Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống
1
/ 43 trang
THÔNG TIN TÀI LIỆU
Thông tin cơ bản
Định dạng
Số trang
43
Dung lượng
1,22 MB
Nội dung
INTERACTION OF MITOCHONDRIAL SSC1 DOMAINS WITH N‐TERMINAL DOMAIN OF TIM44 LIANG CHEN (B.Sc., Tsinghua University) A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SCIENCE DEPARTMENT OF BIOLOGICAL SCIENCES NATIONAL UNIVERSITY OF SINGAPORE 2015 i D Declaratio on I hereby declarre that this tthesis is myy original wo ork and it has been wrritten by me e in urces of info ormation w which have its eentirety. I haave duly acknowledgedd all the sou beeen used in th he thesis. Thiss thesis has also not be een submittted for any d degree in an ny universitty previously. ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Liang Chen n 2 2015/1/22 ii Acknowledgement Foremost, I would like to express my sincere gratitude to my advisor Prof. Yang Daiwen for the continuous support of my study and research, for his patience, motivation, enthusiasm, and immense knowledge. His guidance helped me in all the time of research and writing of this thesis. His encouragement and tolerance warmed my heart with selfless love during the most difficult time in my study. In addition, I would like to thank Prof. Henry Mok for his encouragement, insightful comments, and hard questions during my research. My special thanks go to Dr. Fan jingsong for his help of the NMR structure calculation of Ssc1 LID sub‐domain. I am grateful my fellow labmates in NUS Structural biology labs: Dr. Lin Zhi, Dr. Zhang jingfeng, Dr. Lim Jack Wee, Dr. Tan Kang Wei, Dr. Wang shujing, etc. for their guidance, discussion and help on my study and research. Last but not the least; I would like to thank my parents, my wilfe and my son for supporting me spiritually throughout my life. iii Table of Contents Title page i Declaration ii Acknowledgements iii Table of Contents iv Summary vi List of Figures vii List of Tables viii Chapter 1 Background and Objectives 1 1.1 Protein import into the yeast mitochondria 1 1.2 The pathway of mitochondrial matrix protein translocation 2 1.3 The regulated cycling of Ssc1 by its interaction with Tim44 6 1.4 LID sub‐domain 9 1.5 Objective 10 Chapter 2 Study of the interaction of Tim44 N‐terminal domain and different Ssc1 domains 11 2.1 Materials and Methods 11 2.1.1 Media: LB broth, LB agar and M9 minimal media 11 2.1.2 Purification and characterization of Tim44 N terminal segment 11 iv 2.1.3 Purification and characterization of Ssc1 NBD, PBD and PBD 12 LID sub‐domain 2.1.4 Testing the interaction of Tim44N with both NBD and PBD of Ssc1 by His‐Tag pull down method 13 2.1.5 Testing the interaction of Tim44N with both NBD and PBD 13 of Ssc1 by NMR titration method 2.1.6 NMR spectroscopy 14 2.1.7 NMR data analysis 14 2.1.8 Structure calculation 14 2.1.9 Circular dichroism spectroscopy 15 2.2 Results and Discussion 15 2.2.1 Purification of the N‐terminal domain of Tim44 15 2.2.2 Structure of the N‐terminal domain of Tim44 16 2.2.3 Purification of the NBD and PBD of Ssc1 18 2.2.4 Secondary structure of the NBD and PBD of Ssc1 19 2.2.5 Interaction of Tim44N with the NBD and PBD of Ssc1 20 2.2.6 NMR structure of LID sub‐domain in Yeast SSC1 24 Chapter 3 Conclusion and Future works 29 3.1 Conclusion 29 3.2 Future work 29 References 31 v Summary Ssc1 is a heat shock protein 70 in the mitochondrial matrix of yeast and is essential for cell viability. It is recruited by Tim44 under the TIM23 channel. Ssc1 can bind to the presequence or N‐terminal signal sequence and then dissociate from Tim44. Although the interaction of Tim44 and Ssc1 is important for the function, how they interact still remains unknown. In this thesis, we used NMR spectroscopy and other biochemical and biophysical techniques to characterize the structures of Ssc1 domains and N‐terminal domain of Tim44 and study their interactions. We found that the N‐terminal domain of Tim44 adopts a disordered structure and the nucleotide binding domain and peptide binding domain of Ssc1 adopt well folded structures. The peptide binding domain can be divided into two sub‐domains: BETA and LID. We solved the three‐dimensional structure of the LID –domain by NMR. Although it shares low sequence identity with the LID domain of E. coli heat shock protein 70, their structures are highly similar. Interestingly, the disordered N‐terminal domain of Tim44 interacts with both the nucleotide