This item is only available as the following downloads:
1 CHARACTERIZATION OF THE MOBILE LOOPS IN GM2 ACTIVATOR PROTEIN AND THE FLAP REGION OF HIV 1 PROTEASE VIA ELECTRON PARAMAGNETIC RESONANCE TECHNIQUES By JEFFREY DAVID CARTER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012
2 2012 Jeffrey David Carter
3 To my grandmother, Nancy, whose love and support I will always cherish; to my Mom and Dad for their encouragement throughout my years of studies.
4 ACKNOWLEDGMENTS I would like to especially thank my grandmother Nancy Como for her unconditional love and support, and my parents Dia ne and David Carter for being wonderful role mod els. Next, I would like to express thanks to my advisor, Dr. Gail Fanucci for her diligence in training me to become a scientist. I would also like to thank all of members of my doctoral committee, Doctors Nicole Horenstein, Mavis McKenna Nick Polfer, and David Powell. I would like to thank Alex Angerhofer for time on the E 580 instrument. I would like to thank all the past and present member s of the Fanucci group for their help, friendship, and most importantly keeping me sane through the graduate schoo l experience. In particular, I like to express sincere thanks to Dr. Jamie Kear, Dr. Natasha Pirman, Dr. Jared Lynch, Adam Smith, Dr. Austin Turner, Ian deVera, and Dr. Mandy Blackburn. Each you made graduate school much more enjoyable, and I will always cherish your friendships. I would also like to thank my undergraduate mentee baffling experiments. I wish to express my gratitude to Drs. Alan G. Marshall and Jeremiah Tipton of the National High Magnetic Field Laboratory in Tallahassee Florida and Chr istine Schubert Wright of the Department o f Pharmacology, X ray Laboratory and University of Vir ginia Health System for early help with the GM2AP project. Christine and her post doc, Fang Xu, generat ed six of the single CYS mutants along with the two double mutants. Finally, I would like to thank NIH AIDS Reagents Program for free access to HIV 1 protease inhibitors, and the National High Magnetic Field Lab IHRP, NIH, NSF, a nd UF Startup for funding. The work described herein was supported by NSF MBC 0746533 and ARI DMR 9601864, NIH R37 AI28571, the UF Center for AIDS Research, NHMFL IHRP, NIH R01 GM077232 awarded to Gail E. Fanucci.
5 TABLE OF CONTENTS page ACKNOWLEDGM ENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 15 ABSTRACT ................................ ................................ ................................ ................... 23 CHAPTER 1 INTRODUCTION TO THE BIOLOGICAL SYSTEMS: GM2 ACTIVATOR PROTEIN AND HIV 1 PROTEASE ................................ ................................ ......... 25 Scope of the Dissertation ................................ ................................ ........................ 25 Introduction to GM2 Activator Protein ................................ ................................ ..... 30 Monosaccharide Naming and Abbreviation ................................ ...................... 30 Glycolipids, Glycosphingolipids, and Gangliosides ................................ ........... 30 Glycosphingolipid Degradation ................................ ................................ ......... 32 Sphingolipid Activator Proteins ................................ ................................ ......... 32 Saposin A ................................ ................................ ................................ ... 33 Saposin B ................................ ................................ ................................ ... 34 SAP C ................................ ................................ ................................ ........ 35 SAP D ................................ ................................ ................................ ........ 35 GM2 activator protein ................................ ................................ ................. 36 Hexosaminidase A ................................ ................................ ........................ 37 Lipid Storage Diseases and Diseases Associated with Abnormal Sphingolipid Metabolism ................................ ................................ ............... 37 Tay Sachs disease ................................ ................................ .................... 37 Sandhoff disease ................................ ................................ ....................... 38 Introduction to HIV 1 Protease ................................ ................................ ................ 38 Introduction to Human Immunodeficiency Virus ................................ ............... 38 Genetic Variability ................................ ................................ ............................ 39 HIV 1 Protease ................................ ................................ ................................ 40 Conformational Sampling and the Conformational Ensemble of the HIV 1 Protease Flaps ................................ ................................ .............................. 41 Inhibitors of HIV 1 Protease ................................ ................................ ............. 41 2 INTRODUCTION TO BIOPHYSICAL METHODOLOGIES ................................ ..... 57 Introduction ................................ ................................ ................................ ............. 57 Mass Spectrometry ................................ ................................ ................................ 57 Fluorescence Spectroscopy ................................ ................................ .................... 59 Circular Dichroism Spectroscopy ................................ ................................ ............ 60
6 Electron Paramagnetic Resonance Spectroscopy ................................ .................. 61 Site Directed Spin Labeling ................................ ................................ .............. 61 (1 Oxyl 2,2,5,5 Tetramethyl Pyrroline 3 Methyl) Methane thiosulfonate (MTSL) ................................ ................................ .............. 62 4 Maleimido TEMPO (MSL) ................................ ................................ ....... 63 Non native amino aci d p acetyl L phenylalanine spin label ....................... 63 Continuous Wave Electron Paramagnetic Resonance Spectroscopy (CW EPR) ................................ ................................ ................................ ............. 64 Introduction and history ................................ ................................ .............. 64 CW EPR sample requirements ................................ ................................ .. 65 CW EPR theory ................................ ................................ .......................... 65 Nitroxid e spectral line shapes ................................ ................................ .... 66 Pulsed Electron Paramagnetic Resonance Spectroscopy ............................... 68 Introduction and history ................................ ................................ .............. 68 Sample requirements ................................ ................................ ................. 69 DEER theory ................................ ................................ .............................. 69 E580 spectrometer DEER setup ................................ ................................ 