluorescence intensity ¼
ðK D þ ½PFD0 þ ½Cy3-CS0 Þ
2.5. Measurement of dissociation rate constants of substrate proteins from PhPFD in the presence of chaperonin Dissociation of substrate proteins from PhPFD in the presence of aN472C was quantified by TIRFM (Fig. 3A). First, complex of substrate protein (Cy3-CS or Cy5-GFP) and PhPFD was formed as described in Section 2.3 and incubated at 50 C for 10 min. Then, 200 nM unlabeled substrate protein and 400 nM aN472C (or 2.6 lM IPMDH as a negative control) were added to the sample and further incubated at 50 C. After the various incubation times, 20 ll of the sample solution were taken and infused into a flow cell. koff of substrate proteins from PhPFD were obtained as described above. A PhPFD mutant tc8atc5b which had reduced affinity with chaperonin was also used as a control. 2.6. Single molecule imaging of chaperonin and substrate proteins by TIRFM A flow cell of which glass surface was coated with streptavidin was prepared beforehand to immobilize 488bio-CPN as follows. Buffer A containing 43 lM biotinylated BSA was infused into the flow cell, being washed, then 17 lM streptavidin was infused, and washed again. On one hand, the solution containing 100 nM PhPFD and 50 nM Cy5CS in buffer A was incubated at 50 C for 10 min, followed by the addition of 100 nM 488bio-CPN and additional 10 min incubation. The solution was diluted approximately 1000-fold with buffer A, and immediately infused into the flow cell coated with streptavidin to immobilize 488bio-CPN at 23 C. After washing free protein molecules with 100 ll buffer A, the flow cell was filled with 20 ll buffer A containing the oxygen scavenger system (Fig. 4A). Images were taken as described in Microscopy. At least 5 fields of images were recorded for each assay, and statistical analysis was performed for two independent experiments. The positions of 488bio-CPN and Cy5-CS in the same field were marked individually. Complexes of 488bio-CPN and Cy5-CS were revealed as positions where the two fluorescence signals were superimposed coincidently. The agreement ratio (%) was calculated by dividing the number of Cy5-CS molecules bound to 488bio-CPN by the total number of 488bio-CPN molecules. Similar experiments were performed by changing the order of the addition of PhPFD, CS and 488bio-CPN.
3. Results and discussion 3.1. Dissociation of substrate proteins from PhPFD complexes Previously we have performed kinetic studies of the interaction between PhPFD and group II chaperonin using SPR sensor [8,17]. However, we failed to measure binding and dissociation constants between prefoldin and unfolded substrate proteins by such conventional methods because nonspecific binding of unfolded substrate proteins to the supports or instruments was significant and unavoidable. In this study, we have developed a method based on fluorescence microscopy taking advantages of the following characteristics of PhPFD. First, PhPFD binds to a glass slide by electrostatic interaction so strong that amount of immobilized prefoldin did not decrease after washing. Secondly, PhPFD retained its binding activity to substrate protein after immobilization judging from the fact that the amount of substrate protein bound to PhPFD did not decrease after immobilization (data not shown). In addition, importantly, non-specific binding of the substrate proteins on the glass slide was negligible (5%). The dissociation of substrate proteins from PhPFD complexes was measured as shown in Fig. 1A. After mixing with
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðK D þ ½PFD0 þ ½Cy3-CS0 Þ2 4 ½PFD0 ½Cy3-CS0 2
.
ð1Þ
3720
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
Fig. 1. Measurement of koff of substrate proteins from PhPFD (A) Experimental procedure for assaying the dissociation of substrate proteins from PhPFD using fluorescence microscopy. First, fluorescent substrate proteins and PhPFD were mixed and incubated in a test tube for complex formation. Subsequently, excess amount of unlabeled substrate proteins were added, and aliquots of the solution taken at various incubation times were introduced into a flow cell. After washing out free proteins, the amount of substrate proteins bound to PhPFD attached to a glass surface was quantified by TIRFM. (B,C) Dissociation of substrate proteins (B, Cy3-CS; C, Cy5-GFP) from PhPFD at 23 C (triangles) and 50 C (circles). Broken lines and solid lines indicate the results of exponential fitting to the data at 23 and 50 C, respectively.
fluorescently labeled substrate proteins (Cy3-CS or Cy5-GFP) and subsequent incubation with excess unlabeled substrates, PhPFD was immobilized on the slide glass surface. Then the unbound substrate proteins were washed out and the amount of remaining fluorescently labeled substrate proteins bound to the glass via PhPFD was quantified. Fig. 1B and C show the dissociation of Cy3-CS and Cy5-GFP from PhPFD, respectively. The values of koff determined by single exponential fitting to the data were 1.0 · 103 s1 (23 C) and 5.9 · 103 s1 (50 C)
for Cy3-CS, and 5.5 · 104 s1 (23 C) and 3.0 · 103 s1 (50 C) for Cy5-GFP, respectively. Dissociation of substrate proteins from PhPFD at room temperature (23 C) was much slower than that at 50 C, suggesting that dissociation of substrate proteins from PhPFD during the preparation and observation of fluorescence microscopy at 23 C was negligible. Therefore, for further experiments, proteins were incubated at 50 C where the archaeal proteins are active, and microscopic observation was performed at room temperature (23 C).
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
3721
Fig. 2. Equilibrium binding of Cy3-CS to PhPFD. (A) Experimental procedure for measurement of KD of a substrate protein to PhPFD using fluorescence microscopy. Substrate proteins were incubated with PhPFD at 50 C. Then an aliquot of the solution was introduced into a flow cell. After washing out free proteins, the amount of complexes of substrate proteins and PhPFD attached to a glass surface were quantified by TIRFM. (B) The amount of Cy3-CS molecules bound to PhPFD at 50 C was measured with various concentrations of Cy3-CS ([Cy3-CS]0) and fixed concentrations of PhPFD ([PFD]0) (20 nM: squares; 50 nM: triangles; 100 nM: circles). The molar ratio of Cy3-CS to PhPFD varied from 0.5 to 6. KD values were obtained by fitting the data using Eq. (1) and fitted results are shown in lines.
3.2. KD and the second-order association rate constants between substrate proteins and PhPFD Dissociation constants (KD) of substrate proteins to PhPFD was measured using TIRFM (Fig. 2A). Denatured Cy3-CS was incubated with fixed concentrations of PhPFD (20, 50 or 100 nM) in buffer A at 50 C for 10 min, and the solution was infused into a flow cell. After washing out free substrate proteins with buffer A, the amount of substrate proteins bound to PhPFD was measured by TIRFM. KD was determined by fitting the data with Eq. (1) (Fig. 2B). KD values obtained at PhPFD concentrations of 20, 50 and 100 nM were 15, 20 and 20 nM, respectively. The average value of KD was 19 ± 3 nM. KD of Cy5-GFP to PhPFD was also determined and was 5 nM (data not shown). Using the obtained values of KD and koff, the second-order association rate constants (kon) were calculated, and these were 3.0 · 105 M1 s1 for Cy3-CS and 6.0 · 105 M1 s1 for Cy5GFP. Taking these results into account, we conclude that the substrate protein–PhPFD complexes are not stable, in other words, the complex is in a dynamic equilibrium of association and dissociation.
3.3. Chaperonin enhances dissociation of substrate proteins from PhPFD complexes Dissociation of substrate proteins from PhPFD complex in the presence of chaperonin (aN472C) was measured (Fig. 3). aN472C facilitated the release of Cy3-CS from PhPFD, which was demonstrated by 5-fold increase in koff in the presence of aN472C (Fig. 3B and G, koff = 2.8 · 102 s1). aN472C also accelerated the release of Cy5-GFP from PhPFD about 5-fold (Fig. 3C, koff = 1.1 · 102 s1). This effect was absent when IPMDH was used, indicating that the accelerated release of substrate protein was aN472C specific (Fig. 3D and G, koff = 5.4 · 103 s1). A PhPFD mutant, tc8atc5b, which was found to have similar affinity with substrate proteins as that of the wild-type, but had impaired affinity with chaperonin [8], is used to further confirm the effect of prefoldin–chaperonin interaction (Fig. 3E). koff of Cy3-CS from tc8atc5b without chaperonin is 9.0 · 103 s1 (triangles and a broken line of Fig. 3E), which is similar to that of wild-type. As expected, chaperonin did not promote the release of substrate protein from tc8atc5b (circles and a solid line of Fig. 3E, koff = 4.7 · 103 s1). Taken all together, we conclude that facilitated release of substrate
3722
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
Fig. 3. Dissociation of substrate proteins from PhPFD in the presence (circles and fitted solid lines) or absence (triangles and fitted broken lines) of chaperonin. koff values were calculated by exponentially fitting to data points and fitted results are shown in lines. Error bars indicate standard deviations. (A) Experimental procedure for measurement of koff in the presence of chaperonin. See also Fig. 1A for the procedure in the absence of chaperonin. (B) Dissociation of Cy3-CS from PhPFD in the presence or absence of aN472C. (C) Dissociation of Cy5-GFP from PhPFD in the presence or absence of aN472C. (D) Dissociation of Cy3-CS from PhPFD in the presence or absence of IPMDH (negative control for aN472C). (E) Dissociation of Cy3-CS from PhPFD mutant (tc8atc5b) in the presence or absence of aN472C. (F) Dissociation of Cy3-CS from PhPFD in a solution containing 5 mM ATP in the presence or absence of aN472C. (G) Relative koff values of Cy3-CS from the complex with PhPFD.
protein from PhPFD is caused by a specific interaction between prefoldin and the chaperonin. The acceleration effect of aN472C on Cy3-CS dissociation from PhPFD was reduced when 5 mM ATP was added (Fig. 3F and G, koff = 1.8 · 102 s1), suggesting that the promoted dissociation of substrate proteins from prefoldin is related to the chaperonin conformational state since we have recently shown that the conformation of chaperonin changes to the closed conformation in the presence of ATP [15].
This facilitated release of substrate protein would be caused by the formation of ternary complex of prefoldin, substrate protein and chaperonin, as suggested by MartinBenito et al. [10] for eukaryotic prefoldin, actin and CCT complex. In other words, our observation supports the idea that the ternary complex is also formed for archaeal prefoldin system, which is reasonable since the structural similarities of prefoldins and type II chaperonins from archaea and eukaryote are high.
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
3723
Fig. 4. Single molecule imaging of Cy5-CS and 488bio-CPN complexes (A) The schematic drawing of the experiment. Complex of PhPFD and Cy5CS was mixed with 488bio-CPN and incubated at 50 C for 10 min in a test tube. The solution was infused into a flow cell and 488bio-CPN was attached to a glass slide via biotinylated BSA and streptavidin. The fluorescence of individual 488bio-CPN and Cy5-CS molecules were observed by TIRFM. (B) Single molecule imaging of 488bio-CPN immobilized on the glass slide (top) and Cy5-CS bound to 488bio-CPN (bottom). Individual complexes of 488bio-CPN bound to Cy5-CS were revealed as spots where the two fluorescence signals are superimposed (arrow heads). Scale bar, 5 lM.
One possible reason for the facilitated substrate release is conformational change of prefoldin. Outward conformational change of archaeal prefoldin upon substrate protein binding was observed by Lundin et al. [6]. So, it is possible to assume that the similar outward conformational change of the prefoldin tentacles upon chaperonin binding results in a faster release of the substrate protein. Another possible reason is a loss of cooperativity of prefoldin toward substrate protein. Since the six tentacles of prefoldin have been shown to bind to substrate proteins in a concerted manner [6,8,9], it is likely that the cooperative binding ability of prefoldin to substrate proteins decrease upon chaperonin binding to one(s) of prefoldin tentacles, which would facilitate a release of substrate proteins from prefoldin. 3.4. Single molecule imaging of chaperonin–substrate protein complexes To confirm the delivery of the substrate protein arrested by PhPFD to the chaperonin, single molecule observation was carried out by TIRFM (Fig. 4A). The solution containing complex of PhPFD and Cy5-CS was mixed with aN472C modified with Alexa Fluor 488 and biotin (488bio-CPN) and incubated at 50 C for 10 min. After capturing 488bio-CPN on the glass slide via streptavidin in a flow cell, the fluorescence of individual 488bio-CPN and Cy5-CS molecules were observed by TIRFM. The positions of 488bio-CPN and Cy5-CS were identified independently using illumination by blue and red laser, respectively. As shown in Fig. 4B, fluorescent spots of Cy5-CS were observed at the same positions of the fluorescence spots of 488bio-CPN. These superpositions revealed the binding of Cy5-CS to 488bioCPN. The agreement ratio of Cy5-CS molecules bound to 488bio-CPN was about 40% (142/358). Co-incubation with another protein, IPMDH, did not affect the ratio (42%, 286/666), indicating the specific binding of protein substrates to 488bioCPN. Addition of PhPFD after the formation of the substrate protein and chaperonin complexes did not decrease the ratio of Cy5-CS bound to 488bio-CPN (43%, 353/813). This result indicates that PhPFD cannot plunder the substrate protein bound to ‘‘open form’’ of chaperonin without ATP [15,20]. It is also possible that substrate protein trapped by chaperonin is unreachable from PhPFD by steric hindrance of helical protrusion, since the substrate protein binding site of chaperonin is
now thought to be located inside face of the apical domain, just below the helical protrusion [14,20]. Further experiments to examine the effect of ATP, would be necessary to understand the mechanisms for binding of prefoldin to chaperonin and for substrate hand-off in vivo.
4. Conclusion Although it has been postulated that unfolded proteins are transferred from prefoldin to the chaperonin, the detailed hand-off mechanism has remained unsolved to date. Our kinetic study could partly explain the detail of the mechanism. Prefoldin and substrate proteins are in a dynamic equilibrium of association and dissociation. Upon addition of chaperonin, release of the substrate from prefoldin is facilitated by about 5fold. This observation indirectly supports the presence of ternary complex of prefoldin, substrate protein and chaperonin. After the facilitated release of substrate proteins from prefoldin, they should enter the cavity of the chaperonin since prefoldin binds to the chaperonin as ‘‘the lid of a pot’’ [10]. Our results support the idea that substrate proteins are efficiently folded by the type II chaperonin in cooperation with prefoldin. Acknowledgements: We thank T. Yoshida and K. Watanabe for helpful suggestions and critical reading of the manuscript. The IPMDH expression plasmid was a kind gift from A.Yamagishi (Tokyo University of Pharmacy and Life Science). The work reported here is a part of the 21st Century COE (Center of Excellence) program for the ‘‘Physics of Self-organized Systems’’ (T.F.) and the ‘‘Future Nano-Materials’’ (M.Y.) research and education projects, which are supported financially by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT). This study was also supported by grants-in-aid for Specially Promoted Research (T.F.) and for scientific research on priority areas (14037266 and 15032254 to H.T., and 13033008, 14037216 and 15032212 to M.Y.), and a grant from the National Project on Protein Structural and Functional Analyses (M.Y) from MEXT.
References [1] Geissler, S., Siegers, K. and Schiebel, E. (1998) A novel protein complex promoting formation of functional alpha- and gammatubulin. EMBO J. 17, 952–966.
3724 [2] Hansen, W.J., Cowan, N.J. and Welch, W.J. (1999) Prefoldinnascent chain complexes in the folding of cytoskeletal proteins. J. Cell Biol. 145, 265–277. [3] Siegers, K., Waldmann, T., Leroux, M.R., Grein, K., Shevchenko, A., Schiebel, E. and Hartl, F.U. (1999) Compartmentation of protein folding in vivo: sequestration of non-native polypeptide by the chaperonin–GimC system. EMBO J. 18, 75– 84. [4] Vainberg, I.E., Lewis, S.A., Rommelaere, H., Ampe, C., Vandekerckhove, J., Klein, H.L. and Cowan, N.J. (1998) Prefoldin, a chaperone that delivers unfolded proteins to cytosolic chaperonin. Cell 93, 863–873. [5] Leroux, M.R., Fandrich, M., Klunker, D., Siegers, K., Lupas, A.N., Brown, J.R., Schiebel, E., Dobson, C.M. and Hartl, F.U. (1999) MtGimC, a novel archaeal chaperone related to the eukaryotic chaperonin cofactor GimC/prefoldin. EMBO J. 18, 6730–6743. [6] Lundin, V.F., Stirling, P.C., Gomez-Reino, J., Mwenifumbo, J.C., Obst, J.M., Valpuesta, J.M. and Leroux, M.R. (2004) Molecular clamp mechanism of substrate binding by hydrophobic coiled-coil residues of the archaeal chaperone prefoldin. Proc. Natl. Acad. Sci. USA 101, 4367–4372. [7] Okochi, M., Yoshida, T., Maruyama, T., Kawarabayasi, Y., Kikuchi, H. and Yohda, M. (2002) Pyrococcus prefoldin stabilizes protein-folding intermediates and transfers them to chaperonins for correct folding. Biochem. Biophys. Res. Commun. 291, 769–774. [8] Okochi, M., Nomura, T., Zako, T., Arakawa, T., Iizuka, R., Ueda, H., Funatsu, T., Leroux, M. and Yohda, M. (2004) Kinetics and binding sites for interaction of prefoldin with group II chaperonin: contiguous non-native substrate and chaperonin binding sites in archaeal prefoldin. J. Biol. Chem. 279, 31788– 31795. [9] Siegert, R., Leroux, M.R., Scheufler, C., Hartl, F.U. and Moarefi, I. (2000) Structure of the molecular chaperone prefoldin: unique interaction of multiple coiled coil tentacles with unfolded proteins. Cell 103, 621–632. [10] Martin-Benito, J., Boskovic, J., Gomez-Puertas, P., Carrascosa, J.L., Simons, C.T., Lewis, S.A., Bartolini, F., Cowan, N.J. and Valpuesta, J.M. (2002) Structure of eukaryotic prefoldin and of its
T. Zako et al. / FEBS Letters 579 (2005) 3718–3724
[11] [12]
[13]
[14]
[15]
[16] [17] [18]
[19] [20]
complexes with unfolded actin and the cytosolic chaperonin CCT. EMBO J. 21, 6377–6386. Sakikawa, C., Taguchi, H., Makino, Y. and Yoshida, M. (1999) On the maximum size of proteins to stay and fold in the cavity of GroEL underneath GroES. J. Biol. Chem. 274, 21251–21256. Iizuka, R., Yoshida, T., Maruyama, T., Shomura, Y., Miki, K. and Yohda, M. (2001) Glycine at the 65th position plays an essential role in ATP-dependent protein folding by Archael group II chaperonin. Biochem. Biophys. Res. Commun. 289, 1118–1124. Numata, K., Muro, M., Akutsu, N., Nosoh, Y., Yamagishi, A. and Oshima, T. (1995) Thermal stability of chimeric isopropylmalate dehydrogenase genes constructed from a thermophile and a mesophile. Protein Eng. 8, 39–43. Shomura, Y., Yoshida, T., Iizuka, R., Maruyama, T., Yohda, M. and Miki, K. (2004) Crystal structures of the group II chaperonin from thermococcus strain KS-1: steric hindrance by the substituted amino acid, and inter-subunit rearrangement between two crystal forms. J. Mol. Biol. 335, 1265–1278. Iizuka, R., Yoshida, T., Shomura, Y., Miki, K., Maruyama, T., Odaka, M. and Yohda, M. (2003) ATP binding is critical for the conformational change from an open to closed state in archaeal group II chaperonin. J. Biol. Chem. 278, 44959–44965. Taguchi, H., Ueno, T., Tadakuma, H., Yoshida, M. and Funatsu, T. (2001) Single-molecule observation of protein-protein interactions in the chaperonin system. Nat. Biotechnol. 19, 861–865. Zako, T., Funatsu, T. and Yohda, M. (2004) Kinetic analysis of interactions between archaeal prefoldin and chaperonin. Recent Res. Dev. Biophys. 3, 475–483. Harada, Y., Sakurada, K., Aoki, T., Thomas, D.D. and Yanagida, T. (1990) Mechanochemical coupling in actomyosin energy transduction studied by in vitro movement assay. J. Mol. Biol. 216, 49–68. Burston, S.G., Ranson, N.A. and Clarke, A.R. (1995) The origins and consequences of asymmetry in the chaperonin reaction cycle. J. Mol. Biol. 249, 138–152. Iizuka, R., So, S., Inobe, T., Yoshida, T., Zako, T., Kuwajima, K. and Yohda, M. (2004) Role of the helical protrusion in the conformational change and molecular chaperone activity of the archaeal group II chaperonin. J. Biol. Chem. 279, 18834–18839.