and peptide binding domains of Ssc1. However, the LID sub‐domain and BETA‐domain with an extra short helix from the LID domain could not interact with Tim44. Because the interaction significantly reduces NMR signal intensities, we are unable to determine the exact binding sites. On the basis of our results, we propose that the overall conformation of the full‐ length peptide binding domain is required for binding Tim44, which may involve special orientation of the BETA and LID sub‐domains. vi List of Tables Table 1 NMR data and structure determination details for LID sub‐domain in Yeast SSC1 vii List of Figures Figure 1 Overview of mitochondrial protein import pathways. Figure 2 Model of mitochondrial matrix protein translocation Figure 3 The solution structure of ADP and substrate‐bound DnaK, an Hsp70 homolog in E.coli. Figure 4 SDS PAGE of Purified Tim44N Figure 5 Circular dichroism spectrum of Tim44N in 20 m phosphate (pH 7.0) Figure 6 1D 1H NMR spectrum of Tim44N in 20 m phosphate (pH 7.0) Figure 7 2D 1H‐15N HSQC NMR spectrum of Tim44N in 20 m phosphate (pH 7.0). Figure 8 SDS PAGE of Purified Ssc1 NBD, PBD, PBDmin‐YE Figure 9 Figure 10 Circular dichroism spectra of Ssc1 NBD and PBD in 20 m phosphate (pH 7.0) H 1D NMR of Ssc1 PBD Figure 11 His‐Tag Pull‐down of Tim44N with NBD and PBD Figure 12 Comparison of HSQC spectra of Tim44N in the free formand complex form Figure 13 Comparison of HSQC spectra of PBD in the free form and mixture form Figure 14 Comparison of 1H‐15N HSQC spectra of PBD Lid sub‐domain in the free form and mixture form Figure 15 Comparison of 1H‐15N HSQC spectra of PBDmin‐YE domain: free form and mixture form (red) Figure 16 The stereo‐view of the superposition of 20 energy‐minimized conformers representing the 3‐D NMR structure for LID sub‐domain in Yeast Ssc1 Figure 17 Ribbon diagram of the lowest‐energy conformer representing the 3‐D NMR structure of LID sub‐domain in Yeast SSC1 Figure 18 Sequence alignment for LID sub‐domain in yeast SSC1, E. Coli DNAK and human HSP110 viii Chapter 1 Background and Objectives 1.1 Protein import into the yeast mitochondria Yeast mitochondria contain about 800~1000 kinds of proteins, but only a small fractions of them are made in situ. The rest (about 99%) are encoded by the genome in the nucleus, synthesized in the cytosolic ribosomes.[1] These mitochondrial proteins are then recognized by the receptors on the outer membrane of mitochondria, and delivered into one of the four mitochondrial compartments: outer membrane, inner membrane, inter membrane space and the matrix. These four protein translocation pathways leading to different destinations are accomplished by the coordinated actions of several machineries located in both outer and inner mitochondrial membranes (Figure 1). The TOM (translocase of the outer membrane) complex is the major gateway for protein import. It recognizes the precursor mitochondrial proteins by their signal sequences and then collaborates with other mitochondrial translocases (SAM (sorting and assembly machinery), MIA (mitochondrial intermembrane space import and assembly), TIM22 (carrier translocase of the inner mitochondrial membrane), TIM23 (presequence translocase of the inner mitochondrial membrane), OXA (insert/export machinery of the inner membrane), PAM (presequence translocase associated motor)) for further sorting and delivery. [2‐3] 1 Figure 1. Overview O off mitochonddrial protein n import pa athways. Itt is adapted d ffrom [3]. 1.2 The patthway of mitochondriaal matrix prrotein transslocation One important pathway to impo rt preprote eins into the e mitochon drial matrixx involves delicate coo operation oof TOM co omplex, TIM23 compplex at the e brane, and PAM (pressequence‐asssociated motor) m com mplex in the e innermemb matrix (Figgure 2).[2‐3 3] The mattrix‐destine ed proteins contain aa positively‐ charged N‐‐terminal signal sequeence (preseq quence), fo orming an aamphipathicc helix. [4] W When the prresequencee emerges aat the outlett of the TO M complexx, it is recoggnized by Tim50, T a TTIM23 complex compo onent that t exposes a a hydrophilicc domain to o the interm membrane space. Tim50 then intteracts with h the intermembrane space domaain of Tim23, leading to the opeening of the e TIM23 translocation channel aat the inn ner membrrane. [5‐6] After the e presequencce enters th his channel, it is driven by the inne er membranne potentiaal △ψ to reaach the miitochondriaal matrix. The T final translocationn process iss 2 implies that the interaction may bbe not stron ng because exchange bbetween the e bound and d unbound d forms in an interm mediate NM MR time rregime can n significantlyy broaden tthe NMR siggnals. In this case, we a are unable tto use NMR R to determin ne the binding sites. Figure 12. Comparison n of HSQC spectra of Tim44N in the free foorm (green) and compleex form (red d). Red: HSSQC of the m mixture of 15 N‐labled TTim44N and d unlabled PB BD (molar ratio 1:3) We also trried to add d un‐labeledd Tim44N into 15N‐lab beled PBD. Under the e condition of o 1:1 ratio of Tim4 4N to PBD D, many HSQC peakss from PBD D disappeareed (Figure 1 13). In addittion, severaal cross pea aks shifted ssignificantlyy (Figure 13), further sh howing the interaction n of the two o proteins. Due to the e disappearance of maany peaks, we are cu urrently nott able to aanalyze the e spectra to ffind out which residuess are involvved in the biinding sites 21 1 rred: 15N‐PBD 15 g green: N‐PB BD:Tim44N=11:1 Figure 13. Comparison of HSQC C spectra off PBD in th he free form m (red) and d 15 mixture forrm (green). Green: HS HSQC of thee mixture of o N‐lableed PBD and d unlabled Tiim44N (mollar ratio 1:11). In order to o know if th he LID‐subddomain in PBD P is involved in the interaction n with Tim44 4N, we carrried out ttitration exxperiments by adding un‐labeled d Tim44N intto 15N‐labelled LID. Diffferent from m PBD, the HSQC specttrum of LID D was not influenced by Tim44N N (Figure even when n the mol ar ratio of LID:Tim44N N reached 2 2:1. The ressult shows tthat the LID D sub‐domaain does nott interact witth Tim44N. 22 2 Green: 15N‐LID Red: 15N‐LID: Tim m44N=1:1 Figure 14. Comparison n of H‐ N HSQC specctra of PBD Lid sub‐doomain in thee ffree form (green) ( and d mixture foorm (red). Red: R HSQC of the mixtture of 15N‐ labled LID a and un‐labeeled Tim44N N (molar rattio 1:1). 15 In addition n, PBDmin‐‐YE, a PBD truncated mutant which w lacks the last 3 predicted helix h and exxists as a m monomer in n high prote ein concenttration, wass found not to interact with Tim444N either (Figure ( 15). Since Tim444N cannott bind eitherr the LID subdomain o r the PBD ttruncated m mutant (whiich containss the BETA su ub‐domain and the firsst short helix in the LID D domain), w we propose e that the ovverall confo ormation off full‐length PBD in the e apo‐form is required d for Tim44N N binding, which w may involve spe ecial orientation of th e BETA and d LID sub‐do omain. Whe en the subbstrate bind ds to the BETA B sub‐ddomain, the e conformational change e of the PBD D domain w will be trigge ered, whichh lead to the e dissociation n of the Tim m44‐Ssc1 coomplex, an nd cause the leaving oof Ssc1 from m 23 3 the TIM23 complex. Grreen: 15N‐PBD Dmin‐YE 15 Re ed: N‐PBDm min‐YE: Tim444N=1:2 Figure 15. Comparison of H‐ N N HSQC speectra of PB BDmin‐YE doomain: freee fform (green n) and mixtture form (rred). Red: H HSQC of the e mixture of of 15N‐labled d PBDmin‐YEE and un‐lab beled Tim444N (molar ra atio of PBD‐‐YE:Tim44N N is 1:2) 15 of LID sub‐ddomain in Ye Yeast SSC1 2.2.6 NMR structure o ertiary fold of LID sub‐ domain, we e In order to gain structural insightss into the te solved the solution NM MR structurre. The NMR R structure of LID sub‐ddomain wass refined to ffinal RMSD of 0.92 ± 0.19 Å (baackbone) (Ta able 1) for the 20 bestt structures (Figure 16 and Figure 17). The NMR N structu ure of LID ssub‐domain n revealed a monomeric fold, com mprising thrree anti‐parrallel arrangged helixess. dle structu re is very similar to o the LID ssub‐domain n This three‐‐helix bund structures in E. coli homologues h s Dnak solvved by NM MR [23] andd X‐ray [24] thought thee sequence homogeneeity is very lo ow (Figure 18). 24 4 Table 1. NMR data and structure determination details for LID sub‐domain in Yeast SSC1 Parameters a All NOE distance restraints Intra‐residue Inter‐residue 997 261 Sequential (׀i–j׀ = 1) 313 Medium‐range ( 1 [...]... dissociation upon 6 signal peptide addition. (1) Domains of Ssc1 and Tim44 that participate in their interaction Tim44 can co‐immunoprecipitate with individual NBD and PBD of Ssc1. [16] In contrast, only the N terminal segment of Tim44 can form stable complex with the full length Ssc1. The interaction between Ssc1 and the well‐folded C terminal domain of Tim44 is very weak. [14] ... beads in 0.5mL PBS buffer. The boiled solution was analyzed by SDS‐PAGE. 2.1.5 Testing the interaction of Tim4 4N with both NBD and PBD of Ssc1 by NMR titration method 2D 1H‐1 5N HSQC comparison was used to confirm the interaction between Tim4 4N and Ssc1 PBD and between Tim4 4N and LID sub domain. 1 5N labled protein was synthesized by bacteria grown in M9 media with 1g/L 15NH4Cl, and then purified according to the same protocol as the unlabled protein. 1H‐... Based on the discoveries mentioned above, we propose the following model about the structure of the complex of Tim44 Ssc1: The N terminal of Tim44 has its binding sites on both NBD and PBD of Ssc1, probably in a way that the N terminal region of Tim4 4N binds to NBD and its C terminal region binds to PBD. When Ssc1 is in the ATP state without the presequence (substrate), Tim44 can interact ... mutation in Tim4 4N s second half region was also reported to weaken its interaction with Ssc1. [17] (3) Domains of Ssc1 that are required to cause the dissociation of Tim44 Ssc1 complex upon presequence addition The binding of the signal peptide to the PBD causes quick dissociation of Tim44 Ssc1 complex. However, in the absence of the NBD, Tim44 can be co‐immunoprecipitated with the Ssc1 PBD after signal peptide addition, indicating that NBD is involved in this dissociation process.[16] ... His‐Tag Pull‐down of TTim4 4N with NBD (leftt) and PBD ((right). S /N N: the superna atant after incubation of NBD or PBD with im mmobilized Tim4 4N via a a His‐tag; W W1‐W3: Wa ash of the N Ni‐NTA agarrose; E: the e elution of TTim4 4N with h its binding partners. We furtherr tried to identify wh ich residue es in Tim44 4N are invoolved in the e interaction with PBD. Thus NMR R titration ... N terminal domains of Tim44 (Tim4 4N) . As the first step, I aim to characterize interactions of Tm4 4N and different domains of Ssc1 in this thesis. In addition, I also aim to solve the three‐dimensional structure of the LID domain since it is critical for the function and understanding of the structural characters. 10 Chapter 2 Study of the interaction of Tim44 N terminal domain and ... Mutations of Ssc1 and Tim44 which weaken their interaction Co‐immunoprecipitation studies showed Ssc1 2 (P442S mutant of Ssc1) has decreased interaction with Tim44, but a second mutation of Ssc1 (D519E or V524I) can restore the reduced interaction to a normal level. These mutated residues are located in PBD of Ssc1. [16] An E67A mutation in Tim4 4N s second half region was also reported to weaken its ... and an extended α‐helical LID sub domain that covers the substrate binding site. Generally, the two domains of Hsp70 protein are closely interrelated in function and structure. When ATP is bound to the NBD, the PBD is in the open conformation with high on and off substrate binding rates. In contrast, the ADP state of NBD corresponds to a close state of PBD with high affinity to the substrate and low on and off ... implies that the interaction may bbe not stron ng because exchange bbetween the e bound and d unbound d forms in an interm mediate NM MR time rregime can n significantlyy broaden tthe NMR siggnals. In this case, we a are unable tto use NMR R to determin ne the binding sites. Figure 12. Comparison n of HSQC spectra of Tim4 4N in the free foorm (green) 1 and compleex form (red... 2.2.5 Intera action of Tim m4 4N withh the NBD and PBD of S Ssc1 In order to o determine e if Tim4 4N N interacts with Ssc1 PBD and N NBD, His‐tagg pull‐down eexperimentts were perfformed. Th he results (FFigure 11) inndicate thatt Tim4 4N intteracts with NBD and PPBD, respecttively. S /N W1 W2 W3 E S /N W1 W2 W W3 E NBD Tim4 4N PBD Tim4 4N ... 2.2.1 Purification of the N‐terminal domain of Tim44 15 2.2.2 Structure of the N‐terminal domain of Tim44 16 2.2.3 Purification of the NBD and PBD of Ssc1 18 2.2.4 Secondary structure of the NBD and PBD of Ssc1 ... biochemical and biophysical techniques to characterize the structures of Ssc1 domains and N‐terminal domain of Tim44 and study their interactions. We found that the N‐terminal domain of Tim44 adopts a disordered structure and the nucleotide binding domain and peptide binding domain of Ssc1 adopt well folded ... out the structure and dynamics of the complex between the full length Ssc1 and the N‐terminal domains of Tim44 (Tim44N). As the first step, I aim to characterize interactions of Tm44N and different domains of Ssc1 in this