www.sciencemag.org/cgi/content/full/science.1237908/DC1 Supplementary Materials for Structure of Parkin Reveals Mechanisms for Ubiquitin Ligase Activation Jean-François Trempe, Véronique Sauvé, Karl Grenier, Marjan Seirafi, Matthew Y. Tang, Marie Ménade, Sameer Al-Abdul-Wahid, Jonathan Krett, Kathy Wong, Guennadi Kozlov, Bhushan Nagar, Edward A. Fon,* Kalle Gehring* *Corresponding author. E-mail: [email protected] (K.G.); [email protected] (E.A.F.) Published 9 May 2013 on Science Express DOI: 10.1126/science.1237908 This PDF file includes: Materials and Methods Supplementary Text Figs. S1 to S12 Table S1 Captions for movies S1 and S2 References Other Supplementary Material for this manuscript includes the following: Movies S1 and S2
23
Embed
Supplementary Materials for - Science...NMR spectroscopy 15N-labeled UbcH7 and parkin constructs (full-length, R0-RBR, R0-RBR-W403A, and R0-RBR-W403A-C431S) were buffer-exchanged in
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Other Supplementary Material for this manuscript includes the following:
Movies S1 and S2
2
Materials and Methods Cloning, expression and purification of recombinant proteins
Rattus norvegicus full-length Parkin DNA was codon-optimized for E.coli expression (DNAexpress Inc., Montreal, Canada) and subcloned into pGEX6P-1 (GE Healthcare). Parkin deletions (77-465, 95-465, 141-465, 220-465) were generated by PCR amplification. PCR mutagenesis was used to generate parkin single-point mutants. Protein expression was done in BL21 (DE3) E. coli cells using conditions previously described (13) except for the IPTG concentration that was lowered to 25 μM. The pET28a-LIC-UbcH7 vector was a gift from C. Arrowsmith (Structural Genomics Consortium, Toronto, Canada). 15N-labeled UbcH7 was produced in M9 minimal medium supplemented with 15NH4Cl. All proteins were purified by GST-Sepharose 4B (GE healthcare) and eluted with 20 mM reduced glutathione in 20mM Tris/HCl, 120mM NaCl, 2mM DTT, pH 7.5. Eluted proteins were used without buffer-exchange for Ub assays with GST-parkin. To produce untagged parkin for crystallization, the GST-purified protein was further cleaved with 3c protease overnight at 4ºC and applied onto Superdex 75 16/60 (GE healthcare) in 30 mM Tris-HCl pH 8.0, 150 mM NaCl, 2 mM DTT. The peak containing cleaved parkin R0-RBR was reapplied onto fresh GST-Sepharose 4B and concentrated. To produce untagged parkin constructs for Ub assays, the GST-purified proteins were cleaved with 3C protease overnight at 4ºC and buffer-exchanged in 10 mM HEPES pH 7.0, 50 mM NaCl, 0.5 mM TCEP using Zeba desalting spin column (Thermo). His6-tag UbcH7 was produced in BL21 (DE3) E. coli cells using pET28a-LIC-UbcH7 and purified by Ni-NTA (Qiagen) followed by gel-filtration on Superdex 75 16/60 (GE healthcare).
Crystallization
Crystals of parkin R0-RBR were grown at 4°C using hanging drop vapor diffusion method by mixing 1 μl of protein at 12 mg ml-1 in 15 mM Tris-HCl pH 8.0 , 0.2 M NaCl, 10 mM DTT and 1 μl of mother liquor containing 0.9 M sodium malonate pH 7.0, 0.1 M HEPES pH 7.0, 5% (v/v) glycerol, 56 mM ß-mercaptoethanol. Crystals appeared after a week and reached their maximal size of 50 μm x 50 μm x 200 μm after three weeks. Crystals were cryoprotected in a solution of 2 M sodium malonate pH 7.0, 0.1 M HEPES pH 7.0, 5% (v/v) glycerol before being flash-frozen in liquid nitrogen. Crystals of parkin full-length were grown at 4°C using hanging drop vapor diffusion method by mixing 1 μl of protein at 7 mg ml-1 in 20 mM Tris-HCl pH 7.4, 2 mM DTT and 1 μl of mother liquor containing 22.5% PEG3350, 0.2 M (NH4)2SO4, 0.1 M HEPES pH 7.0, 10% (v/v) glycerol, 3% sorbitol and 25 mM ß-mercaptoethanol.
Data collection and structure determination
Diffraction data for parkin R0-RBR were collected at the CMCF beamline 08ID-1 at the Canadian Light Source (Table S1). A total of 250 images with an oscillation angle of 1° were measured at the Zn K-edge. Reflections were integrated using Mosflm (30) and scaled with SCALA as implemented in the CCP4 package (31). The structure was solved by SAD using AutoSol as implemented in the PHENIX package (32). Twenty-four zinc sites, corresponding to 3 molecules, were found in the asymmetric unit. Model building
3
was performed using the program COOT (33), and was refined using PHENIX. No explicit non-crystallographic symmetry restraints and no TLS were used during refinement. Low-resolution data for full-length parkin were collected at the F1 beamline at the Cornell High Energy Synchrotron Source (Table S1). Reflections were integrated and scaled using HKL2000 (HKL Research Inc.). The structure was determined by molecular replacement (MR) using Phaser (34). A first MR search was performed using trimer from the R0-RBR crystal structure as a search model. The solution (Z-score > 20) consisted of four trimers in each asymmetric unit. In a second MR round, the crystal structure of mouse parkin Ubl (pdb 2zeq (35)) was used as a search model, using the solution of the first MR search. One copy of the Ubl was found (Z-score > 9) and applied to all twelve chains using non-crystallographic symmetry operators. Rigid-body refinement was performed with PHENIX, treating the Ubl, IBR and RING0-RING1-REP-RING2 as distinct rigid bodies. Coordinates and structure factors of the R0-RBR and full-length parkin crystal structures were deposited in the Protein Data Bank under accession codes 4K7D and 4K95.
Small-angle X-ray scattering (SAXS)
Small-angle X-ray scattering data of parkin R0-RBR and full-length were collected on our in-house Anton Paar SAXSess camera equipped with a PANalytical PW3830 X-ray generator and a Roper/Princeton CCD detector. The beam length was set to 18 mm, and the beam profile was recorded using an image plate for subsequent desmearing. Scattering data were collected at 20°C at protein concentrations of: 2.9, 5.5, and 11.0 mg ml-1 for 3, 6 and 3 hours respectively for R0-RBR; 3.0 and 6.0 mg ml-1 for 6 and 6 hours respectively for R0-RBR W403A; and 2.5, 5.0 and 10.0 mg ml-1 for 1 hour each concentration of full-length. Background scattering from the buffer (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 8 mM DTT) was measured for 6 hours. Data were recorded from an image plate for the W403A construct, whereas other data sets were recorded using the CCD camera. Dark current correction, scaling, buffer subtraction, binning, desmearing and merging were performed using SAXSquant 3.0 (Anton Paar). The merged scattering curve was fitted to individual chains in the crystal structure using CRYSOL (36). The pair-distance distribution was calculated using GNOM (37).
Autoubiquitination assays
Ubiquitination assays were performed at 37°C in the presence of 50 mM Tris/HCl pH 7.5, 50 mM NaCl, 0.5 mM DTT, 2 mM ATP, 10 mM MgCl2, 40 nM E1, 2-4
-parkin. Reactions were stopped with 3X sample buffer plus 100 mM DTT and analyzed by SDS-PAGE. Proteins were then transferred to nitrocellulose and stained with Ponceau. Membranes were blocked with 5% milk in PBS-T (0.1% Tween 20) and incubated with mouse anti-ubiquitin (Covance) or rabbit anti-parkin (Abcam ab15954) diluted in PBS-T with 3% bovine serum albumin (BSA). The membrane was washed with PBS-T and incubated with HRP-coupled goat anti-mouse or anti-rabbit (Jackson Immunoresearch). Detection was performed with Western Lightning ECL (Perkin Elmer).
4
Single turnover UbcH7~Ub discharging assay The substrate was generated by charging 1 mg of purified UbcH7 with ubiquitin at
37°C for 1.5 h in a reaction mix containing 10 nM E1, 2 mM ATP, 100 μM ubiquitin, 0.5 mM TCEP in PBDM8 ATP regeneration buffer (38). The reaction was diluted and applied onto a MonoS 10/10 column (GE Healthcare) and eluted with a NaCl gradient in 30 mM HEPES pH 7.0. Single-turnover discharge assays for parkin mutants were
- -parkin, in the presence of 20 mM HEPES pH 7.5, 50 mM NaCl and 0.5 mM TCEP. Reactions were stopped with 3X sample buffer plus 20 mM TCEP and analyzed by SDS-PAGE. Gels were stained either with Coomassie or silver (Thermo Pierce kit) or transferred to a nitrocellulose membrane and probed with mouse anti-ubiquitin as described above. Band intensities were quantified by densitometry using Fiji.
NMR spectroscopy
15N-labeled UbcH7 and parkin constructs (full-length, R0-RBR, R0-RBR-W403A, and R0-RBR-W403A-C431S) were buffer-exchanged in NMR buffer (20 mM Tris·HCl, 100 mM NaCl, 1 mM TCEP, pH 7.4). Titrations were performed by mixing 0.2 mM 15N-UbcH7 with parkin proteins to obtain the concentrations indicated. HSQC spectra were acquired at 293 K on a 800 MHz Varian spectrometer equipped with a triple-resonance (1H, 13C, 15N) cryoprobe. Spectra were processed using NMRpipe (39) and analyzed with NMRView J (One Moon Scientific). 15N-1H backbone assignment for UbcH7 were obtained from the BMRB entry 15498 (40).
Cell culture and plasmid DNA
HeLa cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS, 4 mM L-glutamine, and 0.1% Pen/Strep and maintained in a 37°C incubator with 5% CO2. Homo sapiens full-length parkin was cloned in pEGFP-C2 (Clontech) and mutated by PCR mutagenesis. Construct expression was verified by immunoblotting using rabbit anti-GFP (Invitrogen) and mouse anti-VDAC1 (Abcam, ab14734) as a loading control.
HeLa cells (1.5x105) were seeded onto a 35mm Glass Bottom Microwell Dish -
parkin wild-type, W403A or C431S plasmid DNA using GeneJuice® (EMD Millipore) according to the manufacturer’s protocol. After 24 h, cells were transferred onto a heated stage maintained at 37°C and at 5% CO2 using a Zeiss temperature controller and cell perfusion system (Carl Zeiss, Thornwood, NY). To visualize mitochondria, cells were transduced with CellLight® mitochondria-RFP (Invitrogen) as per the manufacture protocol. Cells were treated with carbonyl cyanide m-chlorophenyl hydrazone (CCCP) (Sigma), a proton ionophore, at a final concentration of 20 μM. Microscopy was performed 24 hours post-transfection on a Zeiss AxioObserver.Z1 inverted fluorescent microscope for a total of 2 hours. Images were acquired with a 20x objective (Plan-Apochromat) with a side-mounted AxiocamMRm camera, using the Zeiss XBO75 Xenon illumination system and detected using the appropriate filters. Images were taken at 1
5
minute intervals for 2 h. Automated multidimensional acquisition control and compilation of movies were performed using the Zeiss Zen 2011 software.
HeLa cells were transfected with 10 nM non-targeting or PINK1 SMARTpool® ON-TARGETplus siRNA (Dharmacon, Thermo Scientific) using Lipofectamine RNAiMAX (Invitrogen) according to the manufacturer’s protocol for reverse transfections. Knockdown efficiency after CCCP treatment was tested separately by immunolotting using rabbit anti-PINK1 (Novus, C8830) and mouse anti-actin (EMD Millipore) as a loading control. After 24 h, cells were transfected with GFP-parkin WT or W403A using jetPRIME® (Polyplus Transfection). After another 24 h, cells were treated for 3 h with 20 μM CCCP and fixed with 4% formaldehyde in PBS. After washing and blocking with 3% BSA, cells were incubated with rabbit anti-Tom20 (Santa Cruz) followed by Alexa Fluor® 555 donkey anti-rabbit (Invitrogen). Images were acquired with a 63x Plan Apo oil objective using a Zeiss LSM710 confocal microscope.
Supplementary Text Author Contributions
J.-F.T. performed molecular cloning, ubiquitination assays, SAXS data processing, assisted R0-RBR structure determination and wrote the manuscript. V.S. performed construct design, protein purification, SAXS data acquisition and R0-RBR crystal structure determination. K. Grenier performed the PINK1-dependent GFP-parkin recruitment assay and data analysis. M.Y.T performed the GFP-parkin recruitment assay with time-lapse fluorescent microscopy and data analysis. M.M. performed protein purification, crystallization trials and SAXS data acquisition. M.S. purified and crystallized full-length parkin and performed NMR titrations. S.A-A-W. performed NMR titrations and data processing. K.W. produced proteins and acquired SAXS data. J.K. cloned the GFP-parkin mutants. S.A. performed NMR titrations. B.N. and G.K. assisted structure determination of full-length parkin. K.Gehring and E.A.F. conceived experiments, performed data analysis and edited the manuscript.
6
Fig. S1. Electron density in the crystal structure of parkin R0-RBR. (A,B,C) Snapshots of regions in the crystal structure of parkin C-terminal domains (2.8 Å resolution). The original SAD-phased, density-modified map is contoured in green at 1.0s. The anomalous difference map is contoured in red at 10.0s. The refined structural model is shown as sticks (RING0, white; RING1, cyan; REP, yellow; RING2, pink). Zinc atoms are shown as grey spheres. (D) Snapshot of a region in the crystal structure of full-length parkin (6.5 Å resolution). The rigid-body refined 2Fo-Fc density map is contoured in green at 1.0s. Domains are shown as Ca trace of different colors (Ubl, red; RING1, cyan; REP, yellow.)
7
Fig. S2. Asymmetric unit composition and analysis of parkin crystal structures. (A) Cartoon representation of three chains in the asymmetric unit. Left, colored by domains (Ubl, red; RING0, green; RING1, cyan; IBR, magenta; REP, yellow; RING2, salmon). Right, colored by structure (full-length in magenta, R0RBR in cyan). The REP adopts the same conformation in all three chains, and mediates weak crystal contacts only in chain C: the REP in chains A and B is exposed to a solvent channel. (B) Superposition of the three chains in the asymmetric unit colored as in (A) and displayed as C trace. The structures were superposed according to the “align” protocol in pymol. The arrow on the left indicates the distance between the IBR domains in chains A and B. The maximum interatomic distance in chain A is 95 Å. (C) Superposition of the IBR domains in chains A,B,C (green, cyan, magenta) with the NMR structure of the human parkin IBR domain (pdb 2JMO, grey). Structures are displayed as Ca trace. Zinc atoms are shown as spheres. .
8
Fig. S3. SAXS analysis of parkin. (A) Guinier plots of SAXS data collected for wild-type parkin full-length (1-465) and R0RBR (141-465), as well as W403A R0RBR and at the highest concentration. The linearity of the plots indicate monodispersity. Molecular weight estimation using different methods are shown on the right. Lysozyme was used as a standard for the I0/c method of molecular weight determination. (B) SAXS curves from all concentrations were merged to calculate pair-distance distribution functions using the program GNOM.
9
Fig. S4. Structural comparison of the RING1 domain. C trace superposition of RING domains in cross-eye stereoview and two different orientations (top, bottom). The RING1 domain from parkin (cyan) was superposed on the RING domains from RNF144A (magenta, PDB 1wim), c-Cbl (orange, PDB 1fbv), CIAP2 (pale green, PDB 3eb6), TRAF6 (dark green, PDB 3hcu), Bmi1 (hot pink, PDB 3rp6) and RNF4 (brown, PDB 4ap4). The location of the N- and C-termini are indicated, with C corresponding to the C-termini of parkin/RNF144A and C’ to the C-termini of all other RING domains. Note the location of the insertion loop bearing Asp280 as well as the C-terminal helix, both found uniquely in the RBR proteins parkin and RNF144A.
10
Fig. S5. Modeling of the RING:E2 complex. (A) Ca trace superposition of RING:E2 complexes in cross-eye stereoview. The RING1 domain from parkin (cyan) was used as the template to superpose the RING domains from c-Cbl:UbcH7 (orange, PDB 1fbv), CIAP2:UbcH5b (pale green, PDB 3eb6), TRAF6:Ubc13 (dark green, PDB 3hcu), Bmi1:UbcH5c (hot pink, PDB 3rp6) and RNF4:UbcH5a~Ub (brown, PDB 4ap4). The structures were superposed using the “align” protocol in PyMOL. (B) Cross-eye stereoview cartoon representation of the RING:E2 superposition described above, displayed in the context of the crystal structure of parkin. The RING domains in complex with the E2s are not displayed for clarity of presentation. The superposition shows that all E2 structures bind in the same orientation and to the same site on the RING1 domain of parkin. The double-headed arrow indicate the ~50 Å distance between the active site cysteine in E2 and C431 in RING2.
11
Fig. S6. Structure-guided sequence alignment of RBR E3 ligases. Parkin was aligned with HHARI, HOIP and HOIL by matching zinc-coordinating Cys and His residues in RING1 ( ), IBR ( ) and RING2 ( ), as well as the active site Cys ( ). The REP region is boxed and shows little conservation with other RBR proteins. The positions of Cys457 and His461 in parkin are indicated by arrows.
12
Fig. S7. Sequence alignment of parkin from different species. Residue numbers are shown on the left and right. Invariant residues are shown as white letter on black background.
13
Fig. S8. Effect of different mutations on the autoubiquitination activity of parkin. (A) Autoubiquitination assay with 1 μM GST-parkin constructs and 2 μM UbcH7, in the absence or presence of 50 μM wild-type or K0 ubiquitin. Reactions were run for 2 hours at 37°C, stopped with sample buffer (100 mM DTT), and analyzed by SDS-PAGE and Coomassie staining. (B) Autoubiquitination assay performed as described above, performed with wild-type ubiquitin. (C) Autoubiquitination assay with 1 μM GST-parkin constructs and 2 μM UbcH7. Reactions were run for 0, 30, 60 and 120 min at 37°C, stopped with sample buffer (100 mM DTT), and analyzed by SDS-PAGE and Coomassie staining. (D) Summary table of effect of different mutations on the autoubiquitination ligase activity of parkin compared to wild-type.
14
Fig. S9. Activation of untagged parkin by mutagenesis. (A) Autoubiquitination assay with 1 μM parkin constructs and 2 μM UbcH7. Reactions were run for 0, 30, 60 and 120 min at 37°C, stopped with sample buffer (100 mM DTT), and analyzed by SDS-PAGE, Ponceau staining and immunoblotting with an anti-ubiquitin antibody. The results show that the GST tag does not affect the activation of parkin by disruption of the RING0:RING2 or REP:RING1 interactions. (B) Autoubiquitination assays performed as described above for 2h in the absence and presence of 40 nM E1 and analyzed by SDS-PAGE and Coomassie staining. Additional controls were performed without E2 and ubiquitin for the F463A mutant. The assay shows that the high molecular weight species are the results of ubiquitination and not aggregation.
15
Fig. S10. UbcH7~Ub discharging assays. (A) UbcH7~Ub discharge assays with wild-type GST-parkin and mutants. UbcH7~Ub (1.5 μM) was incubated with 0.4 μM GST-parkin at 37°C. Reactions were stopped with sample buffer containing TCEP to reduce disulfide bonds but keep thioester bonds intact. Coomassie-stain of the experiment shown in Fig. 3A. (B) UbcH7~Ub (1.8 μM) was incubated with 0.4 μM GST-parkin at 37°C. Products were resolved by SDS-PAGE and stained with Coomassie. (C) UbcH7~Ub discharging assays with untagged parkin wild-type and single mutants. Reactions were performed at 37°C with 0.4 μM purified ubiquitin-charged UbcH7 and 1 μM parkin. Products were analyzed by SDS-PAGE for immunoblotting with anti-UbcH7 antibody.
16
Fig. S11. UbcH7 binding to parkin by NMR. (A) Portions of NMR HSQC spectra of 200 μM 15N-labeled UbcH7 in the absence (left) and presence of increasing amounts of wild-type full-length parkin. Signals arising from residues (Trp101 and Phe63) that are predicted to be in the UbcH7-RING1 interface based on the Cbl-UbcH7 crystal structure (PDB 1FBV) specifically weaken upon the addition of parkin. (B) HSQC spectra of 15N-labeled UbcH7 (150 μM) in the absence and presence of R0-RBR parkin. The W403A mutant interacts more strongly with UbcH7 than the wild-type R0-RBR fragment. The peaks that are most strongly affected are the same. (C) Peak loss as a function of parkin added. Spectra were normalized for dilution and the number of scans and the peak intensity lost relative to the free UbcH7 spectrum plotted.
17
Fig. S12. Recruitment of GFP-parkin mutants to mitochondria. Parkin recruitment to mitochondria upon membrane depolarization. HeLa cells were transduced with CellLight® mitochondria-RFP and transfected with equal amount of GFP-parkin wild-type, W403A or C431S. Cells were treated with 20 μM CCCP and visualized by time-lapse microscopy. Experiments were performed in a double-blinded fashion with n=4 and over 150 cells analyzed in each condition. Scale bar: 10 μm.
18
Table S1. X-ray crystallography data collection and refinement statistics Data collection Parkin R0-RBR Parkin full-length X-ray source CLS 08ID-1 CHESS F2 Wavelength (Å) 1.2824 0.9183 Space group P212121 P21212 Cell dimensions
a, b, c (Å) 84.7, 106.6, 154.2 208.6, 277.4, 125.9 , ß, (°) 90, 90, 90 90, 90, 90
R.m.s. deviations Bond lengths (Å) 0.010 0.011 Bond angles (°) 0.91 1.174
Ramanchandran statistics Favoured region (%) 95.4 97.2** Allowed region (%) 4.6 2.8** Disallowed region (%) 0 0**
*Data for the highest resolution shell are shown in parentheses. #Highest resolution shell recommended for heavy-atom search (3.37-3.20) **Statistics derived from molecular replacement search models.
19
Movie S1. Time-lapse imaging of parkin recruitment to the mitochondria after treatment with CCCP in HeLa cells transfected with GFP-parkin. Live cell imaging was initiated at 24 hours post-transfection and images were acquired every minute post-CCCP treatment. Cells were visualized using a filter that detects GFP. Images were acquired using a 20× objective for a total of 90 minutes and merged together to create a movie file. The movie shows a rapid and progressive recruitment of GFP-parkin wild-type at the mitochondria after the addition of CCCP. At the beginning of the movie GFP-parkin wild-type is expressed throughout the cell. At 40 minutes, the distribution of GFP-parkin wild-type became more punctuate and gradually co-localized to the mitochondria. The W403A started recruiting at an earlier time point (30 min), while the C431S showed no recruitment.
Movie S2. Single cell time-lapse imaging of parkin recruitment to the mitochondria after treatment with CCCP in HeLa cells transfected with GFP-parkin mutants. To visualize mitochondria, cells were transduced with CellLight® mitochondria-RFP. Live cell imaging was initiated at 24 hours post-transfection and images were acquired every minute post-CCCP treatment. Cells were visualized using filters that detect GFP or RFP. Fluorescent images were merged to create a movie file. Images were acquired using a 20× objective for a total of 70 minutes. The movie shows a rapid and progressive recruitment of GFP-parkin wild-type at the mitochondria after the addition of CCCP. At the beginning of the movie GFP-parkin wild-type is expressed throughout the cell. At 40 minutes, the distribution of GFP-parkin wild-type becomes more punctuate and is co-localized at the mitochondria (seen as the merging of green and red colours to create yellow colour). The W403A started recruiting at an earlier time point (30 min), while the C431S showed no recruitment.
References
1. T. Kitada et al., Mutations in the parkin gene cause autosomal recessive juvenile