75 3 SURFACE LOOP CONFORMATIONAL FLEXIBILITY OF THE GM2 ACTIVATOR PROTEIN IS NOT ALTERED BY LIGAND BINDING BUT IS NECESSARY FOR LIPID LIGAND EXTRACTION FROM BILAYER SURFACES ................................ ................................ ................................ .......... 100 Introduction ................................ ................................ ................................ ........... 100 Materials and Methods ................................ ................................ .......................... 103 Materials ................................ ................................ ................................ ......... 103 Methods ................................ ................................ ................................ .......... 103 Expression of GM2AP constructs ................................ ............................ 104 GM2AP purification protocol ................................ ................................ .... 104 Spin labeling ................................ ................................ ............................ 105 Crosslinking ................................ ................................ ............................. 106 Results and Discussion ................................ ................................ ......................... 107 4 MASS SPECTROMETRIC ANALYSIS AND CONFIRMATION OF DISULFIDE BONDING AND REPORTER SITE LABELING EFFICIENCY IN GM2AP ............ 134 Introduction ................................ ................................ ................................ ........... 134 Methods ................................ ................................ ................................ ................ 135 Direct Infusion and Reversed Phase Capillary Liquid Chromatography ......... 135 Mass Spectrometry ................................ ................................ ........................ 135 Sequential Digestion with Trypsin, GluC, and AspN ................................ ....... 138 Data Analysis and Informatics ................................ ................................ ........ 138 Results and Discussion ................................ ................................ ......................... 139 5 DOUBLE ELECTRON ELECTRON RESONANCE STUDIES OF HIV 1 PROTEASE ................................ ................................ ................................ .......... 154
7 Introducti on and Previous Work ................................ ................................ ............ 154 Materials and Methods ................................ ................................ .......................... 160 Materials ................................ ................................ ................................ ......... 160 Me thods ................................ ................................ ................................ .......... 161 Expression of HIV 1 protease ................................ ................................ .. 161 Purification of HIV 1 protease ................................ ................................ .. 161 Incorporation of the unnatural amino acid ................................ ................ 164 DEER experiments ................................ ................................ .................. 165 DEER data analysis ................................ ................................ ................. 165 Population validation ................................ ................................ ................ 165 Results and Discussion ................................ ................................ ......................... 166 Effect of Sample Freezing Conditions on the Conformational Ensemble of HIV 1PR ................................ ................................ ................................ ...... 166 Single Point Drug Induced Mutations Confer Alterations in the Conformational Ensemble of HIV 1 Protease ................................ .............. 168 Electron Paramagnetic Resonance Studies of HIV 1 Protease Via Incorporation of the Unnatural Amino Acid Acetyl Phenylalanine ................ 172 6 FUTURE DIRECTIONS ................................ ................................ ........................ 209 APPENDIX: CW AND PULSED EPR PARAMETERS ................................ ................ 212 LIST OF REFERENCES ................................ ................................ ............................. 213 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 227
8 LIST OF TABLES Table page 1 1 Common monosaccharide abbreviations. ................................ .......................... 43 1 2 Representat ive diseases associated with abnormal sphingolipid metabolism. ... 43 1 3 A comparison of HIV 1 and HIV 2. ................................ ................................ ..... 43 1 4 FDA approved drugs f or the treatment of HIV/AIDS. ................................ .......... 43 3 1 Summary of Double CYS R1 GM2AP Construct Distance Experiments .......... 119 3 2 Differential Scanning C alorimetry Values For GM2AP Constructs ................... 119 3 3 GM2AP Cross linking Ligand Extraction Data ................................ .................. 119 5 1 Distance parameters for HIV 1PR C (apo) and various freezing conditions. .... 174 5 2 Distance parameters for HIV 1PR C (IDV) and various freezing conditions. .... 174 5 3 Distance parameters for HIV 1PR (TPV) and various freezing conditions. ....... 175 5 4 Distance parameters for HIV 1PR (CaP2) and various freezing conditions. ..... 175 5 5 Summary of distance parameters obtained from DEER EPR of HIV 1PR. ....... 176 A 1 Typical CW EPR parameters used for work reported within ............................. 212 A 2 Typical pulsed EPR parameters used for DEER reported within. ..................... 212 A 3 4 pulse table used for DEER experiments performed in deuterated matrices. 212
9 LIST OF FIGURES Figure page 1 1 Glycolipid structure, where Y = lipid ................................ ................................ ... 44 1 2 General struct ure of a glyceroglycolipid ................................ .............................. 44 1 3 General structure of a glycosylphosphotidylinositol ................................ ............ 44 1 4 Structure of sphingosine ................................ ................................ ..................... 45 1 5 Glycosphingolipid structure ................................ ................................ ................. 45 1 6 General structure of a ganglioside ................................ ................................ ...... 45 1 7 Ganglioside GM2 structure ................................ ................................ ................. 46 1 8 Glycos phingolipid degradation pathway ................................ ............................. 47 1 9 Crystal structure of Saposin A (PDB ID 2 DOB) ................................ ................. 48 1 10 Unit cell crystal structure of Saposin B (PDB ID 1N69) ................................ ...... 48 1 11 Unit cell crystal structure of Saposin C (PDB ID 2QYP) ................................ ..... 48 1 12 Unit cell crystal structure of Saposin D (PDB ID 2RB3) ................................ ...... 49 1 13 Representation of ganglioside hydrolysis ................................ ........................... 49 1 14 Cartoon structure of GM2AP showing the 4 disulfide bonds in red (PDB 1G13). ................................ ................................ ................................ ................ 50 1 15 Cartoon representation of GM2AP binding lipid membranes and extracting its specific substrate GM2 (PDB 1G13). ................................ ................................ .. 50 1 16 hexosaminidase A (PDB ID 2GJX). ................................ ............. 51 1 17 HIV 1 life cycle. ................................ ................................ ................................ ... 52 1 18 Cartoon representation of the gag and gag pol polyproteins cleaved by HIV 1 protease during viral maturation. ................................ ................................ ........ 53 1 19 Structure of Subtype B HIV 1 Protease (PDB ID 2BPX) ................................ ..... 53 1 20 Flap conformations of HIV 1 protease captured by molecular dynamics simulations ................................ ................................ ................................ .......... 54 1 21 Points of inhibition within the HIV 1 viral life cycle. ................................ ............. 55
10 1 22 Structures of the FDA approved protease inhibitors. ................................ .......... 56 2 1 Schematic diagram of a simple mass spectrometer with a sector type mass analyzer. ................................ ................................ ................................ ............. 91 2 2 Schematic diagram of a sta ndard fluorescence spectro meter ............................ 91 2 3 Schematic diagram of a standard circul ar dichroism instrumental ...................... 92 2 4 Effect of choice of spin label on the deriva tive EPR line shape of HIV 1PR Subtype B si ................................ ................................ ................................ ......... 92 2 5 Site directed spin labeling scheme showing the reaction of MTSL with a free thiol group of an engineered CYS residue. ................................ ......................... 93 2 6 Structure of MSL spin label. ................................ ................................ ............... 93 2 7 Spin labeling scheme for the incorporation of the K1 spin label moiety. ............. 94 2 8 Scheme for electron transition in a magnetic field ................................ .............. 94 2 9 Cartoon r epresentation of EPR. ................................ ................................ ........ 95 2 10 Dependence of EPR spectral line shape on motion, correlation time, and degree of anisotropy. ................................ ................................ .......................... 96 2 11 The 2 + 1 pulse DEER sequence. ................................ ................................ ...... 96 2 12 The 3 pulse DEER sequence ................................ ................................ ............. 96 2 13 The 4 pulse DEER sequence ................................ ................................ ............. 97 2 14 Absorption spectra for a nitroxide spin label ................................ ....................... 97 2 15 Sample dipolar evolution curve showing locations of raw dipolar modulation .... 98 2 16 Selection of regularization parameter in TKR DEER analysis method ............... 98 2 17 Example s of processed DEER data. ................................ ............................... 99 3 1 Structures of GM2AP and MTSL. ................................ ................................ ..... 120 3 2 The three unique monomers (A, B, C) of apo GM2AP (PDB ID 1G13). ........... 120 3 3 The red spheres in monomer A of GM2AP (PDB 1G13) represent the C positions of the reporter sites chosen for SDSL EPR experiments. .................. 121 3 4 Stack plot of EPR line shapes from six sites in the flexible loops of GM2AP. ... 121
11 3 5 DEER data from GM2AP double mutants ................................ ....................... 122 3 6 GM2AP construct I72C/S130C l ong pass filtered and background subtracted dipolar echo curve ................................ ................................ ............................ 123 3 7 GM2AP constru ct I72C/S130C bound to GM2 l ong pass filtered and background subtracted dipolar echo curve ................................ ....................... 124 3 8 GM2AP construct I66C/S126C l ong pass filtered a nd background subtracted dipolar echo curve ................................ ................................ ............................ 125 3 9 GM2AP constr uct I66C/L126C bound to GM2 l ong pass filtered and background subtracted dipolar echo curve ................................ ....................... 126 3 10 Differential scanning calorimetry for GM2A P constructs A60R1 and L126R1. 127 3 11 Overlay of are a normalized EPR spectra and MOMD simulated spectra of the ten spin labeled GM2AP constructs ................................ ................................ .. 128 3 12 GM2AP chromatograms from pro tein purification. ................................ .......... 129 3 13 Representative EPR spectra of GM2AP L126R1 and I66R1 as a function of temperature ................................ ................................ ................................ ...... 130 3 14 Linear regression of the ratio of the fractions of mobile and immobile L126R1 populations plotted as a function of inverse temperature ................................ 131 3 15 Representative diagram of the dansyl DHPE fluorescent lipid extraction assay ................................ ................................ ................................ ................ 132 3 16 Tethering results from double mutant GM2AP constructs ................................ 133 3 17 Representative diagram of the gel filtration GM2 lipid extraction assay. Gel filtration assay for GM2 extract ion ................................ ................................ .... 133 4 1 16.5% Tris HCL Biorad pre cast gel of GM2AP constructs purified for MS analysis ................................ ................................ ................................ ............ 146 4 2 Broadband 14.5 T electrospra y positive ion of intact wild type, L126C MSL, I66C MSL, N136C M SL, I66C/L126C MSL ................................ ...................... 147 4 3 Parent ion FTICR data for GM2AP MSL ................................ .......................... 148 4 4 Protease digestion scheme for GM2AP L126 MSL. ................................ ......... 149 4 5 FTICR results for GM2AP L126 MSL peptide. ................................ .................. 150 4 6 Scheme presenting the theoretical disulfide bound peptides after sequential proteolytic digestion ................................ ................................ ......................... 151
12 4 7 CID MS/MS disulfide identification ................................ ................................ ... 152 4 8 ECD MS/MS disulfide identification. ................................ ............................... 153 5 1 H IV crystal structure ................................ ................................ ......................... 176 5 2 DEER results of subtype B HIV 1Pr with and without the FDA approved inhibitor ................................ ................................ ................................ ............. 177 5 3 DEER ex periments are typically run at cryogenic temperatures ....................... 177 5 4 Locations of polymorphisms in PRE and POST, with respect to Subtype B (LAI). ................................ ................................ ................................ ................. 178 5 5 Unnatural amino acid p acetylphenylalanine and structurally similar amino acid tyrosine. ................................ ................................ ................................ .... 178 5 6 Chemical reaction for Incorporation of unnatural amino acid to form spin lab eled side chain K1. ................................ ................................ ...................... 178 5 7 pSUPAR3 plasmid map ................................ ................................ .................... 179 5 8 Sequence alignment for HIV 1 protease LAI with constructs B si C si PRE i I63P si A71V si and POST i ................................ ................................ ................ 179 5 9 HIV 1PR Subtype C apo frozen at 20 ................................ .............................. 180 5 10 HIV 1PR Subtype C apo frozen in liquid nitrogen ................................ ............. 181 5 11 HIV 1PR Subtype C apo frozen in isopentane ................................ ................. 182 5 12 HIV 1PR Subtype C apo frozen in isopentane from 37C ................................ .. 183 5 13 Stacked distance distributi on profiles of HIV 1PR Subtype C, K55R1, apo after various freez ing conditions ................................ ................................ ....... 184 5 14 HIV 1PR Subtype C TPV bound frozen at 20 C ................................ ............. 185 5 15 HIV 1PR Subtype C TPV bound frozen in liquid nitrogen. ................................ 186 5 16 HIV 1PR Subtype C TPV bound fro zen in isopentane ................................ ..... 187 5 17 HIV 1PR Subtype C TPV bound apo heated to 37 C then frozen in isopentane ................................ ................................ ................................ ........ 188 5 18 Stacked distance distribution profiles of HIV 1PR Subtype C K55R1 bound to TPV ................................ ................................ ................................ .................. 189
13 5 19 HIV 1PR Subtype C IDV bound frozen at 20. Background subtracted dipolar echo cu rve ................................ ................................ ................................ ........ 190 5 20 HIV 1PR Subtype C IDV bound frozen in liquid nitrogen ................................ .. 191 5 21 HIV 1PR Su btype C IDV bound frozen in isopentane ................................ ...... 192 5 22 HIV 1PR Subtype C IDV bound apo heated to 37 C then frozen in isopentane ................................ ................................ ................................ ........ 193 5 23 Stacked distance distribution profiles of HIV 1PR Subtype C, K55R1, bound with IDV ................................ ................................ ................................ ............ 194 5 24 HIV 1PR Subtype C CA p2 bound frozen at 20 C ................................ ........... 195 5 25 HIV 1PR Subtype C CA p2 bound frozen in liquid nitrogen ............................. 196 5 26 HIV 1PR Subtype C CA p2 bound frozen in isopentane ................................ .. 197 5 27 HIV 1PR Subtype C CA p2 bound apo heated to 37 C then frozen in isop entane ................................ ................................ ................................ ........ 198 5 28 Stacked distance distribution profiles of HIV 1PR Subtype C, K55R1, bound with CA p2 ................................ ................................ ................................ ........ 199 5 29 DEER resu lts for drug pressure mutations ................................ ....................... 1 99 5 30 Population analysis and molecular dyna mics simulations for HIV 1 PR .......... 200 5 31 HIV 1PR PRE apo ................................ ................................ ............................ 201 5 32 HI V 1PR POST apo ................................ ................................ .......................... 202 5 33 HIV 1PR Subtype B L63P apo ................................ ................................ .......... 203 5 34 HIV 1PR Subtype B A71V Apo ................................ ................................ ......... 204 5 35 The relative populations (%) for closed, semi open, tucked/curled, and wide open conformations of PRE apo and POST apo ................................ ............. 205 5 36 Mas s spectra of spin labeled PRE ................................ ................................ .... 205 5 37 Mass spectra of spin labeled POST ................................ ................................ 206 5 38 Mass spectra of spin labeled L63P ................................ ................................ ... 206 5 39 Mass spectra of spin labeled A71V ................................ ................................ .. 207 5 40 Analysis of mass spectra for HIV 1 protease constructs. ................................ 207
14 5 41 Bsi Amber. Long pass filtered and background subtra cted dipolar echo curve overlaid with the TKR regenerated echo curve from DeerAnalysis ................... 208 6 1 HSQC NMR spectra of uniformly labeled 15N GM2AP upon addition of 4x molar excess GM2 micelles ................................ ................................ .............. 210 6 2 Crystallization trials of wild type GM2AP ................................ .......................... 211
15 LIST OF ABBREVIATION S (+)mRNA plus sense messenger RNA angstrom AIDS Acquired Immunodeficiency Syndrome AmpR ampicillin resistance APV Amprenavir ARV antiretroviral drugs AT adenine thymine ATV Atazanavir BEV baculovirus BME Mercaptoethanol, 2 Mercaptoethanol BM(PEG) 2 1,8 bis maleimidotriethlyeneglycol BM(PEG) 3 1 ,11 bis maleimidotriethyleneglycol bp base pair BSA bovine serum albumin B si subtype B stabilized inactivated C Celsius CD circular dichro ism CD4 Cluster of Differentiation 4 CI chemical ionization CID collision induced dissociation CRF circulating recombinant form C si subtype C stabilized inactivated
16 CW continuous wave CYS cysteine Da Dalton dB decibel DEER do uble electron electron resonance DNA deoxyribonucleic acid DRV Darunavir DTME dithio bismaleimidoethane E. coli Escherichia coli EI entry inhibitor Env envelope EPR electron paramagnetic resonance ER endoplasmic reticulum ESEEM elec tron spin echo envelope modulation ESI electrospray ionization ESR electron spin resonance FDA Food and Drug Administration FI fusion inhibitors FIV feline immunodeficiency virus FPLC fast protein liquid chromatography FPV Fosamprenav ir FRET fluorescence resonance energy transfer FT ICR Fourier transform ion cyclotron resonance
17 FWHM full width at half maximum G Gauss Gal g alactose GalN g alactosamine GalNAc N acetylgalactosamine GHz gigahertz Glc g lucose GlcA glucuronic acid GlcCer glucosylceramide GlcN g lucosamine GlcNac N acetylglucosamine gp 120 glycoprotein 120 gp 41 glycoprotein 41 HAART Highly Active Antiretroviral Therapy HIV Human Immunodeficiency Virus HIV 1 Hum an immunodeficiency virus type 1 HIV 1PR HIV 1 protease HIV 2 Human immunodeficiency virus type 2 HPLC high pressure liquid chromatography IAP 3 (2 Iodoacetamido) PROXYL IASL 4 (2 Iodoacetamido) TEMPO IDA iminodiacetic acid IDV Indinavi r
18 II integrase inhibitor IHRP In house research program IMAC immobilized metal ion affinity chromatography Int integrase IP isopentane IPTG Isopropyl D 1 thiogalactopyranoside ITC isothermal titration calorimetry J Joule K Kelv in kD kilodalton LB Luria Bertani media LC liquid chromatography LN 2 liquid nitrogen LPV Lopinavir LTR long terminal repeat LTQ linear qu adrupole M major group MALDI matrix assisted laser desorption ionization MC Monte Carlo MCS multiple cloning site MD molecular dynamics MDR769 Multi drug resistant variant 769 min minute
19 mL milliliter mM millimolar mol mole Mol. wt. molecular weight MS mass spectrometry mS milli Siemens ms milli second MSL 4 Maleimido TEMPO MTSL (1 oxyl 2,2,5,5 tetramethyl 3 pyrroline 3 methyl) methanethiosulfonate MHZ megahertz NaOAc sodium acetate NHLBI National Heart, Lung, and Blood Institute NHMFL National High Magnetic Field Lab NIH National Institute of H ealth NFV Nelfinavir Ng nanogram NK natural killer nL nanoliter Nle norleucine nm nanometer NMR nuclear magnetic resonance NNRTI non nucleoside reverse transcriptase inhibitor NOE Nuclear Overhauser Effect
20 NRTI nucleoside reve rse transcriptase inhibitor NSF National Science Foundation NTA nitrilotriacetic acid O outlier group OD optical density OI Opportunistic infection ORF open reading frame PDB Protein Data Bank PEG polyethylene glycol PI protease inhibitors pmol picomol ppm part per million PMPR pentamutated protease R1 MTSL RNA ribonucleic acid rms root mean square rpm rotations per minute RT reverse transcriptase RTV Ritonavir Sap A Saposin A Sap B Saposin B Sap C Saposin C Sap D Saposin D
21 SDS sodium dodecyl sulfate SDSL site directed spin labeling SDS PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis SIV Simian Immunodeficiency virus SIVcpz Simian Immunodeficiency virus, chimpanzee SIVsm Simian Immunodeficiency virus, sooty mangabey SNR signal to noise ratio SPR surface plasmon resonance SU surface SQV Saquinavir T Tesla TKR Tikhonov Regularization T m melting temperature T M phase memory time TM transmemb rane TPV Tipranavir tRNA transfer RNA TSD Tay Sachs disease UF University of Florida L microliter UNAIDS Joint United Nations Programme on HIV/AIDS UV ultraviolet UV Vis ultraviolet visible spectroscopy
22 VMD visual molecular dynamics WHO World Health Organization WT wild type
23 Abstract of Dissertation Presented to the Grad uate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHARACTERIZATION OF THE MOBILE LOOPS IN GM2 ACTIVATOR PROTEIN AND THE FLAP REGION OF HIV 1 PROTEASE VIA ELECTRON PAR AMAGNETIC RESONANCE TECHNIQUES By Jeffrey David Carter August 2012 Chair: Gail E. Fanucci Major: Chemistry The work in this dissertation focuses on two unrelated proteins, namely GM2 activator protein (GM2AP) and HIV 1 protease (HIV 1PR) each playing a role in health and disease. GM2AP malfunction can be involved in the lipid storage disease Tay Sachs, and HIV 1PR is the enzyme responsible for maturation of the human immunodeficiency virus (HIV) Site directed spin labeling (SDSL) paired with electron paramagnetic resonance spectroscopy (EPR) is an excellent technique for examining the conformations and conformational ch anges of a protein, and was utilized here, mobile region. GM2 AP is thought to sample three distinct conformations, although the functional importance of these conformers has not yet been defined. Various X ray structures with different lipid substrates reveal multiple conformations of a mobile loop region and t hese conformations suggest a functional conformation change that may occur upon lipid ligand or vesicle binding Here, multiple conformations of the mobile loop in solution were detected by EPR, indicating that the conformations identified in various cryst al
24 structures are present in solution and may play a functional role in ligand binding or interactions with lipid vesicles. Distance measurements from EPR, however, did not lead to detection of local conformational changes in the loop regions as a result o f ligand binding. Furthermore, functional assays showed that immobilization of the loops inhibited the ability of GM2AP to bind its ligand, indicating that although GM2AP may not undergo a global conformational change as a result of ligand binding, loop pl asticity is essential to its role as a general lipid transfer protein. HIV 1PR functions as a homodimer with two mobile loops regulating access of substrate to the active site cavity. The double electron electron resonance ( DEER ) results rep orted here for HIV 1PR show most notably, that single dru g induced mutations even outsid e of the active site pocket are sufficient to cause a substantial shift in the flap conformations likely play ing an important role in drug resistance and efficacy The conformationa l ensemble for the parental subtype B suggests that the flaps favor the semi open conformation. The addition of the single mutation A71V, and closed conformation. The re sults indicate a possible mechanism for combining primary and secondary mutations to effect drug resistance, where conformational ensemble equilibria are altered which may then impact enzyme efficiency and altered inhibitor binding affinities.
25 CHAPTER 1 INTRODUCTION TO THE BIOLOGICAL SYSTEMS: GM2 ACTIVATOR PROTEI N A ND HIV 1 PROTEASE S cope of the Dissertation This dissertation describes experiments carried out with two protein systems, namely GM2 activator protein and HIV 1 protease. Chapter 1 is entitl ed Introduction to Biological Systems, and it serves as an introduction to the biological aspects of the research included within. First, the biological system of the GM2 activator protein is described in detail. This includes first describing and provid ing structures of the glycolipids, glycosphingolipids, and gangliosides. Next, the sphingolipid degradation pathway is described, including relevant enzymes and accessory proteins, and lipid storage diseases that can occur as a result of a genetic malfunc tion in any number of proteins. Following this is a section describing the structure and function of the sphingolipid activator proteins, incl uding saposins A D, and the GM2 activator protein, hexosaminidase A. The crystal structu res of each of these proteins are shown and examined with respect to space group and unit cell. Tay Sachs and Sandhoff disease are then discussed as two relevant GM2AP related lipid storage diseases. Following the introduction to the GM2 activator protein is an introduction to HIV 1 protease. Starting with a discussion of the worldwide impact, human immunodeficiency virus (HIV) and the autoimmune deficien cy syndrome (AIDS) are discussed. A brief introduction to the virus and its life cycle are provided, followed by a discussion regarding points of inhibition within the viral life cycle. A summary of current FDA approved drugs is given, including structur es and details about each of the nine FDA
26 approved protease inhibitors, namely Indinavir, Ritonavir, Amprenavir (Fosamprenavir), Atazanavir, Darunavir, Saquinavir, Tipranavir, Nelfinivir, and Lopinavir. Chapter 2, entitled Introduction to Biophysical Me thodologies provides a summary of the biophysical techniques utilized here. Several different methodologies are described including mass spectrometry, fluorescence spectroscopy, circular dichroism spectroscopy (CD), site directed spin labeling, continuo us wave (CW) and pulsed electron paramagnetic resonance spectroscopy (EPR) In ad dition to the introductions given in Chapter 2, each research chapter will have a detailed materials and methods section for all experiments contained within that respective chapter Chapter 3 is entitled Surface Loop Conformational Flexibility of the GM2 Activator Protein is not Altered by Ligand Binding but is Necessary for Lipid Ligand Extraction from Bilayer Surfaces Site directed spin labeling was utilized to invest igate the flexible loop regions of GM2AP. We were able to engineer a ninth and tenth cysteine in the protein and isolate homogeneous samples of spin labeled GM2AP. Analysis of the EPR spectral line shapes and MOMD simulations for spin labels in the disord ered chain segment and both loops are consistent with the various conformations seen in X ray structures. Multiple conformations of the mobile loop in solution were detected by EPR, indicating that the conformations seen in the crystal structure are presen t in solution and may play a functional role in ligand binding or interactions with lipid vesicles. Although, distance measurements from EPR were not able to detect local conformational changes in the loop regions as a result of ligand binding. The halo state of R1 mutants was confirmed by differential scanning calorimetry, which observed a 0.9 C increase in the thermotropic unfoldi ng of GM2AP constructs with GM2 bound. Furthermore, functional
27 assays showed that immobilization of the loops inhibited th e ability of GM2AP to bind its ligand, indicating that although GM2AP may not undergo a global conformational change as a result of ligand binding, loop plasticity is essential to its role as a general lipid transfer protein. Chapter 4 is entitled Mass Spe ctrometric Analysis and Confirmation of Disulfide Bonding and Reporter Site Labeling Efficiency in GM2AP Mass spectrometry enables direct identification of posttranslational modifications, including disulfide bonds ( Tipton, Carte et al 2009; Bean, Carr e t al, 1992; Badock, Raida et al, 1998; Gorman, Wallis, 2002; Xu, Zhang et al, 2008; Yen, Joshi et al, 2002; Nair, Nilson et al,2006; Popolo, Ragni et al, 2008). Because of the analytical challenge associated with verifying multiple disulfide bonds, a multi tier assay was developed (Tipton, Carter 2009) Following the identification of the intact protein a systematic approach was used to verify proper folding and labeling of a recombinantly overexpressed and MSL labeled L126C GM2AP construct. Furthermore, ver ification of disulfide connectivity was per formed from sequential trypsin, Glu C, and Asp N enzymatic digestions. The data acquired from all three sequential digests by accurate mass measurement MS and MS/MS sequence tags resulted in quick identification o f all four disulfide peptides as well as the sequence location of the MSL Cys. The result is a high confidence MS assay that will be easy to apply to other spin labeled proteins containing multiple disulfide linkages. Chapter 5 is entitled Double Electron Electron Resonance Studies of HI V 1 Protease SDSL DEER was used to further the examination of HIV 1PR flap conformational ensembles DEER data are generally collected at cryogenic temperatures due to a phase memory time that is too short for detection of the spin
28 echo at room temperature. The question is often raised on whether the conformations trapped at 65K accurately represent the thermo dynamic conformations sampled. Distances between spin labeled sites in the flap region of an apo HIV 1PR Subtype C co nstruct, as well as CA p2, Indinavir (IDV) and Tipranavir (TPV) bound HIV 1PR Subtype C construct s, were measured following four different freezing conditions : a slow 20 C freeze, an intermediate liquid nitrogen freezing freeze, and rapid freezes from bo th 25 C and 37 C using isopentane in c ombination with liquid nitrogen Re s ults indicated very small differences in the captured conformational ensemble of HIV 1PR with varying freezing methods. Additional studies were done to determine the conformationa l ensembles for four protease constructs, namely PRE and POST, and L63P and A71V secondary mutants. PRE and POST were derived from a pediatric subject infected via maternal transmission with HIV (Ho, Coman et al 2008) before and after PI treatment; resp ectively. T he gag pol alleles were isolated from serial blood samples obtained over 7 years, starting before therapy initiation (PRE) and after the development of multiple drug resistance following 77 weeks of PI therapy with Ritona vi r (RTV) and an additio nal 16 weeks with Indinavir (IDV) (POST) (Barrie, Perez et al 1996). PRE contains polymorphisms caused by the norma l process of genetic drift. Results indicate that PRE adopts a predominately closed flap conformation, but that the emergence of drug press ure selected mutations after (POST) causes the conformational sampling to favor a semi open population. Also investigated were the effects of two second ary mutations L63P and A71V on the conformational ensemble L63P is hypothesized to alter the conformat ional sampling resulting in a more closed state in the PRE construct. A71V a common secondary mutation seen in mu lt i drug
29 resistant constructs (Clemente, Hemraini et al 2003), is thought to induce a more closed conformation. A minimal amount of early w ork was done to incorporate and site directed spin lab el an un natural amino acid into HIV 1PR at position 55 of the flaps. This method differs from the traditional approach to site directed spin labeling, in which a thiol reactive spin label is attached to a native or non native CYS residue. Previous work from the lab of Wayne Hubbel reported an orthogonal labeling strategy that does not label any functional group found within the 20 naturally occurring amino acids (Fleissner, Brustad et al, 2009). In thi s method, Fleissner utilized the genetically encoded unnatural amino acid p acetyl L phenylalanine (p AcPhe) and reacted it with a hydroxylamine reagent to yield a nitroxide side chain termed K1 They were able to demonstrate success using seven different mutants of T4 lysozyme, each containing a single p AcPhe at a solvent exposed helix site. It was shown that the EPR spectra of the K1 mutants had higher nitroxide mobiliti es than the spectra of mutants containing the R1 side chain Here we were successful in expressing HIV 1 protease K55 p acetyl L phenylalanine in large scale inclusion bodies, purifying, refolding (with mediocre success) in acid ic solution, and spin labeling; finally, a low qual ity DEER dataset was obtained. Chapter 6 is called Future Dir ections and discusses early protein crystallization trials and NMR work on GM2AP The results from SDSL EPR, DSC, and fluorescence experiments of GM2AP CYS constructs indicate that the flexible loops do not undergo conformational exchange as a result of ligand binding. But, loop mobility is an important attribute in the ability of GM2AP to extract its multiple ligands. T o further investigate how ligand binding alters the average conformation of GM2AP, uniformly 15 N labeled
30 protein was generated for solu tion nuclear magnetic resonance ( NMR ) heteronuclear quantum coherence ( HSQC ) measurements The bind ing of GM2 does not induce a shift i n most of the resonances, only a few peaks are seen to shift or disappear, indicating that GM2AP does not undergo a glob al conformational shi f t as a result of ligand binding. NMR a ssignments for GM2AP are lacking, and future NMR studies are aimed at mapping out the specific regions in GM2AP that a re altered during interactions with ligands. In an attempt to overcome the ove rlapping NMR resonance peaks from the fast conformational exchange sub s timescale, crystallizations trials have been started in attempted to gain amino acid specific assignments from solid state NMR. Introduction to GM2 Activator Protein Monosacchari de Naming and Abbreviation basic form of carbohydrate found in nature. Monosaccharides are named and often abbreviated, and a list of common monosaccharide abbreviations is shown i n Table 1 1. Glycolipids, Glycosphingolipids, and Gangliosides Glycolipids are part of a family of compounds called the glycoconjugates. They are glycosyl derivatives of lipids with a carbohydrate group attached, as shown in the basic glycolipid struct ure in Figure 1 1. For more information on glycolipids, the reader is referred to several excellent reviews (Stults, Sweeley et al. 1989) The term glycolipid refers broadly to a compound that contain s at least one monosaccharide residue often more, bound by a glycosidic linkage to a hydrophobic moiety such as a sphingoid or ceramid e, among others These molecules are often found on the surface of cells to provide surface markers for cellular recognition. There exist several subtypes of glycolipids, including glyceroglycolipids, glycosphingolipids, and
31 glycosylphosphotidylinositols. Glyceroglycolipids, which include galactolipids and sulfolipids, are glycolipids with a glycerol backbone. A representative structure is shown in Figure 1 2. G lycosylphosphatidyl inositols (GPIs) are glycolipid molecules that contain saccharides glycosidically linked to an inositol moiety of a phosphatidylinositol or glycoproteins (Ferguson and Williams 1988; Low and Saltiel 1988) The general structure of a GPI is shown in Figure 1 3. GPIs are widespread in nature, and often serve as protein anchors on cell surfaces. Generally, GPI anchors are covalently linked to the C terminal end of a protein or peptide via an amide linkage (Ferguson and Williams 1988) Glycosphingolipids, which include cerebrosides, gangliosides, globosides, sul fatides, and glycophosphosphingolipids, are a subtype of glycolipids containing at least one monosaccharide and the amino alcohol sphingosine. The structure of sphingosine is shown in Figure 1 4, and the general structure of a glycosphingolipid is shown i n Figure 1 5. Gangliosides, also known as sialoglycosphingolipids are the most complex of the glycolipids. They are composed of glycosphingolipids with one or more sialic acid groups linked to the sugar group. Found predominantly on the surface of nerv e cells, many different gangliosides have been identified, whose structural differences tend to involve the stru cture of the sugar The oligosaccharide chains of gangliosides actually contain N acetylneuraminic acid (NeuNAc), an acetylated derivative of s ialic a cid, which imparts a negative charge. The ceremide is linked via its C glucosyl residue, which galactosyl residue. The general structure of a ganglioside is shown in Figure 1 6. Examples of gangliosides include GM1 and G M2. The names given for gangliosides are a description of their structure. The letter G refers to ganglioside, and the subscripts
32 M, D, T and Q indicate s the number of sialic acids ( mono di tri and quatra sialic acid ) The numerical subscripts 1, 2 an d 3 refer to the carbohydrates attached to the ceramide, where 1 equals GalGalNAcGalGlc ceramide, 2 equals GalNAcGalGlc ceramide and 3 equals GalGlc ceramide The structure of ganglioside GM2 is shown in Figure 1 7. Glycosphingolipid Degradation Glycosph ingolipids are degraded by exoglycosidases and hydrolases. In addition to these enzymes, non catalytic sphingolipid activator proteins are required to activate and deliver the substrate to the enzyme for cleavage. Some activator proteins function by bind ing to GSLs and forming aqueous soluble complexes, which provide aqueous soluble enzymes access to a lipid that would otherwise be in a hydrophobic environment. Figure 1 8 shows a diagram of part of the glycosphingolipid degradation pathway. For more inf ormation, readers are referred to many excellent reviews (Sandhoff and Kolter 1996) Degradation of the glycosphingolipids takes place in lysosomal compartments within the cell (Sand hoff and Kolter 1995) GSLs are endocytosed, trafficked and sorted through early and late endosomal compartments on the way to the lysosome, where specific enzymes sequentially cleave sugar groups, eventually producing ceramide, as shown in Figure 1 8, w hich is finally deacylated to sphingosine (Griffiths, Hoflack et al. 1988) More than ten different enzymes are involved in this degradation process. Sphingolipid Activator Proteins Saposins, an acronym for sp hingolipid activator proteins (SAPs), function to activate glycosphingolipid degradation by interacting with lipid molecules and presenting them for hydrolysis. In other words, SAPs are non enzymatic accessory proteins
33 required for sphingolipid hydrolysis by specific hydrolases. There exist five known SAPs, including saposins A D and GM2 activator protein (Darmoise, Maschmeyer et al. 2010) For an excellent review of the sphingolipid activator proteins, one can refer to a 2010 Ad vances in Immunology chapter called The Immunological Functions of Saposins by Alexandre Darmoise et al. (Darmoise, Maschmeyer et al. 2010) The PSAP gene encodes for a highly conserved 524 amino acid precursor glycoprotein ca lled prosaposin, which is proteolytically processed to liberate four cleavage products called SAPs A D (Furst, Machleidt et al. 1988; O'Brien, Kretz et al. 1988) These proteins have a high level of homology. Eac h domain contains approximately 80 amino acid residues, each with two prolines, six cysteine residues, nearly identical disulfide connectivity, and similar sites of glycosylation, with one in saposins B, C, and D, and two in saposin A. Saposins A D are mo stly helical in structure, as seen in the ribbon diagrams shown in Figures 1 9 1 12. GM2AP is encoded by different gene. With a cup topology GM2AP is not structurally homologous to SAPs A D (Darmoise, Maschmeyer et al. 2010 ) Saposin A SAP A activates the hydrolysis of 4 methlyumbelliferyl glucoside, glucocerebroside, and galactocerebroside (Morimoto, Martin et al. 1989) SAP A deficiency leads to the late onset lipid storage disease called gl oboid cell leukodystrophy that is very similar to Krabbe disease, a condition caused by a deficiency in galactosylceramide galactosidase (Morimoto, Martin et al. 1989; Matsuda, Vanier et al. 2001; Spiegel, Bach et al. 2005; Ahn, Leyko et al. 2006; Locatelli Hoops, Remmel et al. 2006; Darmoise, Maschmeyer et al. 2010) SapA has been crystallized and the crystal structure solved by X Ray diffraction to a resolution of 2.00
34 Angstroms and a space group described as P 2 1 2 1 2 A ribbon diagram of the crystal structure is shown in Figure 1 9. Saposin B Saposin B was the first of the saposin activator proteins to be discovered, and is often considered a non specific activator protein (Mehl and Jatzkewitz 1964) It is necessary for the hydrolysis of sulfatides by arylsulfatase A as well as Gb3 and galactosidase (Wenger, DeGala et al. 1989; Schlote, Harzer et al. 1991; Ahn, Faull et al. 2003; Darmoise, Maschmeyer et al. 2010) SapB also works with GM2AP for degradation by hexosaminadase A (Wilkening, Linke et al. 2000) As an added function, SapB also binds phosphatidylinositol and transfers the phospholipid between biologic al membranes (Ciaffoni, Tatti et al. 2006) Defects in SapB can lead to several lipid storage disorders, such as metachromatic leukosystrophy (MLD) (Li, Kihara et al. 1985; Sun, Witte et al. 2008) SapB has been crystallized and the crystal structure solved by X Ray diffraction to a resolution of 2.20 Angstroms (Ahn, Faull et al. 2003) A ribbon diagram of the crystal structure is shown in Figure 1 10 The protein is a homodimer with the lipid binding cavity found between the monomers. The space group is described as P 3 1 2 1 (Ahn, Faull et al. 2003) The unit cell contains three independent peptide chains called A, B, and C (depicted as white, grey, and black, respectively, in Figure 1 10) that form two distinct homodimers. The first dimer, called dimer AB is formed by chains A and B (white and grey, respectively), and the second dimer, called the V shaped fold of each monomer is stabilized by two disulfide bonds (Ahn, Faull et al. 2 003)
35 SAP C Saposin C was the second saposin to be discovered and stimulates the hydrolysis of glycocerebroside by glycosylceramidase and galactocerebroside by galactoslyceramidase (Ho and O'Brien 1971) Defects in SapC can lead to a juvenile form of the lipid storage disease called Gaucher disease, where B GlcCer is accumulated (Ho and O'Brien 1971) SapC has been crystallized and the crystal structure solved by X Ray diffraction to a res olution of 2.45 Angstroms (Rossmann, Schultz Heienbrok et al. 2008) A ribbon diagram of the crystal structure is shown in Figure 1 11. The protein is a homodimer wi th boomerang shaped monomers and a space group described as C 2 2 2 1 (Rossmann, Schultz Heienbrok et al. 2008) SAP D Saposin D is arguably the least studied saposin to date but is by far the most abundant saposin in normal tissue (O'Brien and Kishimoto 1991) SapD activates the hydrolysis of ceramide by avid ceramidase. Defects in SapD lead to accumulation of a hydroxyl fatty acid ceremides (Matsuda, Kido et al. 2004) S apD is also a membrane disrupter. By binding membranes which contain anionic lipids such as BMP and PI, SapD facilitates small vesicle formation (Ciaffoni, Salvioli et al. 2001) SapD has been crystallized and the crystal structure solved by X Ray diffraction to a resolution of 2.10 Angstroms (Rossmann, Schultz Heienbrok et al. 2008) A ribbon diagram of the crystal structure is shown in Figure 1 12. The protein is a homodimer that exists as either a substrate free closed helix bundle or in an open ligand bound conformation with the lipid binding cavity found between the monomers (Rossmann, Schultz Heienbrok et al. 2008)
36 GM2 activator protein The GM2 Activator Protein ( GM2AP ) is an 18 kDa accessory protein that plays a fundamental and essential role in the specific catabolism of the ganglioside GM2 to GM3, as shown in Figure 1 13 (Conzelmann and Sandhoff 1979; Furst and Sandhoff 1992) The high affinity of GM2AP for GM2 is base d on recognition of the oligosaccharide moiety and the ceramide tail GM2AP is thought to bind GM2 in intralysosomal vesicles and present the oligosaccharide head group for hydrolytic cleavage. Hexosaminidase A (Hex A) specifically cleaves the GalNAc sugar group from the ceramide, producing GM3. Because GM2AP does not actually catalyze a chemical reaction, it is considered an accessory protein, not an enzyme, in the catabolism of GM2 Mutatio ns in GM2AP lead to an accumulation of GM2 in the lysosomes, causing the lysosomal storage diseases Tay Sachs or the AB variant of (Conzelmann and Sandhoff 1979) GM2AP has been crystallized and the crystal structure solved by X Ray diffraction to a resolution of 2.00 Angstroms (Wright, Li et al. 2000) A ribbon diagram of the crystal structure is shown in Figure 1 14, and the po sitions of the four disulfide bridges are shown in red. The protein functions as a monomer. The space group is described as P 2 1 2 1 2 1 an d the unit cell contains four separate monomers (Wright, Li et al. 2000) The structure consists of an eight stranded antiparallel cup with an accessible hy drophobic cavity large enough for binding large lipid acyl chains (Wright, Li et al. 2000) Two possible mechanisms for GM2AP function have been presented (Sandhoff and Kolter 1996) recognized and lifted from the bilayer thereby presenting it to the water soluble enz yme for cleavage. However, it is also possible that GM2AP binds the GM2 ligand and forms
37 a water soluble complex whereby the enzymatic reaction would take place free in solution. A cartoon model of the latter hypothesis is presented in the cartoon repres entation in Figure 1 15. Note that the red circles denote a conformational change in the mobile loop region from an open conformation to a closed conformation. Hexosaminidase A Hexosaminidase A (HexA) is found in the lysosomes and is enzyme responsible for the degradation of the GM2 ganglioside. HexA has been crystallized and the crystal structure solved by X Ray diffraction to a resolution of 2.80 Angstroms (Lemieux, Mark et al. 2006) A ribbon diagram of the crystal structure is shown in Figure 1 16 The protein functions as a heterodimer. The space group is described as C2 and the unit cell contains four sep arate monomers (Lemieux, Mark et al. 2006) Lipid Storage Diseases and Diseases Associated with Abnormal Sphingolipid Metabolism Lipid storage diseases, also referred to as gangliosidoses, are brought on by mutations i n genes coding for enzymes involved in ganglioside metabolism. The enzyme deficiency causes accumulation of the corresponding sphingolipid in the cell, disrupting cellular function. A list of representative diseases associated with abnormal sphingolipid metabolism is given in Table 1 2 Tay Sachs disease Tay Sachs disease (TSD), also referred to as the B variant (or type 1) of GM2 gangliosidosis, or Hexosaminidase A deficiency, is a very rare, inherited, progressive neuronal disorder. The disease can o nset in infants, adolescence, or adulthood, but is far more common and severe during infancy. Symptoms vary widely, but include slow development, loss of motor skills, vision and hearing loss, mental disability, and
38 paralysis. A cherry red spot in the ey e is characteristic of Tay Sachs disease. Children with Tay Sachs disease usually live only into early childhood. Mutations in the HexA gene, which encodes for an enzyme hexosaminidase A, are responsible for the disease. The condition is inherited in an autosomal recessive pattern (both copies of the gene harbor mutations). These genetic mutations are found most frequently in people of Ashkenazi Jewish heritag hexosaminidase A is responsible for breaking down the GM2 ganglioside. If the function of this lysosomal enzyme is hindered, GM2 will accumulate to toxic levels in the neurons of the brain and spinal cord, leading to neuronal destruction. San dhoff d isease Sandhoff disease, also called Hexosaminidase A and B deficiency or Jatzkewitz Pilz syndrome, is a rare genetic lipid storage disease. The autosomal recessive disorder is linked to a buildup of GM2 ganglioside, which is toxic at high levels, which leads to progressive damage to the central nervous system via destruction of nerve cells Sandhoff disease can onset in infants, juveniles, or adults, but is most common during infancy. