Structure, dynamics and interaction study of Glutaminase Interacting Protein (GIP) and its complex with Glutaminase L and Brain-specific Angiogenesis Inhibitor 2 (BAI2) peptide and characterization of subunit A of Heterodisulfide Reductase (HdrA) from Methanothermobacter marburgensis by Mohiuddin Ovee A dissertation submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Doctor of Philosophy Auburn, Alabama August 3, 2013 Keywords: Glutaminase L, GIP, NMR, Hdr, FPLC Copyright 2013 by Mohiuddin Ovee Approved by Smita Mohanty, Major Professor & Co-Chair of dissertation committee, Associate Professor of Chemistry and Biochemistry Eduardus Duin, Co-Chair, Associate Professor of Chemistry and Biochemistry Holly Ellis, Associate Professor of Chemistry and Biochemistry Michael Squillacote, Associate Professor of Chemistry and Biochemistry Orlando Acevedo, Associate Professor of Chemistry and Biochemistry ?Reproduced in part with permission from [Zoetewey, D. L., Ovee, M., Banerjee, M., Bhaskaran, R., and Mohanty, S. (2011) Promiscuous Binding at the Crossroads of Numerous Cancer Pathways: Insight from the Binding of Glutaminase Interacting Protein with Glutaminase L, Biochemistry-Us 50, 3528-3539.] Copyright [2011] American Chemical Society.? iii Abstract Glutaminase interacting protein (GIP) is a 124 amino acid long protein containing a single PDZ domain. This protein intersects a number of important biological pathways. In many of these pathways, the mechanism of function of this domain is still unknown. Its involvement in cancer pathways makes it a good target for drug development. We resolved the solution structures of both free GIP and GIP in complex with the C-terminal peptide analog of Glutaminase L to shed light on the mechanism of binding with the goal of future development of a potential inhibitor for GIP. To understand more of GIP?s function, interactions with two target peptides were investigated using different biophysical methods. One of the peptides was homologous to the C-terminus of brain-specific angiogenesis inhibitor 2 (BAI2) and the other one used had a consensus PDZ class I binding motif. Both of the peptides showed moderate binding affinity toward GIP with the BAI2 peptide having comparatively higher affinity. Elucidating the mechanism of interactions for different target partners would help to lay out the network of function for GIP. In a separate project, to understand the mechanism of electron bifurcation in methanogenic archaea, efforts were made to purify either heterodisulfide reductase (Hdr) or the subunit A of Hdr (HdrA) from Methanothermobacter marburgensis or Methanococcus maripaludis. We were able to purify HdrA with limited purity and showed the presence of [4Fe-4S] clusters in HdrA through EPR studies. However, efforts to purify Hdr from both organisms were with limited success. It is important to continue the efforts to obtain pure iv Hdr/HdrA to investigate the mechanism by which electron bifurcation takes place within this enzyme complex. v Acknowledgments First of all, I would like to show my deepest gratitude to Dr. Smita Mohanty, major professor and co-chair, for her guidance, support, encouragement, and help throughout my graduate study. I would like to convey my heartfelt gratitude to Dr. Eduardus Duin, co-chair of my dissertation committee, for his relentless support, guidance and unbiased help and encouragement for the completion of my graduate study. I would like to thank Dr. Holly Ellis, Dr. Michael Squillacote and Dr. Orlando Acevedo for their help, support and constructive suggestion to my dissertation. I also want to thank my former colleague Dr. David Zoetewey, a postdoc with Dr. Smita Mohanty, with whom I had the privilege to work on several projects and learned a lot about NMR techniques. I would like to thank my friend Suman Mazumder for his continued support. I would like to thank former lab member Dr. Monimoy Banerjee and a very dear colleague Dr. Rajagopalan Bhaskaran who is the former NMR facilities manager of AU Department of Chemistry. I would like to thank Selamawit Ghebreamlak, Dr. Shigeki Saito, Dr. Chengdong Huang, Dr. Janarthanan Krishnamoorthy, Xiao Xiao, Divya Prakash, and Dr. Uma Katre for their meaningful discussions and help. I would like to thank NMR director Dr. Michael Meadows for his help. I would like to thank the chairman of the Department of Chemistry and Biochemistry Dr. J. V. Ortiz for his consistent support and help. I would also like to thank graduate program officer Dr. Rik Blumenthal, Dean of the Graduate School Dr. George Flowers and COSAM interim dean Dr. Charles Savrda for their commitment to the successful completion of my Ph.D. dissertation. I would like to thank all professors and the administrative staff at the vi Department of Chemistry and Biochemistry, and all my friends at Auburn who directly or indirectly contributed to the success of my study. I would like to thank U.S. Department of Agriculture PECASE Presidential Early Career Award for Scientists and Engineers Award 2003- 35302-12930, National Science Foundation Grant IBN-0628064, and National Institutes of Health Grant DK082397 for their funding to Dr. Smita Mohanty, which financially supported my research in her lab. I would like to thank my parents, all my family members, and relatives for their support and help. Last but not the least; I would like to thank my wife for her continuous encouragement, love and moral support. vii Table of contents Abstract .......................................................................................................................................... iii Acknowledgements ..........................................................................................................................v List of Tables .................................................................................................................................xv List of Figures .............................................................................................................................. xvi Chapter 1 Introduction .....................................................................................................................1 1.1 NMR .......................................................................................................................................1 1.1.1 Principles of NMR ......................................................................................................1 1.1.2 Relaxation processes ...................................................................................................4 1.1.2.1Spin-lattice relaxation ...................................................................................5 1.1.2.2 Spin-spin relaxation .....................................................................................5 1.1.3 Chemical shift .............................................................................................................7 1.1.4 Spin-spin coupling ......................................................................................................8 1.1.4.1 Scalar coupling (J-coupling) ........................................................................8 1.1.4.2 Dipolar coupling ..........................................................................................9 1.1.5 Nuclear Overhauser effect (NOE) ..............................................................................9 1.1.6 Multidimensional NMR spectroscopy ......................................................................12 viii 1.1.6.1 Two-dimensional NMR (2D NMR)...........................................................12 1.1.6.1.1 COSY ..........................................................................................13 1.1.6.1.2 TOCSY .......................................................................................13 1.1.6.1.3 HSQC ..........................................................................................13 1.1.6.1.4 NOESY .......................................................................................15 1.1.6.2 Three-dimensional NMR (3D NMR).........................................................15 1.1.7 Protein NMR .............................................................................................................18 1.1.8 Study of protein-ligand interaction by NMR ............................................................21 1.2 PDZ domain and Glutaminase Interacting Protein (GIP) ....................................................24 1.3 Glutaminase Interacting Protein (GIP)/Tax Interacting Protein-1 (TIP-1) ..........................26 1.4 References ............................................................................................................................29 Chapter 2 Characterization of Glutaminase Interacting Protein (GIP): a PDZ domain ................36 2.1 Background ..........................................................................................................................36 2.1.1 Protein-protein interaction network ..........................................................................36 2.1.2 PDZ domain and its classes ......................................................................................38 2.1.3 Glutaminase interacting protein as a class I PDZ domain ........................................39 2.1.4 Objective of the study ...............................................................................................41 2.2 Materials and Methods .........................................................................................................42 2.2.1 Cloning, over-expression and purification of 15N, 13C-labeled GIP .........................42 2.2.2 NMR data collection .................................................................................................42 ix 2.2.3 Analysis of dynamics data ........................................................................................43 2.2.4 Structure calculation and refinement ........................................................................43 2.3 Results ..................................................................................................................................45 2.3.1 NMR structure determination of free GIP ................................................................45 2.3.1.1 Introduction ................................................................................................45 2.3.1.2 Backbone assignments of free GIP ............................................................47 2.3.1.3 Side-chain assignments of free GIP ...........................................................53 2.3.1.4 NOE assignments of free GIP ....................................................................65 2.3.1.5 Structure calculation of free GIP ...............................................................68 2.3.1.6 Refinement of structures by ARIA ............................................................72 2.3.1.7 NMR structure of free GIP ........................................................................73 2.3.1.8 Accession codes .........................................................................................77 2.3.2 Dynamics of free GIP from 15N relaxation measurements .......................................77 2.4 Conclusion ............................................................................................................................81 2.5 References ............................................................................................................................82 Chapter 3 Study of the mechanism of interaction between GIP and the Glutaminase L peptide ..90 3.1 Introduction ..........................................................................................................................90 3.2 Objective of the study ..........................................................................................................91 3.3 Materials and Methods .........................................................................................................92 3.3.1 Cloning, over-expression and purification of 15N, 13C-labeled GIP .........................92 x 3.3.1.1 Transformation of E. coli BL21DE3pLysS cells with the recombinant plasmid pET-3c/GIP .................................................................................92 3.3.1.2 Preparation of overnight culture ................................................................93 3.3.1.3 Expression of the protein in batch culture .................................................94 3.3.1.4 Cell harvest and lysis .................................................................................94 3.3.1.5 Protein purification ....................................................................................95 3.3.1.6 NMR sample preparation ...........................................................................95 3.3.2 NMR data collection .................................................................................................95 3.3.3 Analysis of dynamics data ........................................................................................97 3.3.4 Structure calculation and refinement ........................................................................97 3.4 Results ..................................................................................................................................99 3.4.1 Protein expression .....................................................................................................99 3.4.2 Protein purification .................................................................................................104 3.4.3 NMR structure determination of GIP-Glutaminase L peptide complex ................104 3.4.3.1 Effect of peptide binding to the resonances of GIP protein .....................104 3.4.3.2 Backbone and side-chain assignments .....................................................111 3.4.3.2.1 For protein .................................................................................111 3.4.3.2.1 For peptide ................................................................................115 3.4.3.3 NOE assignments .....................................................................................121 3.4.3.4 Structure calculation ................................................................................126 3.4.4 Comparison of the structure of free GIP with that of the GIP-Glutaminase L peptide complex ...................................................................................................................130 xi 3.4.5 Binding and specificity of the Glutaminase L peptide ............................................132 3.4.6 Dynamics of the GIP-Glutaminase L peptide complex from 15N relaxation measurements ........................................................................................................136 3.4.7 Intermediate chemical exchange within GIP due to the binding of the Glutaminase L peptide .................................................................................................................140 3.5 Discussion ..........................................................................................................................141 3.5.1 Specificity in the binding interaction between GIP and Glutaminase L peptide ....141 3.5.2 The effects of the Glutaminase L peptide binding on the dynamics of GIP ...........144 3.5.3 Comparison to other GIP-peptide complex structures ............................................145 3.5.4 Comparison between NMR and crystal structures..................................................146 3.5.5 Potential for drug design .........................................................................................147 3.6 Accession codes .................................................................................................................148 3.7 References ..........................................................................................................................149 Chapter 4 Determination of the mode of interaction of Glutaminase Interacting Protein (GIP) with two different interacting partners .......................................................................155 4.1 Introduction ........................................................................................................................155 4.1.1 PDZ domain and its functions.................................................................................155 4.1.2 Binding pocket of PDZ domain ..............................................................................155 4.1.3 GIP as a PDZ domain .............................................................................................156 4.1.4 GIP in the brain .......................................................................................................157 4.1.5 Identification of interacting partners in brain .........................................................157 4.2 Materials and Methods .......................................................................................................158 4.2.1 Expression and purification of 15N- and unlabeled GIP .........................................158 xii 4.2.2 Fluorescence ...........................................................................................................159 4.2.3 Circular Dichroism (CD) ........................................................................................159 4.2.4 Nuclear Magnetic Resonance (NMR) .....................................................................160 4.3 Results and Discussion .......................................................................................................161 4.3.1 Protein expression ...................................................................................................161 4.3.2 Protein purification .................................................................................................161 4.3.3 Interaction of BAI2 peptide with GIP .....................................................................164 4.3.3.1 Characterization by Fluorescence spectroscopy ......................................164 4.3.3.2 Characterization by CD spectroscopy ......................................................167 4.3.3.3 Characterization by 1H,15N-HSQC NMR ................................................169 4.3.3.4 Chemical shift perturbations of GIP upon binding to the BAI2 peptide .175 4.3.4 Interaction of the control peptide with GIP ............................................................180 4.3.4.1 Characterization by Fluorescence spectroscopy ......................................180 4.3.4.2 Characterization by CD spectroscopy ......................................................183 4.3.4.3 Characterization by 1H,15N-HSQC NMR ................................................185 4.3.4.4 Chemical shift perturbations of GIP upon binding to the control peptide ......................................................................................................190 4.3.5 Comparison of interaction between GIP and BAI2 peptide with interaction between GIP and control peptide ..........................................................................................193 4.4 References ..........................................................................................................................196 Chapter 5 Characterization of subunit A of Heterodisulfide Reductase (HdrA) from Methanothermobacter marburgensis ......................................................................202 5.1 Introduction ........................................................................................................................202 xiii 5.1.1 Electron bifurcation ................................................................................................202 5.1.2 History of electron bifurcation ................................................................................202 5.1.3 Mechanism of electron bifurcation .........................................................................205 5.1.4 Electron bifurcation in other systems .....................................................................206 5.1.5 Models for electron bifurcation ..............................................................................207 5.1.5.1 Model I .....................................................................................................207 5.1.5.2 Model II ...................................................................................................207 5.1.5.3 Model III ..................................................................................................208 5.2 Objective of this study ........................................................................................................210 5.3 Materials and methods .......................................................................................................212 5.3.1 Purification of hydrogenase:heterodisulfide reductase complex (MvhADG/HdrABC) from M. marburgensis ...................................................212 5.3.1.1 Growth of M. marburgensis cells ............................................................212 5.3.1.2 Harvest and sonication of M. marburgensis cells ....................................213 5.3.1.3 Purification of hydrogenase:heterodisulfide reductase complex (MvhADG/HdrABC) ............................................................................213 5.3.2 Purification of heterodisulfide reductase (Hdr) from M. maripaludis cells ............214 5.3.2.1 M. maripaludis cells....................................................................................214 5.3.2.2 Purification of heterodisulfide reductase ...................................................214 5.3.3 Purification of HdrA from M. maripaludis HdrAmarburgensis cells ............................215 5.3.3.1 M. maripaludis HdrAmarburgensis cells ...........................................................215 xiv 5.3.3.2 Purification of HdrA ...................................................................................215 5.3.4 Iron determination ...................................................................................................216 5.3.5 UV-vis absorption analysis .....................................................................................217 5.3.6 EPR measurements .................................................................................................217 5.4 Results ................................................................................................................................218 5.4.1 Purification of the hydrogenase:heterodisulfide reductase complex (MvhADG/HdrABC) from M. marburgensis ....................................................218 5.4.2 Purification of heterodisulfide reductase (Hdr) from M. maripaludis cells ............223 5.4.3 Purification of HdrA from M. maripaludis HdrAmarburgensis cells ............................229 5.4.3.1 Purification of HdrA ...................................................................................229 5.4.3.2 Protein and iron determination ....................................................................232 5.4.3.3 UV-vis absorption of the protein sample ....................................................232 5.4.3.4 EPR measurement of the HdrA protein sample ..........................................234 5.5 Conclusions and future direction ........................................................................................236 5.6 References ..............................................................................................................................238 Appendix Tables ..........................................................................................................................241 xv List of Tables Table 1.1 Relationship of I to atomic number and mass number ................................................2 Table 1.2 Correlation observed for some of the most commonly used 3D NMR experiments .20 Table 1.3 Classification of PDZ domains ..................................................................................28 Table 2.1 Sequential alignment of C-terminal binding partners of GIP ....................................40 Table 2.2 Statistics of side-chain assignments of free GIP ........................................................61 Table 2.3 NMR structural statistics for the 20 selected lowest energy structures of free GIP ...75 Table 3.1 Statistics of side-chain assignments of the GIP-Glutaminase L peptide complex ... 114 Table 3.2 Statistics of available proton assignments of the Glutaminase L peptide ................ 116 Table 3.3 NMR structural statistics for the 20 selected lowest energy structures of the GIP-Glutaminase L Peptide Complex .....................................................................128 Table 4.1 Dissociation constants of various residues of GIP upon binding with the BAI2 peptide by NMR ......................................................................................................174 Table 4.2 Dissociation constants of various residues of GIP upon binding with the control peptide by NMR ......................................................................................................189 xvi List of Figures Figure 1.1 Nuclear spin states in a magnetic field .......................................................................3 Figure 1.2 ?Precession? of nucleus in a magnetic field ...............................................................4 Figure 1.3 The two components of a spinning nucleus in an applied magnetic field ..................6 Figure 1.4 Energy diagram for a two-spin system .....................................................................11 Figure 1.5 General scheme for a 2D experiment........................................................................12 Figure 1.6 A) COSY and B) TOCSY spectra ............................................................................14 Figure 1.7 3D experiment as a combination of two sets of 2D experiments .............................16 Figure 1.8 Schematic presentation of how with the addition of another evolution time to the 2D experiment (Figure 1.7) can result in another frequency dimension for a 3D experiment ................................................................................................................17 Figure 1.9 Effect of chemical exchange on NMR spectra .........................................................22 Figure 1.10 Examples of PDZ domain containing proteins .......................................................25 Figure 2.1 Different experiment tools to define protein interactions .........................................37 Figure 2.2 Flowchart of structure determination by NMR .........................................................46 Figure 2.3 Sequential assignments of V13-K20 from (1H, 13C)-strips of HNCACB experiment.................................................................................................................49 Figure 2.4 Sequential assignments of V13-K20 from (1H, 13C)-strips of HNCA experiment ...50 Figure 2.5 HNCACB and CBCA(CO)NH strips of I18 residue ................................................51 Figure 2.6 1H, 15N-HSQC spectrum of free GIP ........................................................................52 xvii Figure 2.7 Proline cis/trans isomerization ..................................................................................53 Figure 2.8 HC(CO)NH and HSQC-TOCSY spectra showing side-chain assignments of V60 and R59 residues .......................................................................................................55 Figure 2.9 HCCH-COSY spectrum of H??proton of N26 residue showing non-degeneracy of H??protons ................................................................................................................56 Figure 2.10 HCCH-TOCSY spectrum of H?3 proton of P45 residue .......................................57 Figure 2.11 HNHA spectrum of non-degenerate H? protons of G70 residue ...........................58 Figure 2.12 CC(CO)NH spectrum of Q23 residue .....................................................................59 Figure 2.13 Aliphatic region of the 1H, 13C-HSQC spectrum of free GIP .................................62 Figure 2.13 A Part of the aliphatic region of the 1H, 13C-HSQC spectrum with assignments ...63 Figure 2.14 Aromatic region of the 1H, 13C-HSQC spectrum with assignments .......................64 Figure 2.15 15N-edited HSQC-NOESY spectrum of I33 residue ...............................................66 Figure 2.16 13C-edited HSQC-NOESY spectrum of I33 QD1 proton .......................................67 Figure 2.17 Input and output files for CYANA 1.0.6 ................................................................70 Figure 2.18 Iterative cycle of CYANA 1.0.6 run .......................................................................70 Figure 2.19 Input and output files for CYANA 2.1 ...................................................................71 Figure 2.20 Iterative cycle of CYANA 2.1 run ..........................................................................71 Figure 2.21 Ribbon diagrams of the ensemble of the 20 superimposed lowest energy structures of free GIP..............................................................................................................74 Figure 2.22 The S2 values derived using the modelfree analysis from the steady state 1H-15N NOE, R1 and R2 relaxation times of free GIP for each non-overlapping well defined residue .......................................................................................................79 Figure 2.23 Residues with S2 values below the threshold of 0.85 are mapped in red onto the structure of free GIP colored blue ..........................................................................80 Figure 3.1 Expression of GIP analyzed by SDS-PAGE ...........................................................101 xviii Figure 3.2 1D NMR spectrum of non-homogeneously labeled GIP sample ...........................102 Figure 3.3 1D NMR spectrum of homogeneously labeled GIP sample ...................................103 Figure 3.4 Combined 1H and 15N backbone amide chemical shift perturbations (?HN) are plotted as a function of residue number in GIP .....................................................106 Figure 3.5 The magnitudes of ?HN presented in Figure 3.4 are represented as different colors on a ribbon diagram of free GIP .............................................................................107 Figure 3.6 Combined HA and CA backbone chemical shift perturbations (?HC) are plotted as a function of residue number in GIP .......................................................................108 Figure 3.7 The magnitudes of ?HC presented in Figure 3.6 are represented as different colors on a ribbon diagram of free GIP .............................................................................109 Figure 3.8 An overlay of free GIP is shown in red and GIP-Glutaminase L peptide at a ratio of 1:3 in blue ...............................................................................................................110 Figure 3.9 Sequential assignments of V13-K20 in the GIP-Glutaminase L peptide complex from (1H, 13C)-strips of HNCACB experiment.......................................................112 Figure 3.10 1H, 15N-HSQC spectrum of the GIP-Glutaminase L peptide complex .................113 Figure 3.11 1H, 15N-HSQC spectrum of the Glutaminase L peptide .......................................117 Figure 3.12 1H, 13C-HMQC spectrum of the Glutaminase L peptide ......................................118 Figure 3.13 Homonuclear 2D TOCSY spectrum of the Glutaminase L peptide .....................119 Figure 3.14 2D selectively filtered NOESY spectrum of the Glutaminase L peptide .............120 Figure 3.15 13C-edited HSQC-NOESY spectrum of I33 QD1 proton of GIP in its bound form ......................................................................................................................123 Figure 3.16 1H, 13C-HSQC spectrum of GIP in the bound form ..............................................124 Figure 3.17 Three different HSQC-NOESY spectra of I33 residue of the GIP .......................125 Figure 3.18 Ribbon diagrams of the ensemble of the 20 superimposed lowest energy structures of complexed GIP in blue with the Glutaminase L peptide in red ........................127 Figure 3.19 An overlay of free GIP is shown in green with the complexed GIP protein in blue and the Glutaminase L peptide in red ....................................................................131 Figure 3.20 Heavy atom details from the binding site of GIP with the Glutaminase L peptide ..................................................................................................................135 xix Figure 3.21 A plot of ?S2 as a function of residue number where ?S2 refers to S2 of the GIP- Glutaminase L peptide complex minus that of free GIP ......................................138 Figure 3.22 The magnitude of ?S2 upon binding to the Glutaminase L peptide was mapped onto the structure of free GIP...............................................................................139 Figure 4.1 Expression of GIP analyzed by SDS-PAGE ...........................................................162 Figure 4.2 Purification of GIP analyzed by SDS-PAGE .........................................................163 Figure 4.3 Fluorescence emission spectrum of GIP with the BAI2 peptide ............................165 Figure 4.4 Non-linear curve fitting assuming 1:1 binding between GIP and the BAI2 peptide where (F0 - FC)/( F0 - Fmin) was plotted against peptide concentration ..................166 Figure 4.5 Changes in the CD spectra of GIP upon binding with increasing concentrations of the BAI2 peptide for the wavelength range of 194 nm to 250 nm .........................168 Figure 4.6 Changes of 2D 1H,15N-HSQC spectra upon addition of the BAI2 peptide to 100 ?M of 15N-labeled GIP ..................................................................................................171 Figure 4.7 Expanded region of the spectra demonstrating the chemical shift perturbations of residue E17 upon titration of GIP with the BAI2 peptide ......................................172 Figure 4.8 The NMR titration binding curve for the titration of GIP with the BAI2 peptide ..173 Figure 4.9 Chemical shift perturbations (??) of the GIP backbone amide groups upon binding with the BAI2 peptide .............................................................................................178 Figure 4.10 Fluorescence emission spectrum of GIP with the control peptide ........................181 Figure 4.11 Non-linear curve fitting assuming 1:1 binding between GIP and the control peptide where (F0 - FC)/( F0 - Fmin) was plotted against peptide concentration .182 Figure 4.12 Changes in the CD spectra of GIP upon binding with increasing concentrations of the control peptide for the wavelength range of 194 nm to 250 nm ....................184 Figure 4.13 Changes of 2D 1H,15N-HSQC spectra upon addition of the control peptide to 100 ?M of 15N-labeled GIP ........................................................................................186 Figure 4.14 Expanded region of the spectra demonstrating the chemical shift perturbations of residue N81 upon titration of GIP with the control peptide.................................187 Figure 4.15 The NMR titration binding curve for the titration of GIP with the control peptide ..................................................................................................................188 Figure 4.16 Chemical shift perturbations (??) of the GIP backbone amide groups upon binding with the control peptide .......................................................................................192 Figure 4.17 The Chemical shift perturbations (??) of the GIP backbone amide groups upon binding with the BAI2 (red) and the control (black) peptide ...............................195 xx Figure 5.1 Model of the structure of the hydrogenase:heterodisulfide reductase complex from Methanothermobacter marburgensis ......................................................................204 Figure 5.2 Models for electron bifurcation ..............................................................................209 Figure 5.3 Chromatography profile of DEAE-Sepharose column for purification of MvhADG/HdrABC complex .............................................................................219 Figure 5.4 Chromatography profile of Q-Sepharose column for purification of MvhADG/HdrABC complex ............................................................................220 Figure 5.5 Chromatography profile of Superdex 200 column for purification of MvhADG/HdrABC complex .............................................................................221 Figure 5.6 15%SDS-PAGE analysis of the fractions from Superdex 200 column for MvhADG/HdrABC complex purification ...........................................................222 Figure 5.7 EPR spectra of iron-sulfur clusters .........................................................................224 Figure 5.8 EPR spectrum of Hdr complex from M. maripaludis .............................................225 Figure 5.9 Chromatography profile of Superdex 200 column for purification of Hdr from M. maripaludis .............................................................................................................226 Figure 5.10 17% SDS-PAGE analysis of the Hdr complex from M. maripaludis cells ..........227 Figure 5.11 8% native PAGE analysis of the Hdr complex from M. maripaludis cells ..........228 Figure 5.12 Chromatography profile of nickel column for purification of HdrA from M. maripaludis HdrAmarburgensis .................................................................................230 Figure 5.13 12% SDS-PAGE analysis of the HdrA sample from M. maripaludis HdrAmarburgensis cells ...........................................................................................231 Figure 5.14 UV-vis absorption of the HdrA protein sample ....................................................233 Figure 5.15 EPR spectra of HdrA protein sample at the temperature of 8 K and at different microwave frequencies ........................................................................................235 1 Chapter 1 Introduction 1.1 NMR 1.1.1 Principles of NMR Nuclear magnetic resonance (NMR) was first introduced in 1938. From then on, NMR has become one of the most powerful analytical techniques, widely used in many different fields. NMR is based on the fact that certain nuclei possess spin angular momentum and a resulting magnetic moment. Since a nucleus is positively charged, it would act as a spinning charged particle like a current flowing in a circle. If the nucleus has an angular momentum, P, then such spinning would produce a magnetic field parallel to the spinning axis, and the nucleus would have a magnetic moment, ?. From quantum mechanics, it is known that angular momentum is quantized in half-integral or integral multiples of h/2?, where h is Planck?s constant. If I denotes the nuclear spin quantum number, the maximum observable component of angular momentum can be given as: P = Ih/2? (1.1) The spin quantum number can be different for different nuclei such as 0, 12, 1, 32 etc. If I is zero, there will not be any angular momentum for the nucleus, examples are 12C and 16O. The spin quantum number I is related to the atomic number and mass number (Table 1.1). 2 Atomic number Mass number Spin quantum number (I) Even Even 0 Even or odd Odd 1 2, 3 2, 5 2, . . . . . Even Odd 1, 2, 3, . . . . . Table 1.1: Relationship of I to atomic number and mass number. Adapted from reference 1. Since both protons and neutrons spin in the nucleus, they will pair with other protons and neutrons in the same nucleus but with opposite spin and, thus, such a relationship between I and atomic number and mass number can be established. The angular momentum of the nucleus will follow the (2I + 1) rule to acquire the orientation with respect to the external magnetic field when placed in a uniform magnetic field. A nucleus that has half spin angular momentum (I = 12), such as 1H or 13C, will have two orientations, i.e., a lower energy and a higher energy orientation (Figure 1.1). In the lower energy orientation the magnetic moment of the nucleus will be aligned along the external magnetic field whereas in the higher energy orientation it will be aligned against the magnetic field. The potential energy of the nucleus in each orientation equals to ?B0cos???where B0 is the strength of the external magnetic field and ? is the angle between the axis of the spin and the direction of the magnetic field. The energy difference, ?E, between the two energy states is proportional to the external magnetic field. 3 Energy Figure 1.1: Nuclear spin states in a magnetic field. Adapted from reference 1. Due to the influence of the external magnetic field the spinning nucleus ?precesses?, i.e., the two ends of the spinning axis follow a circular path but opposite in direction to each other (Figure 1.2). For the transition of the nucleus from the lower energy state to the higher energy state, a radio frequency wave that has the exactly equal frequency to that of the ?precession? needs to be applied perpendicular to the external magnetic field. The Larmor equation states the relationship between the frequency of this electromagnetic radiation ? and the strength of the magnetic field B0 ????????????????????????????????????????????????????????????????????????B0/2? (1.2) where ??is the gyromagnetic ratio. Each nucleus has its own characteristic gyromagnetic ratio, for example, 1H has a gyromagnetic ratio (42.576 MHzT-1) that is approximately 10 times that of 15N (4.316 MHzT-1) and 4 times that of 13C (10.705 MHzT-1) (1). No field With magnetic field I = -12 I = +12 0 ?E 4 Figure 1.2: ?Precession? of nucleus in a magnetic field. Adapted from reference 1. 1.1.2 Relaxation processes According to the Boltzmann distribution, there is a slight population difference between the two energy states since the nuclei are slightly in excess of number in the lower energy state than in the higher energy state. When the radio frequency wave is applied, it causes the transition of these excess nuclei from the lower energy state to the higher energy state until the population difference becomes zero, as the populations at both energy levels become equal. Such a state is referred to as ?saturation? state. To regain the Boltzmann distribution of the higher number of nuclei in the ground state, various relaxation processes take place that allow the nuclei from the higher energy state to come back to the lower energy state. This results in an equilibrium state at an intermediate level between restorations of the initial Boltzmann distribution and complete elimination of that distribution. Such a state can continue to produce an NMR signal. Precessional orbit B0 Spinning axis Nucleus 5 There are mainly two relaxation processes: 1. Spin-Lattice relaxation (T1) 2. Spin-Spin relaxation (T2) 1.1.2.1 Spin-lattice relaxation (T1) The precessing nucleus under the influence of an external magnetic field will also face the fluctuating fields generated by the lattice. If the orientation of the field of the lattice is correct and its frequency equals the precession frequency of the nuclei of the higher energy level, then the energy of the nuclei can release energy to the lattice in the form of thermal energy and the nuclei can relax back to the lower energy state along the Z-axis. T1 depends both on molecular motion of the lattice and the gyromagnetic ratio of the nucleus. The external magnetic field has a very strong influence on T1 and the higher the magnetic field the slower the T1 value (which means more efficient relaxation). 1.1.2.2 Spin-spin relaxation (T2) This relaxation process is also known as transverse relaxation. In this relaxation process the excited magnetization vector decays in the direction of X-Y plane which is perpendicular to the external magnetic field. The magnetic field of a precessing nucleus has two components; one that is aligned with the external magnetic field and another one is spinning at processional frequency in the X-Y plane. The component parallel with the applied field is its static component and the other one is rotating component (Figure 1.3). 6 B0 Static Component Rotating Component Figure 1.3: The two components of a spinning nucleus in an applied magnetic field. The rotating component present in flipping orientation is also shown here. Adapted from reference 1. 7 The static component of the magnetic field of the nucleus will add up to the main field as experienced by any neighboring nucleus resulting in the broadening of the resonance signals. If the neighboring nucleus is precessing at the same frequency as that of the rotating component and at the correct orientation, then it would cause a mutual exchange of energy between the two nuclei. Such exchange of energy would cause the spin to relax back to its original state. The time needed for such relaxation is known as spin-spin relaxation or T2. This would also cause the broadening of the resonance signal. T2 is shorter than T1 and can be determined by NMR. 1.1.3 Chemical shift Depending upon the magnitude of the external magnetic field, gyromagnetic ratio of the nucleus and the molecular environment of the nucleus, a nucleus comes to resonance at a certain frequency. It is the third factor which gives rise to the notion of ?chemical shift?. Chemical shift of a particular nucleus can be defined by the following equation: ???????????????????????????????????????????????????????????s - ?Standard)/Z (1.3) where ?s is the resonance frequency of the nucleus in Hz, ?Standard is the resonance frequency, in Hz, of an internal standard (that usually gives a sharp signal at a high value of the magnetic field) while recording NMR spectra, and Z is the frequency of the instrument in MHz (megahertz=106 Hz). Thus the unit for chemical shift (?) is parts per million or ppm. Nuclei with different molecular environment show different chemical shifts. This is very useful for structure determination by NMR. 8 1.1.4 Spin-spin coupling 1.1.4.1 Scalar coupling (J-coupling) Interaction between nuclei connected through the network of chemical bonds results in scalar coupling. This results in splitting of NMR peaks. This happens due to the two possible spin states for any given nucleus (Figure 1.1). Two chemically bonded nuclei influence each other?s magnetic field by their different spin states. If the nucleus remains aligned parallel to the external magnetic field in the lower energy state, the bonded nucleus will need a slightly lower magnetic strength to come to resonance. Whereas, in the other case, it will experience a lower total magnetic field and will require a little higher value of magnetic field to come to resonance. However, the reason behind the transmission of the influence of the magnetic state to a nucleus to another lies in the fact of the changed electronic spin states due to the existing nuclear spin states. If another nucleus overlaps with the same affected bonding orbital then the changed electronic spin states affect the nuclear spin states of the second one. It results in a slight change in resonance frequency for the second nucleus. These two nuclei are called J-coupled. Scalar coupling has many uses in NMR including the three-bond J-coupling for the measurement of dihedral angle, understanding the structural make-ups of atoms in a molecule and, very importantly, coherence transfer or magnetization transfer through scalar couplings. Another important feature of scalar coupling is that it is always constant for a certain set of bonds in a certain molecular structure independent of the external magnetic field. This J-coupling constant property can be very useful in the investigation of various small molecules including drugs. 9 1.1.4.2 Dipolar coupling Dipolar coupling is through space interaction between nuclear spins. Dipolar coupling is involved in most spin-spin relaxation. The Nuclear Overhauser effect (NOE) is also an important outcome from dipolar coupling, which results from the change in the intensity of the resonance signal of a nucleus when the signal of dipolar coupled another nucleus is changed. NOEs are very important for the investigation of the structure of various bio-macromolecules (such as protein, DNA, and RNA) and large organic compounds and also the interaction between different molecules. 1.1.5 Nuclear Overhauser effect (NOE) When a nucleus is irradiated with radio frequency, relaxation processes only through scalar coupling are not enough for that nucleus to reach the equilibrium state. Then, through a dipolar coupling relaxation mechanism, this nucleus can transfer some of its energy to another nucleus that is close enough in space. The second nucleus behaves as if it had been irradiated and relaxes back to the ground state. It causes the population of the ground state to increase and, thus, the intensity of the second nucleus is enhanced. This phenomenon is called nuclear Overhauser effect (NOE). It can be illustrated by considering two nuclei A and B that are close enough in space for the relaxation process to affect each other. Both of the nuclei can exist in two different spin states, ? or ?, where ? is the lower energy state. Thus, these two nuclei can be represented by four energy states: ??, ??, ??, and ?? (Figure 1.4). The allowed transitions here are between 10 adjacent levels, such as from ?? to ???or from ???to ?? (W1). When a radiation frequency is applied to irradiate one of the two nuclei, then the populations between sates ???and ???or between states ???and ???become equal. However, there is still a population difference that remains between spin states ?? and ??. Dipolar coupling relaxation process (W2) allows restoration of this difference to some extent resulting in the increase of intensity of the NMR resonance line for the second nucleus. This results in a positive NOE which is prevalent for small molecules that tumble in solution fast. For larger molecules, which slowly tumble in the solution, another type of relaxation process operates between ?? and ?? (W0). As a result, a decrease in the population difference occurs between ???and??? (or ?? and ??). This produces lesser intensity in the lines known as negative NOE. NOE difference measurements can be used to determine the distance between two nuclei. Its intensity is inversely proportional to r6, where r is the distance between two nuclei. Thus, with the increase in the distance of the two nuclei, there will be a proportional decrease of NOE intensity. 11 Figure 1.4: Energy diagram for a two-spin system. Adapted from reference (2). 12 1.1.6 Multidimensional NMR spectroscopy 1.1.6.1 Two-dimensional NMR (2D NMR) With the development of various multidimensional NMR spectroscopic methods in the last few decades it was possible to observe the growth in the successful application of NMR to biological studies. It started off with the introduction of a time period known as the evolution period between preparation and detection periods by Jeener in 1971 which formed the basis for the two-dimensional (2D) NMR spectroscopy. Thus, the time-axis of any 2D experiment can be divided into three (or four) segments (Figure 1.5). These are the preparation period, the evolution period and the detection period. The preparation period allows the nuclei to reach thermal equilibrium. Also, it helps to produce the same starting condition each time. The evolution period t1 is gradually increased. After each t1, the magnetization is detected in the form of a FID during the detection period t2. As a result, a series of FIDs are obtained. Fourier transformation of the t2 dimension yields a set of 1D spectra with the varying intensities of the lines due to the changes in the t1 duration. A desired 2D spectrum is possible to obtain with a subsequent Fourier transformation of t1 dimension. Figure 1.5: General scheme for a 2D experiment. Adapted from reference (3). 13 Some of the very important 2D experiments are discussed below. 1.1.6.1.1 COSY In this experiment, the magnetization is transferred between protons that are chemically bonded (up to 3 bonds) on adjacent nuclei (Figure 1.6A). Thus, it provides the information on the protons that are 3J-coupled. This is one of the first and simplest multi-dimensional experiments (4). 1.1.6.1.2 TOCSY In this experiment, information on all the protons attached to nuclei within a given spin system (Figure 1.6 B) is obtained. This includes protons that are beyond 3J chemical bonds. In this experiment, following the evolution period, during the mixing period, the spin is locked in the transverse plane for some time. Scalar coupling results in the transfer of coherence during this mixing period. 1.1.6.1.3 HSQC In NMR, proton is more sensitive (has higher gyromagnetic ratio) than any other heteronuclei. To get a good signal of the heteronuclei, an HSQC (Heteronuclear Single-Quantum Coherence) (5) experiment utilizes the INEPT (Insensitive Nuclei Enhancement by Polarization Transfer) sequence to transfer the magnetization of the proton to its bonded heteronuclei (13C or 15N). This is then transferred back to the magnetization of the proton by a second INEPT sequence for the detection. An HSQC spectrum has two axes; one is for the proton chemical shift and another one is for the heteronuclear chemical shift. 14 Figure 1.6: A) COSY and B) TOCSY spectra. Adapted from reference (6). 15 1.1.6.1.4 NOESY In this experiment, dipolar coupling is involved between two spatially close (nearer than 5 angstrom) nuclei. Magnetization is transferred through the J-coupling during the mixing time. For structure calculations, NOESY is one of the most useful information since it connects the nuclei through space. The distance information comes as a function of the intensity of the peaks. 1.1.6.2 Three-dimensional NMR (3D NMR) For the determination of the structure of small proteins, 2D experiments have been used quite successfully over time (7). However, with the increase in the size (more than 100 residues) of the proteins, 2D experiments alone were not enough anymore to get the structure. There are two basic reasons for this limitation of 2D experiments: i. For the larger protein, due to the large volume of information for the high number of residues of the protein, only the space of two-dimension becomes insufficient. As a result, too much overlap of the peaks within the spectrum makes it impossible to interpret the data. ii. As the size of the protein increases, the rotational correlation time increases. This results in slower movement of the protein in the solution leading to the broadening of the line-width of the resonance which can become larger than the J-coupling constant (7). In order to improve the chance of determination of the structure of larger proteins using NMR, the dimension needs to be increased to get rid of the overlap and also heteronuclear coupling is essentially utilized to make use of the scalar coupling which is larger than the line- 16 widths. That is why, larger proteins are routinely over-expressed by growing heterologous expression systems in minimal media containing 15N or 13C- labeled component as their sole source of nitrogen or carbon (7). 3D NMR, principally, can be easily constructed by combining two sets of 2D NMR experiments (Figure 1.7). As illustrated in Figure 1.7, by removing the detection period of the first set of 2D experiment and preparation period for the second, a 3D experiment combining two evolution periods (t1 and t2), independent of each other, is originated. Figure 1.7: 3D experiment as a combination of two sets of 2D experiments. Adapted from reference (3). 17 Figure 1.8: Schematic presentation of how with the addition of another evolution time to the 2D experiment (Figure 1.7) can result in another frequency dimension for a 3D experiment. The black dots representing NOE cross peaks in 2D spectrum are hard to attribute to the correct proton destination in the F2 dimension. But, in the 3D spectrum, the expansion of another frequency dimension arising from the heteronucleus allows the determination of NOEs involving protons that lie in three different planes. Now, in the 3D spectrum, each plane corresponds to the specific chemical shift of the heteronucleus whereby NOE peaks arising from the interaction between protons attached to the heteronucleus and the other protons on F3 dimension can be detected. Some of the peaks cannot be seen because of the overlapped planes in the presentation used in this figure. Adapted from reference (8). 18 1.1.7 Protein NMR There are two principal methods to determine the structure of a protein; one is X-ray crystallography and another one is NMR. Both of these techniques have their own pros and cons. The focus on NMR as an alternative tool to X-ray crystallography to determine the structure of proteins has grown over the years for several reasons. Not all proteins can be crystallized and even, if it is crystallized, it might not produce good enough diffraction data to get the structure. Also, with proteins in crystals one could be missing some important dynamic information that the proteins in solution might possess, something that can be detected by NMR. However, NMR has an intrinsic disadvantage of larger line-width attributed to longer tumbling time with increasing size of the protein. Also, it needs a very high concentration of protein as a sample (~300-600 ?L protein of 0.1-3mM) and concentrated protein tends to aggregate. To determine a 3D structure of a protein by NMR, the first step is to assign the back-bone of the protein. Various heteronuclear 3D experiments are employed for this purpose. Among these, the most common ones are HNCA (9), HN(CO)CA (9, 10), HNCACB (11), CBCA(CO)NH (11), HNCO (9), and HN(CA)CO (12). All these experiments are composed of a 2D HSQC plane of 15N and 1H in X and Y axes while in the Z-dimension 13C chemical shifts are placed. In the HNCA experiment, the amide proton is correlated with the C? atom of its own residue (residue i) and of the residue preceding it (residue i-1). On the other hand, the HN(CO)CA experiment allows the correlation between the amide proton (residue i) and C? atom of its preceding residue (residue i-1). Assignments from these two experiments can be accomplished in parallel to match the chemical shifts. Similarly, HNCO and HN(CA)CO spectra can be examined together to determine the correlation between amide proton and carbonyl 19 carbons and HNCACB and CBCA(CO)NH spectra for both C? and C? atoms? correlation with amide protons. All these experiments are simultaneously assigned to correctly obtain the chemical shifts of each possible nucleus without any ambiguity (Table 1.2). Once the backbone assignment is done, the next step will be to assign the side chains of the protein using various 3D heteronuclear experiments such as 15N-edited HSQC-TOCSY (13, 14) and HCC(CO)NH (9). The basic principle of any NMR structure determination is to assign a specific resonance to each proton and then to identify the NOE interactions between a pair of protons. A number of experiments are used to determine these NOE interactions such as 15N/13C-edited HSQC- NOESY (13-15). Initially, the sequential and short-range inter-residual NOEs are assigned as they are comparatively easier to assign. Then, the long-range NOEs are dealt with which are much harder to assign but are the most important for determining the global fold of the protein structure. The structure calculation of the protein can then be initiated by using the obtained NOE restraints and dihedral angle restraints which are entered into computer programs like CYANA (16) or XPLOR-NIH (17, 18). For soluble proteins, the energy function of the structure is further lowered by using a water refinement module in ARIA (19). The output of these computational calculations consists of a set of best structures from all the probable calculated structures characterized by good convergence of the well defined parts of the protein. A good set of 20 structures should have low RMSD, a low energy function and few angle and distance violations. A software program named PROCHECK (20) is used to assess these attributes of the structure. Table 1.2: Correlation observed for some of the most commonly used 3D NMR experiments. Adapted from reference (6). 21 1.1.8 Study of protein-ligand interaction by NMR The study of the interaction between proteins and other molecules (such as DNA, RNA, sugars, or even another protein/peptide) in solution has become more common (21). Such studies hold key to understanding various biological processes, for example, the interaction between enzyme and its substrate/inhibitor or the binding of various transcription factors to DNA. NMR is a very useful and powerful tool to investigate such interactions. One important aspect of the NMR study of protein-ligand interactions is to determine the effect of chemical exchange on NMR spectra, that is, to determine whether the bound and free form of the protein coexist in the fast or slow exchange regime on the NMR time scale. In fast exchange, a single average resonance peak is observed, whereas in slow exchange, two different resonance peaks are observed for a single nucleus (Figure 1.8). For intermediate exchange the two resonance peaks will appear to coalesce together into one, and if there is no exchange, then there would hardly be any line broadening (Figure 1.9). As previously discussed, line-width of the resonance peaks is inversely proportional to spin-spin relaxation (T2). This phenomenon can easily be correlated with the strength of the interaction for the protein-complex. The faster the chemical exchange indicates, the looser the interaction between protein and its interacting partner. 15N-HSQC experiment is routinely used to observe the effect of chemical exchange on the protein for its specific residues. These observations can sometimes even lead to a basic idea on what part of the protein is actually involved in binding. Also, the determination of dissociation constant (KD) values for that specific interaction is possible through these experiments. However, though, such a method for determination of KD values is not very accurate (21). 22 Figure 1.9: Effect of chemical exchange on NMR spectra. Drawn according to http://web.nmsu.edu/~snsm/classes/chem435/Lab8/. 23 To determine the structure of the protein-interacting partner complex, it is important to assign the resonances of protons for both the protein and interacting partner individually. For the protein, as for its free form, the same different types of 3D heteronuclear experiments are done to obtain the resonances of the protons of the protein in its complex form. If the interacting partner is a protein and isotopically labeled, then the same experimental procedures can be followed for the assignment of protons of the partner as well. However, if the interacting partner is not isotopically labeled (such as unlabeled peptide), then along with other conventional experiments, a unique experiment, known as filtered NOESY experiment, is used. This experiment can be designed in such a way that any resonance that arises from labeled nuclei should be eliminated. Thus, only those resonances that originate from the unlabeled peptide are detected and assigned. In this way, successful assignment of the nuclei (mainly protons) present in the unlabeled peptide is possible. Additionally, filtered NOESY experiments can also be employed to determine the intermolecular NOEs between the protein and the peptide. This allows building up NOEs necessary to dock the peptide onto the protein in NMR calculation. 24 1.2 PDZ domain and Glutaminase Interacting Protein (GIP) To maintain an efficient and active cellular physiology, it is important for the cells to maintain effective signaling systems and protein-protein interaction is at the root of these signaling systems. There are several motifs/domains/modules that are involved in such interactions. PDZ domains (Post Synaptic Density 95 (PSD-95), Discs Large (Dlg) and Zonula Occludentes (ZO-1)) (22-24) are one of the most ubiquitous and well known domains involved in protein-protein interactions. The PDZ domain is widespread in the nature. It is involved in multiple processes and possibly numerous others are yet to be discovered. Protein scaffolding (25, 26), maintaining cell polarities (27), localizing and clustering of ion-channels are to name only a few of myriad of the processes it plays a role in. It is an 80-100 amino acid long motif. Usually, PDZ-domain containing proteins, having more than one PDZ domains, are involved in the formation of multimeric protein complexes (Figure 1.10). This domain is primarily found in eukaryotic organisms (28), but can be found in a slightly different form in prokaryotes and plants as well (29-31). Based on its specificity toward the sequence of its binding partner, PDZ domain can be classified into three major broad classes: a. Class I PDZ domain (binding motif S/T-X-?-COOH, where ? is a hydrophobic amino acid and X is any amino acid) b. Class II PDZ domain (binding motif ?-X-?-COOH) c. Class III PDZ domain (binding motif X-X-C-COOH) 25 Figure 1.10: Examples of PDZ domain containing proteins. Adapted from reference (28). 26 However, some PDZ domains cannot be categorized into any of the above classes (Table 1.3) (32). PDZ domains are involved in various cancer pathways (33-40). G-protein coupled receptors and ion channels are very important candidates for drug development. It was found that GIP interacts with these proteins for its proper function. Thus, developing a drug molecule that will compete with GIP for binding with these targets could prove promising (33). Because of its diverse functions and implications in various diseases including cancer, it is very important to investigate the interacting partners of PDZ domains and gain structural insight into the mode of binding of PDZ domains with their partner protein, which are critical for the development of drug candidates (41). 1.3 Glutaminase Interacting Protein (GIP)/Tax Interacting Protein-1 (TIP-1) GIP, also known as Tax Interacting Protein 1 (TIP-1) is a small PDZ domain containing protein. This protein is 124 amino acid long with a molecular mass of 13.7 kDa. GIP contains a single PDZ domain which is unique among PDZ containing proteins. GIP is also an excellent protein for structure-function studies by solution NMR, since, it is a small globular protein having good solubility properties and stability (NMR sample can be stored even up to several months without any aggregation). Additionally, research methods have been developed in our laboratory to over-express and purify the recombinant protein in milligram quantities in a single step (42). Additionally, GIP is implicated in many cancer pathways due to its interactions with a growing list of partner proteins all with different roles in the cell. The role of GIP in many of 27 these processes is not yet understood at the molecular level. Thus, to understand the functions of GIP, it is important to characterize the interaction between GIP with different binding partner proteins to gain an insight into its mechanism of interaction and mode of recognition. In this dissertation work, we have solved the solution structures of GIP both in the free state and also bound to a substrate, the C-terminal octa peptide of Glutaminase L (KENLESMV- COOH) using solution NMR. This is the first NMR structure of a complex of GIP. With this structural information, essential knowledge can be obtained on the mechanism of interactions and mode of recognition between GIP and those interacting partners that contain C-terminal recognition motif. This knowledge will be essential for structure-based drug design with either GIP as target or its partner proteins. Further, we also characterized the interaction between GIP and a peptide mimic of a human Brain-specific Angiogenesis Inhibitor 2 (BAI2) using various biophysical techniques. Discovery of the complete network of interacting partners for GIP is necessary to comprehend fully the function of GIP in the human brain and other parts of the body. 28 PDZ Domain Consensus binding sequence Ligand protein Class I P-3 -P-2 - P-1 -P0 S/T-X-?-COOH Syntrophin (43) E-S-T-V-COOH Voltage-gated Na+ channel PSD-95 (26) E-T-D-V-COOH Shaker-type K+ channel GIP (44) E-S-M-V-COOH Glutaminase-L Class II ?-X-?-COOH hCASK (45) E-Y-Y-V-COOH Neurexin Erythrocyte p55 (46) E-Y-F-I-COOH Glycophorin C Class III X-X-C-COOH Mint-1 (47) D-H-W-C-COOH N-type Ca +2 Channel SITAC (48) Y-X-C-COOH L6 antigen Other nNOS (49) G-D-X-V-COOH MelR MAGI PDZ2 (50) S/T-W-V-COOH Phage display Engineered from SF6 (51) K/R-Y-V-COOH Synthesized peptide Table 1.3: Classification of PDZ domains. Adapted from reference (32). 29 1.4 References 1. Rahman, A. U. (1986) Nuclear Magnetic Resonance: Basic Principles, Springer-Verlag, New York. 2. Prestegard, J. H., Bougault, C. M., and Kishore, A. I. (2004) Residual dipolar couplings in structure determination of biomolecules, Chemical reviews 104, 3519-3540. 3. Cavanagh, J., Fairbrother, W.J., Palmer, A.G., Skelton, N.J., and Rance, M. (2006) Protein NMR spectroscopy: principles and practice., second ed., Academic Press. 4. Aue, W. P., Bartholdi, E., and Ernst, R. R. (1976) Two dimensional spectroscopy. Application to nuclear magnetic resonance., J. Chem. Phys. 64, 2229-2246. 5. Bodenhausen, G., and Ruben, D. J. (1980) Natural abundance nitrogen-15 NMR by enhanced heteronuclear spectroscopy, Chemical Physics Letters 69, 185-189. 6. Huang, C. (2010) Production, characterization and structure determination of the C- terminal domain of Stt3p: the catalytic subunit of yeast oligosaccharyl transferase, In Department of Chemistry and Biochemistry, p 253, Auburn University, USA. 7. Clore, G. M., and Gronenborn, A. M. (1993) NMR of proteins, CRC Press Inc., USA. 8. Clore, G. M., and Gronenborn, A. M. (1991) Structures of larger proteins in solution: three- and four-dimensional heteronuclear NMR spectroscopy, Science 252, 1390-1399. 9. Grzesiek, S., and Bax, A. (1992) Correlating backbone amide and side chain resonances in larger proteins by multiple relayed triple resonance NMR, Journal of the American Chemical Society 114, 6291-6293. 10. Bax, A., and Ikura, M. (1991) An efficient 3D NMR technique for correlating the proton and 15N backbone amide resonances with the alpha-carbon of the preceding residue in uniformly 15N/13C enriched proteins, Journal of biomolecular NMR 1, 99-104. 30 11. Muhandiram, D. R., and Kay, L. E. (1994) Gradient-Enhanced Triple-Resonance Three- Dimensional NMR Experiments with Improved Sensitivity, Journal of Magnetic Resonance, Series B 103, 203-216. 12. Clubb, R. T., Thanabal, V., and Wagner, G. (1992) A constant-time three-dimensional triple-resonance pulse scheme to correlate intraresidue 1HN, 15N, and 13C? chemical shifts in 15N/13C-labelled proteins, Journal of Magnetic Resonance (1969) 97, 213-217. 13. Norwood, T. J., Boyd, J., Heritage, J. E., Soffe, N., and Campbell, I. D. (1990) Comparison of techniques for 1H-detected heteronuclear 1H/15N Spectroscopy, Journal of Magnetic Resonance (1969) 87, 488-501. 14. Palmer Iii, A. G., Cavanagh, J., Wright, P. E., and Rance, M. (1991) Sensitivity improvement in proton-detected two-dimensional heteronuclear correlation NMR spectroscopy, Journal of Magnetic Resonance (1969) 93, 151-170. 15. Zhang, O., Kay, L. E., Olivier, J. P., and Forman-Kay, J. D. (1994) Backbone 1H and 15N resonance assignments of the N-terminal SH3 domain of drk in folded and unfolded states using enhanced-sensitivity pulsed field gradient NMR techniques, J Biomol NMR 4, 845-858. 16. Guntert, P. (2004) Automated NMR structure calculation with CYANA, Methods Mol Biol 278, 353-378. 17. Schwieters, C. D., Kuszewski, J. J., Tjandra, N., and Clore, G. M. (2003) The Xplor-NIH NMR molecular structure determination package, J Magn Reson 160, 65-73. 18. Schwieters, C. D., Kuszewski, J. J., and Marius Clore, G. (2006) Using Xplor?NIH for NMR molecular structure determination, Progress in Nuclear Magnetic Resonance Spectroscopy 48, 47-62. 31 19. Linge, J. P., Habeck, M., Rieping, W., and Nilges, M. (2003) ARIA: automated NOE assignment and NMR structure calculation, Bioinformatics 19, 315-316. 20. Laskowski, R. A., Rullmannn, J. A., MacArthur, M. W., Kaptein, R., and Thornton, J. M. (1996) AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR, J Biomol NMR 8, 477-486. 21. Roberts, G. C. K. (1993) NMR of macromolecules : a practical approach, IRL Press at Oxford University Press, Oxford ; New York. 22. Ponting, C. P., and Phillips, C. (1995) DHR domains in syntrophins, neuronal NO synthases and other intracellular proteins, Trends in biochemical sciences 20, 102-103. 23. Kennedy, M. B. (1995) Origin of PDZ (DHR, GLGF) domains, Trends in biochemical sciences 20, 350. 24. Sheng, M., and Sala, C. (2001) PDZ domains and the organization of supramolecular complexes, Annual review of neuroscience 24, 1-29. 25. Kornau, H. C., Schenker, L. T., Kennedy, M. B., and Seeburg, P. H. (1995) Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD- 95, Science 269, 1737-1740. 26. Kim, E., Niethammer, M., Rothschild, A., Jan, Y. N., and Sheng, M. (1995) Clustering of Shaker-type K+ channels by interaction with a family of membrane-associated guanylate kinases, Nature 378, 85-88. 27. Bhat, M. A., Izaddoost, S., Lu, Y., Cho, K. O., Choi, K. W., and Bellen, H. J. (1999) Discs Lost, a novel multi-PDZ domain protein, establishes and maintains epithelial polarity, Cell 96, 833-845. 32 28. Lee, H. J., and Zheng, J. J. (2010) PDZ domains and their binding partners: structure, specificity, and modification, Cell communication and signaling : CCS 8, 8. 29. Pallen, M. J., and Ponting, C. P. (1997) PDZ domains in bacterial proteins, Molecular microbiology 26, 411-413. 30. Ponting, C. P. (1997) Evidence for PDZ domains in bacteria, yeast, and plants, Protein science : a publication of the Protein Society 6, 464-468. 31. Liao, D. I., Qian, J., Chisholm, D. A., Jordan, D. B., and Diner, B. A. (2000) Crystal structures of the photosystem II D1 C-terminal processing protease, Nature structural biology 7, 749-753. 32. Harris, B. Z., and Lim, W. A. (2001) Mechanism and role of PDZ domains in signaling complex assembly, Journal of cell science 114, 3219-3231. 33. Wang, N. X., Lee, H. J., and Zheng, J. J. (2008) Therapeutic use of PDZ protein-protein interaction antagonism, Drug news & perspectives 21, 137-141. 34. Cushing, P. R., Fellows, A., Villone, D., Boisguerin, P., and Madden, D. R. (2008) The relative binding affinities of PDZ partners for CFTR: a biochemical basis for efficient endocytic recycling, Biochemistry 47, 10084-10098. 35. Wong, H. C., Bourdelas, A., Krauss, A., Lee, H. J., Shao, Y., Wu, D., Mlodzik, M., Shi, D. L., and Zheng, J. (2003) Direct binding of the PDZ domain of Dishevelled to a conserved internal sequence in the C-terminal region of Frizzled, Molecular cell 12, 1251-1260. 36. Aarts, M., Liu, Y., Liu, L., Besshoh, S., Arundine, M., Gurd, J. W., Wang, Y. T., Salter, M. W., and Tymianski, M. (2002) Treatment of ischemic brain damage by perturbing NMDA receptor- PSD-95 protein interactions, Science 298, 846-850. 33 37. Dev, K. K. (2004) Making protein interactions druggable: targeting PDZ domains, Nature reviews. Drug discovery 3, 1047-1056. 38. Wolde, M., Fellows, A., Cheng, J., Kivenson, A., Coutermarsh, B., Talebian, L., Karlson, K., Piserchio, A., Mierke, D. F., Stanton, B. A., Guggino, W. B., and Madden, D. R. (2007) Targeting CAL as a negative regulator of DeltaF508-CFTR cell-surface expression: an RNA interference and structure-based mutagenetic approach, The Journal of biological chemistry 282, 8099-8109. 39. Georgescu, M. M., Morales, F. C., Molina, J. R., and Hayashi, Y. (2008) Roles of NHERF1/EBP50 in cancer, Current molecular medicine 8, 459-468. 40. Georgescu, M. M. (2008) NHERF1: molecular brake on the PI3K pathway in breast cancer, Breast cancer research : BCR 10, 106. 41. Banerjee, M. (2011) Human Glutaminase Interacting Protein (GIP): a Potential Candidate for Anti-Cancer Drug Design, In Department of Chemistry and Biochemistry, p 172, Auburn University, USA. 42. Banerjee, M., Huang, C., Marquez, J., and Mohanty, S. (2008) Probing the structure and function of human glutaminase-interacting protein: a possible target for drug design, Biochemistry-Us 47, 9208-9219. 43. Schultz, J., Hoffmuller, U., Krause, G., Ashurst, J., Macias, M. J., Schmieder, P., Schneider-Mergener, J., and Oschkinat, H. (1998) Specific interactions between the syntrophin PDZ domain and voltage-gated sodium channels, Nature structural biology 5, 19-24. 34 44. Olalla, L., Aledo, J. C., Bannenberg, G., and Marquez, J. (2001) The C-terminus of human glutaminase L mediates association with PDZ domain-containing proteins, FEBS Lett 488, 116-122. 45. Songyang, Z., Fanning, A. S., Fu, C., Xu, J., Marfatia, S. M., Chishti, A. H., Crompton, A., Chan, A. C., Anderson, J. M., and Cantley, L. C. (1997) Recognition of unique carboxyl-terminal motifs by distinct PDZ domains, Science 275, 73-77. 46. Marfatia, S. M., Byron, O., Campbell, G., Liu, S. C., and Chishti, A. H. (2000) Human homologue of the Drosophila discs large tumor suppressor protein forms an oligomer in solution. Identification of the self-association site, The Journal of biological chemistry 275, 13759-13770. 47. Maximov, A., Sudhof, T. C., and Bezprozvanny, I. (1999) Association of neuronal calcium channels with modular adaptor proteins, The Journal of biological chemistry 274, 24453-24456. 48. Borrell-Pages, M., Fernandez-Larrea, J., Borroto, A., Rojo, F., Baselga, J., and Arribas, J. (2000) The carboxy-terminal cysteine of the tetraspanin L6 antigen is required for its interaction with SITAC, a novel PDZ protein, Molecular biology of the cell 11, 4217- 4225. 49. Stricker, N. L., Christopherson, K. S., Yi, B. A., Schatz, P. J., Raab, R. W., Dawes, G., Bassett, D. E., Jr., Bredt, D. S., and Li, M. (1997) PDZ domain of neuronal nitric oxide synthase recognizes novel C-terminal peptide sequences, Nature biotechnology 15, 336- 342. 35 50. Fuh, G., Pisabarro, M. T., Li, Y., Quan, C., Lasky, L. A., and Sidhu, S. S. (2000) Analysis of PDZ domain-ligand interactions using carboxyl-terminal phage display, The Journal of biological chemistry 275, 21486-21491. 51. Schneider, S., Buchert, M., Georgiev, O., Catimel, B., Halford, M., Stacker, S. A., Baechi, T., Moelling, K., and Hovens, C. M. (1999) Mutagenesis and selection of PDZ domains that bind new protein targets, Nature biotechnology 17, 170-175. 36 Chapter 2 Characterization of Glutaminase Interacting Protein (GIP): a PDZ domain 2.1 Background 2.1.1 Protein-protein interaction network One of the challenging tasks to understand cells and diseases is to know how within the cells a network of proteins is connected. This knowledge of protein networks will help to shed light on the inner machinery of the cells. It would also allow scientists to specifically target a protein within that network to treat a disease and, thus, help narrowing down the potential targets to combat diseases. Proteins can interact with other proteins, metabolites and DNA or RNA in a cell. Several experimental tools have been employed to determine the protein-protein interaction either as a direct approach such as yeast two-hybrid screening, mass spectrometry (MS) and immunoprecipitation or on a genome-wide level such as chromatin immunoprecipitation (ChIP) on chip assays and double knockout assays in yeast (Figure 2.1) (1). However, constructing a comprehensive protein-protein interaction network is well beyond the scope of this thesis. Besides establishing an interaction network between different proteins, it is also very important to investigate the mechanisms by which the proteins interact with each other. With an insight into the structure, binding mechanisms and mode of interactions between different proteins, it is possible to design the most effective and selective drug 37 Figure 2.1: Different experiment tools to define protein interactions. A. Yeast two-hybrid screening B. ChIP on chip assay C. Double knockout assay in yeast. Adapted from reference (1). 38 compounds that may have an expected therapeutic effect as a result of their interactions with the targeted proteins. These protein-protein interaction networks are important for maintaining a continuous and ordered communication within cells. The interactions between proteins are facilitated via a number of different interacting domains such as SH2 (Src Homology 2) (2), SH3 (Src Homology 3) (3), PH (Pleckstrin-homology) (4), PDZ (Post synaptic density 95, Discs large and Zonula occludentes) (5, 6) and others (7) which can be present either as a single domain or multiple domains in a single protein. 2.1.2 PDZ domain and its classes In nature, there are many protein-protein interaction modules present. One of the most important of these interaction modules is the PDZ domain (8). These domains are small and contain 80-100 amino acid residues. 1-2 ?-helices and 5-6 ?-strands comprise these domains. Animals contain many PDZ domain/s containing proteins. However, in yeast and plants, ?PDZ- like? domains that are structurally similar, but not exactly same, have been found (9, 10). PDZ- like domain consists of 5 ?-strands (?1-?5) capped by 2 ?-helices (?2 and ?3) and also two short ?-strands at the N and C termini (?N and ?C). Also, a well-defined ?-helix (?1) is formed between the ?1 and ?2 loop (11). PDZ domains are involved in various important cellular functions, including signaling pathways and acting as scaffolds to organize multimeric complexes often with the help of other protein-protein interaction modules (7). PDZ domains usually recognize the unstructured C-terminal end of their interacting partner proteins (12). But, 39 in rare cases, proteins with internal motifs that structurally mimic the C-terminus can bind to PDZ domains (13, 14). PDZ domains can be categorized mainly into three classes according to the sequence specificity of their binding partners (15). They are class I (X-S/T-X-?-COOH) (16), class II (X-?-X-?-COOH) (7), class III (X-E/D-X-?-COOH) (17) and, also, various other minor classes (18) where ? is any hydrophobic residue and X is any residue (19). 2.1.3 Glutaminase interacting protein as a class I PDZ domain Glutaminase Interacting Protein (GIP) is a PDZ domain containing protein that has a number of important functions (20). It is also known as Tax Interacting Protein-1 (TIP-1) (21). GIP is a very small protein containing only 124 amino acid residue. Also, it is unique among PDZ containing proteins since the whole protein is composed solely of a single PDZ domain without any other additional domain/s. All other PDZ domain containing proteins usually have either more than one PDZ domains and/or contain other domains such as SH2, SH3 etc. (7). Over the last several years, there has been an increasing number interacting partner proteins reported for GIP including Glutaminase L (20), ?-Catenin (22, 23), Fas (24, 25), HTLV Tax (Human T-lymphotropic virus Tax) (21), HPV E6 (Human papillomavirus E6) (26), Rhotekin (27) and Kir 2.3 (28, 29). All these interacting partners contain the PDZ class I (X-S/T-X-I/L/V- COOH) binding motif. To get an insight into the mechanism of GIP?s recognition and mode of interaction with such a wide range of proteins, it is critical to investigate these binding events to understand the molecular basis of the functions that these proteins carry out in the cells. For example, ?-Catenin and Rhotekin are important in the Wnt and Rho signaling pathways, respectively. Fas is a member of the Tumor Necrosis Factor (TNF) family of receptors, while 40 HTLV Tax and HPV E6 are both viral proteins from oncogenic viruses. Lastly, GIP regulates the inward rectifier potassium channel Kir 2.3 in renal epithelial cells. GIP has been shown to be involved in a variety of different cancer and cell signaling pathways with its numerous binding- partner proteins. Also, GIP is involved in the regulation of Glutaminase L, which has been shown to be up-regulated in various cancers (30-32). By doing sequence alignment of all of these discovered interacting partners; it is possible to identify the optimal consensus sequence for GIP binding as to be E-S-X-V-COOH (Table 2.1) (19). Table 2.1: Sequential alignment of C-terminal binding partners of GIP. Adapted from reference (19). 41 2.1.4 Objective of the study As a first step towards understanding the mechanism of recognition and mode of interaction of GIP with its various partner proteins, it is imperative to solve the high resolution structure of GIP at atomic level. The atomic structures of proteins both in the free-state and bound with their substrate provide snapshots of many complex features of the biological event including residues involved in the binding, site of interaction etc. Solution-state NMR enables us to investigate the protein under biological condition and also allows examining the dynamics of these processes in the timescale of picoseconds to seconds. In this chapter, the NMR experiments and analysis method are described that were used to determine the atomic structure and the dynamics of free GIP in solution. We determined the NMR structure of free GIP in solution with a backbone RMSD of 0.45 ?. We also investigated the dynamics of the free GIP. Comparison of this structural and dynamic information of free GIP with those of GIP bound with a surrogate peptide that mimics the C-terminus of Glutaminase L (Chapter 3 of this dissertation paper) yielded insight into the mechanism of interaction of GIP with its binding partners. 42 2.2 Materials and Methods The research work described here was carried out in the laboratory of Dr. Smita Mohanty. 2.2.1 Cloning, over-expression and purification of 15N, 13C-labeled GIP According to the protocol developed in Dr. Smita Mohanty?s laboratory, the double- labeled free GIP protein was prepared by Dr. Smita Mohanty and other group members (23). 2.2.2 NMR Data collection All NMR data were collected on a Bruker Avance 600 MHz spectrometer with a triple resonance 1H/13C/15N TCI cryoprobe equipped with z-axis pulsed field gradients at either the Department of Chemistry and Biochemistry, Auburn University, Auburn, AL, or the New York Structural Biology Center (NYSBC), New York, NY. The data were processed using NMRPipe (33) and analyzed using Sparky (34). For structure determination of free GIP, samples between 500 ?M and 1 mM of uniformly 15N/13C-labeled GIP in 50 mM phosphate buffer containing 5% D2O pH 6.5, 1 mM EDTA and 0.01% (w/v) NaN3 were prepared. All NMR experiments were performed at 298 K. To determine the 15N T1 values, NMR spectra were recorded with relaxation delays of 10, 600, 50, 500, 100, 400, 200, 300 and 10 ms. To determine 15N T2 values, NMR spectra were recorded with delays of 17, 153, 34, 17, 136, 51, 119, 68, 102, 85 and 34 ms. The relaxation times were randomized and some points repeated in order to avoid any systematic errors that may arise when the data are collected sequentially. The relaxation rates were 43 calculated by least squares fitting of peak heights versus relaxation delay to a single exponential decay. Steady state 1H-15N NOE values were calculated from the ratio of peak heights in a pair of NMR spectra acquired with and without proton saturation. These dynamics data were analyzed in collaboration with Dr. David Zoetewey in Dr. Smita Mohanty?s research group. For backbone and side-chain assignments of free GIP, the following NMR experiments were recorded at 298 K: 2D 1H,15N-HSQC (35), 3D HNCACB (36), 3D CC(CO)NH (37), 3D CBCA(CO)NH (36), 3D 15N-edited HSQC-TOCSY (38, 39) with an 80 ms mixing time, 3D HC(CO)NH (37), 3D HNHA (40), 3D HNCO (37) and 3D HN(CA)CO (41) at NYSBC by Dr. Smita Mohanty. NOE distance restraints were obtained from 3D 15N-edited HSQC-NOESY (38, 39, 42) and 3D 13C-edited HSQC-NOESY (38, 39, 42) spectra collected both at NYSBC and also again at AU with the 13C carrier frequency in the aliphatic (44 ppm) and aromatic (125 ppm) regions and mixing times of 140 for 15N and 110 ms for 13C, respectively (19). 2.2.3 Analysis of dynamics data Measured relaxation parameters R1, R2 and the steady-state 1H-15N NOE for each residue were used as inputs in the Modelfree 4.15 program developed by Palmer et al (43, 44) to analyze 15N-backbone dynamics. The ?c value for free GIP was calculated using the program Tensor2 for the core region A11-Q112 (45, 46). Of five different models, the best one was chosen according to the selection criteria (43) to get the order parameter (S2) that represents the degree of spatial restriction within the 1H-15N bond vector. These values range from zero for completely isotropic internal motions to unity for totally restricted motion and represent dynamics in the picosecond to nanosecond time scale (19). 44 2.2.4 Structure calculation and refinement A total of 4303 cross peaks were assigned manually using Sparky (34) for free GIP. The assignments were corrected or confirmed with both the CANDID module of CYANA 1.0.6 and NOEASSIGN module of CYANA 2.1 (47), using the standard protocol of eight iterative cycles of NOE assignment and structure calculation. A total of 118 dihedral angles restrains were derived from the TALOS (48) program based on the chemical shift index (CSI) and primary sequence of GIP for free protein calculations. Additionally, a total of 64 hydrogen bond distance restraints (two restraints per bond) for the free protein were derived from the CSI by TALOS. During the iterative NOE assignments, a total of 1134 assignments for free GIP were removed due to overlap, redundancy, or unresolved ambiguity that resulted from low stringency in the initial peak picking phase and high stringency in the final assignments. The final assignments averaged over 25 NOEs per residue for free protein. Final refinement of the 100 lowest energy structures of the 200 total calculated structures was performed with the water refinement protocol implemented in ARIA (49). The 20 structures with the lowest potential energy and best Ramachandran statistics as assessed by PROCHECK (50) were selected for analysis. The structures were visualized with VMD and figures were created using Pymol (51, 52). (Table 2.3 shows the complete structural statistics for structure of GIP alone (19).) 45 2.3 Results 2.3.1 NMR Structure determination of free GIP 2.3.1.1 Introduction De Novo structure determination of small (MW <25 kDa) water soluble molecules such as proteins by NMR spectroscopy is very useful to understand the mechanisms of function of the protein under study. Several steps need to be followed to determine the structure of a protein by NMR. These steps can be summarized in a flowchart (Figure 2.2). Up to this point, the first two steps of the flowchart have been discussed both in the materials and methods section. The next step is to assign the resonance for each individual spin-active nucleus to ultimately utilize those resonances to establish a spatial relationship between these spin-active nuclei through NOE assignments. In the series of steps for the atomic structure determination by NMR, the sequential assignment is the initial step whereby the resonances of the backbone nuclei (15N, 1HN, 13C?, 13C?, 13CO) of the protein chain are assigned. Once this crucial step is accomplished, then the side-chain nuclei attached to these backbone nuclei can be assigned comparatively more easily. When most of the resonances of the nuclei within the protein are assigned, then these resonances would allow assigning NOE resonances which is one of the distance constraints for initial structure calculation by computational method. Hydrogen bonds and dihedral angles (both derived from CSI analysis by TALOS program) are finally used in the three-dimensional structure calculation process. 46 Figure 2.2: Flowchart of structure determination by NMR. Sample preparation. 13C, 15N-labeled NMR data collection. 2D and 3D homo or heteronuclear NMR experiments Resonance assignments. Backbone, side-chain and NOE assignments Generation of constraints. NOE, dihedral angles and hydrogen bonds. Structure calculation by computer in an iterative cycle. 47 2.3.1.2 Backbone assignments of free GIP Backbone assignment of free GIP was carried out previously in our laboratory (24). However, I carried out the backbone assignment as described below again with previously collected NMR data and with some new data to proceed further with the side chain and 3D NOESY data assignments of free GIP. Sequential assignments were accomplished by HNCAB and HNCA experiments. HNCACB experiment allowed the sequential assignments of both C? and C? atoms of both i and (i-1) residues of the peptide chain (Figure 2.3). In HNCA experiment, only C? atoms of both i and (i-1) residues were assigned (Figure 2.4). This sequential assignment was continuous as long as there was no ambiguity or absence of the peaks occurred. To further resolve any ambiguity, HNCO and CBCA(CO)NH experiments were helpful. In the CBCA(CO)NH experiment, only C? and C? atoms of the (i-1) residues were assigned. This helped to reconfirm the assignments of the HNCACB experiment (Figure 2.5). Although, in the figure (Figure 2.5) the peak intensity of the (i-1) residue of HNCACB spectrum is almost same as that of the (i-1) residue of CBCA(CO)NH spectrum, more often than not, the peak intensity of the (i-1) residue of the HNCACB experiments is less than that of the (i-1) residue of CBCA(CO)NH experiment due to the difference in the transfer of magnetization in these two different experiments. This feature also gives an added advantage during the assignments of C? and C? atoms of i and (i-1) residues of HNCACB. HNCO experiment also helped to remove ambiguities and reconfirm the assignments. All these assignments were done by continuously referring to the table of Statistics Calculated for All Chemical Shifts from Atoms of the 20 Common Amino Acids (Biological 48 Magnetic Resonance Data Bank, BMRB, http://www.bmrb.wisc.edu/) which is always updated. Microsoft excel was used to facilitate the sequential assignment. Initially, some unknown numbers were given to each residue. Later on, with the help of excel sheet and the above mentioned table, specific amino acid types and their sequence were obtained. Once sequential assignment was done, it was quite easy to assign the (1H, 15N)-HSQC spectrum (Figure 2.6). N-terminus (M1 residue) of the protein was absent from the (1H, 15N)- HSQC spectrum due to the exchange of the free amide protons with the deuterated solvent (5- 10% D2O). Also, five proline residues were absent from the spectrum due to its unique cyclic structure. However, due to the cis- to trans-isomerization of the proline residues (Figure 2.7), the neighboring residues experienced two different chemical environments, consequently appearing at two different chemical shift positions. Glycine at position 6 and Valine at position 9 were affected by the cis- to trans-isomerization of P5 and P8 and were assigned as G6A and V9A (Figure 2.6). Another noticeable thing in the 1H, 15N-HSQC spectrum was that, the disordered regions within the protein such as N- and C-termini have higher peak intensity with a corresponding higher data height in the Sparky program than the regions that are ordered such as ?-helix and ?- sheet. This happened due to the fast exchange of the flexible regions within the NMR timescale. 49 Figure 2.3: Sequential assignments of V13-K20 from (1H, 13C)-strips of HNCACB experiment (19, 24). Only the C? atoms of the residues were connected with red lines to show the sequential assignment. Positive signals are green and negative signals are red. C? appeared as positive signal and C? appeared as negative signal. 50 Figure 2.4: Sequential assignments of V13-K20 from (1H, 13C)-strips of HNCA experiment (19, 24). C? atoms of the residues were connected with red lines to show the sequential assignment. 51 Figure 2.5: HNCACB and CBCA(CO)NH strips of I18 residue (19, 24). Red lines were used to connect the C? and C? atoms of the (i-1) residue on both spectra. 52 Figure 2.6: 1H, 15N-HSQC spectrum of free GIP (19, 24). Red lines connected the non- degenerate protons of the side-chain amide groups of Asparagine and Glutamine residues. Two red arrows located G6A and V9A. 53 Figure 2.7: Proline cis/trans isomerization. 2.3.1.3 Side-chain assignments of free GIP With the completion of the backbone assignments (24), the next step is to assign the side- chain nuclei of the amino acid residues of the protein. This step is relatively straight-forward. Having the assigned resonances of the backbone atoms, to assign side-chains attached to these backbone atoms (e.g. amide protons), one has to start with a specific amide proton and attached nitrogen resonance of a specific amino acid residue to find the resonances of the side-chain atoms of that residue or the one preceding it from the different spectra. 54 In HC(CO)NH experiment, assignments of side-chain protons of (i-1) residues were accomplished. Side-chain protons of i-residues were assigned in an HSQC-TOCSY experiment. In the later experiment, side-chain protons of the (i-1) residue could also appear as a negative signal (Figure 2.8). Thus, it is a good practice to use these two spectra side-by-side for assigning side-chain protons as a tool for reconfirmation of the assignments. Other spectra used to assign side-chain protons include HCCH-COSY (Figure 2.9) and HCCH-TOCSY (Figure 2.10). These spectra helped to assign the non-degenerate protons of the side-chains. For the assignment of non-degenerate protons of Glycine C?, an HNHA experiment was useful (Figure 2.11). Sometimes, it was also possible to determine non-degenerate protons in 15N-edited HSQC- NOESY or in 13C-edited HSQC-NOESY. For the unambiguous assignments of the NOEs, detection of non-degenerate protons was very important. To assign side-chain carbons, a CC(CO)NH experiment was used which gives resonances of the side-chain carbon atoms of the (i-1) residues (Figure 2.12). 55 Figure 2.8: HC(CO)NH and HSQC-TOCSY spectra showing side-chain assignments of V60 and R59 residues. 56 Figure 2.9: HCCH-COSY spectrum of H? proton of N26 residue showing non-degeneracy of H? protons. 57 Figure 2.10: HCCH-TOCSY spectrum of H?3 proton of P45 residue. 58 Figure 2.11: HNHA spectrum of non-degenerate H? protons of G70 residue. 59 Figure 2.12: CC(CO)NH spectrum of Q23 residue. 60 The statistics of the assignments of the side-chains are summarized in the Table 2.2. In summary, around 92, 95 and 90 percent of all carbon, hydrogen and nitrogen nuclei, respectively, were unambiguously assigned. Although, the peaks in the 1H, 13C-HSQC spectrum are overlapping more than those in the 1H, 15N-HSQC spectrum, this amount of assignments was sufficient to assign most of the peaks of the aliphatic region of the 1H, 13C-HSQC spectrum (Figure 2.13 and Figure 2.13A) and all of the peaks of the aromatic region of the 1H, 13C-HSQC spectrum (Figure 2.14). Assignment of the full 1H, 13C-HSQC spectrum was instrumental in the assignments of the cross-peaks in the 13C-edited HSQC-NOESY spectrum later on. 61 Atom C(CO) C?? C?? C?? C?? C?? C?? C?? Total C % of Assignment 100 100 99 81 73 58 50 100 92.3 Found vs. Expected 124/124 124/124 111/112 74/91 33/45 11/19 2/4 1/1 466/520 Atom HN H?? H?? H?? H?? H?? H?? H?? Total H % of Assignment 99 100 99 87 92 79 100 100 95.2 Found vs. Expected 118/119 124/124 111/112 79/91 44/48 31/39 4/4 1/1 512/538 Atom N N?? N?? Total N % of Assignment 99 40 56 90 Found vs. Expected 118/119 2/5 13/23 133/147 Table 2.2: Statistics of side-chain assignments of free GIP. 62 Figure 2.13: Aliphatic region of the 1H, 13C-HSQC spectrum of free GIP. Inset A contains methyl groups. This inset is blown up in Figure 2.13A. 63 Figure 2.13A: Part of the aliphatic region of the 1H, 13C-HSQC spectrum with assignments. 64 Figure 2.14: Aromatic region of the 1H, 13C-HSQC spectrum with assignments. 65 2.3.1.4 NOE assignments of free GIP NOE assignment is the last and most crucial stage of the resonance assignment step. The sole purpose of all the previous resonance assignments (both backbone and side-chain) was to utilize those resonances in this step so that appropriate and unambiguous NOEs between different protons could be determined. For the structure calculation, NOE constraints are the most important distance constraints. Both 15N-edited HSQC-NOESY (Figure 2.15) and 13C- edited HSQC-NOESY (Figure 2.16) were used to find out NOEs between the protons. Though it sounds quite straight-forward for NOE assignment having most of the protons assigned, this stage is quite challenging due to the fact of the overlap between peaks or absence of expected peaks. That is why proper care was taken during NOE assignment to maintain a balance between not picking up a useful NOE and assigning NOEs of data heights which are not actually representative of those NOEs. To find the global fold of the protein, enough long-range NOEs are needed for the calculation. But, unfortunately, long-range NOEs are usually weak and can easily be shadowed by the intra-residue, short- and medium-range NOEs. Due to these factors, NOE assignment requires an iterative process of manual assignment and correction with the concomitant run and check of structure calculation. Initially, using Sparky a total of 4303 NOE cross peaks were assigned manually. These assignments were then either corrected or confirmed by the process of structure calculation using both CYANA1.0.6 and CYANA 2.1. A total of 1134 of those NOE cross peaks were removed and the rest of the peaks were used in the final structure calculation. Among the finally used cross peaks, a total of 1824 were either sequential, medium or long-range and the rest were just intra-residue NOEs (Table 2.3) (19). 66 Figure 2.15: 15N-edited HSQC-NOESY spectrum of I33 residue. The assignments shown here were manually picked in Sparky which were later confirmed, removed or corrected in the iterative process. 67 Figure 2.16: 13C-edited HSQC-NOESY spectrum of I33 QD1 proton. The assignments shown here were manually picked in Sparky which were later confirmed, removed or corrected in the iterative process. 68 2.3.1.5 Structure calculation of free GIP To calculate the structure of free GIP, both the CANDID module of CYANA 1.0.6 and NOEASSIGN module of CYANA 2.1 were used. The reason for using both of the versions of CYANA is that somehow only the older version of CYANA allowed the use of both defined upper limits of distance constraints (as *.upl files) and undefined (as *.peaks files) distance constraints. The newer version would only allow the usage of *.upl files. The advantage of using both types of files in this case of structure calculation is that it helps to initially determine a structure based on the given upper limits of the distance constraints (*.upl files), then the other undefined peaks can be defined with respect to this initial structure through an iterative process (Figure 2.18). In CYANA 1.0.6, besides *.peaks files dihedrals and hydrogen bonds are also given as *.aco and *.upl files. During each of the CYANA run, the output files were checked for the possible indication for the improvement in the next run, for example, by examining listed violations in *.ovw files or suggested peak assignments by the program itself in *.ass files. After several runs of checking and correcting, an enriched *.upl file was constructed which can then be used in the NOEASSIGN module of CYANA 2.1. In CYANA 2.1, the latest refined *.upl file was used as an input along with the dihedrals and hydrogen bonds. After each run, the violations and energy functions were checked from *.ovw files along with the close examination of output *.pdb files (Figure 2.20). Suspicious upper limits of distance constraints from the *.upl file were removed to achieve lesser violations (distance and angle) and lower energy functions. An initial structure was also used as an input at a later stage of the CYANA 2.1 run. The addition of an initial structure as an input file in the run 69 helped lowering root mean square deviation (RMSD) of the output structures. Final water refinement was done to get 100 lowest energy structures from 200 calculated structures. Of these, 20 structures of lowest potential energy and best Ramachandran statistics found from PROCHECK were used for analysis. Their structural statistics were summarized in the Table 2.3 (19). The ensemble of these 20 structures is shown in Figure 2.21 (19). 70 Figure 2.17: Input and output files for CYANA 1.0.6. * denotes possible preceding or following letters. A structure can also be used as an input file as *.cor/*.pdb file. In all cases, calculation parameter file (*.cya) needs to be changed accordingly. Figure 2.18: Iterative cycle of CYANA 1.0.6 run. Input files; NOE constraints (.peaks) Dihedrals (.aco) Hydrogen bonds (.upl) Sequence file (.seq) Calculation parameter (.cya) Output files; *cycle7.peaks *cycle7-ref.peaks *cycle7.upl Cycle7.pdb Cycle7.ovw *cycle*.ass Several runs Checking the *.ovw and *.ass files with the Sparky Eliminating suspicious peaks Procuring an enriched .upl file from the several runs 71 Figure 2.19: Input and output files for CYANA 2.1. A structure can also be used as an input file as *.cor/*.pdb file. In all cases, calculation parameter file (*.cya) needs to be changed accordingly. Figure 2.20: Iterative cycle of CYANA 2.1 run. Run with .upl files (NOE, H-bond and dihedral) Checking the *.ovw files along with the output *.pdb structure file Eliminating suspicious *.upl files Output structure: ? Lower RMSD ? Lower energy function ? Lower distance and angle violations Input files; NOE constraints (.upl) Dihedrals (.aco) Hydrogen bonds (.upl) Sequence file (.seq) Calculation parameter (.cya) Output files; *.pdb *.ovw 72 2.3.1.6 Refinement of structures by ARIA It has been reported that NMR structures can be significantly improved by using the refinement protocol in explicit solvent (53-55). ARIA is computer software that allows the refinement of NMR structures using a Water refinement protocol. ARIA utilizes slightly modified OPLS (Optimized Potentials for Liquid Simulations) force field to involve Lennard-Jones van der Waals and electrostatic interactions during the water refinement. For the water refinement, the structures are immersed in a 7.0 ? shell of water molecules while keeping the distance between a heavy atom of the protein and oxygen atom of water at 4.0 ? (56). Using the same distance and angle restraints files, as those used for the final structural calculation by CYANA, ARIA 1.2 employs seven cycles of simulated annealing (SA) protocol for structure calculation followed by a final cycle of water refinement protocol. SA protocol is composed of four phases (55): i. 1100 steps of torsion angle simulated annealing at 10,000 K ii. 550 steps of first torsion angle dynamics cooling phase from 10,000 K to 2000 K iii. 5000 steps of second Cartesian dynamics cooling phase from 2000 K to 1000 K iv. 2000 steps of third Cartesian dynamics cooling phase from 1000 K to 0 K In the simulated annealing protocol, the 200 best structures are calculated and arranged according to their total energy. Among these 200 structures, only the 100 best structures are then used in the final cycle of water refinement. 73 2.3.1.7 NMR structure of free GIP A PDZ domain is usually composed of six ?-strands (?1-?6) and two ?-helices (?1 and ?2) (16). Being composed solely of a single PDZ domain, the NMR structure of GIP resembles the characteristic PDZ domain with six ?-strands (?1-?6) and two ?-helices (?1 and ?2) (Figure 2.21). However, GIP contains two additional ? strands (?a and ?b), between ?1 and ?2, anti- parallel to each other connected by a turn (19). The C- and N-termini of the protein are very disordered signifying their free movement in the solution. Another region of the protein, quite unstructured and flexible, is the loop region between the ?2 and the ?3 strand (?2-??3 loop). Apart from these regions, the free GIP protein appears quite structured in the core and illustrated by the convergence of those parts of the structures in the ensemble of 20 superimposed lowest energy structures (Figure 2.21). 74 Figure 2.21: Ribbon diagrams of the ensemble of the 20 superimposed lowest energy structures of free GIP. Adapted from reference (19). ?1 ?a ?b ?1 ?2 ?2 ?3 ?? ?6 75 Assignments Free GIP Sequential |i-j|=1 871 Medium 2?|i-j|?4 331 Long |i-j|>4 622 Intermolecular 0 Hydrogen Bonds a 64 Dihedral Constraints b 118 Ensemble Average c Total energy -3625 ? 125 NOE energy 1131 ? 189 VDW energy -937 ? 75 Bonds energy 85 ? 5 Dihedral energy 657 ? 10 Angle energy 318 ? 22 Improper energy 963 ? 78 Electrostatic energy -4712 ? 67 Ramachandran Plot d Favorable 68.6 Additionally Allowed 26.6 Generously Allowed 3.4 Disallowed 1.5 RMSD (?) e Well-ordered Backbone 0.45 Well-ordered Sidechain 0.92 Table 2.3: NMR structural statistics for the 20 selected lowest energy structures of free GIP. Adapted from reference (19). 76 a Hydrogen bonds were defined by a set of two distance restraints per bond for residues of predicted secondary structure based on TALOS (48) predictions from CSI. b Dihedral constraints were derived from TALOS (48) predictions from CSI. c Energy terms were calculated by the water refinement module of ARIA 1.2 (49). d Ramachandran plot statistics were calculated by PROCHECK (50). e Well ordered regions included residues 11-19, 29-36 and 54-112. 77 2.3.1.8 Accession codes The accession codes for free GIP in the BioMagnetic Resonance Bank (BMRB) and the Protein Data Bank (PDB) are 17254 and 2L4S, respectively. In BMRB, the chemical shifts of the resonances and, in PDB, the atomic coordinates for free GIP have been deposited (19). 2.3.2 Dynamics of free GIP from 15N relaxation measurements Using the Lipari-Szabo formalism based model-free analysis (57), the order parameters (S2) for free GIP were calculated with the data collected by another member in our research group, using steady-state 1H-15N NOE intensities, R1 and R2 relaxation rates. Those residues that could not be analyzed as a result of low intensity or absence from the HSQC spectra due to the overlapping were excluded from the data analysis. Excluded residues include M1, P5, P8, V12, K20, L21, L29, G30, P41, P45, K50, D52, V57, R59, P65, I68, A69, I73, D75, V80, M87, K95, V105 and V118. Aside the N-terminus and five proline residues, S2 values for other residues could not be measured mainly for two reasons: spectral overlap and line broadening. In total, 100 of 118 residues (excluding the N-terminus and 5 prolines) were analyzed to determine the S2 values. It is important to remember here that, the higher the S2 value, the lesser mobile it is. Well-defined secondary structure of the protein should be more ordered and less mobile. Analysis of the dynamics data reveals the same pattern of mobility for free GIP protein. The defined secondary structure of free GIP showed relatively restricted mobility of 0.85 or above (Figure 2.22), whereas, C-and N-termini of the protein and various loops including the ?a-?b hairpin, the ?2-?3 loop and a few other short loops between secondary structural elements exhibited greater flexibility (Figure 2.22 & Figure 2.23). When the RMSD values for individual 78 residues obtained from structural calculation were plotted on the same graph containing information on S2 values, a correlation was found between the order parameters and the overall RMSD values (Figure 2.22). Higher RMSD values corresponded to lower S2 values. An average high S2 value of 0.89 for the core region (A11-Q112) of free GIP was calculated from the modelfree analysis. This high value corresponds to the restricted backbone mobility of a well folded protein. However, as we go toward the termini of the protein this value drops low very suddenly (Figure 2.22) (19). 79 Figure 2.22: The S2 values derived using the modelfree analysis from the steady state 1H-15N NOE, R1 and R2 relaxation times of free GIP for each non-overlapping well defined residue. Residues with order parameters above the threshold 0.85 were colored in blue while those below were colored in red. The backbone RMSD of free GIP for each residue was overlaid on this plot in black. Adapted from reference (19). 80 Figure 2.23: Residues with S2 values below the threshold of 0.85 are mapped in red onto the structure of free GIP colored blue. Adapted from reference (19). 81 2.4 Conclusion We solved the solution structure of free GIP using NMR and determined the dynamics of free GIP protein. The global structure of GIP is consistent with that of the canonical PDZ domain although there are small differences. The dynamics corresponds coherently to the structure of GIP. The more structured the region of the protein is, the lower is its mobility and randomness. This structural and dynamics study of free GIP would allow us to compare and contrast with those of bound GIP that forms complex with a substrate (Chapter 3). Such a comparative study should shed light on the mechanism of interaction between GIP and its binding partner. 82 2.5 References 1. Pellegrini, M., Haynor, D., and Johnson, J. M. (2004) Protein interaction networks, Expert Rev Proteomics 1, 239-249. 2. Neel, B. G. (1993) Structure and function of SH2-domain containing tyrosine phosphatases, Semin Cell Biol 4, 419-432. 3. Kardinal, C., Posern, G., Zheng, J., Knudsen, B. S., Moarefi, I., and Feller, S. M. (1999) Rational development of cell-penetrating high affinity SH3 domain binding peptides that selectively disrupt the signal transduction of Crk family adapters. Amgen Peptide Technology Group, Ann N Y Acad Sci 886, 289-292. 4. Falke, J. J. (2007) Membrane Recruitment as a Cancer Mechanism: A Case Study of Akt PH Domain, Cellscience 4, 25-30. 5. Fanning, A. S., and Anderson, J. M. (1996) Protein-protein interactions: PDZ domain networks, Curr Biol 6, 1385-1388. 6. Woods, D. F., and Bryant, P. J. (1993) ZO-1, DlgA and PSD-95/SAP90: homologous proteins in tight, septate and synaptic cell junctions, Mechanisms of development 44, 85- 89. 7. Garcia-Mata, R., and Burridge, K. (2007) Catching a GEF by its tail, Trends Cell Biol 17, 36-43. 8. Fanning, A. S., and Anderson, J. M. (1996) Protein-protein interactions: PDZ domain networks, Current Biology 6, 1385-1388. 9. Jelen, F., Oleksy, A., Smietana, K., and Otlewski, J. (2003) PDZ domains - common players in the cell signaling, Acta Biochim Pol 50, 985-1017. 83 10. Spaller, M. R. (2006) Act globally, think locally: systems biology addresses the PDZ domain, ACS Chem Biol 1, 207-210. 11. Lee, H. J., and Zheng, J. J. (2010) PDZ domains and their binding partners: structure, specificity, and modification, Cell communication and signaling : CCS 8, 8. 12. Zhang, Q., Fan, J. S., and Zhang, M. (2001) Interdomain chaperoning between PSD-95, Dlg, and Zo-1 (PDZ) domains of glutamate receptor-interacting proteins, J Biol Chem 276, 43216-43220. 13. Christopherson, K. S., Hillier, B. J., Lim, W. A., and Bredt, D. S. (1999) PSD-95 assembles a ternary complex with the N-methyl-D-aspartic acid receptor and a bivalent neuronal NO synthase PDZ domain, J Biol Chem 274, 27467-27473. 14. Lemaire, J. F., and McPherson, P. S. (2006) Binding of Vac14 to neuronal nitric oxide synthase: Characterisation of a new internal PDZ-recognition motif, FEBS Lett 580, 6948-6954. 15. Harris, B. Z., and Lim, W. A. (2001) Mechanism and role of PDZ domains in signaling complex assembly, Journal of cell science 114, 3219-3231. 16. Schultz, J., Hoffmuller, U., Krause, G., Ashurst, J., Macias, M. J., Schmieder, P., Schneider-Mergener, J., and Oschkinat, H. (1998) Specific interactions between the syntrophin PDZ domain and voltage-gated sodium channels, Nat Struct Biol 5, 19-24. 17. Wiedemann, U., Boisguerin, P., Leben, R., Leitner, D., Krause, G., Moelling, K., Volkmer-Engert, R., and Oschkinat, H. (2004) Quantification of PDZ domain specificity, prediction of ligand affinity and rational design of super-binding peptides, J Mol Biol 343, 703-718. 84 18. Tonikian, R., Zhang, Y., Sazinsky, S. L., Currell, B., Yeh, J. H., Reva, B., Held, H. A., Appleton, B. A., Evangelista, M., Wu, Y., Xin, X., Chan, A. C., Seshagiri, S., Lasky, L. A., Sander, C., Boone, C., Bader, G. D., and Sidhu, S. S. (2008) A specificity map for the PDZ domain family, PLoS Biol 6, e239. 19. Zoetewey, D. L., Ovee, M., Banerjee, M., Bhaskaran, R., and Mohanty, S. (2011) Promiscuous Binding at the Crossroads of Numerous Cancer Pathways: Insight from the Binding of Glutaminase Interacting Protein with Glutaminase L, Biochemistry-Us 50, 3528-3539. 20. Olalla, L., Aledo, J. C., Bannenberg, G., and Marquez, J. (2001) The C-terminus of human glutaminase L mediates association with PDZ domain-containing proteins, FEBS Lett 488, 116-122. 21. Rousset, R., Fabre, S., Desbois, C., Bantignies, F., and Jalinot, P. (1998) The C-terminus of the HTLV-1 Tax oncoprotein mediates interaction with the PDZ domain of cellular proteins, Oncogene 16, 643-654. 22. Kanamori, M., Sandy, P., Marzinotto, S., Benetti, R., Kai, C., Hayashizaki, Y., Schneider, C., and Suzuki, H. (2003) The PDZ protein tax-interacting protein-1 inhibits beta-catenin transcriptional activity and growth of colorectal cancer cells, Journal of Biological Chemistry 278, 38758-38764. 23. Zhang, J., Yan, X., Shi, C., Yang, X., Guo, Y., Tian, C., Long, J., and Shen, Y. (2008) Structural basis of beta-catenin recognition by Tax-interacting protein-1, J Mol Biol 384, 255-263. 85 24. Banerjee, M., Huang, C., Marquez, J., and Mohanty, S. (2008) Probing the structure and function of human glutaminase-interacting protein: a possible target for drug design, Biochemistry-Us 47, 9208-9219. 25. Saras, J., Engstrom, U., Gonez, L. J., and Heldin, C. H. (1997) Characterization of the interactions between PDZ domains of the protein-tyrosine phosphatase PTPL1 and the carboxyl-terminal tail of Fas, J Biol Chem 272, 20979-20981. 26. Hampson, L., Li, C., Oliver, A. W., Kitchener, H. C., and Hampson, I. N. (2004) The PDZ protein Tip-1 is a gain of function target of the HPV16 E6 oncoprotein, Int J Oncol 25, 1249-1256. 27. Reynaud, C., Fabre, S., and Jalinot, P. (2000) The PDZ protein TIP-1 interacts with the Rho effector rhotekin and is involved in Rho signaling to the serum response element, J Biol Chem 275, 33962-33968. 28. Le Maout, S., Welling, P. A., Brejon, M., Olsen, O., and Merot, J. (2001) Basolateral membrane expression of a K+ channel, Kir 2.3, is directed by a cytoplasmic COOH- terminal domain, Proc Natl Acad Sci U S A 98, 10475-10480. 29. Alewine, C., Olsen, O., Wade, J. B., and Welling, P. A. (2006) TIP-1 has PDZ scaffold antagonist activity, Mol Biol Cell 17, 4200-4211. 30. Perez-Gomez, C., Campos-Sandoval, J. A., Alonso, F. J., Segura, J. A., Manzanares, E., Ruiz-Sanchez, P., Gonzalez, M. E., Marquez, J., and Mates, J. M. (2005) Co-expression of glutaminase K and L isoenzymes in human tumour cells, Biochem J 386, 535-542. 31. Gao, P., Tchernyshyov, I., Chang, T. C., Lee, Y. S., Kita, K., Ochi, T., Zeller, K. I., De Marzo, A. M., Van Eyk, J. E., Mendell, J. T., and Dang, C. V. (2009) c-Myc suppression 86 of miR-23a/b enhances mitochondrial glutaminase expression and glutamine metabolism, Nature 458, 762-765. 32. Gallagher, F. A., Kettunen, M. I., Day, S. E., Lerche, M., and Brindle, K. M. (2008) 13C MR spectroscopy measurements of glutaminase activity in human hepatocellular carcinoma cells using hyperpolarized 13C-labeled glutamine, Magn Reson Med 60, 253- 257. 33. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes, J Biomol NMR 6, 277-293. 34. Goddard, T. D., and Kneller, D. G. SPARKY 3, University of California, San Francisco. 35. Kay, L., Keifer, P., and Saarinen, T. (1992) Pure absorption gradient enhanced heteronuclear single quantum correlation spectroscopy with improved sensitivity, Journal of the American Chemical Society 114, 10663-10665. 36. Muhandiram, D. R., and Kay, L. E. (1994) Gradient-Enhanced Triple-Resonance Three- Dimensional NMR Experiments with Improved Sensitivity, Journal of Magnetic Resonance, Series B 103, 203-216. 37. Grzesiek, S., and Bax, A. (1992) Correlating backbone amide and side chain resonances in larger proteins by multiple relayed triple resonance NMR, Journal of the American Chemical Society 114, 6291-6293. 38. Norwood, T. J., Boyd, J., Heritage, J. E., Soffe, N., and Campbell, I. D. (1990) Comparison of techniques for 1H-detected heteronuclear 1H---15N Spectroscopy, Journal of Magnetic Resonance (1969) 87, 488-501. 87 39. Palmer, A. G., Cavanagh, J., Wright, P. E., and Rance, M. (1991) Sensitivity improvement in proton-detected two-dimensional heteronuclear correlation NMR spectroscopy, Journal of Magnetic Resonance (1969) 93, 151-170. 40. Vuister, G. W., and Bax, A. (1993) Quantitative J correlation: a new approach for measuring homonuclear three-bond J(HNH.alpha.) coupling constants in 15N-enriched proteins, Journal of the American Chemical Society 115, 7772-7777. 41. Clubb, R. T., Thanabal, V., and Wagner, G. (1992) A constant-time three-dimensional triple-resonance pulse scheme to correlate intraresidue 1HN, 15N, and 13C' chemical shifts in 15N---13C-labelled proteins, Journal of Magnetic Resonance (1969) 97, 213- 217. 42. Zhang, O., Kay, L. E., Olivier, J. P., and Forman-Kay, J. D. (1994) Backbone 1H and 15N resonance assignments of the N-terminal SH3 domain of drk in folded and unfolded states using enhanced-sensitivity pulsed field gradient NMR techniques, J Biomol NMR 4, 845-858. 43. Mandel, A. M., Akke, M., and Palmer, A. G., 3rd. (1995) Backbone dynamics of Escherichia coli ribonuclease HI: correlations with structure and function in an active enzyme, J Mol Biol 246, 144-163. 44. Palmer, A. G., Rance, M., and Wright, P. E. (1991) Intramolecular motions of a zinc finger DNA-binding domain from Xfin characterized by proton-detected natural abundance carbon-13 heteronuclear NMR spectroscopy, Journal of the American Chemical Society 113, 4371-4380. 88 45. Dosset, P., Hus, J.-C., Blackledge, M., and Marion, D. (2000) Efficient analysis of macromolecular rotational diffusion from heteronuclear relaxation data, Journal of Biomolecular NMR 16, 23-28. 46. Tsan, P., Hus, J.-C., Caffrey, M., Marion, D., and Blackledge, M. (2000) Rotational Diffusion Anisotropy and Local Backbone Dynamics of Carbon Monoxide-Bound Rhodobacter capsulatus Cytochrome c?, Journal of the American Chemical Society 122, 5603-5612. 47. Guntert, P. (2004) Automated NMR structure calculation with CYANA, Methods Mol Biol 278, 353-378. 48. Cornilescu, G., Delaglio, F., and Bax, A. (1999) Protein backbone angle restraints from searching a database for chemical shift and sequence homology, J Biomol NMR 13, 289- 302. 49. Linge, J. P., Habeck, M., Rieping, W., and Nilges, M. (2003) ARIA: automated NOE assignment and NMR structure calculation, Bioinformatics 19, 315-316. 50. Laskowski, R. A., Rullmannn, J. A., MacArthur, M. W., Kaptein, R., and Thornton, J. M. (1996) AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR, J Biomol NMR 8, 477-486. 51. Humphrey, W., Dalke, A., and Schulten, K. (1996) VMD: visual molecular dynamics, J Mol Graph 14, 33-38, 27-38. 52. Schrodinger, LLC. (2010) The PyMOL Molecular Graphics System, Version 1.3r1. 53. Linge, J. P., and Nilges, M. (1999) Influence of non-bonded parameters on the quality of NMR structures: a new force field for NMR structure calculation, J Biomol NMR 13, 51- 59. 89 54. Spronk, C. A., Linge, J. P., Hilbers, C. W., and Vuister, G. W. (2002) Improving the quality of protein structures derived by NMR spectroscopy, J Biomol NMR 22, 281-289. 55. Xia, B., Tsui, V., Case, D. A., Dyson, H. J., and Wright, P. E. (2002) Comparison of protein solution structures refined by molecular dynamics simulation in vacuum, with a generalized Born model, and with explicit water, J Biomol NMR 22, 317-331. 56. Linge, J. P., Williams, M. A., Spronk, C. A., Bonvin, A. M., and Nilges, M. (2003) Refinement of protein structures in explicit solvent, Proteins 50, 496-506. 57. Lipari, G., and Szabo, A. (1980) Effect of librational motion on fluorescence depolarization and nuclear magnetic resonance relaxation in macromolecules and membranes, Biophysical Journal 30, 489-506. 90 Chapter 3 Study of the mechanism of interaction between GIP and the Glutaminase L peptide 3.1 Introduction GIP has been shown to be important as a scaffolding protein in the mammalian brain by demonstrating its association with Glutaminase L in astrocytes and neurons (1). Activated Glutaminase catalyzes the production of glutamate and ammonia from the substrate glutamine, which is an important energy generation reaction in mammalian tissues (2). Various other functions of Glutaminase have been reported including involvement in synaptic transmission, hepatic ureagenesis, renal ammoniagenesis and regulation of cerebral concentrations of glutamine and glutamate (3, 4). Two different gene loci in two different chromosomes encode two different isoforms of the enzyme. They are kidney-type (K) isozyme (encoded by a gene located in chromosome 2) and liver-type (L) isozyme (encoded by a gene located in chromosome 12) (5). Localization of these two isozymes has been demonstrated by immunostaining (6). For Glutaminase L, the compartment is neuronal nuclei and Glutaminase K has been found in mitochondria. This suggests that GIP plays a role in the determination of the subcellular distribution of Glutaminase L and, also, in possible interactions with other nuclear proteins (6). The presence of the class I binding motif (ESMV-COOH) at the C-terminal end of Glutaminase L, but not in Glutaminase K, allows these two isozymes to be differentially regulated and spatially localized, even when they are present in the same tissue (5). Glutamine catabolism is a key pathway in the energy generation processes of both tumor cells and normally dividing cells 91 (7-9). Several PDZ domain-containing proteins such as alpha-1-syntrophin (SNT) and GIP has been reported to interact with the C-terminus of Glutaminase L (10). 3.2 Objective of the study To understand the mechanism by which Glutaminase L interacts with GIP, it is important to determine the structure of GIP in complex with Glutaminase L. PDZ domains interact with the C-terminus of the interacting partner, and it has been reported that peptides representing the C- terminal end of the binding partner can act as surrogates for the corresponding partner proteins in vitro (11). Thus, the study of the binding of GIP was carried out with a peptide mimic of the C- terminus of Glutaminase L that would essentially reflect the real binding between GIP and Glutaminase L. In this chapter, we determined the first solution NMR structure of GIP bound to a C-terminal peptide used as a surrogate for Glutaminase L. The C-terminal Glutaminase L peptide, hereinafter referred to as the Glutaminase L peptide has the KENLESMV sequence. Also, to understand how the addition of Glutaminase L peptide affects the dynamics of the protein, the dynamics of the GIP-Glutaminase L peptide complex has been investigated and compared with that of the free GIP. Important insights into the binding mechanism have been gained by demonstration of perturbation of both NMR chemical shifts and backbone dynamics within GIP through ligand binding. Comparison of the structural analysis between the free and bound states of GIP enables to learn the mechanism of interaction between GIP and Glutaminase L peptide. With this information, it is possible to design a small molecule inhibitor for GIP as a potential drug candidate for the treatment of cancer. In addition, because of its promiscuity for 92 having many binding partners, such an inhibitor could prove to be effective against a number of class I PDZ domains with the possibility of treatment of other diseases (10). 3.3 Materials and Methods The research work described here was carried out in the laboratory of Dr. Smita Mohanty. 3.3.1 Cloning, over-expression and purification of 15N, 13C-labeled GIP Following the method developed previously in Dr. Smita Mohanty?s laboratory (11), transformation, over-expression and purification of 15N, 13C-labeled GIP described below was carried out. 3.3.1.1 Transformation of E. coli BL21DE3pLysS cells with the recombinant plasmid pET- 3c/GIP SOC (Super Optimal broth with Catabolite repression) medium and LB (Lysogeny Broth) agar medium was incubated at 37 0C. Both the competent cells (E. coli BL21DE3pLysS) and the plasmid (pET-3c/GIP) were thawed on ice ~30 min. 1 ?L of plasmid was added to the cells and mixed gently with the pipette tip. The competent cells with the added plasmid were kept on ice for ~20 minutes. The cells were then heat shocked by putting them in the water bath set exactly at 42 0C for 45 seconds. To reduce the shocks to the cells, they were transferred to water mixed 93 with ice and kept in there for an additional 20 minutes. After that, 200 ?L of SOC medium was added to the cells under sterile conditions. The cells were then incubated in the 37 0C shaker for 15 min. 50 ?L of the cells containing SOC medium were spread-plated on the LB-agar plate. The plate was incubated overnight at 37 0C. After completion of ~ 16 hours, when the colonies grew visibly, the plate was sealed with parafilm and kept at 4 0C. 3.3.1.2 Preparation of overnight culture 500 mL of the M9 minimal medium was prepared having the following composition: KH2PO4 6.5 g K2HPO4 5 g Na2HPO4 (anhydrous) 4.5 g K2SO4 1.2 g 15NH4Cl 0.6 g The volume was adjusted to 500 mL and the medium was sterilized by autoclaving. To the M9 minimal medium, the following nutrients and antibiotics were added aseptically: 20% 13C glucose 10 mL 5mg/mL Thiamine 2.5 mL 1M MgSO4 1 mL Yeast extract 1 mL 0.1 M CaCl2 250 ?L 94 Trace elements 2.5 mL 100 mg/mL ampicillin 520 ?L Three flasks of 15 mL of sterile M9 minimal media were prepared. To each of the flask, one single colony of the transformed cells from LB-agar plate was added using sterile tips. All the three flasks were transferred immediately to the 37 0C shaker for overnight incubation. 3.3.1.3 Expression of the protein in batch culture After measuring the OD600 of all three overnight cultures, the culture with the maximum OD600 was used to inoculate the M9 medium to a final OD600 of ~0.1 in 500 mL medium. The flask was immediately transferred to the 37 0C shaker. The OD600 of the undiluted culture was checked intermittently for every 1.5-2 hours. Once the OD600 reached 0.4-0.5, the culture was induced by adding 500 ?l of 1M IPTG. The flask was transferred immediately to the shaker to continue incubation at 30 0C for ~15 hrs. 3.3.1.4 Cell harvest and lysis The cell culture was kept on ice for chilling for ~30 min. Centrifugation of the cell culture was done at 8000 rpm for 30 min at 4 0C. The supernatant was discarded. The cells were frozen in liquid nitrogen for 5 minutes. They were subsequently incubated on ice to thaw for ~1.5-2 hours. Once the cell mass becomes fluid, the sample was again frozen in liquid nitrogen for 5 minutes. This procedure was repeated 5 to 6 times. A lysis mixture was prepared with a crushed half-tablet of cocktail protease inhibitor, 10 mL lysis buffer (50 mM phosphate buffer at 95 pH 8, 4 mM Ethylene Diamine Tetra Aceticacid (EDTA), 200 mM NaCl and 4 % glycerol) and 150 ?L of 0.1 M phenylmethylsulfonyl fluoride (PMSF). The cells were mixed with this mixture and sonicated with a 10 second pulse for 8-10 times. The cell lysate was then centrifuged at 12,000 rpm for 25 min at 4 0C. The supernatant was carefully stored at 4 0C. 3.3.1.5 Protein purification The protein was purified from the supernatant by a single-step FPLC method of purification using a Sephacryl S-100 column (11). The buffer used for this size-exclusion chromatography was 20 mM sodium phosphate buffer at pH 6.5 containing 150 mM NaCl, 1 mM EDTA and 0.01% (w/v) sodium azide. The protein was collected from the fraction no. 41- 44. These fractions were pooled together. 3.3.1.6 NMR sample preparation The pooled fraction was concentrated down to ~1 mL, to which, then 10 mL of NMR buffer (50 mM phosphate buffer at pH 6.5 containing 5% D2O, 1 mM EDTA and 0.01% (w/v) NaN3) was added. This was again concentrated down to 1 mL and another 10 mL of NMR buffer was added. It was then concentrated down to ~ 1 ml. The OD280 of the sample was checked to determine the protein concentration and it was stored at 4 0C. Finally, 50 ?L of D2O was added to the sample to make a final concentration of 5 % D2O for NMR experiments. 3.3.2 NMR Data collection 96 All NMR data were collected on a Bruker Avance 600 MHz spectrometer with a triple resonance 1H/13C/15N TCI cryoprobe equipped with z-axis pulsed field gradients at either the Department of Chemistry and Biochemistry, Auburn University, Auburn, AL, Bruker BioSpin Corporation, Billerica, MA, or the New York Structural Biology Center, New York, NY. The data were processed using NMRPipe (12) and Sparky (13). For structure determination, samples between 500 ?M and 1 mM of uniformly 15N/13C-labeled GIP in 50 mM phosphate buffer containing 5% D2O pH 6.5, 1 mM EDTA and 0.01% (w/v) NaN3 were prepared with addition of the Glutaminase L peptide (Chi Scientific, Maynard, MA, USA) at a 1:3 protein to peptide ratio. All NMR experiments were performed at 298 K. Dynamics data were collected by Mohiuddin Ovee and David Zoetewey. To determine the 15N T1 values, NMR spectra were recorded with relaxation delays of 10, 600, 50, 500, 100, 400, 200, 300 and 10 ms. To determine 15N T2 values, NMR spectra were recorded with delays of 17, 153, 34, 17, 136, 51, 119, 68, 102, 85 and 34 ms. The relaxation times were randomized and some points repeated in order to avoid any systematic errors that may arise when the data are collected sequentially. The relaxation rates were calculated by least squares fitting of peak heights versus relaxation delay to a single exponential decay. Steady state 1H-15N NOE values were calculated from the ratio of peak heights in a pair of NMR spectra acquired with and without proton saturation. For backbone and side-chain assignments of the GIP-Glutaminase L peptide complex the following spectra were recorded at 298 K: 2D 1H,15N-HSQC (14), 3D HNCACB (15), 3D CC(CO)NH (16), 3D CBCA(CO)NH (15), 3D 15N-edited HSQC-TOCSY (17, 18) with an 80 ms mixing time, 3D HC(CO)NH (16), 3D HNHA (19), 3D HNCO (16) and 3D HN(CA)CO (20). NOE distance restraints were collected from 3D 15N-edited HSQC-NOESY (17, 18, 21) and 3D 13C-edited HSQC-NOESY (17, 18, 21) with the 13C carrier frequency in the aliphatic (44 ppm) and aromatic (125 ppm) 97 regions and mixing times of 140 for 15N and 110 ms for 13C, respectively. For complex structure determination of GIP with the Glutaminase L peptide, selectively filtered 2D NOESY(22) with a mixing time of 100 ms, 3D 15N-filtered and 3D 13C-filtered NOESY experiments, each with mixing times of 120 ms, were performed (23). The backbone and side-chain assignments of the Glutaminase L peptide were obtained with an unlabeled peptide sample (~4mM) from the following spectra: 2D 1H,15N-HSQC, 2D 1H-13C-HMQC, homonuclear 2D TOCSY (24) and ROESY (25) each with a mixing time of 60 ms (10). 3.3.3 Analysis of dynamics data Dynamics data were analyzed by Mohiuddin Ovee along with Dr. David Zoetewey in Dr. Smita Mohanty?s laboratory. Measured relaxation parameters R1, R2 and the steady-state 1H-15N NOE for each residue were used as inputs in the Modelfree 4.15 program developed by Palmer et al (26, 27) to analyze 15N-backbone dynamics. The ?c value for GIP-Glutaminase L peptide complex was calculated using the program Tensor2 for the core region A11-Q112 (28, 29). Of five different models, the best one was chosen according to the selection criteria (26) to get the order parameter (S2) that represents the degree of spatial restriction within the 1H-15N bond vector. These values range from zero for completely isotropic internal motions to unity for totally restricted motion and represent dynamics on the picosecond to nanosecond time scale (10). 3.3.4 Structure calculation and refinement 98 A total of 2866 NOE cross peaks were assigned manually using Sparky (13) for the GIP- Glutaminase L peptide complex. The assignments were corrected or confirmed with both the CANDID module of CYANA 1.0.6 and NOEASSIGN module of CYANA 2.1 (30), using the standard protocol of eight iterative cycles of NOE assignment and structure calculation. The CANDID module of CYANA 1.0.6 was used on the complex to initially fit the Glutaminase L peptide into the binding pocket of GIP because it allowed the intermolecular assignments to be fixed separately from the intramolecular assignments. To calculate the complex structure, 36 glycine residues were added as a flexible linker between the protein and the peptide. A total of 118 dihedral angles restrains were derived from the TALOS (31) program based on the chemical shift index (CSI) and primary sequence of GIP for protein-peptide complex calculations. Additionally, a total of 66 hydrogen bond distance restraints (two restraints per bond) for the protein-peptide complex were derived from the CSI by TALOS. During the iterative NOE assignments, a total of 490 assignments for the GIP-Glutaminase L peptide complex were removed due to overlap, redundancy, or unresolved ambiguity that resulted from low stringency in the initial peak picking phase and high stringency in the final assignments. The final assignments averaged over 18 and 12 NOEs per residue for protein in the complex, and for the peptide in the complex, respectively. Final refinement of the 100 lowest energy structures of the 200 total calculated structures was performed with the water refinement protocol implemented in ARIA (32). The 20 structures with the lowest potential energy and best Ramachandran statistics as assessed by PROCHECK (33) were selected for analysis. The structures were visualized with VMD and figures were created using Pymol (34, 35). Table 3.3 shows the complete structural statistics for structure of GIP in complex with the Glutaminase L peptide (10). 99 3.4 Results 3.4.1 Protein Expression As described above, 15N, 13C-labeled GIP was expressed in E. coli cells grown in minimal media supplemented with 15N-labeled ammonium chloride and 13C-labeled glucose (Figure 3.1). As seen in NMR studies later on, the isotope labeling of the protein GIP was successful since both the isotope labeled nitrogen and carbon nuclei provided good signals in NMR. However, the initial efforts to isotopically label the protein using the Lysogeny Broth (LB) medium as a growth medium for overnight cell cultures produced inhomogeneous labeling of the nucleus (carbon or nitrogen) even though the starter culture was diluted 25 times in the minimal media. Still use of such a small percentage of LB media was sufficient for the dilution of the isotope labeling; making it impossible to carry out isotope filtered experiments. A simple 1D NMR experiment (36) was carried out to check the homogeneity of the isotope labeling (Figure 3.2). Methyl (-CH3) protons of Leucine 108 of GIP appears in a 1D NMR spectrum at a value of less than zero in the ppm scale which is completely separate from any other peaks of the spin-active nuclei. Thus, observing the splitting patterns of the methyl (-CH3) protons of that residue would help to determine the homogeneity of the isotope labeling of the protein. If there is non-homogeneous isotope labeling, then there would be still spin-inactive 12C present in the protein which would cause no splitting of the (-CH3) protons resulting in a single proton peak. However, the available 13C present in such case, would still split (-CH3) protons, thus, the resulting 1D spectrum should have three peaks for the (-CH3) protons (Figure 3.2). But, if the isotope labeling is homogeneous, then there should only be two peaks for the (-CH3) protons resulting from the splitting by 13C (Figure 3.3). Protein samples prepared from the earlier 100 protocol (using LB medium) gave three peaks for the (-CH3) protons of leucine 108 of the protein. Of which, the intermediate peak is due to the contribution from the (-CH3) protons attached to 12C, whereas, two adjacent peaks on both sides of the middle peak is produced by the splitting of (-CH3) protons attached to 13C (Figure 3.2). The comparatively higher intensity of the middle peak compared to the two shoulder peaks indicated presence of a higher percentage of unlabeled spin-inactive 12C nucleus in the protein (Figure 3.2). This suggested that the GIP proteins prepared using LB medium as the growth medium for the overnight culture was not homogeneously isotope labeled. To achieve homogeneous isotope labeling of GIP protein, the protocol was changed. The growth medium used for the starter culture was changed to M9 minimal medium as well. The purified GIP protein from such expression was checked for the homogeneity of the isotope labeling and this protein sample showed almost 100% isotope labeling (Figure 3.3). For the purpose of the structure determination of GIP-Glutaminase L peptide complex, all the 13C, 15N-labeled GIP protein was produced following the latest protocol. The production of homogeneously labeled protein was a prerequisite for the successful operation of the filtered NOESY experiments. Thus, confirmation of an available homogeneously isotope labeled GIP protein was a very important step in the determination of complex structure. This also wonderfully shows a practical application of the spin-spin coupling having an impact on the research. 101 Figure 3.1: Expression of GIP analyzed by SDS-PAGE. Both lane 1 and 2 show the expression of GIP prior to purification (the red rectangle spots the protein of expected size). The lane MW is for protein marker. 250 150 100 75 50 37 25 20 15 10 MW 1 2 kDa 102 Figure 3.2: 1D NMR spectrum of non-homogeneously labeled GIP sample. p3919gp was the name for the pulse program used for this NMR experiment. Three peaks of (-CH3) protons of L108 103 Figure 3.3: 1D NMR spectrum of homogeneously labeled GIP sample. p3919gp was the name for the pulse program used for this NMR experiment. Two peaks of (-CH3) protons of L108 104 3.4.2 Protein Purification Using size-exclusion chromatography as a single step, GIP was purified on a Sephacryl S-100 column (GE Healthcare). The production of the 15N, 13C-labeled recombinant GIP is around 15.2 mg per liter of bacterial culture. 3.4.3 NMR Structure determination of GIP-Glutaminase L peptide complex 3.4.3.1 Effect of peptide binding to the resonances of GIP protein As GIP was titrated against Glutaminase L peptide, it was possible to track the movement of the resonance peaks in the 1H, 15N-HSQC spectra, because most of the resonances of the protein residues were in the fast chemical exchange on the NMR time scale. However, there were some exceptions. The amino acid residues I18, L21, I28-G35, Q39, D40, Q43, N44, E48, I55, E62, A66, E67, A69 and R96 had peak intensities that either hugely decreased or were below the level of the noise threshold, presumably due to intermediate to slow exchange on the NMR time scale. But, as GIP reaches saturation and the predominant state becomes the bound state, then these undetectable resonances reappeared often in remote regions of the HSQC spectrum relative to their initial positions. For the assignments of the residues that were assumed to be critical to complex formation, such phenomenon produced considerable uncertainty. Such residues include I28-E48 and R96, which are located within the ?2 strand, the ?2-?3 loop and the ?2 helix. This is evidence that GIP interacts with the Glutaminase L peptide primarily through the ?-strand addition mechanism (37) instead of a direct interaction with the ?2 helix. Residues that were not predicted to be part of the binding region also underwent intermediate to slow chemical 105 exchange, such as I18, I55, and E62-A69, which belong to the ?1 and ?3 strands and the ?1 helix, respectively. This observation points to the fact that, due to the binding interaction between the protein and peptide, there are some long range allosteric interactions within the protein. Like 1H, 15N-HSQC spectra, the 1H, 13C-HSQC spectra were also significantly different when compared between free GIP and the GIP- Glutaminase L peptide complex. It was quite impossible to assign a number of key protein side-chain nuclei purely based on free GIP assignments, since there were a couple of factors that created this uncertainty. Firstly, there was severe overlap of carbon and proton chemical shifts and, secondly, the protein-peptide interaction resulted in large chemical shift perturbations. Therefore, it was necessary to reassign the whole protein in its complexed state using the following 3D experiments: HNCACB, CBCA(CO)NH, HCC(CO)NH, CC(CO)NH, HSQC-TOCSY and HCCH-TOCSY. This helped to reassign even the residues, such as L27 ? G35, which initially disappeared but reappeared in distant locations with the course of the titration of GIP with Glutaminase L peptide. This re-assignment of the protein in the complex was very essential considering the amount of chemical shift perturbations for all of the resonances, both backbone and side-chain. Such significant changes in chemical shifts are nicely illustrated from Figure 3.4 to Figure 3.8. Thus, to proceed with structure calculations, each resonance must be reassigned with accuracy (10). 106 Figure 3.4: Combined 1H and 15N backbone amide chemical shift perturbations (?HN) are plotted as a function of residue number in GIP by the equation ?HN={(Hf-Hb)2+((Nf-Nb)/10)2}?, with 10 as a scaling factor. Hf, Hb, Nf and Nb are the chemical shifts of each residue?s amide 1H and 15N in the free (GIP alone) and bound (GIP-Glutaminase L peptide complex) states, respectively. Adapted from reference (10). 107 Figure 3.5: The magnitudes of ?HN presented in Figure 3.4 are represented as different colors on a ribbon diagram of free GIP. White is < 0.1 ppm, yellow is < 0.2 ppm, orange is < 0.5 ppm and red is > 0.5 ppm. Only residues A11-Q112 are shown as residues M1-T10 and A113-S124 are highly disordered and have chemical shifts perturbations of < 0.05 ppm. Adapted from reference (10). ?3 ?2 ?4 ?? ?1 ?a ?b ?1 ?? 108 Figure 3.6: Combined HA and CA backbone chemical shift perturbations (?HC) are plotted as a function of residue number in GIP by the equation ?HC={(Hf-Hb)2+((Cf-Cb)/4)2}?, with 4 as a scaling factor. Hf, Hb, Cf and Cb are the chemical shifts of each residue?s alpha 1H and 13C in the free (GIP alone) and bound (GIP-Glutaminase L peptide complex) states, respectively. Adapted from reference (10). 109 Figure 3.7: The magnitudes of ?HC presented in Figure 3.6 are represented as different colors on a ribbon diagram of free GIP. White is < 0.05 ppm, yellow is < 0.1 ppm, orange is < 0.2 ppm, red-orange is < 0.5 ppm and red is > 0.5 ppm. Only residues A11-Q112 are shown as residues M1-T10 and A113-S124 are highly disordered and have chemical shifts perturbations of < 0.05 ppm. Adapted from reference (10). 110 Figure 3.8: An overlay of free GIP is shown in red and GIP-Glutaminase L peptide at a ratio of 1:3 in blue, but at a lower contour threshold to highlight L29. Arrows indicate the dramatic chemical shift perturbations of L29 and G30. Adapted from reference (10). 1 15 111 3.4.3.2 Backbone and side-chain assignments 3.4.3.2.1 For protein With the available sequential assignments for free GIP, assigning the backbone for GIP in its bound form was not hard. A 3D HNCACB experiment was used to perform sequential assignments of the GIP in complexed state (Figure 3.9). A CBCA(CO)NH spectrum was quite useful in the confirmation of the assignments of the HNCACB spectrum. However, there were certain peaks in the 1H, 15N-HSQC spectrum (Figure 3.10) which required some efforts to identify them for the purpose of acquiring a complete sequential assignment such as L29. This peak goes into an intermediate exchange from fast exchange as the protein goes from the free to the bound state. At a higher concentration of Glutaminase L peptide (1:3 protein to peptide ratio), it re-appears barely at a high contour level in a completely different location (Figure 3.8). After assigning certain side-chain and NOESY experiments, assignments of peaks like this one were confirmed. Several experiments were used to assign side-chains of the protein in its bound form such as HCC(CO)NH, CC(CO)NH, HSQC-TOCSY and HCCH-TOCSY. The statistics of the assignments of the side-chains were summarized in the Table 3.1. In summary, around 93, 95 and 92 percent of all carbon, hydrogen and nitrogen nuclei, respectively, were unambiguously assigned. 112 Figure 3.9: Sequential assignments of V13-K20 in the GIP-Glutaminase L peptide complex from (1H, 13C)-strips of the HNCACB experiment. Only the C? atoms of the residues were connected with red lines to show the sequential assignment. Positive signals are green and negative signals are red. C? appears as positive signal and C? appears as negative signal. 113 Figure 3.10: 1H, 15N-HSQC spectrum of the GIP-Glutaminase L peptide complex. Red lines connected the non-degenerate protons of the side-chain amide groups of Asparagine and Glutamine residues. 114 Atom C (CO) C?? C?? C?? C?? C?? C?? C?? Total C % of Assignment 0 99 99 87 82 84 50 100 93.2 Found vs. Expected 0/124 123/124 111/112 79/91 37/45 16/19 2/4 1/1 369/396 Atom HN H?? H?? H?? H?? H?? H?? H?? Total H % of Assignment 98 99 99 86 98 85 100 100 95.5 Found vs. Expected 117/119 123/124 111/112 78/91 47/48 33/39 4/4 1/1 514/538 Atom N N?? N?? Total N % of Assignment 98 40 70 92 Found vs. Expected 117/119 2/5 16/23 135/147 Table 3.1: Statistics of side-chain assignments of the GIP-Glutaminase L peptide complex. 115 3.4.3.2.2 For peptide To assign the resonances for the residues of the Glutaminase L peptide in its bound form, initially, resonances of the peptide residues were assigned from its free from. Subsequently, those resonances were used as a guiding reference for the assignment of the residues of the peptide in the bound form. Assignment of free Glutaminase L peptide was done by 1H, 15N-HSQC, homonuclear 2D TOCSY, 1H, 13C-HMQC and ROESY experiments. The 1H, 15N-HSQC experiment (Figure 3.11) was used to assign the amide protons of the Glutaminase L peptide based on the usual chemical shifts for the amide protons of respective amino acids (Biological Magnetic Resonance Data Bank, BMRB, http://www.bmrb.wisc.edu/) and on the assignments from other experiments such as homonuclear 2D TOCSY (Figure 3.12), 1H, 13C-HMQC (Figure 3.13) and ROESY. Among the eight residues of the peptide, in addition to K301, two (E302 and M307) did not give any peaks in the spectrum for some unknown reasons. Amide protons of K301 and E302 were never assigned. However, amide proton of M307 was assigned from other spectra. 2D TOCSY experiment helped to assign non-degenerate protons of the side-chains of the peptide. Most of the assignments of the side-chains of the free peptide were done in this experiment. To assign the resonances of the residues of the peptide in its bound state, a special 2D selectively filtered NOESY experiment that results into four different 2D NOESY spectra (22), was used. In this experiment, NOEs that arise from protons attached to either 12C/14N (peptide) or 13C/15N (protein) can be selectively filtered. Thus, there should be one spectrum among the resulting four spectra, which would allow only NOEs that originate from protons attached to 12C 116 or 14N. Through the comparison with the resonances of the residues of the peptide in its free from, assignment of those of the peptide in its bound form, from such spectrum, was achieved (Figure 3.14). Moreover, such an assignment process also helped to determine the structure of the Glutaminase L peptide in its bound form (10). The statistics of the side-chain assignments for Glutaminase L peptide are summarized in Table 3.2. Assignment of the available protons of the peptide was solely considered here since only these protons would produce any possible NOE relationship with the protons of the protein. In summary, about 95% of all the possible protons of the peptide were assigned unambiguously. Atom HN H?? H?? H?? H?? H?? Total H % of Assignment 75 100 100 100 100 100 95.2 Found vs. Expected 6/8 8/8 12/12 7/7 5/5 2/2 40/42 Table 3.2: Statistics of available proton assignments of the Glutaminase L peptide. 117 Figure 3.11: 1H, 15N-HSQC spectrum of the Glutaminase L peptide. 118 Figure 3.12: 1H, 13C-HMQC spectrum of the Glutaminase L peptide. 119 Figure 3.13: Homonuclear 2D TOCSY spectrum of the Glutaminase L peptide. Red line crosses the diagonal peaks. Notice the duplicate peaks on either side of the red line. 120 Figure 3.14: 2D selectively filtered NOESY spectrum of the Glutaminase L peptide. 121 3.4.3.3 NOE assignments Traditional 3D 15N- and 13C-edited HSQC-NOESY experiments were used to assign the NOEs for GIP in the bound form. Interestingly, several intermolecular NOEs between GIP and Glutaminase L peptide were also assigned in these two experiments. To assign the 13C-edited HSQC-NOESY spectrum (Figure 3.15), a 2D 1H, 13C-HSQC spectrum was constructed from the 13C-edited HSQC-NOESY spectrum itself by compressing all the data from the proton z- dimension into a single plane. Although this resulted in a much overlapped spectrum (Figure 3.16), the presence of such a base spectrum was extremely helpful in the assignment of the 13C- edited HSQC-NOESY spectrum. To find intermolecular NOEs between the unlabeled peptide and the 13C, 15N-labeled protein in the complex, F1-filtered/F3-selected NOESY experiments with both 15N/14N and 13C/12C filtering methods were used. Although these filtered experiments were supposed to have only NOEs from unlabeled peptide, it appeared that the experiment was not that stringent and a lot of intramolecular NOEs ?bleed through? to add up the ambiguities. To remove ambiguities in the assignments of 15N-filtered HSQC-NOESY experiment, one approach was to do a control experiment with the same pulse sequence on a free GIP sample (Figure 3.17). Theoretically, such a spectrum should not have any NOEs. But, since there were ?bleeding through?, this spectrum was helpful to establish NOEs only from the unlabeled peptide in the filtered NOESY spectrum with the simultaneous comparison to the controlled spectrum (Figure 3.17). This way, a good number of possible intermolecular NOEs were manually assigned in both traditional 3D 15N-edited HSQC-NOESY and 15N-filtered HSQC-NOESY spectra. 122 Usually, standard 13C-filtered NOESY is the experiment that is most often used for the determination of the structure of a complex. When compared with most of the other complexes of PDZ domains, the GIP-Glutaminase L peptide complex appears to have much fewer observable NOEs in the 13C-filtered NOESY spectrum. The reason behind the lack of observable NOEs is due to line broadening resulting from intermediate to slow exchange of residues in the entire ?2 strand. Thus, only the strongest NOEs were seen which are very important in the ligand binding. Initially, the assignments of intermolecular NOEs done on the traditional 3D (unfiltered) NOESY spectrum were ambiguous. However, with the establishment of the peptide?s relative position in the binding site, those ambiguities could be sorted out. These additional unambiguous assignments were very instrumental for the final structure calculation as they added up to the total number of intermolecular NOEs (10). 123 Figure 3.15: 13C-edited HSQC-NOESY spectrum of the I33QD1 proton of GIP in its bound form. The assignments shown here were manually picked in Sparky which were later confirmed, removed or corrected in the iterative process. 124 Figure 3.16: 1H, 13C-HSQC spectrum of GIP in the bound form. Top- Full spectrum, Bottom- Part of the spectrum was blown up and shown with assignments. 125 Figure 3.17: Three different HSQC-NOESY spectra of I33 residue of the GIP. Left- Traditional 3D 15N-edited HSQC-NOESY spectrum; middle- 15N-filtered HSQC-NOESY control spectrum; right- 15N-filtered HSQC-NOESY spectrum. The assignments shown here were manually picked in Sparky which were later confirmed, removed or corrected in the iterative process. 126 3.4.3.4 Structure calculation Initially, a total of 2866 NOE cross peaks were assigned manually for the GIP- Glutaminase L peptide complex. But, as with the free GIP structure calculation, during the iterative process of GIP-Glutaminase L peptide complex structure calculation, a total of 490 assignments were removed. The selective formation of specific hydrogen bonds between the negatively charged C-terminal Val carboxyl oxygens from the Glutaminase L peptide to the amide protons of L29 and G30 from GIP could be directly identified from their very large induced chemical shift perturbations (Figure 3.8) (38). These hydrogen bonds greatly enhanced the iterative assignment process in fitting the Glutaminase L peptide into the structure of GIP. Final water refinement was done to get the 100 lowest energy structures from 200 calculated structures. Of these, 20 structures of lowest potential energy and best Ramachandran statistics found with PROCHECK were used for analysis. Their structural statistics were summarized in the Table 3.3 (10). The ensemble of these 20 structures is shown in Figure 3.18 (10). 127 Figure 3.18: Ribbon diagrams of the ensemble of the 20 superimposed lowest energy structures of complexed GIP in blue with the Glutaminase L peptide in red. Adapted from reference (10). 128 Assignments GIP-Glutaminase L complex Sequential |i-j|=1 718 Medium 2?|i-j|?4 241 Long |i-j|>4 360 Intermolecular 37 Hydrogen Bonds a 66 Dihedral Constraints b 118 Ensemble Average c Total energy -4816 ? 175 NOE energy 1586 ? 302 VDW energy -1096 ? 67 Bonds energy 170 ? 8 Dihedral energy 749 ? 13 Angle energy 434 ? 26 Improper energy 1009 ? 89 Electrostatic energy -6082 ? 123 Ramachandran Plot d Favorable 71.2 Additionally Allowed 24.3 Generously Allowed 2.7 Disallowed 1.8 RMSD (?) e Well-ordered Backbone 0.67 Well-ordered Sidechain 1.28 Table 3.3: NMR structural statistics for the 20 selected lowest energy structures of the GIP-Glutaminase L Peptide Complex. Adapted from reference (10). 129 a Hydrogen bonds were defined by a set of two distance restraints per bond for residues of predicted secondary structure based on TALOS (31) predictions from CSI. b Dihedral constraints were derived from TALOS (31) predictions from CSI. c Energy terms were calculated by the water refinement module of ARIA 1.2 (32). d Ramachandran plot statistics were calculated by PROCHECK (33). e Well ordered regions included residues 11-19, 29-36 and 54-112. 130 3.4.4 Comparison of the structure of free GIP with that of the GIP-Glutaminase L peptide complex Overall, the structures of both free GIP and the GIP-Glutaminase L peptide complex were somewhat similar, containing the same fold. However, to accommodate the additional ?-strand of the Glutaminase-L peptide, the protein underwent changes in an allosteric manner in the complex. Binding with the peptide made the ?2 helix of GIP move away from ?2 by 0.95 ? to accommodate the additional ?-strand (Figure 3.19). In both free GIP and the complex, the ?2-?3 loop was largely unstructured. However, this loop appeared to have a few NOEs with the Glutaminase L peptide in the complex. This observation is in accordance with the report that GIP interacts with the C-terminal ?-catenin peptide through its PFS loop (residues 45-47) (39). This suggests specificity in the nature of the interaction of GIP with different binding partners. Due to the closeness of the ?1 helix to the binding site, significant chemical shift perturbations were observed in that region (Figure 3.4 to 3.7). But, such changes in chemical shifts were not reflected on the three-dimensional structure of the complex (Figure 3.19). Without complete structure determination, it could be misleading to infer any direct protein-ligand interactions simply based on the chemical shift perturbation map. This fact is illustrated by our observation of significant changes in chemical shifts of the residues that are not part of the binding pocket. Also, through structural comparison of free and bound protein, it was not easy to determine the specific interactions that caused the relatively large changes in the chemical shifts for residues that are located away from the binding site (10). 131 Figure 3.19: An overlay of free GIP is shown in green with the complexed GIP protein in blue and the Glutaminase L peptide in red. Adapted from reference (10). 132 3.4.5 Binding and specificity of the Glutaminase L peptide The C-terminus of a binding partner binds in the binding pocket of the PDZ domain, in a process called ?-strand addition, as an additional antiparallel ?-sheet to the ?2 strand of the protein (37). The binding pocket is created by the groove formed between the ?2 helix and ?2 strand of the protein. Specificity of this binding interaction comes from the sequences of the C- terminus of the interacting protein. Traditionally, the last four residues of the C-terminus of the peptide/ligand are numbered as positions -3, -2, -1 and 0 starting with the C-terminal residue as P0 (10). There is a consensus GLGF loop located at the beginning of the ?2 strand of PDZ domain that forms a series of hydrogen bonds between the backbone amides of the protein and the COO- of the C-terminal peptide. In addition, a hydrophobic interaction is facilitated by this loop to allow the sequence selectivity for the C-terminal residue of the substrate peptide. A more detailed picture of the peptide bound to GIP is shown in Figure 3.20. In GIP, the canonical GLGF motif of PDZ domain is replaced by I28LGF31 motif, suggesting that while G28 is the consensus amino acid in the binding motif of PDZ domains, the mutation to Ile is tolerated perhaps due to the structural role it plays in forming the ?a-?b hairpin. Whereas, G30 of this motif could be deemed as an absolute requirement, since it is the only amino acid that can accommodate the geometry needed for the formation of hydrogen bonds from L29 and G30 of GIP to the COO- at position P0 of the C-terminal peptide. The charged carboxyl group from the C-terminal Val (P0) of the Glutaminase L peptide formed two hydrogen bonds to the backbone amide protons of L29 and G30 of GIP. The hydrophobic side-chain of Val (P0) of the peptide ligand buries itself in the hydrophobic pocket formed by L29, F31, L97 and I33 as well as T98 at 133 the periphery (Figure 3.20). As the protein binds to the peptide, the above mentioned two hydrogen bonds formed between the ligand and the protein caused unusually large chemical shift changes of up to 2.5 ppm for the amides of L29 and G30 in the 1H, 15N-HSQC spectra (Figure 3.8). When chemical shift perturbations of both HN/N and HA/CA pairs were mapped onto the structure of GIP (Figure 3.4 to Figure 3.7), we observed that the regions near the binding site, including the ?2, ?2 and the ?2-?3 loop were generally the most perturbed, however, ?1, which did not appear to be directly involved in the binding was also significantly affected. This clearly demonstrated the allosteric mode of binding for GIP with Glutaminase L peptide. The residue H90 at the beginning of ?2 (?2:1 in PDZ nomenclature) was oriented into the binding pocket and made a specific hydrogen bond with the Ser at P-2 of the peptide (Figure 3.20). This is a general feature of class I PDZ domains as the residue at position ?2:1 provides the sequence selectivity that distinguishes between different classes (40). Generally, there is no specificity at P- 1 (Table 2.1). The Glutaminase L peptide has Met at P-1, which was oriented away from the binding pocket toward the solvent. Some class I PDZ domains have specificity towards E/D or a small amino acid at P-3 (40). This interaction comes from hydrogen bonds between E at P-3 from the Glutaminase L peptide with Y56 and T58 of GIP. Alternately, a transient salt-bridge could potentially exist, but did not appear to be formed with R59 (Figure 3.20) of GIP. This particular salt-bridge has been observed in the crystal structures of GIP with ?-catenin (39) and Kir 2.3 (41). However, no NOEs were observed to support the formation of a salt bridge between E at P-3 of the Glutaminase L peptide with R59 of GIP. In contrast to the static nature of a crystal environment, the dynamic flexibility of the protein side chains in solution contributed to the above observation. It is possible that the flexibility of these side chains would allow them to come close enough to form a transient salt-bridge. However, these results demonstrated that both 134 E at P-3 and R59 were solvent-exposed, thus decreasing the strength of such an interaction in solution. Thus, the salt-bridges observed in the two crystal structures could be due to packing artifacts of crystallization, while the true nature of the salt-bridge in solution is more dynamic (10). 135 Figure 3.20: Heavy atom details from the binding site of GIP with the Glutaminase L peptide. The Glutaminase L peptide was colored in yellow and GIP in green. Potential hydrogen bonds (marked as dashed lines) could be seen from H90 with S at P-2, the COO- from V at P0 with the L29 & G30 amide nitrogens, and E at P-3 with Y56 and T58. V at P0 buries its side chain into a hydrophobic pocket created by L29, F31, I33, L97 and partially T98. Adapted from reference (10). T58 R59 S32 I 33 S -2 H90 L97 E -3 M -1 V 0 F31 G30 L29 T98 Y56 R94 I28 136 3.4.6 Dynamics of the GIP-Glutaminase L peptide complex from 15N relaxation measurements Study on the dynamics of the GIP-Glutaminase L peptide complex was carried out to elucidate the binding mechanism of the Glutaminase L peptide to GIP. Using the Lipari-Szabo formalism-based model-free analysis (42), the order parameters (S2) for GIP-Glutaminase L were calculated using steady-state 1H-15N NOE intensities, R1 and R2 relaxation rates. Those residues that could not be analyzed as a result of low intensity or absence from the HSQC spectra due to the overlapping were excluded from the data analysis. Excluded residues include M1, P5, P8, V12, V13, L21, N26, F31, G35, I37, D40, P41, Q43, P45, E48, D49, K50, D52, Y56, S61, P65, Q72, D75, V80, W83, M85, T86 and A93. Of these, L29, G30, F31, G35, D40, Q43, E48, D49 were from residues that form part of the binding pocket including the ILGF motif (canonical GLGF) and the ?2-?3 loop, and they could not be measured as a result of being too close to the intermediate exchange regime to provide sufficient intensity required for observation in the NMR dynamics data. Aside the N-terminus and five proline residues, S2 values for rest of the excluded residues could not be measured mainly for two reasons: spectral overlap and line broadening. In total, 96 of the 118 residues (excluding the N-terminus and 5 prolines) were analyzed to determine the S2 values. Additionally, ?S2 values between bound and free states were determined for 84 residues. The generalized order parameters, S2, were broadly similar for both the free and complexed states, but exhibited certain differences as explained below. The core region (A11-Q112) of the GIP-Glutaminase L peptide complex had an average S2 value of 0.87 (0.89 for free GIP) as calculated based on the model-free analysis. Although, in general, the core of the protein maintained its structure and flexibility upon binding to the Glutaminase L peptide, however, specific residues exhibited either an increase or decrease in flexibility. Among the 137 residues for which ?S2 could be calculated, G36, G54, A66 and T98 showed a substantial (?S2 > 0.06) decrease in flexibility. Furthermore, residues I4, T51, G74, R96 and K99 showed smaller but still significant increases in S2 (0.03 < ?S2 < 0.06) where the average variance in ?S2 was ?0.015 for all measured residues. Twelve other residues showed positive, but statistically insignificant increases in S2. Likewise twenty-four residues showed statistically insignificant decreases in S2 upon binding. However, residues Q14, H19, I28, D38, N44, F46, T58, G63, G70, D91 and V109 showed a small but statistically significant (-0.03 > ?S2 > -0.06) increase in flexibility. Additionally, residues R15, I18, G24, E25, L27, G34, K76, I77, H90, Q92, E103, R106, L107, R111 and many of the measured residues in the unstructured termini (M1-T10, S113-S124) showed a substantial increase in flexibility (?S2 < -0.06) as shown in Figure 3.21. When these residues were mapped onto the structure of free GIP (Figure 3.22), the biggest decreases in flexibility were displayed by the residues at the C-terminal end of the ?2 helix near the binding site and at the hinge points of the ?2-?3 loop. However, residues, located either on the ?4 and ?6 strands that are distal to the binding site or in the flexible loops such as the ?a-?b hairpin and the ?2-?3 loop, showed the biggest increases in backbone flexibility (10). 138 Figure 3.21: A plot of ?S2 as a function of residue number where ?S2 refers to S2 of the GIP- Glutaminase L peptide complex minus that of free GIP. Positive values are indicated with increasing blue intensity while negative values are indicated with increasing red intensity Adapted from reference (10). 139 Figure 3.22: The magnitude of ?S2 upon binding to the Glutaminase L peptide was mapped onto the structure of free GIP and was indicated by darker intensity for red (increased flexibility) or blue (decreased flexibility). Residues were colored white for one of the following reasons: they could not be measured in both structures due to overlap, they had ?S2 values between the threshold values 0.06 and -0.06, or the residue was a proline. Adapted from reference (10). -0.25 0.15 140 3.4.7 Intermediate chemical exchange within GIP due to the binding of the Glutaminase L peptide Most of the residues of GIP were in fast exchange regime while being titrated with the Glutaminase L peptide. But, the residues that are located within the binding pocket appeared to be in intermediate exchange. Residues L27, I28, L29, G30, F31, S32, I33, G34, and G35 had disappeared or greatly diminished in intensity due to intermediate to slow exchange at a low protein to peptide ratio. But, as the protein approached the saturation point, these residues reappeared in new locations in the 1H, 15N-HSQC spectrum at a higher concentration of Glutaminase L peptide. In addition, residues L29 and G30 had lower intensity in the 2D 1H, 15N- HSQC spectrum compared to all other residues of the protein due to line broadening caused by intermediate exchange, both in the free and complexed states of GIP. Based on the dynamics and chemical shift perturbations studies, we observed that, residues lining the binding pocket showed significant chemical shift perturbations along with substantial changes in the measurable order parameters. It appeared that both ends of the binding pocket experienced opposite effects in S2 values. One end of the binding pocket that is near the C-terminus of the peptide is composed of the ILGF loop and the other C-terminal half of the ?2 helix (K95-R100). The residues L29, G30 and F31 of the IGLF loop were in intermediate exchange, which precluded the measurement of ?S2. The residues R96, T98 and K99 from the ?2 helix) experienced a decreased flexibility upon binding the C-terminal end of the peptide. On the other hand, at the opposite end of the binding pocket, residues from both ?2 and ?2 (G34, H90 and Q92) experienced an increase in flexibility. This observation was consistent with the relatively high RMSD for the N-terminal end of the Glutaminase L peptide. This increased or decreased flexibility in the binding pocket of the 141 protein and that of the peptide suggests that the substrate specificity is limited to the C-terminal four residues of Glutaminase L (10). 3.5 Discussion 3.5.1 Specificity in the binding interaction between GIP and Glutaminase L peptide A number of interactions between GIP and the Glutaminase L peptide i.e. E-S/T-X-I/L/V- COOH (Table 2.1) provides specificity for the recognition process. The amide protons of residues L29 and G30 in the ILGF loop are uniquely positioned in such a way that allows them to form a pair of hydrogen bonds to both carboxyl oxygen atoms of V at P0 from the Glutaminase L peptide (Figure 3.20). Very large chemical shift perturbations observed for these two residues reflect on the nature of these interactions (Figure 3.4 & 3.8). The proximity of the negatively charged carboxyl oxygens at P0 position of the C-terminus of the Glutaminase L peptide to L29 and G30 of the protein caused dramatically different chemical environment at the binding site with large chemical shift perturbations although the protein structure is not significantly affected globally. When compared to the dynamics of free GIP, the effect of peptide binding on the dynamics of the protein appears to be dramatic. The disappearance of residues L27-G35 during the course of the titration due to the intermediate to slow exchange and their reappearance at saturation of the binding site, illustrates the dramatic effect of ligand binding on protein dynamics. 142 The specificity for a hydrophobic residue at P0 of the ligand comes from the hydrophobic pocket created by L29, F31, I33 and L97 of the protein. Val seems to be preferred at P0 more than Leu or Ile possibly due to the steric hindrance on this hydrophobic interaction with the longer side chain of these amino acids. This phenomenon was also observed and discussed in chapter 4 of this dissertation, where binding affinity of the interactions of two ligands (RDGDFQTEV-COOH and RGGSRL-COOH) with GIP was compared. The observance of a high affinity and stronger interaction (almost 7 times) for the ligand with V at its P0 position than the one with L at P0 position, is very likely due to the steric hindrance caused by the long side chain of L. The steric nature of these hydrophobic interactions could be confirmed through point mutation of one or more of the following residues in the binding pocket of GIP: L29V, L97V or T98A. Residue L97, located at position ?2:8, is highly conserved across class I PDZ domains and is known to confer specificity at P0 (40). The side-chains of L29 and L97 interact to form the majority of the surface area of this hydrophobic pocket. These mutations would likely change the selectivity at P0 from Val to Ile, Leu or potentially a larger hydrophobic amino acid currently not allowed such as Phe or Trp. Specificity for S/T at P-2 is due to H90 at position ?2:1 of GIP. However, there is no specificity at the P-1 position. The likely reasons for the lack of specificity could be steric in nature. Firstly, the geometry of G30 is a prerequisite to sterically allow the binding of the C- terminus of the target protein to a PDZ protein. Thus, it could be an evolutionary trade-off between specificity for the C-terminus and sequence specificity at P-1. Secondly, because the 143 binding occurs through ?-strand addition, alternating amino acids are oriented away from the binding site. To identify and distinguish between common and unique features of binding for each ligand, the first NMR structure of the GIP-Glutaminase L peptide complex was compared to the crystal structures of GIP bound to other target proteins. The mode of binding between GIP and each of its ligands is unique and specific. For example, unlike the specific interactions seen between the PFS loop of GIP with ?-catenin (39), there are only a few interactions that occur between the ?2-?3 loop of GIP and the Glutaminase L peptide. Additionally, the E at P-3 of the Glutaminase L peptide makes specific hydrogen bonds to Y56 and T58 of GIP rather than the salt-bridge observed between the D or E at P-3 of ?-catenin or Kir 2.3 respectively with R59 of GIP (39, 41). Thus, it is necessary to experimentally determine the structure of GIP in complex with each of its known ligands to understand the mechanism of interactions for each binding partner. By maximizing the common features and taking advantage of the unique features of ligand binding, we should be able to efficiently design a competitive inhibitor with higher affinity than any of the natural ligands. Specificity for E at P-3 of the peptide is due to the formation of a hydrogen bond with Y56 and/or T58 of GIP. Since Y56 and T58 can each act both as hydrogen bond donors or acceptors, this explains why P-3 can also accommodate multiple side-chains. Furthermore, the lack of side-chains in three glycines in a row: G34, G35 and G36 of GIP render the ability to the protein to bind multiple partners. The lack of side-chains in such a stretch of three residues allows enough space for the different side-chains of the residues of the interacting partners that are located close to this region of the ?2 sheet of the GIP but some 144 residues apart from those involved in the binding interactions. Finally, at positions beyond P-3, GIP shows some specificity such as those observed in the interaction with ?-catenin. During molecular recognition of ?-catenin by GIP, a hydrogen bond is formed between the main-chain oxygen atom of tryptophan residue at P-5 and NE2 atom of Q43 (39). Thus, for any future drug design effort, an aromatic residue at P-5 or P-6 (Table 2.1) could provide additional specificity to GIP (10). 3.5.2 The effects of the Glutaminase L peptide binding on the dynamics of GIP When the dynamics of free GIP was compared to that of the GIP-Glutaminase L complex, in general, residues at the binding site tend to become more ordered, while residues peripheral to the binding site in GIP become more disordered, with a few exceptions. One such exception is residue G34, which is part of the ?2 strand that forms an antiparallel ?-sheet with the Glutaminase L peptide. Although, the dynamic nature of the residue is expected to be more stabilized, yet it actually becomes more disordered. While it is part of the binding site, it is located on the opposite end of the ??-strand from the ILGF binding loop and is near the hinge- point between the ?2 strand and the ?2-?3 loop (residues G36-G54). Additionally, H90, D91 and Q92 show increased flexibility. While H90 makes a direct H-bond to the S at P-2, (Figure 3.20) the specificity of the Glutaminase L peptide is limited to the four C-terminal residues, while the N-terminal four residues are disordered with higher RMSD values. However, overall, the region of GIP, where the peptide directly interacts, becomes more rigid. But, this decrease in flexibility in those regions is apparently offset by an increase in flexibility that is distributed throughout the rest of the protein including core regions of the protein that are distal to the binding site such as 145 ?1, ?4 and ?6 strands and flexible regions of the protein such as the ?a-?b hairpin and ?2-?3 loops as well as both termini (10). 3.5.3 Comparison to other GIP-peptide complex structures Both the N-terminal (M1-T10) and C-terminal (S113-S124) regions of GIP are completely unstructured both in the free form and in the bound form with very few observed NOEs and correspondingly high RMSDs in our structural ensembles (Figure 2.21 & 3.18). The dynamics study further supports this observation, indicating that these regions are completely unstructured (Figure 2.22). Previously, it has been reported that the C-terminal truncation of GIP leads to a decreased affinity for full length ?-catenin in vivo (43). However, the binding modes of the ?-catenin and Glutaminase L peptides to GIP were found generally to be similar (11). Therefore, it is unlikely that the reported decrease in full length ?-catenin affinity to a C- terminally truncated GIP is due to an interaction between the canonical C-terminal binding motif of ?-catenin and the C-terminus (113-124) of GIP. Moreover, upon binding with Glutaminase L, ?-catenin or FAS peptide, the C-terminal region of GIP showed very little change in the chemical shifts (11). Therefore, a possible explanation for the above observation is the decrease in the affinity for the full length ?-catenin upon C-terminal truncation of GIP could be the interaction of the C-terminus of GIP with either a different region of full length ?-catenin or another interacting partner protein in vivo. An in vivo 2-hybrid interaction studies between various deletion mutants for both GIP and ?-catenin supports this hypothesis (43). From these studies, it was observed that a central core region of ?-catenin (173-483) lacking the class I C-terminus still 146 maintained some affinity for GIP (43). In light of our structural and dynamics characterization, the best plausible explanation is that the central core region of ?-catenin interacts directly and specifically with the C-terminus of GIP. Thus, apparently, ?-catenin and GIP each bind to the other protein?s C-terminus (10). 3.5.4 Comparison between NMR and crystal structures To comprehend the dynamic nature of a protein in solution, NMR is the technique of choice for structure determination. While there is good agreement between NMR and crystal structures of free GIP, there are a few key differences. First, in both the free and bound state NMR structures of GIP, both the N- and C-termini (regions 1-10 and 113-124) are highly dynamic and unstructured. Whereas, in the crystal structure of free GIP, the C-terminus forms a helix. This is very likely an artifact of crystallization. Second, in the NMR structures, the ?2-?3 loop from G36-G54 is considerably more flexible in comparison to the crystal structures where this region has a defined structure (39, 41). Flexibility in this loop is also supported by the dynamics data, where significantly lower order parameters compared to the rest of the central core region were observed. Also, relatively few NOEs were observed compared to other regions of the protein. Moreover, all of the observed NOEs were medium range (|i-j|<5) or shorter, but there were no unambiguously defined long-range NOEs (|i-j|>5). This was the case for both free GIP as well as the GIP-Glutaminase L peptide complex. However, for the complex, there were some intermolecular NOEs between the loop and the peptide, indicating a conformational change in this flexible loop upon binding. This conformational change is observed from the decrease in flexibility of G36 and G54 near the hinge-point of the ?2-?3 loop while flexibility increases on 147 either side of the hinge point. Although, a distinct conformational change is observed (Figure 3.19), the loop still remains relatively unstructured compared to the rest of the core protein in both free and bound states. Third, in comparison to crystal structures, the non-canonical ?- hairpin formed by residues L21-I28 has a higher relative backbone RMSD of around 0.85 ? in the free form of GIP compared to the rest of the core structured portion of the protein at 0.45 ?. In the GIP-Glutaminase L complex the corresponding RMSD values are 2.73 ? and 0.67 ?. Like the ?2-?3 loop, this ?-hairpin structure also has mostly medium or short-range NOEs. Since it is exposed to the solvent, it does not make as many contacts with the rest of the protein. This results in very few long-range NOEs for this region and, therefore, this hairpin structure remains relatively unconstrained during the structural calculation. Comparatively, there were more long- range NOEs for this hairpin loop in free GIP than in the complex. That is why; there is an increase in RMSD for this structure within the complex compared to free GIP. The above observation is further supported by the dynamics study as increases in flexibility is observed for residues G24, E25, L27 and I28 in the complex (10). 3.5.5 Potential for drug design Because GIP is very specific for certain types of molecular interactions, designing a drug that would target this protein is a promising endeavor. Since, cells contain literally hundreds of PDZ domains, if a drug is intended to target only the PDZ domain within GIP or broadly other PDZ domains that may share the same specificity as GIP, it is essential that the design of the drug molecule be very specific toward its desired target. Thus, the structural insights gained in 148 this chapter could prove very useful for the future design of a very specific drug molecule. Also, targeting GIP could lead to promising anticancer therapeutics. 3.6 Accession codes The accession codes for GIP-Glutaminase L peptide complex in the BioMagnetic Resonance Bank (BMRB) and the Protein Data Bank (PDB) are 17255 and 2L4T, respectively. In BMRB, the chemical shifts of the resonances and, in PDB, the atomic coordinates for GIP- Glutaminase L peptide complex have been deposited (10). 149 3.7 References 1. Olalla, L., Gutierrez, A., Jimenez, A. J., Lopez-Tellez, J. F., Khan, Z. U., Perez, J., Alonso, F. J., de la Rosa, V., Campos-Sandoval, J. A., Segura, J. A., Aledo, J. C., and Marquez, J. (2008) Expression of the scaffolding PDZ protein glutaminase-interacting protein in mammalian brain, J Neurosci Res 86, 281-292. 2. Krebs, H. A. (1935) Metabolism of amino-acids: The synthesis of glutamine from glutamic acid and ammonia, and the enzymic hydrolysis of glutamine in animal tissues, Biochem J 29, 1951-1969. 3. Olalla, L., Aledo, J. C., Bannenberg, G., and Marquez, J. (2001) The C-terminus of human glutaminase L mediates association with PDZ domain-containing proteins, FEBS Lett 488, 116-122. 4. Curthoys, N. P., and Watford, M. (1995) Regulation of glutaminase activity and glutamine metabolism, Annu Rev Nutr 15, 133-159. 5. Aledo, J. C., Gomez-Fabre, P. M., Olalla, L., and Marquez, J. (2000) Identification of two human glutaminase loci and tissue-specific expression of the two related genes, Mamm Genome 11, 1107-1110. 6. Olalla, L., Gutierrez, A., Campos, J. A., Khan, Z. U., Alonso, F. J., Segura, J. A., Marquez, J., and Aledo, J. C. (2002) Nuclear localization of L-type glutaminase in mammalian brain, J Biol Chem 277, 38939-38944. 7. Medina, M. A., Sanchez-Jimenez, F., Marquez, J., Rodriguez Quesada, A., and Nunez de Castro, I. (1992) Relevance of glutamine metabolism to tumor cell growth, Mol Cell Biochem 113, 1-15. 150 8. Brand, K. (1985) Glutamine and glucose metabolism during thymocyte proliferation. Pathways of glutamine and glutamate metabolism, Biochem J 228, 353-361. 9. Kovacevic, Z., and McGivan, J. D. (1983) Mitochondrial metabolism of glutamine and glutamate and its physiological significance, Physiol Rev 63, 547-605. 10. Zoetewey, D. L., Ovee, M., Banerjee, M., Bhaskaran, R., and Mohanty, S. (2011) Promiscuous Binding at the Crossroads of Numerous Cancer Pathways: Insight from the Binding of Glutaminase Interacting Protein with Glutaminase L, Biochemistry-Us 50, 3528-3539. 11. Banerjee, M., Huang, C., Marquez, J., and Mohanty, S. (2008) Probing the structure and function of human glutaminase-interacting protein: a possible target for drug design, Biochemistry-Us 47, 9208-9219. 12. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes, J Biomol NMR 6, 277-293. 13. Goddard, T. D., and Kneller, D. G. SPARKY 3, University of California, San Francisco. 14. Kay, L., Keifer, P., and Saarinen, T. (1992) Pure absorption gradient enhanced heteronuclear single quantum correlation spectroscopy with improved sensitivity, Journal of the American Chemical Society 114, 10663-10665. 15. Muhandiram, D. R., and Kay, L. E. (1994) Gradient-Enhanced Triple-Resonance Three- Dimensional NMR Experiments with Improved Sensitivity, Journal of Magnetic Resonance, Series B 103, 203-216. 151 16. Grzesiek, S., and Bax, A. (1992) Correlating backbone amide and side chain resonances in larger proteins by multiple relayed triple resonance NMR, Journal of the American Chemical Society 114, 6291-6293. 17. Norwood, T. J., Boyd, J., Heritage, J. E., Soffe, N., and Campbell, I. D. (1990) Comparison of techniques for 1H-detected heteronuclear 1H---15N Spectroscopy, Journal of Magnetic Resonance (1969) 87, 488-501. 18. Palmer, A. G., Cavanagh, J., Wright, P. E., and Rance, M. (1991) Sensitivity improvement in proton-detected two-dimensional heteronuclear correlation NMR spectroscopy, Journal of Magnetic Resonance (1969) 93, 151-170. 19. Vuister, G. W., and Bax, A. (1993) Quantitative J correlation: a new approach for measuring homonuclear three-bond J(HNH.alpha.) coupling constants in 15N-enriched proteins, Journal of the American Chemical Society 115, 7772-7777. 20. Clubb, R. T., Thanabal, V., and Wagner, G. (1992) A constant-time three-dimensional triple-resonance pulse scheme to correlate intraresidue 1HN, 15N, and 13C' chemical shifts in 15N---13C-labelled proteins, Journal of Magnetic Resonance (1969) 97, 213- 217. 21. Zhang, O., Kay, L. E., Olivier, J. P., and Forman-Kay, J. D. (1994) Backbone 1H and 15N resonance assignments of the N-terminal SH3 domain of drk in folded and unfolded states using enhanced-sensitivity pulsed field gradient NMR techniques, J Biomol NMR 4, 845-858. 22. Otting, G., and W?thrich, K. (1989) Extended heteronuclear editing of 2D 1H NMR spectra of isotope-labeled proteins, using the X([omega]1, [omega]2) double half filter, Journal of Magnetic Resonance (1969) 85, 586-594. 152 23. Zwahlen, C., Legault, P., Vincent, S. J. F., Greenblatt, J., Konrat, R., and Kay, L. E. (1997) Methods for Measurement of Intermolecular NOEs by Multinuclear NMR Spectroscopy:? Application to a Bacteriophage ? N-Peptide/boxB RNA Complex, Journal of the American Chemical Society 119, 6711-6721. 24. Bax, A., and Davis, D. G. (1985) MLEV-17-based two-dimensional homonuclear magnetization transfer spectroscopy, Journal of Magnetic Resonance (1969) 65, 355-360. 25. Bax, A., and Davis, D. G. (1985) Practical aspects of two-dimensional transverse NOE spectroscopy, Journal of Magnetic Resonance (1969) 63, 207-213. 26. Mandel, A. M., Akke, M., and Palmer, A. G., 3rd. (1995) Backbone dynamics of Escherichia coli ribonuclease HI: correlations with structure and function in an active enzyme, J Mol Biol 246, 144-163. 27. Palmer, A. G., Rance, M., and Wright, P. E. (1991) Intramolecular motions of a zinc finger DNA-binding domain from Xfin characterized by proton-detected natural abundance carbon-13 heteronuclear NMR spectroscopy, Journal of the American Chemical Society 113, 4371-4380. 28. Dosset, P., Hus, J.-C., Blackledge, M., and Marion, D. (2000) Efficient analysis of macromolecular rotational diffusion from heteronuclear relaxation data, Journal of Biomolecular NMR 16, 23-28. 29. Tsan, P., Hus, J.-C., Caffrey, M., Marion, D., and Blackledge, M. (2000) Rotational Diffusion Anisotropy and Local Backbone Dynamics of Carbon Monoxide-Bound Rhodobacter capsulatus Cytochrome c?, Journal of the American Chemical Society 122, 5603-5612. 153 30. Guntert, P. (2004) Automated NMR structure calculation with CYANA, Methods Mol Biol 278, 353-378. 31. Cornilescu, G., Delaglio, F., and Bax, A. (1999) Protein backbone angle restraints from searching a database for chemical shift and sequence homology, J Biomol NMR 13, 289- 302. 32. Linge, J. P., Habeck, M., Rieping, W., and Nilges, M. (2003) ARIA: automated NOE assignment and NMR structure calculation, Bioinformatics 19, 315-316. 33. Laskowski, R. A., Rullmannn, J. A., MacArthur, M. W., Kaptein, R., and Thornton, J. M. (1996) AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR, J Biomol NMR 8, 477-486. 34. Humphrey, W., Dalke, A., and Schulten, K. (1996) VMD: visual molecular dynamics, J Mol Graph 14, 33-38, 27-38. 35. Schrodinger, LLC. (2010) The PyMOL Molecular Graphics System, Version 1.3r1. 36. Hwang, T. L., and Shaka, A. J. (1995) Water Suppression That Works - Excitation Sculpting Using Arbitrary Wave-Forms and Pulsed-Field Gradients, J Magn Reson Ser A 112, 275-279. 37. Jelen, F., Oleksy, A., Smietana, K., and Otlewski, J. (2003) PDZ domains - common players in the cell signaling, Acta Biochim Pol 50, 985-1017. 38. Schultz, J., Hoffmuller, U., Krause, G., Ashurst, J., Macias, M. J., Schmieder, P., Schneider-Mergener, J., and Oschkinat, H. (1998) Specific interactions between the syntrophin PDZ domain and voltage-gated sodium channels, Nat Struct Biol 5, 19-24. 154 39. Zhang, J., Yan, X., Shi, C., Yang, X., Guo, Y., Tian, C., Long, J., and Shen, Y. (2008) Structural basis of beta-catenin recognition by Tax-interacting protein-1, J Mol Biol 384, 255-263. 40. Tonikian, R., Zhang, Y., Sazinsky, S. L., Currell, B., Yeh, J. H., Reva, B., Held, H. A., Appleton, B. A., Evangelista, M., Wu, Y., Xin, X., Chan, A. C., Seshagiri, S., Lasky, L. A., Sander, C., Boone, C., Bader, G. D., and Sidhu, S. S. (2008) A specificity map for the PDZ domain family, PLoS Biol 6, e239. 41. Yan, X., Zhou, H., Zhang, J., Shi, C., Xie, X., Wu, Y., Tian, C., Shen, Y., and Long, J. (2009) Molecular mechanism of inward rectifier potassium channel 2.3 regulation by tax- interacting protein-1, J Mol Biol 392, 967-976. 42. Lipari, G., and Szabo, A. (1980) Effect of librational motion on fluorescence depolarization and nuclear magnetic resonance relaxation in macromolecules and membranes, Biophysical Journal 30, 489-506. 43. Kanamori, M., Sandy, P., Marzinotto, S., Benetti, R., Kai, C., Hayashizaki, Y., Schneider, C., and Suzuki, H. (2003) The PDZ protein tax-interacting protein-1 inhibits beta-catenin transcriptional activity and growth of colorectal cancer cells, Journal of Biological Chemistry 278, 38758-38764. 155 Chapter 4 Determination of the mode of interaction of Glutaminase Interacting Protein (GIP) with two different interacting partners 4.1 Introduction 4.1.1 PDZ domain and its functions (1) Glutaminase interacting protein (GIP) (2), also known as tax interacting protein-1 (TIP-1) (3), is a 13.7 kDa PDZ domain-containing protein. PDZ domains are one of the most important protein-protein interaction modules found in nature (4). PDZ domain-mediated interactions contribute to cell signaling pathways, adhesion and receptor and ion transporter function (5). PDZ domains often act as scaffolds, specifying protein interactions required for the formation of multimeric complexes (6). The diversity of PDZ domain-protein interactions and their involvement in maintenance of normal physiological functions of the body are significant in the context of clinical disorders. Several human diseases are known to occur as a result of inappropriate protein-protein interactions, which in turn affect gene expression and regulation, transport of biomolecules across the membranes, cell adhesion, antigen recognition and signal transduction (7). 4.1.2 Binding pocket of PDZ domain (1) The binding pocket of PDZ domains and the mode of binding to the interacting partner proteins are each well characterized (5, 8-10). The GLGF motif present in the binding pocket of 156 PDZ domains plays a major role in the binding interactions with the target protein. PDZ domains were therefore previously referred to as GLGF repeat domains (11). PDZ domains exhibit sequence specificity towards the unstructured C-terminal ends of their interacting protein partners. Peptides representing these C-terminal recognition motifs have been shown to act as surrogates for their corresponding partner proteins in vitro (12). Several classes of PDZ domains have been reported based on this specificity: class I {X-S/T-X-?-COOH}, class II {X-?-X-?- COOH}(6), class III {X-E/D-X-?-COOH}(13) and other minor classes (14) where ? is any hydrophobic residue and X is any residue. The interacting peptide forms an additional anti- parallel ?-strand between the ?2 strand and the ?2 helix of PDZ domain (5). 4.1.3 GIP as a PDZ domain (1) GIP is an unusual class I PDZ domain protein in the sense that it is solely composed of a single PDZ domain (6). Structurally, GIP is made up of two ?-helices (?1 and ?2) and six ?- strands (?1, ?a, ?b, ?2???3, ?4, ?5 and ?6) (10, 15). GIP is also striking for the promiscuity of its binding profile. A number of different binding partners have been identified with roles in diverse cellular processes. Some of the reported interacting proteins include Glutaminase L, ?-Catenin, Fas, HTLV (Human T-lymphotropic virus) Tax and HPV (Human papillomavirus) E6, which are involved in signaling pathways, energy generation pathways or oncogenic processes (2, 3, 8, 10, 12, 16-24). 157 4.1.4 GIP in the brain (1) GIP is known to function as a key scaffolding protein in the mammalian brain (25), contributing to the bioenergetics of both normal and cancer cells through its interaction with Glutaminase L (2, 16-18). GIP may also mediate normal brain cellular functions through interactions with other as yet unidentified partner proteins. To fully understand the mechanism of function of GIP in the brain, it is necessary to identify the proteins that interact with GIP in brain cells. 4.1.5 Identification of interacting partners in brain (1) Among the various methods available for the investigation of novel protein-protein interactions, the yeast two-hybrid genetic selection system (Y2H) is a powerful technique with several advantages over traditional biochemical approaches (7). This method was developed by Song and Fields in Saccharomyces cerevisiae (baker?s yeast) and involves the expression within the yeast cell nucleus of two proteins being assessed for interaction (26). Each protein is expressed as a chimera, fused to one domain of the yeast Gal4 transcription factor. Interaction of the two fusion proteins brings the two domains of Gal4 into close enough proximity to restore transcription factor function, detected by activation of Gal4-responsive reporter genes. In this study, our collaborators in Ege University, Izmir, Turkey used the yeast two-hybrid system to screen a human fetal brain cDNA library for GIP-interacting proteins. From that screening, Brain-specific angiogenesis inhibitor 2 (BAI2) was identified as a novel interacting partner of GIP. Here, CD, fluorescence and NMR techniques were used to further confirm BAI2 as an interacting partner of GIP by using a peptide RDGDFQTEV-COOH representing the BAI2 C- 158 terminus. To compare the interaction between GIP and RDGDFQTEV-COOH, another arbitrary peptide RGGSRL-COOH hereinafter termed as control peptide was designed based on the peptide sequence specificity for PDZ domain (Table 2.1) and used to determine the comparative strength of interaction by CD, fluorescence and NMR techniques. 4.2 Materials and Methods (1) The research work described here was carried out in the laboratory of Dr. Smita Mohanty. 4.2.1 Expression and purification of 15N- and unlabeled GIP GIP protein was expressed in E. coli and purified according to our lab protocol (12), E coli (strain BL21DE3pLysS) was transformed with plasmid pET-3c/GIP and cells were cultured in M9 minimal media containing 15N-labeled ammonium chloride for 15N-labeled GIP and in LB-ampicillin media for unlabeled GIP. An overnight culture was diluted 1:25, {v/v} in minimal media (or LB-ampicillin media for unlabeled protein) and grown at 37 ?C to an OD600 of 0.4-0.5. Expression was induced with 1 mM IPTG at 30? C, and after 12 h. incubation (for unlabeled GIP, after 4 hours), cells were harvested by centrifugation. The harvested cells were lysed by sonication using lysis buffer containing 50 mM phosphate buffer at pH 8, 200 mM NaCl, 4 mM EDTA, 4% glycerol, and 1 mM PMSF. After centrifugation of the lysed cells, the supernatant was retained for further purification. 15N- and unlabeled GIP were each purified in a single-step using size exclusion chromatography with a Sephacryl S-100 column {GE Healthcare} according to our lab protocol (12). Pooled fractions of pure protein were concentrated. 159 4.2.2 Fluorescence All fluorescence spectra were recorded on a PerkinElmer Precisely LS 55 Luminescence spectrofluorometer at 25 ?C (?ex 280 nm). Emission spectra were recorded over a range of 300- 500 nm with 1 nm steps. All experiments were carried out in 20 mM phosphate buffer, pH 6.5, 150 mM NaCl, 0.1 mM EDTA and 0.01% NaN3. Stock solutions of the synthetic peptide sequence RDGDFQTEV-COOH, hereafter known as BAI2 peptide and control peptide were prepared in water at a concentration of 10 mM. The target peptides were obtained with >95% purity from Chi Scientific (MA). The stock solutions were then diluted to 1 mM. Aliquots of the 1 mM peptide solutions were directly added to a cuvette containing 2 mL of 1 ?M unlabeled GIP. All titration experiments were corrected to take the dilution effect into account. Emission from the control was corrected by recording subtraction spectra between sample and control probes. 4.2.3 Circular Dichroism (CD) All circular dichroism (CD) experiments were performed on a Jasco J-810 automatic recording spectropolarimeter. Far-UV CD spectra were measured in a 0.05 cm quartz cell at room temperature. The buffer used was 20 mM phosphate buffer (pH 6.5). The protein concentration was 30 ?M. Data were averaged over 100 scans for each protein sample and over 50 scans for each control sample. Response time was 1 s, and scan speed was 100 nm min-1. 160 4.2.4 Nuclear Magnetic Resonance (NMR) All NMR data were collected at 298 K on a Bruker Avance 600 MHz spectrometer equipped with a triple resonance H/C/N TCI cryoprobe at the Department of Chemistry and Biochemistry, Auburn University, Auburn, AL. The data were processed using NMRPipe (27) and analyzed using Sparky (28). The ligand titration experiments were performed and monitored by a series of 2D 15N-edited HSQC experiments. The interaction study was carried out by titration of 100 ?M 15N-labeled GIP with the BAI2 peptide and control peptide. The amide chemical shift perturbations (??) were calculated as ?? =?[{|?? 15N|/10}2 + {|?? 1H|}2]. In the equation, ?? 15N was divided by 10 to account for the difference in the gyromagnetic ratio of the 15N and 1H nuclei to give roughly equal weighting for both types of chemical shift changes. The program ModelTitr (29) was used to calculate the dissociation constant values for various residues of GIP. 161 4.3 Results and Discussion 4.3.1 Protein expression As described above, unlabeled GIP was expressed in bacterial cells growing in LB media and 15N-labeled GIP was expressed in M9 minimal media containing 15N-labeled ammonium chloride. The SDS-PAGE analysis of expression of both unlabeled and 15N-labeled GIP upon induction is given in the Figure 4.1. GIP as a 13.7 kDa size protein appeared as a prominent band in both of the lanes for labeled and unlabeled protein at its due place in the gel (Figure 4.1). 4.3.2 Protein purification Using size-exclusion chromatography as a single step, GIP was purified in a Sephacryl S-100 column (GE Healthcare). The production of the unlabeled and 15N-labeled recombinant GIP is around 46 mg and 12 mg per liter of bacterial culture (Figure 4.2). Comparing the expression profile for unlabeled and 15N-labeled GIP in the Figure 4.1, higher amount of production for unlabeled GIP than 15N-labeled GIP is observed. Recently, Turck et al. at Max Planck Institute of Psychiatry has demonstrated that, when E. coli cells were grown in 15N- labeled media, consistent lower level of protein expression and alteration of growth rates and metabolite levels were observed as compared to when cells grown on unlabeled media (30). 162 Figure 4.1: Expression of GIP analyzed by SDS-PAGE. Lane 1- Unlabeled GIP expression in T0 cells before induction. Lane 2- Unlabeled GIP expression in T5 cells after complete induction. Lane 3- 15N-labeled GIP expression in T0 cells before induction. Lane 4- 15N-labeled GIP expression in T12 cells after complete induction. The red rectangle spots the protein of expected size. The lane MW is for protein marker. 100 75 50 37 25 20 15 10 150 250 MW 1 2 kDa 3 4 163 Figure 4.2: Purification of GIP analyzed by SDS-PAGE. Lane 1 shows the purified 15N-labeled GIP without any impurities. The red rectangle spots the protein of expected size. The lane MW is for protein marker. 250 150 75 100 50 37 25 20 15 10 MW 1 kDa 164 4.3.3 Interaction of BAI2 Peptide with GIP (1) 4.3.3.1 Characterization by Fluorescence spectroscopy When the peptide was titrated against unlabeled GIP, it showed a small but consistent decrease in fluorescence intensity (Figure 4.3). The dissociation constant KD (KD = 1/Ka) was determined using the OriginPro 6.1 software. The decrease in the fluorescence intensity was calculated as (F0 - FC)/(F0 - Fmin), where F0 is the initial fluorescence intensity of free GIP; FC is the corrected fluorescence intensity at a ligand concentration [C], and Fmin is the fluorescence intensity at the saturating concentration of the peptide. The data were fitted to a nonlinear regression of the plot of (F0 - FC)/(F0 - Fmin) against [C] with the equation corresponding to a single binding site (Figure 4.4). The titration of the BAI2 peptide with GIP yielded a dissociation constant of 0.71 ?M. To determine the thermodynamic nature of the interaction, the free energy change of the association was calculated using the following equation: ?G = -RT ln Ka, where Ka is the association constant, T is the temperature and R is the universal gas constant. By putting the experimentally determined Ka (Ka = 1/KD) value into this equation, the ?G value for binding of the BAI2 peptide to GIP was calculated to be-35.08 kJ mol-1, which reflects the spontaneous binding of the peptide to GIP. 165 Figure 4.3: Fluorescence emission spectrum of GIP with the BAI2 peptide. Fluorescence emission plots corresponding to (top to bottom) 0 to 20 ?M concentrations of the peptide to 1 ?M protein sample. In the legend, protein to peptide ratios are indicated with the respective color codes. The black arrow indicates the quenching of fluorescence of GIP upon peptide binding in a downward fashion. 166 Figure 4.4: Non-linear curve fitting assuming 1:1 binding between GIP and the BAI2 peptide where (F0 - FC)/( F0 - Fmin) was plotted against peptide concentration. 167 4.3.3.2 Characterization by CD spectroscopy CD spectroscopy is another powerful tool to investigate the effect of any ligand binding on the secondary structure of the protein. The phosphate buffer, as well as the BAI2 peptide alone, showed minimal signal in the CD measurements. However, any contribution from the peptide and buffer was subtracted from the CD spectrum obtained in subsequent analyses of GIP with peptide. The secondary structure of GIP showed significant changes in the CD spectrum with the titration of different concentrations of the peptide (Figure 4.5). CD data of the GIP- peptide complex was deconvoluted using the program CDPro (31) and the secondary structure content was calculated. From the deconvolution results, the helix content was found to be reduced by ~ 47%, random coil content by ~ 8% and the ?-sheet structure content increased by ~ 29%. The changes in the secondary structure of GIP with the addition of increasing concentration of BAI2 peptide is comparable to that observed with other previously reported binding partners of GIP such as Glutaminase L, FAS and ?-catenin (12). Although, the increase in ?-sheet content in all these cases can be explained by the mode of binding of these peptides to the GIP through ?-strand addition, closer examination of the representative complex structure of GIP with its binding partner does not show any change in the helical content but does indicate some displacement of the helical structure in space (8, 10, 32). CD spectroscopy is sensitive enough to detect even slight changes in the secondary structure of the protein upon interaction with the ligand but is not always sufficient to get a complete picture of the structural features of protein-peptide interactions. 168 Figure 4.5: Changes in the CD spectra of GIP upon binding with increasing concentrations of the BAI2 peptide for the wavelength range of 194 nm to 250 nm. The protein to peptide ratios for the corresponding color codes are indicated in the legend. 169 4.3.3.3 Characterization by 1H,15N-HSQC NMR To examine the interaction of GIP with the C-terminal BAI2 peptide more thoroughly, an NMR analysis was undertaken. NMR can be employed as a very powerful technique for monitoring structure-activity relationships (SAR) in protein-protein or protein-ligand interactions studies (33). The chemical shifts of the backbone amides of a folded protein are extremely sensitive to any changes in their chemical environments, such as temperature, pH, ionic strength, or binding to a ligand. For this reason, the 2D 1H,15N-HSQC spectrum is often called the fingerprint region of a protein, as the exact pattern is unique to each protein under a specific set of environmental conditions. Upon ligand binding, the chemical shifts of the residues involved in the binding change, which is reflected in a series of 2D 1H,15N-HSQC spectra (10, 12). However, when the binding is allosteric, which affects the protein globally rather than locally, the chemical environments of most of the residues in a protein experience a change. Thus, residues that are not directly part of the binding pocket may also show change in their chemical shifts (10). Therefore, any perturbation in the chemical shifts from their original positions may indicate a change in the conformation of the protein upon binding with the ligand (34). However, it is important to note that, for GIP, such chemical shift perturbations should not necessarily indicate a drastic conformational change in the protein (10). To investigate whether BAI2 peptide binds to the protein, 15N-labeled GIP protein was titrated with the synthetic BAI2 peptide to excess (~60 times that of the protein) until complete saturation was achieved. During the course of the titration, the fingerprint region of the protein in the 2D 1H,15N-HSQC spectra was monitored. The fingerprint region of the HSQC spectra of GIP was collected in the absence and presence of different concentrations of the peptide and the spectra were overlaid (Figure 4.6). From the overlay, it was evident that most of the residues of GIP showed moderate changes in chemical 170 shifts upon binding with the peptide, while other residues showed more dramatic changes. Using the program ModelTitr (29), the dissociation constant (KD) values for various residues of GIP were calculated (Table 4.1) by non-linear least-squares fitting of the chemical shift data against ligand concentration to the Langmuir isotherm that involved the assumption of a stoichiometry of 1:1 between the ligand and the protein (i.e. one binding site) (Figure 4.8). The dilution effect on the concentration of the protein due to the addition of the ligand was corrected in the program. The calculated dissociation constant (KD) value from NMR technique (97.77 ?M on an average) was different from the value obtained from fluorescence technique. Since the dissociation constant (KD) value varies depending upon techniques and initial protein concentration used (35- 37), such a difference in the KD values obtained from two different techniques is acceptable. From the KD values of both fluorescence and NMR techniques, the dissociation constant (KD) value falls in the range of low to mid ?M, which indicates a moderate affinity of GIP for the BAI2 peptide. 171 Figure 4.6: Changes of 2D 1H,15N-HSQC spectra upon addition of the BAI2 peptide to 100 ?M of 15N-labeled GIP. The 2D 1H,15N-HSQC spectra demonstrating chemical shift perturbations of residues upon titration of the peptide to GIP. Ratios of GIP to the peptide range from 1:0 to 1:60. 172 Figure 4.7: Expanded region of the spectra demonstrating the chemical shift perturbations of residue E17 upon titration of GIP with the BAI2 peptide. Ratios of GIP to the peptide are 1:0 (green), 1:0.2 (tomato), 1:0.4 (blue), 1:0.6 (beige), 1:0.8 (turquoise), 1:1 (gold), 1:2 (coral), 1:3 (purple), 1:5 (maroon), 1:7 (orange), 1:10 (red), 1:20 (cyan), 1:40 (white), 1:60 (magenta). 173 Figure 4.8: The NMR titration binding curve for the titration of GIP with the BAI2 peptide. The plot shows the changes in the chemical shift of E17 induced by the addition of peptide versus the peptide concentration. Dashed line is the titration curve as fit by the program ModelTitr from NMRPipe. The apparent dissociation constant KD corresponding to residue E17 was determined by fitting the chemical shift change of the residue to increasing concentrations of peptide. The determined KD value was 64.8 ? 10.6% ?M. 174 Table 4.1: Dissociation constants of various residues of GIP upon binding with the BAI2 peptide by NMR. Interaction with BAI2 peptide Residues of GIP Dissociation constants, ?M E17 64.78?10.64% R22 101?5.6% D38 92.5?3% F46 137.7?7.57% E62 102.3?5.27% A66 86.58?2.20% L71 85.85?7.82% N81 84.28?7.29% T86 104.3?6.46% E102 118.4?6.26% 175 4.3.3.4 Chemical shift perturbations of GIP upon binding to the BAI2 peptide (1) Mapping the chemical shift perturbation with respect to residue number for a protein is a way to demonstrate the putative interacting portions of a protein with its interacting partner. For the mapping study of GIP with the BAI2 peptide, a series of the 2D 1H,15N-HSQC spectra of GIP while titrating with increasing peptide concentrations were analyzed. The chemical shifts of most of the residues of GIP in both free and complex forms were determined. During analysis of the 2D 1H,15N- HSQC spectra, the amide proton and nitrogen resonances of most residues showed gradual shifts with increasing peptide concentration, indicating that the complex was in the fast exchange regime on the NMR time scale. However, some residues disappeared or decreased in intensity below the noise level threshold with increasing peptide concentrations but reappeared at higher peptide concentrations suggesting that these residues were in intermediate exchange on the NMR time scale. For example, Leu 29 and Gly 30 initially disappeared with increasing peptide concentrations but reappeared at high peptide concentrations. Some of the residues could not be characterized for this mapping study because of the complete absence of the peak from the HSQC spectrum or peak overlap. These residues included Met 1, all five proline residues, Val 12, Leu 21, Phe 31, Glu 48, Lys 50, Val 57, Val 80 and Val 105. Residues that constitute the ?2 strand (residues 31 to 35) and the ?2 helix (residues 90 to 97) of the protein showed the most chemical shift perturbations compared to other residues as seen on the 2D HSQC spectrum and mapping of chemical shift perturbations (Figure 4.6 and Figure 4.9). This observation is consistent with that of interaction of GIP with a canonical C- terminal binding motif recognition peptide (10, 12). Most of the residues located within this 176 region showed greater than 0.1 ppm perturbations except residues Gln 92, Ala 93 and Leu 97 (Figure 4.9). The large perturbations occurred because the peptide directly interacted with most of these residues of the ?2 strand and ?2 helix. Residues Leu29 and Gly30 showed very large perturbations (greater than 1.0 ppm) (Figure 4.9) probably due to the hydrogen bonding formed between these two residues and the C-terminal end of the peptide (38). Such large chemical shift perturbations for Leu29 and Gly30 are reminiscent of our previous work on the interaction of GIP with a C-terminal peptide analog of Glutaminase L that was reported previously (10). Also, another cluster of residues showing prominent perturbations were residues 66 to residues 71 that form the ?1 helix of the protein (Figure 4.9). Within this region, residues Ala 66, Glu 67, Ile 68 and Ala 69 showed greater perturbations (greater than 0.1 ppm). The significant changes in chemical shifts of this region (?1 helix) of the protein were not due to the direct interaction with the peptide but rather due to the change in the surrounding environment of the helix since this helix is in close proximity to the binding pocket of the protein. In the work shown in the previous chapter, several long-range NOEs were observed between Ile 28 and the ?1 helix indicating a close spatial proximity between the ?a-?b loop and the ?1 helix for the free state of the protein but only a very few NOEs were present for that region of the complex form of the protein with Glutaminase L peptide (BMRB entry: 17254 and 17255) (10). Thus, the reason for comparatively higher chemical shift perturbation for residue Ile 28 (greater than 0.5 ppm) (Figure 4.9) could be twofold. First it is very close to the binding pocket. Second the binding of the BAI2 peptide to the protein probably resulted in the disruption of the interaction (NOEs) between residue Ile 28 and ?1 helix. Although there were certain pockets of residues that showed significant chemical shift perturbations, the binding of the peptide to the protein seemed to induce a change in the chemical environment over nearly the entire protein except for the 177 termini. The N- and C-termini of the protein did not show any significant changes in the chemical shifts (Figure 4.9) upon peptide binding. Thus, the mode of BAI2 peptide binding to GIP can be characterized as allosterically driven analogous to the binding of the Glutaminase L peptide to GIP (10). 178 Figure 4.9: Chemical shift perturbations (??) of the GIP backbone amide groups upon binding with the BAI2 peptide. 179 BAI2 is a member of the adhesion-G protein-coupled receptors (GPCRs) (39, 40). It is composed of 521-amino acids and mainly expressed in neurons (41). BAI2 possesses a Src homology 3 (SH3) domain, composed of 50-60 amino acids that mediates protein-protein interactions and was previously reported as interacting with the C-terminus of Brain-Specific Angiogenesis Inhibitor 1 (BAI1) via its SH3 domain as shown by in vitro binding assays (41). This was the first study reporting an interaction between BAI2 and GIP with an extensive biophysical characterization of their interaction (1). 180 4.3.4 Interaction of the control peptide with GIP 4.3.4.1 Characterization by Fluorescence spectroscopy When the control peptide was titrated against unlabeled GIP, it showed a small but consistent decrease in fluorescence intensity (Figure 4.10). The dissociation constant KD (KD = 1/Ka) was determined using the OriginPro 6.1 software. The decrease in the fluorescence intensity was calculated as (F0 - FC)/(F0 - Fmin), where F0 is the initial fluorescence intensity of free GIP; FC is the corrected fluorescence intensity at a ligand concentration [C], and Fmin is the fluorescence intensity at the saturating concentration of the peptide. The data were fitted to a nonlinear regression of the plot of (F0 - FC)/(F0 - Fmin) against [C] with the equation corresponding to a single binding site (Figure 4.11). The titration of the control peptide with GIP yielded a dissociation constant of 1.07 ?M. To determine the thermodynamic nature of the interaction, the free energy change of the association was calculated using the following equation: ?G = -RT ln Ka, where Ka is the association constant, T is temperature and R is universal gas constant. By putting the experimentally determined Ka (Ka = 1/KD) value into this equation, the ?G value for binding of the BAI2 peptide to GIP was calculated to be -34.06 kJ mol-1, which reflects the spontaneous binding of the peptide to GIP. 181 Figure 4.10: Fluorescence emission spectrum of GIP with the control peptide. Fluorescence emission plots corresponding to (top to bottom) 0 to 20 ?M concentrations of the peptide to 1 ?M protein sample. In the legend, protein to peptide ratios are indicated with the respective color codes. Black arrow indicates the quenching of fluorescence of GIP upon peptide binding in a downward fashion. 182 Figure 4.11: Non-linear curve fitting assuming 1:1 binding between GIP and the control peptide where (F0 - FC)/( F0 - Fmin) was plotted against peptide concentration. 183 4.3.4.2 Characterization by CD spectroscopy Like the BAI2 peptide, the interaction between the control peptide and GIP was also characterized by CD spectroscopy. The control peptide did not alter the CD spectrum from that obtained with phosphate buffer. However, any contribution from the peptide and buffer was subtracted from the CD spectrum obtained in subsequent analyses of GIP with peptide. The secondary structure of GIP showed significant changes in the CD spectrum with the titration of different concentrations of the peptide (Figure 4.12). 184 Figure 4.12: Changes in the CD spectra of GIP upon binding with increasing concentrations of the control peptide for the wavelength range of 194 nm to 250 nm. The protein to peptide ratios for the corresponding color codes are indicated in the legend. 185 4.3.4.3 Characterization by 1H,15N-HSQC NMR Like the interaction with the BAI2 peptide, for the investigation of any possible binding and, less importantly, a subsequent conformational change in GIP, 15N-labeled GIP protein was titrated against the control peptide to excess (~60 times that of the protein) until complete saturation was achieved. During the course of the titration, the fingerprint region of the protein in the 2D 1H,15N-HSQC spectra was monitored. The fingerprint region of the HSQC spectra of GIP was collected in the absence and presence of different concentrations of the peptide and the profiles were overlaid (Figure 4.13). From the overlay, it was evident that most of the residues of GIP showed changes in chemical shifts only slightly (if any) upon binding with the peptide. Using the program ModelTitr (29), the dissociation constant (KD) values for various residues of GIP were calculated (Table 4.2) by non-linear least-squares fitting of the chemical shift data against ligand concentration to the Langmuir isotherm that involved the assumption of a stoichiometry of 1:1 between the ligand and the protein (i.e. one binding site) (Figure 4.15). The dilution effect on the concentration of the protein due to the addition of the ligand was corrected in the program. The calculated dissociation constant (KD) value from NMR technique (717.97 ?M on an average) was different from the value obtained from fluorescence technique as was the case with the BAI2 peptide. From the KD values of both fluorescence and NMR techniques, the dissociation constant (KD) value falls in the range of low to mid ?M, which indicates a moderate affinity of GIP for the control peptide. 186 Figure 4.13: Changes of 2D 1H,15N-HSQC spectra upon addition of the control peptide to 100 ?M of 15N-labeled GIP. The 2D 1H,15N-HSQC spectra demonstrating chemical shift perturbations of residues upon titration of the peptide to GIP. Ratios of GIP to the peptide ranged from 1:0 to 1:60. N81 187 Figure 4.14: Expanded region of the spectra demonstrating the chemical shift perturbations of residue N81 upon titration of GIP with the control peptide. Ratios of GIP to the peptide were 1:0 (green), 1:0.2 (tomato), 1:0.4 (blue), 1:0.6 (beige), 1:0.8 (turquoise), 1:1 (gold), 1:2 (coral), 1:3 (purple), 1:5 (maroon), 1:7 (orange), 1:10 (red), 1:20 (cyan), 1:40 (white), 1:60 (magenta). 188 Figure 4.15: The NMR titration binding curve for the titration of GIP with the control peptide. The plot shows the changes in the chemical shift of N81 induced by the addition of peptide versus the peptide concentration. Dashed line is the titration curve as fit by the program ModelTitr from NMRPipe. The apparent dissociation constant KD corresponding to residue N81 was determined by fitting the chemical shift change of the residue to increasing concentrations of peptide. The determined KD value was 510.1 ? 24.8% ?M. 189 Interaction with RGGSRL Residues of GIP Dissociation constants, ?M Q23 527.4?59.23% I28 1169?6.70% F31 734.1?3.10% G34 485.7?66.98% Q39 1049?19.21% I68 458.6?24.61% N81 510.1?24.79% D84 953?31.25% T86 480.6?63.81% R106 812.2?17.21% Table 4.2: Dissociation constants of various residues of GIP upon binding with the control peptide by NMR. 190 4.3.4.4 Chemical shift perturbations of GIP upon binding to the control peptide For the mapping study of GIP with the control peptide, a series of the 2D 1H,15N-HSQC spectra of GIP with increasing peptide concentrations were analyzed. The chemical shifts of most of the residues of GIP in both free and complex forms were determined. During analysis of the 2D 1H,15N- HSQC spectra, the amide proton and nitrogen resonances of most residues showed gradual shifts with increasing peptide concentration, indicating that the complex was mostly in the fast exchange regime in the NMR time scale. Unlike the perturbation study with the BAI2 peptide, Leu 29 and Gly 30 did not reappear at a higher peptide concentration and could not be mapped in the study. Along with these two, some of the residues could not be characterized for this mapping study because of the complete absence of the peak from the HSQC spectrum or peak overlapping. These residues included Met 1, all five proline residues, Val 12, Leu 21, Phe 31, Glu 48, Lys 50, Val 57, Val 80 and Val 105. Residues that constitute the ?2 strand (residues 31 to 35) and the ?1 helix (residues 66 to 71) of the protein showed the most chemical shift perturbations compared to other residues as seen on the mapping of chemical shift perturbations (Figure 4.16). Most of the residues located within this region showed greater than 0.05 ppm perturbations except residues Gly 34, Gly 70 and Leu 71 (Figure 4.16). Residues that form ?2 helix (residues 90 to 97) had perturbations greater than 0.01 ppm. Apparently, these three clusters of residues were most perturbed due to the interaction with the control peptide. This feature is consistent with that of the interaction of GIP with a canonical C-terminal binding motif recognition peptide (10, 12). As in the interaction with the BAI2 peptide, in this perturbation study with the control peptide residue Ile 28 showed 191 perturbation of greater than 0.1 ppm (Figure 4.16) for the same possible reasons listed in case of interaction with the BAI2 peptide (section 4.3.3.4). Although there were certain pockets of residues that showed significant chemical shift perturbations, the binding of the peptide to the protein seemed to induce a change in the chemical environment over nearly the entire protein except for the termini. The N- and C-termini of the protein did not show any significant changes in the chemical shifts (Figure 4.16) upon peptide binding. 192 Figure 4.16: Chemical shift perturbations (??) of the GIP backbone amide groups upon binding with the control peptide. 193 4.3.5 Comparison of interaction between GIP and BAI2 peptide with interaction between GIP and control peptide To determine the comparative strength of the interaction between two interacting partners, NMR would be deemed as the most appropriate experimental tool since it is the most sensitive technique. The interaction between GIP and the two interacting partners was monitored by examining the series of 2D 1H, 15N-HSQC titration spectra corresponding to the increasing concentrations of the interacting peptides. These spectra were then overlaid to reflect the perturbations of the residues upon binding. Comparison of these two overlays for both of the peptides easily revealed that interaction with the BAI2 peptide appeared to cause more perturbations than with the control peptide (Figure 4.6 and Figure 4.13). The dissociation constant (KD) values determined from NMR for the residues of GIP were on an average about 7 times lower for the BAI2 peptide than with the control peptide (Table 4.1 and Table 4.2). This suggests that the binding of BAI2 peptide to GIP is at least 7 times stronger than that of control peptide to GIP. Moreover, the overlay of chemical shift perturbations map for both of the peptides easily reflected the overall greater chemical shift perturbations for the BAI2 peptide (Figure 4.17). The calculated Gibbs? free energy (?G) from the KD values determined from the fluorescence technique also showed an amount of 1.02 kJ mol-1 extra energy released as a result of binding of the BAI2 peptide to GIP compared to that of the control peptide to GIP. In summary, GIP seems to interact with BAI2 peptide more strongly than the control peptide. Such a preference of interaction might lie in the sequence of the peptide. As discussed in the previous chapter, one of the important interaction between GIP and the canonical C-terminus of the peptide is the hydrophobic interaction formed between the hydrophobic residue at the P0 position of the peptide and the hydrophobic pocket created by Leu 29, Phe 31, Leu 97 and Ile 33 as well 194 as Thr 98 at the periphery of GIP (10). The BAI2 peptide (RDGDFQTEV) has a Val at its P0 whereas the control peptide (RGGSRL) has Leu at its P0. Both are hydrophobic, but Leu is one (- CH2-) group long than Val. Larger side chain of Leu might cause a steric hindrance in the hydrophobic pocket of GIP leading to the disruption of interaction between GIP and the control peptide. Whereas, the smaller side chain of Val at P0 of the BAI2 peptide allows a more favorable interaction with GIP. Such a phenomenon could also be observed when a microarray technique was utilized to determine the protein interaction network of mouse PDZ domain with moderate to high affinity (KD ? 10 ?M). Among the 20 interacting peptides used, 16 had Val at their P0 position (42). 195 Figure 4.17: Chemical shift perturbations (??) of the GIP backbone amide groups upon binding with the BAI2 (red) and the control (black) peptide. 196 4.4 References 1. Zencir, S., Ovee, M., Dobson, M. J., Banerjee, M., Topcu, Z., and Mohanty, S. (2011) Identification of brain-specific angiogenesis inhibitor 2 as an interaction partner of glutaminase interacting protein, Biochem Biophys Res Commun 411, 792-797. 2. Olalla, L., Aledo, J. C., Bannenberg, C., and Marquez, J. (2001) The C-terminus of human glutaminase L mediates association with PDZ domain-containing proteins, Febs Letters 488, 116-122. 3. Rousset, R., Fabre, S., Desbois, C., Bantignies, F., and Jalinot, P. (1998) The C-terminus of the HTLV-1 Tax oncoprotein mediates interaction with the PDZ domain of cellular proteins, Oncogene 16, 643-654. 4. Fanning, A. S., and Anderson, J. M. (1996) Protein-protein interactions: PDZ domain networks, Current Biology 6, 1385-1388. 5. Jelen, F., Oleksy, A., Smietana, K., and Otlewski, J. (2003) PDZ domains - common players in the cell signaling, Acta Biochim Pol 50, 985-1017. 6. Garcia-Mata, R., and Burridge, K. (2007) Catching a GEF by its tail, Trends in Cell Biology 17, 36-43. 7. Topcu, Z., and Borden, K. L. B. (2000) The yeast two-hybrid system and its pharmaceutical significance, Pharmaceutical Research 17, 1049-1055. 8. Zhang, J. X., Yan, X. J., Shi, C. W., Yang, X., Guo, Y., Tian, C. L., Long, J. F., and Shen, Y. Q. (2008) Structural Basis of beta-Catenin Recognition by Tax-interacting Protein-1, Journal of Molecular Biology 384, 255-263. 9. Hung, A. Y., and Sheng, M. (2002) PDZ domains: Structural modules for protein complex assembly, Journal of Biological Chemistry 277, 5699-5702. 197 10. Zoetewey, D. L., Ovee, M., Banerjee, M., Bhaskaran, R., and Mohanty, S. (2011) Promiscuous Binding at the Crossroads of Numerous Cancer Pathways: Insight from the Binding of Glutaminase Interacting Protein with Glutaminase L, Biochemistry-Us 50, 3528-3539. 11. Cho, K. O., Hunt, C. A., and Kennedy, M. B. (1992) The Rat-Brain Postsynaptic Density Fraction Contains a Homolog of the Drosophila Disks-Large Tumor Suppressor Protein, Neuron 9, 929-942. 12. Banerjee, M., Huang, C., Marquez, J., and Mohanty, S. (2008) Probing the structure and function of human glutaminase-interacting protein: A possible target for drug design, Biochemistry 47, 9208-9219. 13. Wiedemann, U., Boisguerin, P., Leben, R., Leitner, D., Krause, G., Moelling, K., Volkmer-Engert, R., and Oschkinat, H. (2004) Quantification of PDZ domain specificity, prediction of ligand affinity and rational design of superbinding peptides, Journal of Molecular Biology 343, 703-718. 14. Tonikian, R., Zhang, Y. N., Sazinsky, S. L., Currell, B., Yeh, J. H., Reva, B., Held, H. A., Appleton, B. A., Evangelista, M., Wu, Y., Xin, X. F., Chan, A. C., Seshagiri, S., Lasky, L. A., Sander, C., Boone, C., Bader, G. D., and Sidhu, S. S. (2008) A specificity map for the PDZ domain family, Plos Biology 6, 2043-2059. 15. Durney, M. A., Birrane, G., Anklin, C., Soni, A., and Ladias, J. A. A. (2009) Solution structure of the human Tax-interacting protein-1, Journal of Biomolecular Nmr 45, 329- 334. 198 16. Medina, M. A., Sanchezjimenez, F., Marquez, J., Quesada, A. R., and Decastro, I. N. (1992) Relevance of Glutamine-Metabolism to Tumor-Cell Growth, Molecular and Cellular Biochemistry 113, 1-15. 17. Brand, K. (1985) Glutamine and Glucose-Metabolism during Thymocyte Proliferation - Pathways of Glutamine and Glutamate Metabolism, Biochemical Journal 228, 353-361. 18. Kovacevic, Z., and Mcgivan, J. D. (1983) Mitochondrial Metabolism of Glutamine and Glutamate and Its Physiological Significance, Physiological Reviews 63, 547-605. 19. Kanamori, M., Sandy, P., Marzinotto, S., Benetti, R., Kai, C., Hayashizaki, Y., Schneider, C., and Suzuki, H. (2003) The PDZ protein tax-interacting protein-1 inhibits beta-catenin transcriptional activity and growth of colorectal cancer cells, Journal of Biological Chemistry 278, 38758-38764. 20. Saras, J., Engstrom, U., Gonez, L. J., and Heldin, C. H. (1997) Characterization of the interactions between PDZ domains of the protein-tyrosine phosphatase PTPL1 and the carboxyl-terminal tail of Fas, Journal of Biological Chemistry 272, 20979-20981. 21. Hampson, L., Li, C. G., Oliver, A. W., Kitchener, H. C., and Hampson, I. N. (2004) The PDZ protein Tip-1 is a gain of function target of the HPV16 E6 oncoprotein, International Journal of Oncology 25, 1249-1256. 22. Reynaud, C., Fabre, S., and Jalinot, P. (2000) The PDZ protein TIP-1 interacts with the Rho effector rhotekin and is involved in Rho signaling to the serum response element, Journal of Biological Chemistry 275, 33962-33968. 23. Le Maout, S., Welling, P. A., Brejon, M., Olsen, O., and Merot, J. (2001) Basolateral membrane expression of a K+ channel, Kir 2.3, is directed by a cytoplasmic COOH- 199 terminal domain, Proceedings of the National Academy of Sciences of the United States of America 98, 10475-10480. 24. Alewine, C., Olsen, O., Wade, J. B., and Welling, P. A. (2006) TIP-1 has PDZ scaffold antagonist activity, Molecular Biology of the Cell 17, 4200-4211. 25. Olalla, L., Gutierrez, A., Jimenez, A. J., Lopez-Tellez, J. F., Khan, Z. U., Perez, J., Alonso, F. J., de la Rosa, V., Campos-Sandoval, J. A., Segura, J. A., Aledo, J. C., and Marquez, J. (2008) Expression of the scaffolding PDZ protein glutaminase-interacting protein in mammalian brain, Journal of Neuroscience Research 86, 281-292. 26. Fields, S., and Song, O. K. (1989) A Novel Genetic System to Detect Protein Protein Interactions, Nature 340, 245-246. 27. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) Nmrpipe - a Multidimensional Spectral Processing System Based on Unix Pipes, Journal of Biomolecular Nmr 6, 277-293. 28. Kneller, T. D. G. a. D. G. SPARKY 3, University of California, San Francisco. 29. Johnson, P. E., Tomme, P., Joshi, M. D., and McIntosh, L. P. (1996) Interaction of soluble cellooligosaccharides with the N-terminal cellulose-binding domain of Cellulomonas fimi CenC .2. NMR and ultraviolet absorption spectroscopy, Biochemistry 35, 13895-13906. 30. Filiou, M. D., Varadarajulu, J., Teplytska, L., Reckow, S., Maccarrone, G., and Turck, C. W. (2012) The 15N isotope effect in Escherichia coli: a neutron can make the difference, Proteomics 12, 3121-3128. 200 31. Sreerama, N., and Woody, R. W. (2000) Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set, Anal Biochem 287, 252-260. 32. Yan, X. J., Zhou, H., Zhang, J. X., Shi, C. W., Xie, X. Q., Wu, Y. N., Tian, C. L., Shen, Y. Q., and Long, J. F. (2009) Molecular Mechanism of Inward Rectifier Potassium Channel 2.3 Regulation by Tax-Interacting Protein-1, Journal of Molecular Biology 392, 967-976. 33. Mao, H. Y., Hajduk, P. J., Craig, R., Bell, R., Borre, T., and Fesik, S. W. (2001) Rational design of diflunisal analogues with reduced affinity for human serum albumin, Journal of the American Chemical Society 123, 10429-10435. 34. Katre, U. V., Mazumder, S., Prusti, R. K., and Mohanty, S. (2009) Ligand Binding Turns Moth Pheromone-binding Protein into a pH Sensor EFFECT ON THE ANTHERAEA POLYPHEMUS PBP1 CONFORMATION, Journal of Biological Chemistry 284, 32167-32177. 35. Hu, H. Y., Horton, J. K., Gryk, M. R., Prasad, R., Naron, J. M., Sun, D. A., Hecht, S. M., Wilson, S. H., and Mullen, G. P. (2004) Identification of small molecule synthetic inhibitors of DNA polymerase beta by NMR chemical shift mapping, Journal of Biological Chemistry 279, 39736-39744. 36. Harris, B. Z., and Lim, W. A. (2001) Mechanism and role of PDZ domains in signaling complex assembly, Journal of Cell Science 114, 3219-3231. 37. Jemth, P., and Gianni, S. (2007) PDZ domains: Folding and binding, Biochemistry 46, 8701-8708. 201 38. Schultz, J., Hoffmuller, U., Krause, G., Ashurst, J., Macias, M. J., Schmieder, P., Schneider-Mergener, J., and Oschkinat, H. (1998) Specific interactions between the syntrophin PDZ domain and voltage-gated sodium channels, Nature Structural Biology 5, 19-24. 39. Jeong, B. C., Kim, M. Y., Lee, J. H., Kee, H. J., Kho, D. H., Han, K. E., Qian, Y. R., Kim, J. K., and Kim, K. K. (2006) Brain-specific angiogenesis inhibitor 2 regulates VEGF through GABP that acts as a transcriptional repressor, Febs Letters 580, 669-676. 40. Okajima, D., Kudo, G., and Yokota, H. (2010) Brain-specific angiogenesis inhibitor 2 (BAI2) may be activated by proteolytic processing, Journal of Receptors and Signal Transduction 30, 143-153. 41. Oda, K., Shiratsuchi, T., Nishimori, H., Inazawa, J., Yoshikawa, H., Taketani, Y., Nakamura, Y., and Tokino, T. (1999) Identification of BAIAP2 (BAI-associated protein 2), a novel human homologue of hamster IRSp53, whose SH3 domain interacts with the cytoplasmic domain of BAI1, Cytogenetics and Cell Genetics 84, 75-82. 42. Stiffler, M. A., Grantcharova, V. P., Sevecka, M., and MacBeath, G. (2006) Uncovering quantitative protein interaction networks for mouse PDZ domains using protein microarrays, J Am Chem Soc 128, 5913-5922. 202 Chapter 5 Characterization of subunit A of Heterodisulfide Reductase (HdrA) from Methanothermobacter marburgensis 5.1 Introduction 5.1.1 Electron bifurcation In a recently discovered process, it has been found that enzymes can produce electrons with a very low redox potential without the involvement of ATP-hydrolysis or radical-SAM enzymes. This process has been termed electron bifurcation. In this process, two electrons enter the bifurcation cycle at a certain redox potential. Of these two electrons, one comes out at a much lower potential and the other comes out at a much higher potential. Although, these electrons are physically separated, there is a tight coupling in this process. To produce one type of electron the other needs to be simultaneously generated while at the same time they continue on separate electron paths in the enzyme complex. One electron stays on a high-potential branch and the other on a low-potential branch. 5.1.2 History of electron bifurcation Buckel et al. first developed the concept of electron bifurcation while trying to explain their observations on enzymes from Clostridia(1, 2). This concept was then adopted by Thauer et al. for methanogens, specifically for Methanothermobacter marburgensis(3, 4). In this organism, they identified the methylviologen-reducing hydrogenase (Mvh)/heterodisulfide reductase (Hdr) 203 as the enzyme complex (MvhADG/HdrABC) that performs the electron bifurcation (Figure 5.1) (4). A similar complex was also discovered in Methanobacterium thermoautotrophicum by this group (5). Independently, the phenomenon of electron bifurcation was also successfully identified by the Leigh group in Methanococcus maripaludis(6). 204 Figure 5.1: Model of the structure of the hydrogenase:heterodisulfide reductase complex from Methanothermobacter marburgensis. 205 5.1.3 Mechanism of electron bifurcation For the electron bifurcation to take place, it seems to be essential that at least one flavin or quinone molecule is present. In the Hdr complexes, it has therefore been proposed that only the site of the FAD (flavin adenine dinucleotide) molecule is the site where the electron bifurcation takes place. FAD is very labile and easily lost from the protein complex. However, it is also easily reconstituted back into the enzyme; making it easy to prove the essentiality of the presence of FAD for this process. Thauer and coworkers proposed a model for the electron bifurcation based on the fact that flavoproteins (FP) can exhibit three different redox potentials, namely an Eo? for the FP/FPH2 couple (n = 2), an Eo? for the FP/FPH? couple (n = 1), and an Eo? for the FPH?/FPH2 couple (n = 1). Eo? (FP/FPH?) is generally more positive and Eo? (FPH?/FPH2) more negative than Eo? (FP/FPH2). In the proposed model, the flavin is reduced by two electrons to the FPH2 form with an intermediate potential. Subsequently it first forms FPH?, releasing the low-potential electron, followed by oxidation to FP, releasing the high-potential electron. In M. marburgensis, the hydrogenase:heterodisulfide reductase complex reduces heterodisulfide at a very low rate by using electrons from the oxidation of hydrogen. This activity of heterodisulfide reductase increases many fold when ferredoxin is added to the kinetic assay. This increase in the enzyme activity lies in the fact of the tight coupling of the ferredoxin reduction and the heterodisulfide reduction during the events of the bifurcation process. The midpoint potential of the H2/H+ couple is about -400 mV under cell growth conditions whereas the midpoint potential of the heterodisulfide/(HS-CoM + HS-CoB) couple is about -140 mV. Under this condition, the expected flow of electrons should be automatically from the site of 206 hydrogen oxidation to that of heterodisulfide reduction. But, as reflected by the enzyme activity assay, it does not constitute the major process unless ferredoxin is simultaneously reduced despite the fact that ferredoxinRED/ferredoxinOX couple has a midpoint potential of -500 mV. Thus, high-potential electrons are generated that are used for heterodisulfide reduction, while low-potential electrons are generated that only reduce ferredoxin. Thermodynamically, these latter electrons could also reduce heterodisulfide, but apparently the enzyme prevents this from happening. 5.1.4 Electron bifurcation in other systems It is remarkable that the electrons that reduce ferredoxin have a lower potential than those released by hydrogen oxidation. ATP hydrolysis coupled to an electron transfer step is the more classic way for an enzyme to change redox potentials. Typical examples are the Fe-protein in the nitrogenase systems and archerases (7). Electron bifurcation is widespread in nature, however, in particular, it is found in the electron transport chain. In complex I of the oxidative phosphorylation pathway, NADH delivers two electrons and a proton to an FMN molecule that is bound to the protein. The FMN donates each electron to a separate iron-sulfur cluster, but the two pathways combine into a single path that reaches to the quinone reduction site. Bifurcation of electrons happens at complex III during the oxidation of ubiquinol (QH2). One electron ends up at cytochrome c whereas the other electron follows a path containing cytochrome bH and cytochrome bL and ends up reducing another quinone molecule. Through crystallographic study, the Rieske 2Fe cluster of complex III is proposed to play an important role in the bifurcation by 207 inducing a conformational change upon oxidation or reduction causing the cluster to change its position relative to the position of the cytochrome c. 5.1.5 Models for electron bifurcation Based on the above scenario in the oxidative phosphorylation chain, three models can be postulated to describe the bifurcated flow of electrons in our enzyme system (Figure 5.2). 5.1.5.1 Model I In model I, two electrons are transferred to the FAD. The resultant change in charge causes it to move closer (at least the flavin part) to the high-potential [4Fe-4S] cluster in the same subunit (HdrA). When it releases one electron to the cluster, the now ?red-hot? FADH then moves toward the bound ferredoxin to give up its second electron and be ready to accept two new electrons. 5.1.5.2 Model II In model II, when one electron is transferred to the high-potential [4Fe-4S] cluster, the cluster moves away from the flavin site forcing the next electron to be transferred to the bound ferredoxin. 208 5.1.5.3 Model III In model III, once the iron-sulfur cluster is reduced, the subsequent electron transfers to the heterodisulfide reduction site is so slow that even the highly reactive ?red hot? semiquinone FADH? state is not able to transfer the second electron to the cluster since that would create a [4Fe-4S]0 state, but instead it just has to transfer the electron to the bound ferredoxin. The ?0? state is generally not attainable for 4Fe clusters, since the midpoint potential for the 1+/0 couple is very low. In this model, no movement is essential. 209 Figure 5.2: Models for electron bifurcation. 210 5.2 Objective of this study To get a detailed understanding of the electron bifurcation process in the heterodisulfide reductase enzyme found in methanogenic Archaea we might have to study the whole hydrogenase:heterodisulfide reductase complex or just the heterodisulfide reductase. However to obtain basic information about this process it is important to obtain the smallest protein component or complex that still contains all the essential cofactors needed for electron bifurcation. The major reason behind this is more of a practical aspect rather than a conceptual understanding. The dominant method to be used to characterize the various redox processes involved during the bifurcation and to characterize the role of electron donors and acceptors is electron paramagnetic resonance (EPR) spectroscopy. Therefore there is an inherent requirement for simplicity of the system (less cofactors/iron-sulfur clusters) under investigation. Otherwise, in investigations with the larger enzyme complexes it would be harder (if not impossible) to interpret the EPR spectroscopic data. Also, establishment of the smallest yet functional subunit system should essentially establish the least but absolutely stringent requirements of the parts of the complex to be present for the electron bifurcation process to take place. As a result, such understanding should support or disprove our hypothesis behind this process. So far, the spectroscopic data that is available, is all from either the hydrogenase:heterodisulfide reductase complex or the Hdr from M. marburgensis (8-11). Therefore our first focus was on the Hdr enzyme from this organism. The bifurcation has not been proven for the Hdr enzyme. The hydrogenase:heterodisulfide reductase complex can be purified using a three column-step purification method (4). The two enzymes can be separated 211 and the Hdr purified in another three column steps. To study the events that take place during the bifurcation process, it would be necessary to do mutational studies. Our collaborator John Leigh at the in University of Washington is well-equipped for work with the organism M. maripaludis. They already have a well-developed set of genetic tools available for this organism. Therefore it was also tested if the M. marburgensis Hdr can be overexpressed in M. maripaludis. This, however, was not possible, since the M. maripaludis strains expressing the Hdr subunits would not grow. However, the HdrA subunit by itself was successfully overexpressed. The purification and initial characterization of this subunit is described here. If the HdrA only shows activity inside the completely folded enzyme we have to depend on the M. Maripaludis Hdr for site-directed mutagenesis. The Hdr in this organism is part of an even larger complex. When expressed with a His6-tag on the HdrB subunit, the full complex from M. maripaludis can be obtained in a single purification step using a Ni-NTA column. The full complex obtained this way contains heterodisulfide reductase (Hdr), hydrogenase (Vhu), formylmethanofuran dehydrogenase (Fwd) and formate dehydrogenase (Fdh). Even the polyferredoxin is part of this complex (FwdF subunit). There exist methods, however, to simplify the enzyme complex. When the cells are grown with H2 as the electron source the Fdh is no longer part of the complex (HDR/Vhu/Fwd). We recently found that the Fwd enzyme is lost when an additional size-exclusion column step is performed. When formate is used as an electron source the hydrogenase is absent (HDR/Fwd/Fdh). It has not been tried yet to see if the HDR will separate from the Fwd/Fdh components. 212 The data presented here shows our first efforts in trying to see what enzyme complexes can be obtained and is very much a work in progress. 5.3 Materials and methods The research work described here was carried out in the laboratory of Dr. Evert Duin. For all the experiments anaerobic conditions are required. To achieve this, all purification steps, sample handling and experiments were done in a glove box (Coy Laboratory Products, Inc., Grass Lake, USA) filled with a gas mixture consisting of 95% N2and 5% H2. Also, all buffers and solutions used in the procedures were degassed by boiling them under a nitrogen or argon atmosphere and subsequent cooling down under vacuum for 2 to 12 hours followed by overnight equilibration inside the glove box. In most cases, the buffers were filtered with 0.45 ?m Millipore filter to remove particles that might affect the columns to be used for protein purification. 5.3.1 Purification of hydrogenase:heterodisulfide reductase complex (MvhADG/HdrABC) from M. marburgensis 5.3.1.1 Growth of M. marburgensis cells M. marburgensis was grown at 65 ?C in a 13 L glass fermenter (New Brunswick) containing 10 L of growth medium. The growth medium (12) contained 65 mM KH2PO4, 50 mM NH4Cl, 30 mM Na2CO3, 0.5 mM nitrilotriacetic acid, 2 mM MgCl2, 50 ?M FeCl2, 1 ?M CoCl2, 1 ?M Na2MoO4, 5 ?M NiCl2, and 20 ?M resazurin. It was made anaerobic by gassing with 80% 213 H2/20% CO2/0.1% H2S at a rate of 1,200 ml/min. The resazurin was added as an indicator to the medium so that change in the color of the medium would indicate when sufficient anaerobic conditions were reached. After 1-2 hour of equilibration, when the optimum temperature and anaerobic condition was reached, the medium was inoculated with about 200 ml of fresh cell culture. The medium was agitated at 1000 rpm. After about 13 hour of incubation, at a ?OD568 of ~4.5, the cells were harvested. 5.3.1.2 Harvest and sonication of M. marburgensis cells The cells were harvested anaerobically by centrifugation at 15,000 rpm using a flow- through centrifuge (Hettich, contifuge 17 RS). The rotor was brought into the anaerobic tent. The cells were suspended in buffer A containing 50 mM Tris/HCl at pH 7.6, 2 mM DTT (Dithiothreitol), 2 mM CoM-SH (Coenzyme M), and 20 ?M FAD. The suspended cells were then sonicated on ice 3 times for a total of 7 min (pulsing for 0.5 seconds). The cells were allowed to cool down in between the runs for a couple of minutes and at the end of the procedure. The sonicated cells were then centrifuged anaerobically at 35000 rpm for 20 minutes. The supernatant was carefully decanted into a beaker equilibrated inside the anaerobic tent. 5.3.1.3 Purification of hydrogenase:heterodisulfide reductase complex (MvhADG/HdrABC) According to the protocol of Thauer et al. (4), the supernatant was applied to a DEAE- Sepharose column equilibrated with buffer A containing 50 mM Tris/HCl pH 7.6. According to the protocol, a NaCl step gradient was used in buffer A: 100 mL 0 M NaCl, 100 mL 0.2 M NaCl, 214 100 mL 0.3 M NaCl, and 100 mL 0.4 M NaCl. The last peak was collected and applied to a Q- Sepharose column equilibrated with buffer A. Again, a NaCl step gradient was used in buffer A: 100 mL 0 M NaCl, 100 mL 0.3 M NaCl, 100 mL 0.4 M NaCl, 100 mL 0.45 M NaCl, and 100 mL 0.54 M NaCl. The last peak was collected and concentrated by filtration using 10kDa filter to 2?3 mL, which was then applied to a Superdex 200 column equilibrated with buffer B (buffer A + 150 mM NaCl). The different fractions collected from this run were then analyzed using a 15% SDS PAGE gel. 5.3.2 Purification of heterodisulfide reductase (Hdr) from M. maripaludis cells 5.3.2.1 M. maripaludis cells The Leigh group at the University of Washington has graciously supplied us wild type M. maripaludis cells. 5.3.2.2 Purification of heterodisulfide reductase The M. maripaludis cells were sonicated on ice 3 times for a total of 7 min (pulsing for 0.5 seconds). The cells were allowed to cool down in between the runs for a couple of minutes and at the end of the procedure. The sonicated cells were then centrifuged anaerobically at 35000 rpm for about 20 minutes. The supernatant was carefully decanted into a beaker equilibrated inside the anaerobic tent. The supernatant was then applied to a nickel column equilibrated with buffer A containing 25 mM HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) pH 7.5, 10 mM sodium dithionite, 100 mM NaCl and 10 mM imidazole. The protein was eluted by 215 washing the column with buffer B containing 25 mM HEPES, 10 mM sodium dithionite, 100 mM NaCl and 100 mM imidazole. The fractions were analyzed both with 17% SDS-PAGE and 8% native PAGE. The concentration of the protein content was determined using the method of Bradford with bovine serum albumin (Serva) as standard (13, 14). Also, just to check, EPR measurements were done on the complex at 77 K and 20 dB. To further purify the protein, the eluted protein was concentrated to about 2 mL by filtration using a 10 kDa filter paper and applied to a Superdex 200 column equilibrated with buffer containing 30 mM Tris-HCl pH 8.0 and 100 mM NaCl. The protein was eluted with the same buffer. The major peaks of the chromatography profile were analyzed with 17% SDS-PAGE and 8% native PAGE. 5.3.3 Purification of HdrA from M. maripaludis HdrAmarburgensis cells 5.3.3.1 M. maripaludis HdrAmarburgensis cells The Leigh group at the University of Washington has graciously supplied us with the M. maripaludis strains with the M. marburgensis HdrA gene. 5.3.3.2 Purification of HdrA The M. maripaludis cells were sonicated on ice 3 times for a total of 7 min (pulsing for 0.5 seconds). The cells were allowed to cool down in between the runs for a couple of minutes and at the end of the procedure. The sonicated cells were then centrifuged anaerobically at 35000 rpm for about 20 minutes. The supernatant was carefully decanted into a beaker equilibrated inside the anaerobic tent. The supernatant was then applied to a nickel column equilibrated with 216 buffer A containing 50 mM Tris-HCl pH 7.6 and 100 mM NaCl. The protein was eluted by washing the column with buffer B containing 50 mM Tris-HCl pH 7.6, 100 mM NaCl and 500 mM imidazole. The fractions were analyzed with 12% SDS-PAGE. The major peak was considered to be the peak of our interest. Thus, to check the concentration of the protein content of this peak the Bradford method was used with bovine serum albumin (Serva) as standard (13, 14). Also, to get the iron content of the protein fraction, a rapid colorimetric method was used (15) (see below). In addition, EPR measurements were done on the reduced protein sample. To further purify the protein, the eluted fraction was applied to a Q-Sepharose column equilibrated with buffer A containing 50 mM Tris-HCl pH 7.6. To elute the protein, a NaCl step gradient was used in the buffer A: 100 mL 0 M NaCl, 100 mL 0.3 M NaCl, 100 mL 0.4 M NaCl, 100 mL 0.45 M NaCl, and 100 mL 0.54 M NaCl. The peak of the interest was collected and concentrated by filtration using a 10 kDa filter to about 1 mL, which were then applied to a Superdex 200 column equilibrated with buffer B (buffer A + 150 mM NaCl). The different major fractions collected from each of the purification step were analyzed with 12% SDS PAGE. 5.3.4 Iron determination The iron standard was prepared using 0.0523 M of ferrous ethylenediammonium sulfate in 0.01 M HCl for the calibration curve. 0.25 mL of freshly prepared iron releasing reagent which contained 0.6 M HCl and 0.142 M potassium permanganate (KMnO4) were added to 0.5 mL of the protein sample and the standards. The digested mixture was incubated in a capped tube for 2 hours at 60?C. Following the digestion, 0.1 mL of reducing, iron chelating reagent which contained 6.5 mM ferrozine (disodium 3-(2-pyridyl)-5,6-bis(4-phenyl sulfonate)-1,2,4- triazine), 13.1 mM neocuprine (2,9-dimethyl-1,10-phenanthroline), 2 M ascorbic acid and 5 M 217 ammonium acetate were added to the digested mixture and mixed. The solution was left to stand at room temperature for at least 30 min. After this, the absorbance of standards and protein samples were measured at 562 nm. A standard curve was constructed by plotting the concentration of the standard versus their absorbance. From this curve, the concentration of the iron content was calculated. 5.3.5 UV-vis absorption analysis The UV-vis absorption spectra of the protein samples were recorded under anaerobic conditions by using stoppered cuvettes in an Agilent 8453 UV-visible Spectrophotometer. To check the iron-sulfur cluster signal, absorbance in the 410-420 nm region was observed. 5.3.6 EPR measurements CW EPR spectra were measured at X-band (9 GHz) frequency on a Bruker EMX spectrometer, fitted with the ER-4119-HS high sensitivity perpendicular-mode cavity. General EPR conditions were: microwave frequency, 9.385 GHz; microwave power incident to the cavity, 0.20 mW; field modulation frequency, 100 kHz; microwave amplitude, 0.6 mT. The Oxford Instrument ESR 900 flow cryostat in combination with the ITC4 temperature controller was used for measurements using a helium flow. Samples for EPR were prepared in quartz tubes that were sealed with a closed off rubber tube. The samples were frozen using liquid nitrogen. 218 5.4 Results 5.4.1 Purification of the hydrogenase:heterodisulfide reductase complex (MvhADG/HdrABC) from M. marburgensis M. marburgensis cell supernatant was applied to DEAE-Sepharose column. The last peak (fraction no. 44-48) of the chromatography profile (Figure 5.3) was collected and applied to a Q- Sepharose column. The last peak (fraction no. 60-68) of its chromatography profile (Figure 5.4) was collected. The fractions of this peak were pooled together and applied to a Superdex 200 column. Fraction no. 18-25 was collected (Figure 5.5). Each of these fractions was then analyzed by 15% SDS-PAGE gel. From the gel (Figure 5.6), it appears that, fractions 19, 20 and 21 contain the MvhADG/HdrABC complex but that there are still other impurities present. From the chromatograph (Figure 5.5), it was also evident that the protein complex was present as a shoulder just in front of the most intense peak. 219 Figure 5.3: Chromatography profile of DEAE-Sepharose column for purification of MvhADG/HdrABC complex. D EAE g ra d ie n t H D R 0 0 1 :1 _ U V D EAE g ra d ie n t H D R 0 0 1 :1 _ C o n c D EAE g ra d ie n t H D R 0 0 1 :1 _ F ra ct io n s D EAE g ra d ie n t H D R 0 0 1 :1 _ L o g b o o k 5 0 0 1000 1500 2000 2500 3000 3500 4000 mAU 0 100 200 300 400 ml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 W a st e 220 Figure 5.4: Chromatography profile of Q-Sepharose column for purification of MvhADG/HdrABC complex. Q se p h H D R 0 0 1 :1 _ U V Q se p h H D R 0 0 1 :1 _ C o n c Q se p h H D R 0 0 1 :1 _ F ra ct io n s Q se p h H D R 0 0 1 :1 _ L o g b o o k 0 5 0 0 1000 1500 2000 2500 3000 mAU 0 100 200 300 400 ml 1 2 3 4 5 6 7 8 9 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 W a st e 221 Figure 5.5: Chromatography profile of Superdex 200 column for purification of MvhADG/HdrABC complex. Su p e rd e x H D R 0 0 1 :1 _ U V Su p e rd e x H D R 0 0 1 :1 _ F ra ct io n s Su p e rd e x H D R 0 0 1 :1 _ L o g b o o k 0 5 0 0 1000 1500 2000 2500 mAU 0 5 0 100 150 ml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 222 Figure 5.6: 15%SDS-PAGE analysis of the fractions from Superdex 200 column for MvhADG/HdrABC complex purification. Lane 1-8 corresponds to the fractions 18-25 in order. Lane MW is for the protein marker. The probable bands of fraction 19, 20 and 21 (lane 2, 3 and 4) for the subunits of the complex are enclosed with red rectangle and the corresponding names of the subunits are given in black bold letters to the left of the rectangles. MW 1 2 kDa 3 4 5 6 7 8 97 66 45 31 21 14 HdrA MvhA MvhG HdrB HdrC MvhD 223 5.4.2 Purification of heterodisulfide reductase (Hdr) from M. maripaludis cells The cell supernatant was applied to the nickel column and the protein was eluted using an imidazole gradient. The concentration of the protein content of this sample was determined using the Bradford method and was found to be 8.67 mg/mL. When EPR was used to check the iron- sulfur cluster signal of this protein complex, the signal (Figure 5.8) did not resemble any of the standard type signals (Figure 5.7). The obvious reason behind this is that the enzyme complex contains multiple clusters, a molybdenum/tungsten site and a nickel site since the Hdr is part of a multimeric complex (Hdr/Vhu/Fwd/Fdh). To isolate Hdr from this complex and as already tested by the Leigh group, size-exclusion chromatography was performed. In the chromatograph, the complex appeared to be separated into three major peaks (Figure 5.9). Fractions 22-23, fractions 24-25 and fractions 26-27 were collected separately. When these fractions were analyzed with SDS-PAGE (Figure 5.10) and native PAGE (Figure 5.11), it was not possible to identify the bands that belonged to the Hdr enzyme. Some of the more promising bands were also analyzed using mass spectrometry, but the obtained sequences did not correspond with those of the Hdr subunits. 224 Figure 5.7: EPR spectra of iron-sulfur clusters. Adapted from reference (16, 17). 225 Figure 5.8: EPR spectrum of Hdr complex from M. maripaludis. Magnetic field (Gauss) 226 Figure 5.9: Chromatography profile of Superdex 200 column for purification of Hdr from M. maripaludis. Su p e rd e x BI F ru n 0 0 1 :1 _ U V Su p e rd e x BI F ru n 0 0 1 :1 _ F ra ct io n s Su p e rd e x BI F ru n 0 0 1 :1 _ L o g b o o k 0 5 0 100 150 200 250 300 mAU 0 5 0 100 150 200 ml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 W a st e 227 Figure 5.10: 17% SDS-PAGE analysis of the Hdr complex from M. maripaludis cells. Lane 1: Cell extract, Lane 2: Flow through from the nickel column, Lane 3: Protein sample eluted from the nickel column, Lane 4: Fraction no. 22-23 from Superdex 200 column, Lane 5: Fraction no. 24-25 from Superdex 200 column and Lane 6: Fraction no. 26-27 from Superdex 200 column. The lane MW is for protein marker. 66 200 45 116 97 31 21 14 6 MW 1 2 kDa 3 4 5 6 228 Figure 5.11: 8% native PAGE analysis of the Hdr complex from M. maripaludis cells. Lane 1: Cell extract, Lane 2: Flow through from the nickel column, Lane 3: Protein sample eluted from the nickel column, Lane 4: Fraction no. 22-23 from Superdex 200 column, Lane 5: Fraction no. 24-25 from Superdex 200 column and Lane 6: Fraction no. 26-27 from Superdex 200 column. The lane MW is for protein marker. The overlapping bands on the consecutive lanes are enclosed by red rectangles. MW 1 2 kDa 3 4 5 6 1236 1048 480 242 146 229 5.4.3 Purification of HdrA from M. maripaludis HdrAmarburgensis cells 5.4.3.1 Purification of HdrA The cell supernatant was applied to a nickel column and the protein was eluted using an imidazole gradient. The major peak (fraction no. 9-13) was collected (Figure 5.12). This fraction was analyzed with 12% SDS-PAGE (Figure 5.13). From the gel, it could be seen that, the protein sample is about 70% pure. The HdrA subunit has an estimated weight of 71 kDa. Further attempts (Q-Sepharose and Superdex 200 column) to purify this protein were not successful. The HdrA band could not be detected on SDS-PAGE after these purification steps. 230 Figure 5.12: Chromatography profile of nickel column for purification of HdrA from M. maripaludis HdrAmarburgensis. H d rA N i H is R u n 0 0 2 :1 _ U V H d rA N i H is R u n 0 0 2 :1 _ C o n c H d rA N i H is R u n 0 0 2 :1 _ F ra ct io n s H d rA N i H is R u n 0 0 2 :1 _ L o g b o o k 0 2 0 0 4 0 0 6 0 0 8 0 0 1000 1200 mAU 0 2 0 4 0 6 0 8 0 100 ml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 W a st e 231 Figure 5.13: 12% SDS-PAGE analysis of the HdrA sample from M. maripaludis HdrAmarburgensis cells. Lane 1: Cell extract, Lane 2: Flow through from the nickel column, Lane 3: Protein sample eluted from the nickel column. The lane MW is for protein marker. The band for HdrA protein is enclosed by a red rectangle. MW 1 2 kDa 3 97 66 45 31 21 232 5.4.3.2 Protein and Iron Determination The concentration of the protein sample eluted from the nickel column was determined by using Bradford method and the concentration was determined to be 2.27 mg/mL or 31.79 ?M (considering the molecular weight of HdrA as 71 kDa). Using the colorimetric method, the iron concentration of the sample was found to be about 258 ?M. The EPR measurements (Figure 5.15) are in line with the presence of [4Fe-4S] cluster. Therefore the cluster content is about 2. 5.4.3.3 UV-vis absorption of the protein sample Cubane iron-sulfur clusters display an absorption band at around 410-420 nm for the oxidized form. For HdrA an absorbance was detected in this region (Figure 5.14). When the protein sample was reduced by dithionite, this signal is lowered and after about three minutes, the signal is completely gone (Figure 5.14). The disappearance of the cluster signal upon reduction could either be due to the reduction itself or disintegration of the cluster. Since an EPR signal was obtained for the reduced protein the former appears to be the case. 233 Figure 5.14: UV-vis absorption of the HdrA protein sample. Black arrow indicates the absorption of the iron-sulfur cluster at around 410 nm. Wavelength (nm) Absor ba nc e HdrA as such HdrA reduced - 0 min HdrA reduced - 3 min 234 5.4.3.4 EPR measurement of the HdrA protein sample EPR spectra of the HdrA protein sample in its reduced condition were recorded at different microwave powers: 20 dB, 30 dB, 40 dB and 50 dB (Figure 5.15). All these spectra were recorded at 8 K after the optimization of the temperature for the enhancement of the signal. From the figure, it was apparent that, the EPR signal for the HdrA protein sample is achieved optimally at 8 K and 30 dB. Under these optimal conditions, the EPR signal is comparable to that of the standard [4Fe-4S] cluster (Figure 5.7). 235 Figure 5.15: EPR spectra of HdrA protein sample at the temperature of 8 K and at different microwave frequencies. Magnetic field (Gauss) 236 5.5 Conclusions and future direction The mechanism of electron bifurcation is still unknown. Understanding its mechanism and discovering its existence in the heterodisulfide reductase enzyme complex system in methanogenic archaea would provide the proof of the important role of electron bifurcation in the anaerobic energy transduction. Also, such understanding would also open the doors of harnessing the power of anaerobic energy conversion. Methane production and methane activation processes are performed by methanogenic and methylotrophic Archaea. In this work, we were able to purify (70%) HdrA, a subunit of the Hdr enzyme from M. marburgensis. From preliminary data, it is observed that possibly at least two iron-sulfur clusters were present in the protein sample. Sequence data, however indicates that there could be 4 clusters present in this subunits. Further purification should bring this number up. Additional reconstitution procedures can also be performed. With the availability of the almost 100% pure protein in the future, it would be possible to find out the accurate type and species of iron-sulfur clusters. Also, such pure protein would allow us to investigate the mechanism of electron bifurcation through different approaches such as redox titration, structural characterization and freeze-quench study. This would also eventually prove whether such a subunit is sufficient to necessarily carry out the electron bifurcation or a larger unit of the complex is needed to successfully yield the desired output. For structural characterization, a pure enzyme in high enough amounts should be prepared to do the X-ray crystallization of the protein. From the structural information, it would be possible to derive the specific sites of the protein that are involved in this mechanism and how they are involved. However, it is also important to note 237 that, there is always a possibility of not having a crystal of the protein good enough to get the necessary structural information of the protein. Thus, mutational study can be done in the absence of structural information. Since, genetic tools for the mutational study are readily available for the organism M. maripaludis, it is worth continuing the efforts to purify the Hdr from its complex of the wild type M. maripaludis. Here, we were able to isolate three different parts of the Hdr complex but without being able to identify them. In the future, efforts should be made to purify Hdr from this complex by varying the conditions or increasing the steps of the chromatography and utilizing mass spectrophotometry to determine specific bands from the gels or specific fractions from the purifications. Also, purification of the Hdr from M. marburgensis could prove worthwhile in the future; especially if it is found out that HdrA subunit is not sufficient to be functional and the whole protein is needed. Here, we were able to obtain the MvhADG/HdrABC complex with impurities. In the future, at each step of purification, H2:CoM-S-S-CoB oxidoreductase activity should be measured for each fraction, so that collection of the fractions can be more accurate. A hydrograph for measuring H2 concentrations has recently been purchased but not tested yet. Also, some of the chromatography conditions could be optimized to diminish the overlap of bands within the chromatography profile. 238 5.6 References 1. Li, F., Hinderberger, J., Seedorf, H., Zhang, J., Buckel, W., and Thauer, R. K. (2008) Coupled ferredoxin and crotonyl coenzyme A (CoA) reduction with NADH catalyzed by the butyryl-CoA dehydrogenase/Etf complex from Clostridium kluyveri, Journal of bacteriology 190, 843-850. 2. Herrmann, G., Jayamani, E., Mai, G., and Buckel, W. (2008) Energy conservation via electron-transferring flavoprotein in anaerobic bacteria, Journal of bacteriology 190, 784- 791. 3. Thauer, R. K., Kaster, A. K., Seedorf, H., Buckel, W., and Hedderich, R. (2008) Methanogenic archaea: ecologically relevant differences in energy conservation, Nature reviews. Microbiology 6, 579-591. 4. Kaster, A. K., Moll, J., Parey, K., and Thauer, R. K. (2011) Coupling of ferredoxin and heterodisulfide reduction via electron bifurcation in hydrogenotrophic methanogenic archaea, Proc Natl Acad Sci U S A 108, 2981-2986. 5. Setzke, E., Hedderich, R., Heiden, S., and Thauer, R. K. (1994) H2: heterodisulfide oxidoreductase complex from Methanobacterium thermoautotrophicum. Composition and properties, European journal of biochemistry / FEBS 220, 139-148. 6. Costa, K. C., Wong, P. M., Wang, T., Lie, T. J., Dodsworth, J. A., Swanson, I., Burn, J. A., Hackett, M., and Leigh, J. A. (2010) Protein complexing in a methanogen suggests electron bifurcation and electron delivery from formate to heterodisulfide reductase, Proc Natl Acad Sci U S A 107, 11050-11055. 239 7. Kim, J., Hetzel, M., Boiangiu, C. D., and Buckel, W. (2004) Dehydration of (R)-2- hydroxyacyl-CoA to enoyl-CoA in the fermentation of alpha-amino acids by anaerobic bacteria, FEMS microbiology reviews 28, 455-468. 8. Shokes, J. E., Duin, E. C., Bauer, C., Jaun, B., Hedderich, R., Koch, J., and Scott, R. A. (2005) Direct interaction of coenzyme M with the active-site Fe-S cluster of heterodisulfide reductase, FEBS Lett 579, 1741-1744. 9. Duin, E. C., Bauer, C., Jaun, B., and Hedderich, R. (2003) Coenzyme M binds to a [4Fe- 4S] cluster in the active site of heterodisulfide reductase as deduced from EPR studies with the [33S]coenzyme M-treated enzyme, FEBS Lett 538, 81-84. 10. Madadi-Kahkesh, S., Duin, E. C., Heim, S., Albracht, S. P., Johnson, M. K., and Hedderich, R. (2001) A paramagnetic species with unique EPR characteristics in the active site of heterodisulfide reductase from methanogenic archaea, European journal of biochemistry / FEBS 268, 2566-2577. 11. Duin, E. C., Madadi-Kahkesh, S., Hedderich, R., Clay, M. D., and Johnson, M. K. (2002) Heterodisulfide reductase from Methanothermobacter marburgensis contains an active- site [4Fe-4S] cluster that is directly involved in mediating heterodisulfide reduction, FEBS Lett 512, 263-268. 12. Schonheit, P., Moll, J., and Thauer, R. K. (1980) Growth-Parameters (Ks, Mu-Max, Ys) of Methanobacterium-Thermoautotrophicum, Arch Microbiol 127, 59-65. 13. Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Analytical biochemistry 72, 248-254. 14. Stoscheck, C. M. (1990) Quantitation of protein, Methods in enzymology 182, 50-68. 240 15. Fish, W. W. (1988) Rapid colorimetric micromethod for the quantitation of complexed iron in biological samples, Methods in enzymology 158, 357-364. 16. Rekittke, I., Wiesner, J., Rohrich, R., Demmer, U., Warkentin, E., Xu, W., Troschke, K., Hintz, M., No, J. H., Duin, E. C., Oldfield, E., Jomaa, H., and Ermler, U. (2008) Structure of (E)-4-hydroxy-3-methyl-but-2-enyl diphosphate reductase, the terminal enzyme of the non-mevalonate pathway, J Am Chem Soc 130, 17206-17207. 17. Cammack, R., Patil, D. S., and Fernandez, V. M. (1985) Electron-spin- resonance/electron-paramagnetic-resonance spectroscopy of iron-sulphur enzymes, Biochemical Society transactions 13, 572-578. 241 Appendix Table A-1 Chemical shift assignments for the nuclei of free GIP (BMRB entry: 17254) Residue no. Amino acid Nucleus Chemical shift 1 MET HA 4.054 1 MET HB2 2.012 1 MET HB3 1.862 1 MET HG2 2.169 1 MET HG3 2.169 1 MET HE1 1.987 1 MET HE2 1.987 1 MET HE3 1.987 1 MET C 177.041 1 MET CA 51.058 1 MET CG 34.178 1 MET CE 24.553 2 SER H 8.17 2 SER HA 4.348 2 SER HB2 3.712 2 SER HB3 3.712 2 SER C 174.058 2 SER CA 58.303 2 SER CB 63.868 2 SER N 121.533 3 TYR H 8.125 3 TYR HA 4.561 3 TYR HB2 2.956 3 TYR HB3 2.846 3 TYR HD1 7.006 3 TYR HD2 7.006 3 TYR HE1 6.738 3 TYR HE2 6.738 3 TYR C 174.731 3 TYR CA 57.946 3 TYR CB 38.884 3 TYR CD1 133.117 3 TYR CE1 118.111 3 TYR N 122.382 4 ILE H 8.015 4 ILE HA 4.274 242 4 ILE HB 1.626 4 ILE HG12 1.363 4 ILE HG13 0.993 4 ILE HG21 0.776 4 ILE HG22 0.776 4 ILE HG23 0.776 4 ILE HD11 0.737 4 ILE HD12 0.737 4 ILE HD13 0.737 4 ILE C 173.509 4 ILE CA 57.896 4 ILE CB 39.192 4 ILE CG1 26.835 4 ILE CG2 16.956 4 ILE CD1 12.588 4 ILE N 127.482 5 PRO HA 4.183 5 PRO HB2 2.244 5 PRO HB3 1.853 5 PRO HG2 1.984 5 PRO HG3 1.908 5 PRO HD2 3.62 5 PRO HD3 3.536 5 PRO C 177.404 5 PRO CA 63.439 5 PRO CB 31.974 5 PRO CG 27.338 5 PRO CD 50.87 6 GLY H 8.423 6 GLY HA2 3.953 6 GLY HA3 3.792 6 GLY C 174.082 6 GLY CA 45.146 7 GLN H 7.986 7 GLN HA 4.585 7 GLN HB2 2.045 7 GLN HB3 1.908 7 GLN HG2 2.282 7 GLN HG3 2.282 243 7 GLN HE21 7.517 7 GLN HE22 6.826 7 GLN C 173.973 7 GLN CA 53.522 7 GLN CB 28.955 7 GLN CG 33.526 7 GLN CD 180.521 7 GLN NE2 112.403 8 PRO HA 4.423 8 PRO HB2 2.225 8 PRO HB3 1.836 8 PRO HG2 1.961 8 PRO HG3 1.923 8 PRO HD2 3.75 8 PRO HD3 3.589 8 PRO C 176.778 8 PRO CA 63.117 8 PRO CB 32.034 8 PRO CG 27.453 8 PRO CD 50.58 9 VAL H 8.296 9 VAL HA 4.172 9 VAL HB 2.018 9 VAL HG11 0.893 9 VAL HG12 0.893 9 VAL HG13 0.893 9 VAL HG21 0.893 9 VAL HG22 0.893 9 VAL HG23 0.893 9 VAL C 176.386 9 VAL CA 62.342 9 VAL CB 32.781 9 VAL CG1 20.775 9 VAL N 120.566 10 THR H 8.235 10 THR HA 4.321 10 THR HB 4.157 10 THR HG21 1.124 10 THR HG22 1.124 244 10 THR HG23 1.124 10 THR C 173.499 10 THR CA 61.59 10 THR CB 69.954 10 THR CG2 21.537 10 THR N 117.806 11 ALA H 8.073 11 ALA HA 4.54 11 ALA HB1 1.262 11 ALA HB2 1.262 11 ALA HB3 1.262 11 ALA C 174.666 11 ALA CA 52.007 11 ALA CB 19.875 11 ALA N 126.558 12 VAL H 8.484 12 VAL HA 4.204 12 VAL HB 1.984 12 VAL HG11 0.852 12 VAL HG12 0.852 12 VAL HG13 0.852 12 VAL HG21 0.852 12 VAL HG22 0.852 12 VAL HG23 0.852 12 VAL C 174.99 12 VAL CA 61.618 12 VAL CB 33.104 12 VAL CG1 20.922 12 VAL N 120.498 13 VAL H 8.118 13 VAL HA 4.843 13 VAL HB 1.817 13 VAL HG11 0.769 13 VAL HG12 0.769 13 VAL HG13 0.769 13 VAL HG21 0.817 13 VAL HG22 0.817 13 VAL HG23 0.817 13 VAL C 176.141 245 13 VAL CA 60.986 13 VAL CB 33.144 13 VAL CG1 21.551 13 VAL CG2 20.761 13 VAL N 124.007 14 GLN H 9.054 14 GLN HA 4.614 14 GLN HB2 1.592 14 GLN HB3 1.769 14 GLN HG2 2.045 14 GLN HG3 2.014 14 GLN HE21 7.124 14 GLN HE22 6.719 14 GLN C 173.87 14 GLN CA 53.934 14 GLN CB 32.143 14 GLN CG 33.046 14 GLN CD 179.198 14 GLN N 124.757 14 GLN NE2 110.972 15 ARG H 8.602 15 ARG HA 5.036 15 ARG HB2 1.732 15 ARG HB3 1.618 15 ARG HG2 1.493 15 ARG HG3 1.39 15 ARG HD2 3.089 15 ARG HD3 3.089 15 ARG HE 7.172 15 ARG C 176.081 15 ARG CA 55.278 15 ARG CB 31.198 15 ARG CG 27.71 15 ARG CD 43.277 15 ARG N 123.97 16 VAL H 8.694 16 VAL HA 4.304 16 VAL HB 1.64 16 VAL HG11 0.593 246 16 VAL HG12 0.593 16 VAL HG13 0.593 16 VAL HG21 0.64 16 VAL HG22 0.64 16 VAL HG23 0.64 16 VAL C 173.773 16 VAL CA 61.22 16 VAL CB 35.353 16 VAL CG1 20.673 16 VAL CG2 21.548 16 VAL N 124.82 17 GLU H 9.16 17 GLU HA 4.939 17 GLU HB2 1.907 17 GLU HB3 1.907 17 GLU HG2 1.92 17 GLU HG3 1.764 17 GLU C 173.81 17 GLU CA 55.088 17 GLU CB 31.524 17 GLU CG 37.513 17 GLU N 129.391 18 ILE H 8.83 18 ILE HA 4.134 18 ILE HB 1.78 18 ILE HG12 1.451 18 ILE HG13 1.451 18 ILE HG21 0.779 18 ILE HG22 0.779 18 ILE HG23 0.779 18 ILE HD11 0.707 18 ILE HD12 0.707 18 ILE HD13 0.707 18 ILE C 174.777 18 ILE CA 60.643 18 ILE CB 40.096 18 ILE CG1 25.828 18 ILE CG2 17.09 18 ILE CD1 15.7 247 18 ILE N 124.847 19 HIS H 8.856 19 HIS HA 4.745 19 HIS HB2 3.157 19 HIS HB3 3.104 19 HIS HD2 7.17 19 HIS HE1 8.299 19 HIS C 173.997 19 HIS CA 54.532 19 HIS CB 28.489 19 HIS CD2 119.487 19 HIS CE1 136.51 19 HIS N 128.321 20 LYS H 8.545 20 LYS HA 4.019 20 LYS HB2 2.117 20 LYS HB3 2.117 20 LYS HG2 1.55 20 LYS HG3 1.55 20 LYS HE2 3.139 20 LYS HE3 3.139 20 LYS C 175.603 20 LYS CA 57.826 20 LYS CB 35.025 20 LYS N 123.322 21 LEU H 8.155 21 LEU HA 4.535 21 LEU HB2 2.006 21 LEU HB3 2.006 21 LEU HG 1.19 21 LEU HD11 0.885 21 LEU HD12 0.885 21 LEU HD13 0.885 21 LEU HD21 0.885 21 LEU HD22 0.885 21 LEU HD23 0.885 21 LEU C 175.742 21 LEU CA 53.469 21 LEU CB 37.528 248 21 LEU N 122.807 22 ARG H 8.739 22 ARG HA 4.728 22 ARG HB2 1.764 22 ARG HB3 1.611 22 ARG HG2 1.458 22 ARG HG3 1.458 22 ARG HD2 3.201 22 ARG HD3 3.085 22 ARG HE 7.348 22 ARG C 175.822 22 ARG CA 56.008 22 ARG CB 30.409 22 ARG CG 27.531 22 ARG CD 43.082 22 ARG N 126.951 23 GLN H 8.593 23 GLN HA 4.477 23 GLN HB2 1.935 23 GLN HB3 1.727 23 GLN HG2 2.153 23 GLN HG3 2.153 23 GLN HE21 7.468 23 GLN HE22 6.799 23 GLN C 175.431 23 GLN CA 54.863 23 GLN CB 30.434 23 GLN CG 33.911 23 GLN CD 180.314 23 GLN N 128.95 23 GLN NE2 111.693 24 GLY H 9.009 24 GLY HA2 3.597 24 GLY HA3 3.961 24 GLY C 174.958 24 GLY CA 46.981 24 GLY N 117.269 25 GLU H 9.063 25 GLU HA 4.228 249 25 GLU HB2 2.153 25 GLU HB3 1.795 25 GLU HG2 2.223 25 GLU HG3 2.178 25 GLU C 175.972 25 GLU CA 56.387 25 GLU CB 29.895 25 GLU CG 36.111 25 GLU N 126.35 26 ASN H 8 26 ASN HA 4.79 26 ASN HB2 2.824 26 ASN HB3 2.648 26 ASN HD21 7.631 26 ASN HD22 7.023 26 ASN C 173.634 26 ASN CA 52.577 26 ASN CB 41.154 26 ASN CG 176.558 26 ASN N 117.979 26 ASN ND2 114.416 27 LEU H 8.348 27 LEU HA 4.898 27 LEU HB2 1.585 27 LEU HB3 1.585 27 LEU HG 1.458 27 LEU HD11 1.156 27 LEU HD12 1.156 27 LEU HD13 1.156 27 LEU HD21 1.156 27 LEU HD22 1.156 27 LEU HD23 1.156 27 LEU C 176.417 27 LEU CA 52.631 27 LEU CB 41.098 27 LEU N 122.555 28 ILE H 9.382 28 ILE HA 4.206 28 ILE HB 1.973 250 28 ILE HG12 1.429 28 ILE HG13 1.429 28 ILE HG21 0.848 28 ILE HG22 0.848 28 ILE HG23 0.848 28 ILE HD11 0.738 28 ILE HD12 0.738 28 ILE HD13 0.738 28 ILE C 175.951 28 ILE CA 60.342 28 ILE CB 39.788 28 ILE CG2 17.767 28 ILE CD1 12.913 28 ILE N 125.2 29 LEU H 8.479 29 LEU HA 4.191 29 LEU HB2 2.207 29 LEU HB3 2.207 29 LEU HG 1.471 29 LEU HD11 0.694 29 LEU HD12 0.694 29 LEU HD13 0.694 29 LEU HD21 0.694 29 LEU HD22 0.694 29 LEU HD23 0.694 29 LEU CA 60.29 29 LEU CB 39.885 29 LEU N 125.19 30 GLY H 8.105 30 GLY HA2 4.147 30 GLY HA3 3.955 30 GLY C 175.42 30 GLY CA 45.44 30 GLY N 101.065 31 PHE H 7.368 31 PHE HA 5.178 31 PHE HB2 2.933 31 PHE HB3 2.544 31 PHE C 172.251 251 31 PHE CA 56.738 31 PHE CB 40.304 31 PHE N 115.927 32 SER H 8.478 32 SER HA 4.762 32 SER HB2 3.888 32 SER HB3 3.508 32 SER C 173.334 32 SER CA 56.879 32 SER CB 65.859 32 SER N 114.993 33 ILE H 8.41 33 ILE HA 5.701 33 ILE HB 1.831 33 ILE HG12 1.363 33 ILE HG13 1.363 33 ILE HG21 0.768 33 ILE HG22 0.768 33 ILE HG23 0.768 33 ILE HD11 0.336 33 ILE HD12 0.336 33 ILE HD13 0.336 33 ILE C 175.618 33 ILE CA 58.866 33 ILE CB 43.721 33 ILE CG1 25.547 33 ILE CG2 20.198 33 ILE CD1 14.791 33 ILE N 113.297 34 GLY H 9.26 34 GLY HA2 4.635 34 GLY HA3 3.627 34 GLY C 172.247 34 GLY CA 43.631 34 GLY N 109.829 35 GLY H 8.548 35 GLY HA2 5.269 35 GLY HA3 3.884 35 GLY C 174.137 252 35 GLY CA 44.268 35 GLY N 106.944 36 GLY H 6.664 36 GLY HA2 4.693 36 GLY HA3 3.978 36 GLY C 177.68 36 GLY CA 43.807 36 GLY N 105.448 37 ILE H 8.497 37 ILE HA 4.124 37 ILE HB 2.041 37 ILE HG12 1.282 37 ILE HG13 0.905 37 ILE HG21 0.961 37 ILE HG22 0.961 37 ILE HG23 0.961 37 ILE HD11 0.764 37 ILE HD12 0.764 37 ILE HD13 0.764 37 ILE C 175.326 37 ILE CA 64.61 37 ILE CB 37.81 37 ILE CG1 26.088 37 ILE CG2 17.973 37 ILE CD1 13.746 37 ILE N 114.741 38 ASP H 9.781 38 ASP HA 4.631 38 ASP HB2 2.93 38 ASP HB3 2.575 38 ASP C 175.329 38 ASP CA 52.637 38 ASP CB 39.794 38 ASP N 117.982 39 GLN H 7.383 39 GLN HA 4.432 39 GLN HB2 1.788 39 GLN HB3 1.788 39 GLN HG2 2.249 253 39 GLN HG3 2.12 39 GLN HE21 7.113 39 GLN HE22 6.75 39 GLN C 175.123 39 GLN CA 53.809 39 GLN CB 30.204 39 GLN CG 33.853 39 GLN CD 179.971 39 GLN N 119.074 39 GLN NE2 113.998 40 ASP H 8.62 40 ASP HA 4.838 40 ASP HB2 2.621 40 ASP HB3 2.793 40 ASP C 176.434 40 ASP CA 51.38 40 ASP CB 41.544 40 ASP N 123.108 41 PRO HA 4.565 41 PRO HB2 2.166 41 PRO HB3 1.903 41 PRO HG2 2.019 41 PRO HG3 1.875 41 PRO HD2 4.081 41 PRO HD3 4.049 41 PRO C 178.338 41 PRO CA 64.346 41 PRO CB 31.943 41 PRO CG 26.957 41 PRO CD 51.031 42 SER H 8.411 42 SER HA 4.15 42 SER HB2 3.892 42 SER HB3 3.892 42 SER C 175.292 42 SER CA 61.065 42 SER CB 62.997 42 SER N 115.249 43 GLN H 7.592 254 43 GLN HA 4.2 43 GLN HB2 2.288 43 GLN HB3 1.872 43 GLN HG2 2.267 43 GLN HG3 2.267 43 GLN HE21 7.452 43 GLN HE22 6.772 43 GLN C 174.995 43 GLN CA 54.837 43 GLN CB 28.77 43 GLN CG 33.649 43 GLN N 118.424 43 GLN NE2 113.049 44 ASN H 7.229 44 ASN HA 4.758 44 ASN HB2 2.722 44 ASN HB3 3.095 44 ASN C 174.149 44 ASN CA 49.94 44 ASN CB 38.914 44 ASN N 119.643 45 PRO HA 4.146 45 PRO HB2 1.871 45 PRO HB3 1.033 45 PRO HG2 1.59 45 PRO HG3 0.892 45 PRO HD2 3.92 45 PRO HD3 3.451 45 PRO C 176.973 45 PRO CA 63.394 45 PRO CB 31.815 45 PRO CG 25.836 45 PRO CD 50.421 46 PHE H 7.591 46 PHE HA 4.347 46 PHE HB2 2.242 46 PHE HB3 3.125 46 PHE C 174.993 46 PHE CA 57.466 255 46 PHE CB 34.214 46 PHE N 115.856 47 SER H 6.803 47 SER HA 4.194 47 SER HB2 3.882 47 SER HB3 3.558 47 SER C 175.363 47 SER CA 56.83 47 SER CB 63.808 47 SER N 112.15 48 GLU H 8.996 48 GLU HA 4.124 48 GLU HB2 2.012 48 GLU HB3 1.946 48 GLU HG2 2.275 48 GLU HG3 2.193 48 GLU C 176.307 48 GLU CA 57.98 48 GLU CB 30.314 48 GLU CG 36.446 48 GLU N 124.463 49 ASP H 7.961 49 ASP HA 4.566 49 ASP HB2 2.89 49 ASP HB3 2.658 49 ASP C 176.656 49 ASP CA 52.844 49 ASP CB 41.534 49 ASP N 117.472 50 LYS H 8.454 50 LYS HA 4.191 50 LYS HB2 1.892 50 LYS HB3 1.892 50 LYS HG2 1.704 50 LYS HG3 1.704 50 LYS HD2 1.426 50 LYS HD3 1.426 50 LYS C 176.387 50 LYS CA 56.481 256 50 LYS CB 30.466 50 LYS CG 24.361 50 LYS CD 29.706 50 LYS N 120.286 51 THR H 8.386 51 THR HA 4.333 51 THR HB 4.321 51 THR HG21 1.115 51 THR HG22 1.115 51 THR HG23 1.115 51 THR C 174.7 51 THR CA 61.667 51 THR CB 69.95 51 THR CG2 21.564 51 THR N 108.627 52 ASP H 7.619 52 ASP HA 4.494 52 ASP HB2 3.138 52 ASP HB3 2.776 52 ASP C 176.844 52 ASP CA 54.645 52 ASP CB 41.523 52 ASP N 122.291 53 LYS H 8.771 53 LYS HA 4.621 53 LYS HB2 2.172 53 LYS HB3 2.172 53 LYS HG2 1.379 53 LYS HG3 1.379 53 LYS HD2 1.61 53 LYS HD3 1.61 53 LYS HE2 3.079 53 LYS HE3 3.079 53 LYS C 176.922 53 LYS CA 56.139 53 LYS CB 32.121 53 LYS N 129.103 54 GLY H 8.734 54 GLY HA2 3.545 257 54 GLY HA3 4.137 54 GLY C 172.307 54 GLY CA 45.208 54 GLY N 106.524 55 ILE H 9.016 55 ILE HA 4.78 55 ILE HB 2.086 55 ILE HG12 1.272 55 ILE HG13 1.131 55 ILE HG21 0.729 55 ILE HG22 0.729 55 ILE HG23 0.729 55 ILE HD11 0.455 55 ILE HD12 0.455 55 ILE HD13 0.455 55 ILE C 174.992 55 ILE CA 57.592 55 ILE CB 35.815 55 ILE CG2 18.546 55 ILE CD1 8.457 55 ILE N 120.479 56 TYR H 8.788 56 TYR HA 5.419 56 TYR HB2 2.473 56 TYR HB3 2.473 56 TYR HD1 6.898 56 TYR HD2 6.898 56 TYR HE1 6.607 56 TYR HE2 6.607 56 TYR C 175.689 56 TYR CA 55.444 56 TYR CB 42.81 56 TYR CD1 133.806 56 TYR CE1 117.699 56 TYR N 124.57 57 VAL H 8.764 57 VAL HA 4.362 57 VAL HB 2.134 57 VAL HG11 0.728 258 57 VAL HG12 0.728 57 VAL HG13 0.728 57 VAL HG21 0.75 57 VAL HG22 0.75 57 VAL HG23 0.75 57 VAL C 176.841 57 VAL CA 62.641 57 VAL CB 32.109 57 VAL CG1 22.548 57 VAL CG2 21.981 57 VAL N 121.119 58 THR H 8.654 58 THR HA 4.399 58 THR HB 4.478 58 THR HG21 1.043 58 THR HG22 1.043 58 THR HG23 1.043 58 THR C 175.144 58 THR CA 62.131 58 THR CB 69.162 58 THR CG2 21.221 58 THR N 118.563 59 ARG H 7.166 59 ARG HA 4.346 59 ARG HB2 1.714 59 ARG HB3 1.646 59 ARG HG2 1.578 59 ARG HG3 1.497 59 ARG HD2 3.105 59 ARG HD3 3.105 59 ARG C 175.017 59 ARG CA 56.26 59 ARG CB 34.149 59 ARG CG 27.197 59 ARG CD 43.252 59 ARG N 119.725 60 VAL H 8.469 60 VAL HA 4.316 60 VAL HB 1.831 259 60 VAL HG11 0.775 60 VAL HG12 0.775 60 VAL HG13 0.775 60 VAL HG21 0.534 60 VAL HG22 0.534 60 VAL HG23 0.534 60 VAL C 175.3 60 VAL CA 62.063 60 VAL CB 34.95 60 VAL CG1 21.696 60 VAL CG2 22.037 60 VAL N 122.993 61 SER H 8.112 61 SER HA 4.278 61 SER HB2 3.875 61 SER HB3 3.653 61 SER C 175.357 61 SER CA 59.246 61 SER CB 62.928 61 SER N 122.681 62 GLU H 9.447 62 GLU HA 4.163 62 GLU HB2 1.982 62 GLU HB3 2.015 62 GLU HG2 2.359 62 GLU HG3 2.292 62 GLU C 178.129 62 GLU CA 58.275 62 GLU CB 29.09 62 GLU CG 36.193 62 GLU N 131.457 63 GLY H 9.619 63 GLY HA2 4.044 63 GLY HA3 3.679 63 GLY C 174.146 63 GLY CA 45.47 63 GLY N 116.309 64 GLY H 7.585 64 GLY HA2 4.396 260 64 GLY HA3 3.885 64 GLY C 172.075 64 GLY CA 45.323 64 GLY N 105.751 65 PRO HA 4.233 65 PRO HB2 2.55 65 PRO HB3 2.018 65 PRO HG2 1.995 65 PRO HG3 2.073 65 PRO HD2 3.664 65 PRO HD3 3.35 65 PRO C 178.941 65 PRO CA 64.883 65 PRO CB 32.565 65 PRO CG 27.896 65 PRO CD 49.084 66 ALA H 7.33 66 ALA HA 4.088 66 ALA HB1 1.51 66 ALA HB2 1.51 66 ALA HB3 1.51 66 ALA C 177.609 66 ALA CA 54.231 66 ALA CB 19.389 66 ALA N 119.976 67 GLU H 8.221 67 GLU HA 3.874 67 GLU HB2 2.123 67 GLU HB3 2.063 67 GLU HG2 2.195 67 GLU HG3 2.038 67 GLU C 181.181 67 GLU CA 59.638 67 GLU CB 30.025 67 GLU CG 37.148 67 GLU N 121.916 68 ILE H 8.032 68 ILE HA 3.739 68 ILE HB 1.796 261 68 ILE HG12 1.526 68 ILE HG13 1.169 68 ILE HG21 0.897 68 ILE HG22 0.897 68 ILE HG23 0.897 68 ILE HD11 0.761 68 ILE HD12 0.761 68 ILE HD13 0.761 68 ILE C 177.365 68 ILE CA 64.314 68 ILE CB 38.095 68 ILE CG1 28.864 68 ILE CG2 16.975 68 ILE CD1 13.227 68 ILE N 119.497 69 ALA H 7.125 69 ALA HA 4.309 69 ALA HB1 1.464 69 ALA HB2 1.464 69 ALA HB3 1.464 69 ALA C 177.401 69 ALA CA 52.807 69 ALA CB 21.222 69 ALA N 119.68 70 GLY H 7.57 70 GLY HA2 4.26 70 GLY HA3 3.679 70 GLY C 174.456 70 GLY CA 44.907 70 GLY N 103.907 71 LEU H 7.747 71 LEU HA 3.769 71 LEU HB2 1.282 71 LEU HB3 1.282 71 LEU HG 0.673 71 LEU HD11 -0.289 71 LEU HD12 -0.289 71 LEU HD13 -0.289 71 LEU HD21 -0.289 262 71 LEU HD22 -0.289 71 LEU HD23 -0.289 71 LEU C 174.486 71 LEU CA 55.195 71 LEU CB 41.364 71 LEU CG 27.019 71 LEU N 123.521 72 GLN H 8.436 72 GLN HA 4.565 72 GLN HB2 1.857 72 GLN HB3 1.857 72 GLN HG2 2.177 72 GLN HG3 2.177 72 GLN HE21 7.28 72 GLN HE22 6.687 72 GLN C 175.23 72 GLN CA 54.012 72 GLN CB 31.964 72 GLN CG 33.562 72 GLN CD 180.248 72 GLN N 123.821 72 GLN NE2 110.86 73 ILE H 8.402 73 ILE HA 3.242 73 ILE HB 1.562 73 ILE HG12 0.841 73 ILE HG13 1.404 73 ILE HG21 0.792 73 ILE HG22 0.792 73 ILE HG23 0.792 73 ILE HD11 0.96 73 ILE HD12 0.96 73 ILE HD13 0.96 73 ILE C 177.196 73 ILE CA 63.538 73 ILE CB 38.014 73 ILE CG1 28.351 73 ILE CG2 17.977 73 ILE CD1 13.988 263 73 ILE N 119.852 74 GLY H 8.92 74 GLY HA2 3.846 74 GLY HA3 2.565 74 GLY C 173.551 74 GLY CA 44.655 74 GLY N 116.672 75 ASP H 7.615 75 ASP HA 4.445 75 ASP HB2 2.439 75 ASP HB3 2.439 75 ASP C 174.754 75 ASP CA 55.718 75 ASP CB 40.626 75 ASP N 122.336 76 LYS H 8.444 76 LYS HA 4.348 76 LYS HB2 2.153 76 LYS HB3 2.153 76 LYS HG2 1.937 76 LYS HG3 1.937 76 LYS HD2 1.74 76 LYS HD3 1.74 76 LYS C 176.773 76 LYS CA 54.659 76 LYS CB 33.946 76 LYS N 123.369 77 ILE H 8.695 77 ILE HA 3.853 77 ILE HB 1.51 77 ILE HG12 1.153 77 ILE HG13 1.153 77 ILE HG21 0.678 77 ILE HG22 0.678 77 ILE HG23 0.678 77 ILE HD11 0.697 77 ILE HD12 0.697 77 ILE HD13 0.697 77 ILE C 174.243 264 77 ILE CA 62.219 77 ILE CB 37.871 77 ILE CG2 19.332 77 ILE CD1 13.642 77 ILE N 125.789 78 MET H 9.101 78 MET HA 4.449 78 MET HB2 2.119 78 MET HB3 1.648 78 MET HG2 2.357 78 MET HG3 2.357 78 MET HE1 1.908 78 MET HE2 1.908 78 MET HE3 1.908 78 MET C 177.971 78 MET CA 55.524 78 MET CB 32.469 78 MET CG 31.375 78 MET CE 16.22 78 MET N 125.28 79 GLN H 7.628 79 GLN HA 5.237 79 GLN HB2 1.762 79 GLN HB3 1.828 79 GLN HG2 2.08 79 GLN HG3 2.008 79 GLN HE21 7.612 79 GLN HE22 6.701 79 GLN C 176.886 79 GLN CA 54.88 79 GLN CB 35.022 79 GLN CG 34.353 79 GLN CD 179.161 79 GLN N 116.831 79 GLN NE2 111.553 80 VAL H 8.487 80 VAL HA 4.606 80 VAL HB 1.978 80 VAL HG11 0.866 265 80 VAL HG12 0.866 80 VAL HG13 0.866 80 VAL HG21 0.866 80 VAL HG22 0.866 80 VAL HG23 0.866 80 VAL C 174.671 80 VAL CA 60.612 80 VAL CB 34.705 80 VAL CG2 22.254 80 VAL N 120.584 81 ASN H 9.703 81 ASN HA 4.485 81 ASN HB2 3.265 81 ASN HB3 3.059 81 ASN HD21 7.448 81 ASN HD22 6.813 81 ASN C 174.647 81 ASN CA 54.365 81 ASN CB 36.77 81 ASN CG 177.786 81 ASN N 127.611 81 ASN ND2 111.245 82 GLY H 8.443 82 GLY HA2 3.988 82 GLY HA3 3.411 82 GLY C 173.416 82 GLY CA 45.189 82 GLY N 102.442 83 TRP H 8.367 83 TRP HA 4.566 83 TRP HB2 3.265 83 TRP HB3 3.218 83 TRP HD1 7.347 83 TRP HE1 10.133 83 TRP HE3 7.556 83 TRP HZ2 7.418 83 TRP HZ3 7.051 83 TRP HH2 7.088 83 TRP C 175.926 266 83 TRP CA 56.726 83 TRP CB 29.469 83 TRP CD1 128.059 83 TRP CE3 120.917 83 TRP CZ2 114.443 83 TRP CZ3 121.082 83 TRP CH2 124.323 83 TRP N 123.318 83 TRP NE1 129.705 84 ASP H 8.593 84 ASP HA 4.445 84 ASP HB2 2.672 84 ASP HB3 2.643 84 ASP C 176.543 84 ASP CA 55.801 84 ASP CB 41.902 84 ASP N 125.959 85 MET H 8.046 85 MET HA 4.72 85 MET HB2 2.115 85 MET HB3 1.639 85 MET HG2 2.565 85 MET HG3 2.317 85 MET HE1 1.958 85 MET HE2 1.958 85 MET HE3 1.958 85 MET C 176.746 85 MET CA 54.155 85 MET CB 33.284 85 MET CE 18.93 85 MET N 123.667 86 THR H 8.51 86 THR HA 4.195 86 THR HB 4.165 86 THR HG21 1.342 86 THR HG22 1.342 86 THR HG23 1.342 86 THR C 175.349 86 THR CA 64.954 267 86 THR CB 69.597 86 THR CG2 22.089 86 THR N 115.987 87 MET H 8.539 87 MET HA 4.585 87 MET HB2 2.104 87 MET HB3 1.849 87 MET HG2 2.525 87 MET HG3 2.418 87 MET HE1 2.126 87 MET HE2 2.126 87 MET HE3 2.126 87 MET C 174.638 87 MET CA 54.364 87 MET CB 32.291 87 MET CG 32.317 87 MET CE 23.105 87 MET N 123.127 88 VAL H 7.965 88 VAL HA 4.873 88 VAL HB 2.366 88 VAL HG11 0.89 88 VAL HG12 0.89 88 VAL HG13 0.89 88 VAL HG21 0.913 88 VAL HG22 0.913 88 VAL HG23 0.913 88 VAL C 176.856 88 VAL CA 59.058 88 VAL CB 35.025 88 VAL CG1 18.861 88 VAL CG2 22.54 88 VAL N 113.502 89 THR H 8.506 89 THR HA 4.381 89 THR HB 4.097 89 THR HG21 1.224 89 THR HG22 1.224 89 THR HG23 1.224 268 89 THR C 175.023 89 THR CA 61.116 89 THR CB 70.777 89 THR CG2 22.138 89 THR N 112.612 90 HIS H 9.989 90 HIS HA 3.864 90 HIS HB2 3.546 90 HIS HB3 3.286 90 HIS HD2 6.868 90 HIS HE1 7.73 90 HIS C 177.411 90 HIS CA 61.701 90 HIS CB 29.11 90 HIS CD2 124.364 90 HIS CE1 137.237 90 HIS N 122.908 91 ASP H 9.185 91 ASP HA 4.301 91 ASP HB2 2.652 91 ASP HB3 2.41 91 ASP C 178.552 91 ASP CA 57.525 91 ASP CB 41.698 91 ASP N 115.99 92 GLN H 7.808 92 GLN HA 3.791 92 GLN HB2 2.361 92 GLN HB3 1.868 92 GLN HG2 2.38 92 GLN HG3 2.38 92 GLN HE21 7.456 92 GLN HE22 6.883 92 GLN C 179.313 92 GLN CA 58.899 92 GLN CB 28.859 92 GLN CG 34.801 92 GLN CD 180.317 92 GLN N 118.454 269 92 GLN NE2 111.173 93 ALA H 7.967 93 ALA HA 3.832 93 ALA HB1 1.269 93 ALA HB2 1.269 93 ALA HB3 1.269 93 ALA C 178.754 93 ALA CA 55.28 93 ALA CB 18.986 93 ALA N 121.707 94 ARG H 8.258 94 ARG HA 3.567 94 ARG HB2 1.833 94 ARG HB3 1.676 94 ARG HG2 1.563 94 ARG HG3 1.339 94 ARG HD2 3.182 94 ARG HD3 3.182 94 ARG C 180.137 94 ARG CA 59.887 94 ARG CB 29.895 94 ARG CG 27.187 94 ARG CD 43.276 94 ARG N 117.066 95 LYS H 8.41 95 LYS HA 3.72 95 LYS HB2 1.755 95 LYS HB3 1.65 95 LYS HG2 1.469 95 LYS HG3 1.258 95 LYS HD2 1.508 95 LYS HD3 1.508 95 LYS HE2 2.831 95 LYS HE3 2.831 95 LYS C 179.265 95 LYS CA 59.559 95 LYS CB 32.178 95 LYS CG 25.88 95 LYS CD 28.95 270 95 LYS N 120.092 96 ARG H 7.454 96 ARG HA 3.859 96 ARG HB2 1.707 96 ARG HB3 1.603 96 ARG HG2 1.475 96 ARG HG3 1.355 96 ARG HD2 2.437 96 ARG HD3 2.437 96 ARG C 178.665 96 ARG CA 57.47 96 ARG CB 29.387 96 ARG CG 26.337 96 ARG CD 42.469 96 ARG N 116.31 97 LEU H 7.602 97 LEU HA 3.999 97 LEU HB2 2.433 97 LEU HB3 2.433 97 LEU HG 1.351 97 LEU HD11 1.025 97 LEU HD12 1.025 97 LEU HD13 1.025 97 LEU HD21 1.025 97 LEU HD22 1.025 97 LEU HD23 1.025 97 LEU C 176.33 97 LEU CA 56.818 97 LEU CB 34.197 97 LEU N 117.299 98 THR H 7.185 98 THR HA 4.541 98 THR HB 4.253 98 THR HG21 1.115 98 THR HG22 1.115 98 THR HG23 1.115 98 THR C 174.776 98 THR CA 60.281 98 THR CB 69.126 271 98 THR CG2 22.632 98 THR N 103.957 99 LYS H 7.005 99 LYS HA 4.136 99 LYS HB2 1.766 99 LYS HB3 1.621 99 LYS HG2 1.485 99 LYS HG3 1.293 99 LYS HD2 1.615 99 LYS HD3 1.615 99 LYS C 178.407 99 LYS CA 57.649 99 LYS CB 31.969 99 LYS CG 24.827 99 LYS CD 28.933 99 LYS N 124.162 100 ARG H 8.693 100 ARG HA 3.902 100 ARG HB2 1.828 100 ARG HB3 1.828 100 ARG HG2 1.713 100 ARG HG3 1.678 100 ARG HD2 3.213 100 ARG HD3 3.213 100 ARG C 176.11 100 ARG CA 58.488 100 ARG CB 30.058 100 ARG CG 27.514 100 ARG CD 43.228 100 ARG N 127.484 101 SER H 7.597 101 SER HA 4.233 101 SER HB2 4.064 101 SER HB3 3.716 101 SER C 174.046 101 SER CA 58.167 101 SER CB 62.969 101 SER N 109.086 102 GLU H 7.165 272 102 GLU HA 4.503 102 GLU HB2 1.748 102 GLU HB3 1.748 102 GLU HG2 2.144 102 GLU HG3 2.076 102 GLU C 175.151 102 GLU CA 54.722 102 GLU CB 30.739 102 GLU CG 36.083 102 GLU N 121.519 103 GLU H 8.733 103 GLU HA 4.008 103 GLU HB2 1.931 103 GLU HB3 1.931 103 GLU HG2 2.289 103 GLU HG3 2.145 103 GLU C 174.747 103 GLU CA 57.945 103 GLU CB 30.339 103 GLU CG 37.076 103 GLU N 123.304 104 VAL H 7.589 104 VAL HA 4.821 104 VAL HB 1.704 104 VAL HG11 0.481 104 VAL HG12 0.481 104 VAL HG13 0.481 104 VAL HG21 0.508 104 VAL HG22 0.508 104 VAL HG23 0.508 104 VAL C 175.446 104 VAL CA 60.205 104 VAL CB 34.629 104 VAL CG1 21.194 104 VAL CG2 20.779 104 VAL N 117.592 105 VAL H 8.764 105 VAL HA 4.595 105 VAL HB 1.866 273 105 VAL HG11 0.947 105 VAL HG12 0.947 105 VAL HG13 0.947 105 VAL HG21 0.905 105 VAL HG22 0.905 105 VAL HG23 0.905 105 VAL C 173.303 105 VAL CA 59.725 105 VAL CB 34.292 105 VAL CG1 21.267 105 VAL CG2 22.733 105 VAL N 121 106 ARG H 8.842 106 ARG HA 4.724 106 ARG HB2 1.839 106 ARG HB3 1.792 106 ARG HG2 1.514 106 ARG HG3 1.345 106 ARG HD2 2.898 106 ARG HD3 2.381 106 ARG C 175.423 106 ARG CA 55.515 106 ARG CB 29.879 106 ARG CG 28.355 106 ARG CD 43.559 106 ARG N 126.175 107 LEU H 9.447 107 LEU HA 5.247 107 LEU HB2 1.686 107 LEU HB3 1.686 107 LEU HG 1.236 107 LEU HD11 0.853 107 LEU HD12 0.853 107 LEU HD13 0.853 107 LEU HD21 0.853 107 LEU HD22 0.853 107 LEU HD23 0.853 107 LEU C 176.451 107 LEU CA 53.338 274 107 LEU CB 41.631 107 LEU CG 25.912 107 LEU N 125.682 108 LEU H 7.906 108 LEU HA 5.037 108 LEU HB2 1.369 108 LEU HB3 1.369 108 LEU HG 1.097 108 LEU HD11 0.709 108 LEU HD12 0.709 108 LEU HD13 0.709 108 LEU HD21 0.709 108 LEU HD22 0.709 108 LEU HD23 0.709 108 LEU C 176.295 108 LEU CA 54.777 108 LEU CB 44.285 108 LEU CG 24.282 108 LEU N 123.684 109 VAL H 9.081 109 VAL HA 5.611 109 VAL HB 1.863 109 VAL HG11 0.557 109 VAL HG12 0.557 109 VAL HG13 0.557 109 VAL HG21 0.587 109 VAL HG22 0.587 109 VAL HG23 0.587 109 VAL C 174.671 109 VAL CA 58.007 109 VAL CB 35.316 109 VAL CG1 18.316 109 VAL CG2 21.324 109 VAL N 119.516 110 THR H 8.926 110 THR HA 5.106 110 THR HB 3.941 110 THR HG21 1.072 110 THR HG22 1.072 275 110 THR HG23 1.072 110 THR C 173.001 110 THR CA 60.459 110 THR CB 70.979 110 THR CG2 21.717 110 THR N 113.689 111 ARG H 8.732 111 ARG HA 4.767 111 ARG HB2 1.862 111 ARG HB3 1.525 111 ARG HG2 1.503 111 ARG HG3 1.421 111 ARG HD2 2.982 111 ARG HD3 2.928 111 ARG C 175.195 111 ARG CA 54.677 111 ARG CB 33.352 111 ARG CG 25.826 111 ARG CD 43.189 111 ARG N 125.719 112 GLN H 8.848 112 GLN HA 4.437 112 GLN HB2 2.039 112 GLN HB3 1.98 112 GLN HG2 2.412 112 GLN HG3 2.362 112 GLN HE21 7.605 112 GLN HE22 6.863 112 GLN C 176.206 112 GLN CA 55.931 112 GLN CB 29.528 112 GLN CG 33.905 112 GLN CD 180.088 112 GLN N 123.698 112 GLN NE2 111.936 113 SER H 8.497 113 SER HA 4.372 113 SER HB2 3.782 113 SER HB3 3.782 276 113 SER C 174.299 113 SER CA 58.362 113 SER CB 63.861 113 SER N 117.066 114 LEU H 8.222 114 LEU HA 4.333 114 LEU HB2 1.572 114 LEU HB3 1.572 114 LEU HD11 0.812 114 LEU HD12 0.812 114 LEU HD13 0.812 114 LEU HD21 0.866 114 LEU HD22 0.866 114 LEU HD23 0.866 114 LEU C 175.995 114 LEU CA 55.191 114 LEU CB 42.341 114 LEU CG 27.057 114 LEU CD1 24.635 114 LEU CD2 24.635 114 LEU N 124.369 115 GLN H 8.3 115 GLN HA 4.223 115 GLN HB2 2.023 115 GLN HB3 1.918 115 GLN HG2 2.298 115 GLN HG3 2.298 115 GLN HE21 7.511 115 GLN C 175.897 115 GLN CA 55.995 115 GLN CB 29.328 115 GLN CG 33.838 115 GLN N 121.367 115 GLN NE2 112.635 116 LYS H 8.249 116 LYS HA 4.206 116 LYS HB2 1.756 116 LYS HB3 1.688 116 LYS HG2 1.374 277 116 LYS HG3 1.374 116 LYS HD2 1.616 116 LYS HD3 1.616 116 LYS HE2 2.926 116 LYS HE3 2.926 116 LYS C 176.281 116 LYS CA 56.326 116 LYS CB 33.063 116 LYS CG 24.705 116 LYS CD 29.179 116 LYS CE 42.599 116 LYS N 122.711 117 ALA H 8.237 117 ALA HA 4.263 117 ALA HB1 1.313 117 ALA HB2 1.313 117 ALA HB3 1.313 117 ALA C 177.84 117 ALA CA 52.548 117 ALA CB 19.121 117 ALA N 125.141 118 VAL H 8.052 118 VAL HA 3.992 118 VAL HB 1.998 118 VAL HG11 0.864 118 VAL HG12 0.864 118 VAL HG13 0.864 118 VAL HG21 0.864 118 VAL HG22 0.864 118 VAL HG23 0.864 118 VAL C 175.808 118 VAL CA 62.454 118 VAL CB 32.845 118 VAL CG1 20.981 118 VAL N 119.587 119 GLN H 8.358 119 GLN HA 4.257 119 GLN HB2 2.035 119 GLN HB3 1.967 278 119 GLN HG2 2.303 119 GLN HG3 2.303 119 GLN HE21 7.468 119 GLN HE22 6.811 119 GLN C 174.405 119 GLN CA 56.019 119 GLN CB 29.36 119 GLN CG 33.792 119 GLN CD 180.41 119 GLN N 124.129 119 GLN NE2 112.419 120 GLN H 8.422 120 GLN HA 4.262 120 GLN HB2 2.032 120 GLN HB3 1.977 120 GLN HG2 2.329 120 GLN HG3 2.329 120 GLN HE21 7.458 120 GLN HE22 6.778 120 GLN C 175.928 120 GLN CA 56.204 120 GLN CB 29.475 120 GLN CG 33.779 120 GLN CD 180.187 120 GLN N 122.016 120 GLN NE2 112.351 121 SER H 8.322 121 SER HA 4.366 121 SER HB2 3.82 121 SER HB3 3.82 121 SER C 177.038 121 SER CA 58.585 121 SER CB 63.75 121 SER N 116.821 122 MET H 8.314 122 MET HA 4.459 122 MET HB2 2.075 122 MET HB3 1.977 122 MET HG2 2.557 279 122 MET HG3 2.499 122 MET HE1 2.04 122 MET HE2 2.04 122 MET HE3 2.04 122 MET C 175.929 122 MET CA 55.661 122 MET CB 32.817 122 MET CE 16.856 122 MET N 122.031 123 LEU H 8.128 123 LEU HA 4.357 123 LEU HB2 1.586 123 LEU HB3 1.586 123 LEU HD11 0.838 123 LEU HD12 0.838 123 LEU HD13 0.838 123 LEU HD21 0.838 123 LEU HD22 0.838 123 LEU HD23 0.838 123 LEU C 176.122 123 LEU CA 55.438 123 LEU CB 42.249 123 LEU N 123.409 124 SER H 7.801 124 SER HA 4.192 124 SER HB2 3.774 124 SER HB3 3.774 124 SER C 178.548 124 SER CA 59.9 124 SER CB 64.869 124 SER N 122.045 280 Appendix Table A-2 Chemical shift assignments for the nuclei of GIP-Glutaminase L peptide complex (BMRB entry: 17255) For GIP in bound form: Residue no. Amino acid Nucleus Chemical shift 1 MET HE1 1.969 1 MET HE2 1.969 1 MET HE3 1.969 1 MET CE 24.605 2 SER H 8.157 2 SER HA 4.341 2 SER HB2 3.697 2 SER HB3 3.697 2 SER CA 58.313 2 SER CB 64.004 2 SER N 121.525 3 TYR H 8.112 3 TYR HA 4.546 3 TYR HB2 2.942 3 TYR HB3 2.82 3 TYR HD1 6.992 3 TYR HD2 6.992 3 TYR CA 57.993 3 TYR CB 39.021 3 TYR N 122.397 4 ILE H 8.009 4 ILE HA 4.258 4 ILE HB 1.608 4 ILE HG12 1.349 4 ILE HG13 0.973 4 ILE HG21 0.765 4 ILE HG22 0.765 4 ILE HG23 0.765 4 ILE HD11 0.726 4 ILE HD12 0.726 4 ILE HD13 0.726 4 ILE CA 57.781 4 ILE CB 39.215 4 ILE CG1 26.832 4 ILE CG2 16.826 281 4 ILE CD1 12.858 4 ILE N 127.694 5 PRO HA 4.164 5 PRO HB2 2.235 5 PRO HB3 1.842 5 PRO HG2 1.956 5 PRO HG3 1.888 5 PRO HD2 3.588 5 PRO HD3 3.558 5 PRO CA 63.449 5 PRO CB 31.983 5 PRO CG 27.372 5 PRO CD 50.786 6 GLY H 8.416 6 GLY HA2 3.773 6 GLY HA3 3.938 6 GLY CA 45.191 6 GLY N 110.047 7 GLN H 7.972 7 GLN HA 4.564 7 GLN HB2 2.016 7 GLN HB3 1.884 7 GLN HG2 2.26 7 GLN HG3 2.26 7 GLN HE21 7.503 7 GLN HE22 6.815 7 GLN CA 53.541 7 GLN CB 29.191 7 GLN CG 33.609 7 GLN N 120.547 7 GLN NE2 112.516 8 PRO HA 4.406 8 PRO HB2 2.21 8 PRO HB3 1.822 8 PRO HG2 1.931 8 PRO HG3 1.959 8 PRO HD2 3.728 8 PRO HD3 3.57 8 PRO CA 63.097 8 PRO CB 32.022 8 PRO CG 27.34 282 8 PRO CD 50.506 9 VAL H 8.292 9 VAL HA 4.167 9 VAL HB 1.995 9 VAL HG11 0.874 9 VAL HG12 0.874 9 VAL HG13 0.874 9 VAL HG21 0.846 9 VAL HG22 0.846 9 VAL HG23 0.846 9 VAL CA 62.222 9 VAL CB 32.996 9 VAL CG1 20.97 9 VAL CG2 18.866 9 VAL N 120.591 10 THR H 8.24 10 THR HA 4.31 10 THR HB 4.144 10 THR HG21 1.108 10 THR HG22 1.108 10 THR HG23 1.108 10 THR CA 61.631 10 THR CB 70.009 10 THR CG2 21.537 10 THR N 117.745 11 ALA H 8.052 11 ALA HA 4.528 11 ALA HB1 1.242 11 ALA HB2 1.242 11 ALA HB3 1.242 11 ALA CA 51.899 11 ALA CB 19.927 11 ALA N 126.513 12 VAL H 8.498 12 VAL HA 4.203 12 VAL HB 1.968 12 VAL HG11 0.832 12 VAL HG12 0.832 12 VAL HG13 0.832 12 VAL HG21 0.825 12 VAL HG22 0.825 283 12 VAL HG23 0.825 12 VAL CA 61.544 12 VAL CB 33.505 12 VAL CG1 20.661 12 VAL CG2 21.208 12 VAL N 120.32 13 VAL H 8.091 13 VAL HA 4.846 13 VAL HB 1.802 13 VAL HG11 0.742 13 VAL HG12 0.742 13 VAL HG13 0.742 13 VAL HG21 0.803 13 VAL HG22 0.803 13 VAL HG23 0.803 13 VAL CA 60.976 13 VAL CB 33.285 13 VAL CG1 21.44 13 VAL CG2 20.847 13 VAL N 123.775 14 GLN H 9.018 14 GLN HA 4.587 14 GLN HB2 1.556 14 GLN HB3 1.764 14 GLN HG2 2.043 14 GLN HG3 1.99 14 GLN HE21 7.109 14 GLN HE22 6.723 14 GLN CA 53.958 14 GLN CB 31.895 14 GLN CG 32.911 14 GLN N 124.782 14 GLN NE2 111.111 15 ARG H 8.569 15 ARG HA 5.057 15 ARG HB2 1.715 15 ARG HB3 1.584 15 ARG HG2 1.474 15 ARG HG3 1.39 15 ARG HD2 3.062 15 ARG HD3 3.062 284 15 ARG CA 55.234 15 ARG CB 31.427 15 ARG CG 27.65 15 ARG CD 43.281 15 ARG N 123.905 16 VAL H 8.669 16 VAL HA 4.286 16 VAL HB 1.632 16 VAL HG11 0.578 16 VAL HG12 0.578 16 VAL HG13 0.578 16 VAL HG21 0.624 16 VAL HG22 0.624 16 VAL HG23 0.624 16 VAL CA 61.203 16 VAL CB 35.308 16 VAL CG1 20.648 16 VAL CG2 21.48 16 VAL N 124.788 17 GLU H 9.037 17 GLU HA 5.059 17 GLU HB2 1.868 17 GLU HB3 1.868 17 GLU HG2 1.881 17 GLU HG3 1.755 17 GLU CA 55.121 17 GLU CB 31.717 17 GLU CG 37.584 17 GLU N 129.823 18 ILE H 8.919 18 ILE HA 4.252 18 ILE HB 1.539 18 ILE HG12 1.447 18 ILE HG13 1.447 18 ILE HG21 0.751 18 ILE HG22 0.751 18 ILE HG23 0.751 18 ILE HD11 0.705 18 ILE HD12 0.705 18 ILE HD13 0.705 18 ILE CA 60.507 285 18 ILE CB 40.565 18 ILE CG1 27.611 18 ILE CG2 19.662 18 ILE CD1 15.163 18 ILE N 123.907 19 HIS H 8.807 19 HIS HA 4.8 19 HIS HB2 3.161 19 HIS HB3 3.109 19 HIS HD2 7.191 19 HIS CA 54.686 19 HIS CB 28.869 19 HIS N 127.764 20 LYS H 8.579 20 LYS HA 4.086 20 LYS HB2 1.556 20 LYS HB3 1.556 20 LYS HG2 1.533 20 LYS HG3 1.533 20 LYS HD2 1.005 20 LYS HD3 1.005 20 LYS HE2 3.135 20 LYS HE3 3.135 20 LYS CA 57.803 20 LYS CB 35.415 20 LYS CG 27.333 20 LYS CD 27.562 20 LYS N 122.931 21 LEU H 8.855 21 LEU HA 4.556 21 LEU HB2 1.604 21 LEU HB3 1.452 21 LEU HG 1.378 21 LEU HD11 0.907 21 LEU HD12 0.907 21 LEU HD13 0.907 21 LEU HD21 0.832 21 LEU HD22 0.832 21 LEU HD23 0.832 21 LEU CA 53.47 21 LEU CB 45.533 286 21 LEU CG 26.821 21 LEU CD1 23.627 21 LEU CD2 25.494 21 LEU N 123.076 22 ARG H 8.626 22 ARG HA 4.671 22 ARG HB2 1.753 22 ARG HB3 1.592 22 ARG HG2 1.449 22 ARG HG3 1.449 22 ARG HD2 3.145 22 ARG HD3 3.064 22 ARG HE 7.278 22 ARG CA 56.081 22 ARG CB 30.275 22 ARG CG 27.38 22 ARG CD 42.829 22 ARG N 126.534 22 ARG NE 84.017 23 GLN H 8.628 23 GLN HA 4.446 23 GLN HB2 1.907 23 GLN HB3 1.695 23 GLN HG2 2.097 23 GLN HG3 2.097 23 GLN HE21 7.439 23 GLN HE22 6.776 23 GLN CA 54.76 23 GLN CB 30.51 23 GLN CG 33.866 23 GLN N 129.278 23 GLN NE2 111.736 24 GLY H 8.969 24 GLY HA2 3.962 24 GLY HA3 3.573 24 GLY CA 46.96 24 GLY N 117.183 25 GLU H 9.038 25 GLU HA 4.214 25 GLU HB2 1.772 25 GLU HB3 1.772 287 25 GLU HG2 2.214 25 GLU HG3 2.141 25 GLU CA 56.392 25 GLU CB 30.001 25 GLU CG 36.165 25 GLU N 126.285 26 ASN H 7.931 26 ASN HA 4.823 26 ASN HB2 2.762 26 ASN HB3 2.644 26 ASN HD21 7.628 26 ASN HD22 6.983 26 ASN CA 52.39 26 ASN CB 41.24 26 ASN N 117.868 26 ASN ND2 114.311 27 LEU H 8.284 27 LEU HA 4.949 27 LEU HB2 1.479 27 LEU HB3 1.26 27 LEU HG 1.38 27 LEU HD11 1.288 27 LEU HD12 1.288 27 LEU HD13 1.288 27 LEU HD21 0.739 27 LEU HD22 0.739 27 LEU HD23 0.739 27 LEU CA 54.385 27 LEU CB 44.516 27 LEU CG 27.206 27 LEU CD1 24.553 27 LEU CD2 25.63 27 LEU N 123.213 28 ILE H 8.902 28 ILE HA 4.628 28 ILE HB 1.904 28 ILE HG12 1.382 28 ILE HG13 1.009 28 ILE HG21 0.763 28 ILE HG22 0.763 28 ILE HG23 0.763 288 28 ILE HD11 0.743 28 ILE HD12 0.743 28 ILE HD13 0.743 28 ILE CA 59.684 28 ILE CB 42.522 28 ILE CG1 26.678 28 ILE CG2 18.814 28 ILE CD1 13.632 28 ILE N 119.351 29 LEU H 10.951 29 LEU HA 4.332 29 LEU HB2 1.607 29 LEU HB3 1.607 29 LEU HG 1.501 29 LEU HD11 0.89 29 LEU HD12 0.89 29 LEU HD13 0.89 29 LEU HD21 0.758 29 LEU HD22 0.758 29 LEU HD23 0.758 29 LEU CA 55.691 29 LEU CB 44.463 29 LEU CG 27.32 29 LEU CD1 27.183 29 LEU CD2 27.305 29 LEU N 124.679 30 GLY H 9.334 30 GLY HA2 4.006 30 GLY HA3 4.154 30 GLY CA 46.146 30 GLY N 107.432 31 PHE H 7.408 31 PHE HA 5.002 31 PHE HB2 2.885 31 PHE HB3 3.647 31 PHE CA 56.653 31 PHE CB 39.742 31 PHE N 117.124 32 SER H 8.63 32 SER HA 5.814 32 SER HB2 3.558 289 32 SER HB3 3.558 32 SER CA 56.286 32 SER CB 65.93 32 SER N 112.923 33 ILE H 8.694 33 ILE HA 5.781 33 ILE HB 1.692 33 ILE HG12 1.387 33 ILE HG13 1.387 33 ILE HG21 0.749 33 ILE HG22 0.749 33 ILE HG23 0.749 33 ILE HD11 0.191 33 ILE HD12 0.191 33 ILE HD13 0.191 33 ILE CA 58.253 33 ILE CB 43.397 33 ILE CG1 26.682 33 ILE CG2 20.174 33 ILE CD1 13.604 33 ILE N 113.209 34 GLY H 9.05 34 GLY HA2 3.776 34 GLY HA3 4.817 34 GLY CA 43.545 34 GLY N 108.49 35 GLY H 9.485 35 GLY HA2 3.847 35 GLY HA3 5.379 35 GLY CA 43.984 35 GLY N 107.605 36 GLY H 6.608 36 GLY HA2 3.947 36 GLY HA3 4.696 36 GLY CA 43.893 36 GLY N 105.396 37 ILE H 8.581 37 ILE HA 4.151 37 ILE HB 2.043 37 ILE HG12 1.232 37 ILE HG13 1.272 290 37 ILE HG21 0.95 37 ILE HG22 0.95 37 ILE HG23 0.95 37 ILE HD11 0.748 37 ILE HD12 0.748 37 ILE HD13 0.748 37 ILE CA 64.826 37 ILE CB 37.411 37 ILE CG1 26.094 37 ILE CG2 17.894 37 ILE CD1 13.718 37 ILE N 115.311 38 ASP H 9.911 38 ASP HA 4.615 38 ASP HB2 2.902 38 ASP HB3 2.571 38 ASP CA 52.607 38 ASP CB 40.145 38 ASP N 118.124 39 GLN H 7.504 39 GLN HA 4.414 39 GLN HB2 1.938 39 GLN HB3 1.938 39 GLN HG2 2.333 39 GLN HG3 2.063 39 GLN HE21 6.979 39 GLN HE22 6.939 39 GLN CA 53.483 39 GLN CB 30.343 39 GLN CG 34.236 39 GLN N 119.958 39 GLN NE2 115.609 40 ASP H 8.562 40 ASP HA 4.932 40 ASP HB2 2.618 40 ASP HB3 2.952 40 ASP CA 51.198 40 ASP CB 41.257 40 ASP N 121.473 41 PRO HA 4.535 41 PRO HB2 2.163 291 41 PRO HB3 1.837 41 PRO HG2 2.03 41 PRO HG3 1.875 41 PRO HD2 4.121 41 PRO HD3 4.059 41 PRO CA 64.219 41 PRO CB 31.64 41 PRO CG 26.831 41 PRO CD 50.882 42 SER H 8.294 42 SER HA 4.133 42 SER HB2 3.9 42 SER HB3 3.9 42 SER CA 61.232 42 SER CB 63.076 42 SER N 115.571 43 GLN H 7.289 43 GLN HA 4.202 43 GLN HB2 2.306 43 GLN HB3 1.709 43 GLN HG2 2.185 43 GLN HG3 2.11 43 GLN HE21 7.572 43 GLN HE22 6.853 43 GLN CA 54.826 43 GLN CB 29.029 43 GLN CG 33.862 43 GLN N 118.268 43 GLN NE2 113.059 44 ASN H 7.07 44 ASN HA 4.677 44 ASN HB2 2.674 44 ASN HB3 3.098 44 ASN CA 50.103 44 ASN CB 38.917 44 ASN N 119.825 45 PRO HA 4.114 45 PRO HB2 1.846 45 PRO HB3 0.988 45 PRO HG2 1.542 45 PRO HG3 0.804 292 45 PRO HD2 3.933 45 PRO HD3 3.416 45 PRO CA 63.395 45 PRO CB 31.896 45 PRO CG 25.705 45 PRO CD 50.501 46 PHE H 7.635 46 PHE HA 4.364 46 PHE HB2 2.311 46 PHE HB3 3.205 46 PHE CA 58.119 46 PHE CB 40.103 46 PHE N 115.588 47 SER H 6.711 47 SER HA 4.205 47 SER HB2 3.854 47 SER HB3 3.43 47 SER CA 56.518 47 SER CB 63.853 47 SER N 111.81 48 GLU H 8.761 48 GLU HA 4.117 48 GLU HB2 1.977 48 GLU HB3 1.938 48 GLU HG2 2.262 48 GLU HG3 2.209 48 GLU CA 57.831 48 GLU CB 30.398 48 GLU CG 36.449 48 GLU N 124.099 49 ASP H 7.967 49 ASP HA 4.571 49 ASP HB2 2.872 49 ASP HB3 2.6 49 ASP CA 52.816 49 ASP CB 41.614 49 ASP N 118.167 50 LYS H 8.452 50 LYS HA 4.195 50 LYS HB2 1.857 50 LYS HB3 1.857 293 50 LYS HG2 1.377 50 LYS HG3 1.377 50 LYS HD2 1.415 50 LYS HD3 1.415 50 LYS CA 56.273 50 LYS CB 29.975 50 LYS CG 24.015 50 LYS CD 27.99 50 LYS CE 42.441 50 LYS N 120.392 51 THR H 8.326 51 THR HA 4.308 51 THR HB 4.314 51 THR HG21 1.093 51 THR HG22 1.093 51 THR HG23 1.093 51 THR CA 61.561 51 THR CB 69.861 51 THR CG2 21.495 51 THR N 108.193 52 ASP H 7.642 52 ASP HA 4.459 52 ASP HB2 3.125 52 ASP HB3 2.788 52 ASP CA 54.751 52 ASP CB 41.837 52 ASP N 122.292 53 LYS H 8.844 53 LYS HA 4.65 53 LYS HB2 2.183 53 LYS HB3 1.746 53 LYS HG2 1.33 53 LYS HG3 1.33 53 LYS HD2 1.615 53 LYS HD3 1.615 53 LYS HE2 3.103 53 LYS HE3 3.103 53 LYS CA 56.489 53 LYS CB 32.362 53 LYS CG 25.237 53 LYS CD 29.771 294 53 LYS CE 42.225 53 LYS N 129.965 54 GLY H 8.819 54 GLY HA2 4.168 54 GLY HA3 3.555 54 GLY CA 45.275 54 GLY N 106.511 55 ILE H 9.223 55 ILE HA 4.748 55 ILE HB 2.08 55 ILE HG12 1.043 55 ILE HG13 1.043 55 ILE HG21 0.693 55 ILE HG22 0.693 55 ILE HG23 0.693 55 ILE HD11 0.399 55 ILE HD12 0.399 55 ILE HD13 0.399 55 ILE CA 57.559 55 ILE CB 36.017 55 ILE CG2 18.011 55 ILE CD1 8.017 55 ILE N 120.313 56 TYR H 8.952 56 TYR HA 5.306 56 TYR HB2 2.53 56 TYR HB3 2.44 56 TYR HD1 6.888 56 TYR HD2 6.888 56 TYR CA 55.628 56 TYR CB 42.675 56 TYR N 125.445 57 VAL H 8.656 57 VAL HA 4.502 57 VAL HB 2.07 57 VAL HG11 0.626 57 VAL HG12 0.626 57 VAL HG13 0.626 57 VAL HG21 0.632 57 VAL HG22 0.632 57 VAL HG23 0.632 295 57 VAL CA 62.146 57 VAL CB 31.995 57 VAL CG1 21.59 57 VAL CG2 22.338 57 VAL N 121.052 58 THR H 8.719 58 THR HA 4.307 58 THR HB 4.247 58 THR HG21 1.104 58 THR HG22 1.104 58 THR HG23 1.104 58 THR CA 62.326 58 THR CB 68.681 58 THR CG2 22.48 58 THR N 119.551 59 ARG H 7.23 59 ARG HA 4.302 59 ARG HB2 1.66 59 ARG HB3 1.64 59 ARG HG2 1.546 59 ARG HG3 1.593 59 ARG HD2 3.114 59 ARG HD3 2.951 59 ARG HE 7.493 59 ARG CA 55.987 59 ARG CB 34.173 59 ARG CG 27.264 59 ARG CD 43.391 59 ARG N 119.022 59 ARG NE 85.721 60 VAL H 8.428 60 VAL HA 4.198 60 VAL HB 1.805 60 VAL HG11 0.746 60 VAL HG12 0.746 60 VAL HG13 0.746 60 VAL HG21 0.493 60 VAL HG22 0.493 60 VAL HG23 0.493 60 VAL CA 62.372 60 VAL CB 34.958 296 60 VAL CG1 21.47 60 VAL CG2 21.665 60 VAL N 121.675 61 SER H 8.124 61 SER HA 4.209 61 SER HB2 3.797 61 SER HB3 3.617 61 SER CA 59.379 61 SER CB 62.898 61 SER N 122.728 62 GLU H 9.285 62 GLU HA 4.137 62 GLU HB2 1.955 62 GLU HB3 1.955 62 GLU HG2 2.358 62 GLU HG3 2.219 62 GLU CA 58.156 62 GLU CB 29.133 62 GLU CG 36.177 62 GLU N 130.89 63 GLY H 9.626 63 GLY HA2 3.651 63 GLY HA3 4.03 63 GLY CA 45.414 63 GLY N 116.325 64 GLY H 7.537 64 GLY HA2 3.88 64 GLY HA3 4.362 64 GLY CA 45.295 64 GLY N 106.344 65 PRO HA 4.152 65 PRO HB2 2.623 65 PRO HB3 2.043 65 PRO HG2 2.229 65 PRO HG3 2.229 65 PRO HD2 3.681 65 PRO HD3 3.283 65 PRO CA 65.246 65 PRO CB 32.125 65 PRO CG 28.746 65 PRO CD 48.835 297 66 ALA H 7.668 66 ALA HA 3.864 66 ALA HB1 1.405 66 ALA HB2 1.405 66 ALA HB3 1.405 66 ALA CA 55.058 66 ALA CB 19.158 66 ALA N 120.001 67 GLU H 8.483 67 GLU HA 3.797 67 GLU HB2 2.13 67 GLU HB3 2.13 67 GLU HG2 2.173 67 GLU HG3 2.029 67 GLU CA 59.972 67 GLU CB 29.842 67 GLU CG 37.015 67 GLU N 121.308 68 ILE H 8.068 68 ILE HA 3.656 68 ILE HB 1.814 68 ILE HG12 1.566 68 ILE HG13 1.124 68 ILE HG21 0.878 68 ILE HG22 0.878 68 ILE HG23 0.878 68 ILE HD11 0.725 68 ILE HD12 0.725 68 ILE HD13 0.725 68 ILE CA 64.302 68 ILE CB 38.187 68 ILE CG1 29.075 68 ILE CG2 17.005 68 ILE CD1 13.073 68 ILE N 120.686 69 ALA H 7.584 69 ALA HA 4.262 69 ALA HB1 1.433 69 ALA HB2 1.433 69 ALA HB3 1.433 69 ALA CA 52.936 298 69 ALA CB 21.118 69 ALA N 119.667 70 GLY H 7.533 70 GLY HA2 3.675 70 GLY HA3 4.246 70 GLY CA 44.925 70 GLY N 103.682 71 LEU H 7.806 71 LEU HA 3.696 71 LEU HB2 1.23 71 LEU HB3 0.723 71 LEU HG 0.578 71 LEU HD11 -0.426 71 LEU HD12 -0.426 71 LEU HD13 -0.426 71 LEU HD21 0.532 71 LEU HD22 0.532 71 LEU HD23 0.532 71 LEU CA 55.139 71 LEU CB 44.142 71 LEU CG 26.68 71 LEU CD1 26.063 71 LEU CD2 23.966 71 LEU N 123.765 72 GLN H 8.356 72 GLN HA 4.535 72 GLN HB2 1.86 72 GLN HB3 1.783 72 GLN HG2 2.157 72 GLN HG3 2.157 72 GLN HE21 7.241 72 GLN HE22 6.668 72 GLN CA 54.032 72 GLN CB 31.972 72 GLN CG 33.47 72 GLN N 123.599 72 GLN NE2 110.957 73 ILE H 8.313 73 ILE HA 3.152 73 ILE HB 1.496 73 ILE HG12 1.444 299 73 ILE HG13 1.444 73 ILE HG21 0.724 73 ILE HG22 0.724 73 ILE HG23 0.724 73 ILE HD11 0.956 73 ILE HD12 0.956 73 ILE HD13 0.956 73 ILE CA 63.642 73 ILE CB 38.011 73 ILE CG1 28.791 73 ILE CG2 17.827 73 ILE CD1 14.033 73 ILE N 119.802 74 GLY H 9.063 74 GLY HA2 2.602 74 GLY HA3 3.913 74 GLY CA 44.751 74 GLY N 117.216 75 ASP H 7.614 75 ASP HA 4.407 75 ASP HB2 2.396 75 ASP HB3 1.943 75 ASP CA 55.759 75 ASP CB 40.474 75 ASP N 122.366 76 LYS H 8.322 76 LYS HA 4.292 76 LYS HB2 2.17 76 LYS HB3 2.17 76 LYS HG2 1.923 76 LYS HG3 1.923 76 LYS HD2 1.597 76 LYS HD3 1.597 76 LYS HE2 2.505 76 LYS HE3 2.505 76 LYS CA 54.396 76 LYS CB 33.989 76 LYS CG 25.24 76 LYS CE 40.262 76 LYS N 122.798 77 ILE H 8.673 300 77 ILE HA 3.799 77 ILE HB 1.506 77 ILE HG12 1.18 77 ILE HG13 1.18 77 ILE HG21 0.664 77 ILE HG22 0.664 77 ILE HG23 0.664 77 ILE HD11 0.68 77 ILE HD12 0.68 77 ILE HD13 0.68 77 ILE CA 62.126 77 ILE CB 37.944 77 ILE CG2 19.388 77 ILE CD1 13.628 77 ILE N 125.785 78 MET H 9.09 78 MET HA 4.425 78 MET HB2 1.649 78 MET HB3 1.649 78 MET HG2 2.328 78 MET HG3 2.328 78 MET HE1 1.887 78 MET HE2 1.887 78 MET HE3 1.887 78 MET CA 55.478 78 MET CB 32.441 78 MET CG 31.386 78 MET CE 16.089 78 MET N 125.255 79 GLN H 7.607 79 GLN HA 5.225 79 GLN HB2 1.753 79 GLN HB3 1.784 79 GLN HG2 2.074 79 GLN HG3 2.028 79 GLN HE21 7.581 79 GLN HE22 6.681 79 GLN CA 54.835 79 GLN CB 35.024 79 GLN CG 34.401 79 GLN N 116.394 301 79 GLN NE2 111.454 80 VAL H 8.497 80 VAL HA 4.511 80 VAL HB 1.956 80 VAL HG11 0.826 80 VAL HG12 0.826 80 VAL HG13 0.826 80 VAL HG21 0.839 80 VAL HG22 0.839 80 VAL HG23 0.839 80 VAL CA 60.588 80 VAL CB 34.915 80 VAL CG1 22.066 80 VAL CG2 21.302 80 VAL N 120.39 81 ASN H 9.572 81 ASN HA 4.474 81 ASN HB2 3.238 81 ASN HB3 3 81 ASN HD21 7.4 81 ASN HD22 6.577 81 ASN CA 54.337 81 ASN CB 36.692 81 ASN N 127.486 81 ASN ND2 110.358 82 GLY H 8.479 82 GLY HA2 3.439 82 GLY HA3 4.006 82 GLY CA 45.208 82 GLY N 102.662 83 TRP H 8.355 83 TRP HA 4.547 83 TRP HB2 3.305 83 TRP HB3 3.172 83 TRP HD1 7.338 83 TRP HE1 10.127 83 TRP HE3 7.552 83 TRP HZ2 7.46 83 TRP HZ3 7.063 83 TRP HH2 7.167 83 TRP CA 57.081 302 83 TRP CB 29.505 83 TRP N 123.595 83 TRP NE1 129.833 84 ASP H 8.486 84 ASP HA 4.397 84 ASP HB2 2.642 84 ASP HB3 2.573 84 ASP CA 55.886 84 ASP CB 41.823 84 ASP N 126.297 85 MET H 8.086 85 MET HA 4.647 85 MET HB2 2.142 85 MET HB3 1.611 85 MET HG2 2.614 85 MET HG3 2.261 85 MET HE1 1.931 85 MET HE2 1.931 85 MET HE3 1.931 85 MET CA 54.125 85 MET CB 33.489 85 MET CG 33.162 85 MET CE 18.415 85 MET N 123.789 86 THR H 8.577 86 THR HA 4.202 86 THR HB 4.162 86 THR HG21 1.35 86 THR HG22 1.35 86 THR HG23 1.35 86 THR CA 64.971 86 THR CB 69.76 86 THR CG2 22.199 86 THR N 115.359 87 MET H 8.456 87 MET HA 4.605 87 MET HB2 2.076 87 MET HB3 1.8 87 MET HG2 2.481 87 MET HG3 2.397 87 MET HE1 2.102 303 87 MET HE2 2.102 87 MET HE3 2.102 87 MET CA 54.189 87 MET CB 32.485 87 MET CG 32.224 87 MET CE 22.95 87 MET N 123.097 88 VAL H 7.846 88 VAL HA 4.854 88 VAL HB 2.38 88 VAL HG11 0.857 88 VAL HG12 0.857 88 VAL HG13 0.857 88 VAL HG21 0.872 88 VAL HG22 0.872 88 VAL HG23 0.872 88 VAL CA 59.028 88 VAL CB 34.933 88 VAL CG1 18.871 88 VAL CG2 22.427 88 VAL N 113.068 89 THR H 8.567 89 THR HA 4.33 89 THR HB 4.598 89 THR HG21 1.207 89 THR HG22 1.207 89 THR HG23 1.207 89 THR CA 61.359 89 THR CB 70.8 89 THR CG2 21.994 89 THR N 112.758 90 HIS H 10.017 90 HIS HA 3.776 90 HIS HB2 3.453 90 HIS HB3 3.263 90 HIS HD2 6.949 90 HIS CA 61.727 90 HIS CB 28.469 90 HIS N 122.939 91 ASP H 9.26 91 ASP HA 4.286 304 91 ASP HB2 2.65 91 ASP HB3 2.346 91 ASP CA 57.504 91 ASP CB 41.965 91 ASP N 115.47 92 GLN H 7.792 92 GLN HA 3.777 92 GLN HB2 1.871 92 GLN HB3 1.871 92 GLN HG2 2.423 92 GLN HG3 2.385 92 GLN HE21 7.479 92 GLN HE22 6.88 92 GLN CA 59.038 92 GLN CB 29.071 92 GLN CG 34.912 92 GLN N 117.704 92 GLN NE2 111.261 93 ALA H 7.952 93 ALA HA 3.808 93 ALA HB1 1.249 93 ALA HB2 1.249 93 ALA HB3 1.249 93 ALA CA 55.281 93 ALA CB 19.046 93 ALA N 121.954 94 ARG H 8.233 94 ARG HA 3.396 94 ARG HB2 1.876 94 ARG HB3 1.876 94 ARG HG2 1.524 94 ARG HG3 1.351 94 ARG HD2 3.329 94 ARG HD3 3.133 94 ARG HE 7.373 94 ARG CA 60.056 94 ARG CB 30.159 94 ARG CG 26.791 94 ARG CD 43.024 94 ARG N 117.346 94 ARG NE 82.115 305 95 LYS H 8.53 95 LYS HA 3.632 95 LYS HB2 1.721 95 LYS HB3 1.621 95 LYS HG2 1.469 95 LYS HG3 1.233 95 LYS HD2 1.521 95 LYS HD3 1.521 95 LYS HE2 2.827 95 LYS HE3 2.827 95 LYS CA 59.739 95 LYS CB 32.195 95 LYS CG 26.033 95 LYS CD 29.112 95 LYS CE 41.89 95 LYS N 119.979 96 ARG H 7.587 96 ARG HA 3.842 96 ARG HB2 1.664 96 ARG HB3 1.58 96 ARG HG2 1.488 96 ARG HG3 1.298 96 ARG HD2 2.498 96 ARG HD3 2.205 96 ARG CA 57.556 96 ARG CB 29.221 96 ARG CG 26.247 96 ARG CD 42.228 96 ARG N 117.105 97 LEU H 7.519 97 LEU HA 3.957 97 LEU HB2 1.739 97 LEU HB3 1.669 97 LEU HG 1.26 97 LEU HD11 0.865 97 LEU HD12 0.865 97 LEU HD13 0.865 97 LEU HD21 0.682 97 LEU HD22 0.682 97 LEU HD23 0.682 97 LEU CA 56.692 306 97 LEU CB 42.796 97 LEU CG 27.379 97 LEU CD1 24.261 97 LEU CD2 26.302 97 LEU N 117.133 98 THR H 7.131 98 THR HA 4.681 98 THR HB 4.148 98 THR HG21 1.174 98 THR HG22 1.174 98 THR HG23 1.174 98 THR CA 60.287 98 THR CB 69.202 98 THR CG2 21.366 98 THR N 105.101 99 LYS H 6.889 99 LYS HA 4.131 99 LYS HB2 1.74 99 LYS HB3 1.74 99 LYS HG2 1.463 99 LYS HG3 1.257 99 LYS HD2 1.592 99 LYS HD3 1.592 99 LYS HE2 2.763 99 LYS HE3 2.763 99 LYS CA 57.633 99 LYS CB 32.011 99 LYS CG 24.723 99 LYS CD 29.327 99 LYS CE 41.653 99 LYS N 123.891 100 ARG H 8.682 100 ARG HA 3.876 100 ARG HB2 1.814 100 ARG HB3 1.814 100 ARG HG2 1.725 100 ARG HG3 1.725 100 ARG HD2 3.212 100 ARG HD3 3.212 100 ARG CA 58.589 100 ARG CB 30.071 307 100 ARG CG 27.562 100 ARG CD 43.322 100 ARG N 128.32 101 SER H 7.55 101 SER HA 4.178 101 SER HB2 4.023 101 SER HB3 3.685 101 SER CA 58.202 101 SER CB 62.929 101 SER N 108.729 102 GLU H 7.224 102 GLU HA 4.474 102 GLU HB2 1.715 102 GLU HB3 1.715 102 GLU HG2 2.059 102 GLU HG3 2.059 102 GLU CA 54.703 102 GLU CB 31.086 102 GLU CG 36.134 102 GLU N 121.834 103 GLU H 8.675 103 GLU HA 3.966 103 GLU HB2 1.893 103 GLU HB3 1.893 103 GLU HG2 2.229 103 GLU HG3 2.103 103 GLU CA 57.83 103 GLU CB 30.546 103 GLU CG 36.967 103 GLU N 122.965 104 VAL H 7.516 104 VAL HA 4.663 104 VAL HB 1.648 104 VAL HG11 0.438 104 VAL HG12 0.438 104 VAL HG13 0.438 104 VAL HG21 0.509 104 VAL HG22 0.509 104 VAL HG23 0.509 104 VAL CA 60.264 104 VAL CB 34.457 308 104 VAL CG1 21.078 104 VAL CG2 20.453 104 VAL N 117.462 105 VAL H 8.621 105 VAL HA 4.636 105 VAL HB 1.769 105 VAL HG11 0.482 105 VAL HG12 0.482 105 VAL HG13 0.482 105 VAL HG21 0.86 105 VAL HG22 0.86 105 VAL HG23 0.86 105 VAL CA 59.716 105 VAL CB 34.866 105 VAL CG2 22.253 105 VAL N 120.82 106 ARG H 8.765 106 ARG HA 4.748 106 ARG HB2 1.728 106 ARG HB3 1.777 106 ARG HG2 1.461 106 ARG HG3 1.407 106 ARG HD2 3.076 106 ARG HD3 3.076 106 ARG CA 55.449 106 ARG CB 30.06 106 ARG CG 28.171 106 ARG CD 43.238 106 ARG N 125.489 107 LEU H 9.46 107 LEU HA 5.212 107 LEU HB2 1.186 107 LEU HB3 1.586 107 LEU HD11 0.756 107 LEU HD12 0.756 107 LEU HD13 0.756 107 LEU HD21 0.584 107 LEU HD22 0.584 107 LEU HD23 0.584 107 LEU CA 53.216 107 LEU CB 44.49 309 107 LEU CG 26.967 107 LEU CD1 25.593 107 LEU CD2 26.876 107 LEU N 125.4 108 LEU H 7.88 108 LEU HA 5.019 108 LEU HB2 1.416 108 LEU HB3 1.333 108 LEU HG 1.203 108 LEU HD11 0.693 108 LEU HD12 0.693 108 LEU HD13 0.693 108 LEU HD21 0.67 108 LEU HD22 0.67 108 LEU HD23 0.67 108 LEU CA 54.524 108 LEU CB 44.27 108 LEU CG 26.993 108 LEU CD1 24.317 108 LEU CD2 24.312 108 LEU N 123.53 109 VAL H 9.071 109 VAL HA 5.548 109 VAL HB 1.817 109 VAL HG11 0.541 109 VAL HG12 0.541 109 VAL HG13 0.541 109 VAL HG21 0.577 109 VAL HG22 0.577 109 VAL HG23 0.577 109 VAL CA 57.915 109 VAL CB 35.182 109 VAL CG1 18.294 109 VAL CG2 21.28 109 VAL N 119.5 110 THR H 8.905 110 THR HA 5.085 110 THR HB 3.9 110 THR HG21 1.045 110 THR HG22 1.045 110 THR HG23 1.045 310 110 THR CA 60.651 110 THR CB 70.979 110 THR CG2 21.734 110 THR N 114.109 111 ARG H 8.784 111 ARG HA 4.751 111 ARG HB2 1.842 111 ARG HB3 1.842 111 ARG HG2 1.467 111 ARG HG3 1.409 111 ARG HD2 2.97 111 ARG HD3 2.903 111 ARG HE 7.214 111 ARG CA 54.668 111 ARG CB 33.387 111 ARG CG 25.81 111 ARG CD 43.363 111 ARG N 126.316 111 ARG NE 84.728 112 GLN H 8.861 112 GLN HA 4.414 112 GLN HB2 2.034 112 GLN HB3 1.986 112 GLN HG2 2.406 112 GLN HG3 2.348 112 GLN HE21 7.594 112 GLN HE22 6.857 112 GLN CA 55.824 112 GLN CB 29.451 112 GLN CG 33.858 112 GLN N 123.79 112 GLN NE2 111.755 113 SER H 8.501 113 SER HA 4.366 113 SER HB2 3.751 113 SER HB3 3.751 113 SER CA 58.268 113 SER CB 63.893 113 SER N 117.088 114 LEU H 8.218 114 LEU HA 4.324 311 114 LEU HB2 1.539 114 LEU HB3 1.539 114 LEU HG 1.451 114 LEU HD11 0.802 114 LEU HD12 0.802 114 LEU HD13 0.802 114 LEU HD21 0.845 114 LEU HD22 0.845 114 LEU HD23 0.845 114 LEU CA 55.253 114 LEU CB 42.257 114 LEU CD1 23.667 114 LEU CD2 24.728 114 LEU N 124.389 115 GLN H 8.291 115 GLN HA 4.206 115 GLN HB2 2.012 115 GLN HB3 1.881 115 GLN HG2 2.277 115 GLN HG3 2.277 115 GLN CA 56.065 115 GLN CB 29.517 115 GLN CG 33.738 115 GLN N 121.34 116 LYS H 8.239 116 LYS HA 4.19 116 LYS HB2 1.72 116 LYS HB3 1.665 116 LYS HG2 1.359 116 LYS HG3 1.359 116 LYS HD2 1.602 116 LYS HD3 1.602 116 LYS HE2 2.913 116 LYS HE3 2.913 116 LYS CA 56.336 116 LYS CB 33.131 116 LYS CG 24.747 116 LYS CD 29.078 116 LYS CE 42.129 116 LYS N 122.701 117 ALA H 8.221 312 117 ALA HA 4.238 117 ALA HB1 1.304 117 ALA HB2 1.304 117 ALA HB3 1.304 117 ALA CA 52.538 117 ALA CB 19.14 117 ALA N 125.024 118 VAL H 8.044 118 VAL HA 3.976 118 VAL HB 1.979 118 VAL HG11 0.85 118 VAL HG12 0.85 118 VAL HG13 0.85 118 VAL HG21 0.865 118 VAL HG22 0.865 118 VAL HG23 0.865 118 VAL CA 62.506 118 VAL CB 32.914 118 VAL CG1 20.843 118 VAL CG2 21.131 118 VAL N 119.586 119 GLN H 8.351 119 GLN HA 4.242 119 GLN HB2 2.018 119 GLN HB3 1.938 119 GLN HG2 2.282 119 GLN HG3 2.282 119 GLN HE21 7.471 119 GLN HE22 6.81 119 GLN CA 56.076 119 GLN CB 29.543 119 GLN CG 33.767 119 GLN N 123.994 119 GLN NE2 112.474 120 GLN H 8.415 120 GLN HA 4.226 120 GLN HB2 2.033 120 GLN HB3 1.95 120 GLN HG2 2.295 120 GLN HG3 2.295 120 GLN HE21 7.446 313 120 GLN HE22 6.813 120 GLN CA 56.283 120 GLN CB 29.575 120 GLN CG 33.751 120 GLN N 122.052 120 GLN NE2 112.499 121 SER H 8.308 121 SER HA 4.35 121 SER HB2 3.8 121 SER HB3 3.8 121 SER CA 58.662 121 SER CB 63.771 121 SER N 116.851 122 MET H 8.305 122 MET HA 4.45 122 MET HB2 2.054 122 MET HB3 1.943 122 MET HG2 2.538 122 MET HG3 2.458 122 MET HE1 2.012 122 MET HE2 2.012 122 MET HE3 2.012 122 MET CA 55.51 122 MET CB 32.796 122 MET CG 32.046 122 MET CE 16.893 122 MET N 122.049 123 LEU H 8.121 123 LEU HA 4.334 123 LEU HB2 1.57 123 LEU HB3 1.57 123 LEU HD11 0.787 123 LEU HD12 0.787 123 LEU HD13 0.787 123 LEU HD21 0.851 123 LEU HD22 0.851 123 LEU HD23 0.851 123 LEU CA 55.269 123 LEU CB 42.366 123 LEU CG 26.891 123 LEU CD1 23.163 314 123 LEU CD2 25.049 123 LEU N 123.282 124 SER H 7.793 124 SER HA 4.178 124 SER HB2 3.758 124 SER HB3 3.758 124 SER CA 59.892 124 SER CB 64.881 124 SER N 121.994 For the Glutaminase L Peptide: Residue no. Amino acid Nucleus Chemical shift 1 LYS HA 4.016 1 LYS HB2 1.794 1 LYS HB3 1.794 1 LYS HG2 1.381 1 LYS HG3 1.381 1 LYS HD2 1.641 1 LYS HD3 1.641 2 GLU HA 4.266 2 GLU HB2 1.968 2 GLU HB3 1.843 2 GLU HG2 2.192 2 GLU HG3 2.192 3 ASN H 8.62 3 ASN HA 4.666 3 ASN HB2 2.791 3 ASN HB3 2.669 3 ASN HD21 7.56 3 ASN HD22 6.869 4 LEU H 8.306 4 LEU HA 4.262 4 LEU HB2 1.565 4 LEU HB3 1.565 4 LEU HG 1.537 4 LEU HD11 0.791 4 LEU HD12 0.791 4 LEU HD13 0.791 315 4 LEU HD21 0.842 4 LEU HD22 0.842 4 LEU HD23 0.842 5 GLU H 8.381 5 GLU HA 4.196 5 GLU HB2 1.96 5 GLU HB3 1.838 5 GLU HG2 2.189 5 GLU HG3 2.189 6 SER H 8.146 6 SER HA 4.351 6 SER HB2 3.77 6 SER HB3 3.77 7 MET H 8.318 7 MET HA 4.476 7 MET HB2 2.05 7 MET HB3 1.952 7 MET HG2 2.525 7 MET HG3 2.442 7 MET HE1 2.152 7 MET HE2 2.152 7 MET HE3 2.152 8 VAL H 7.592 8 VAL HA 3.979 8 VAL HB 1.985 8 VAL HG11 0.79 8 VAL HG12 0.79 8 VAL HG13 0.79 8 VAL HG21 0.813 8 VAL HG22 0.813 8 VAL HG23 0.813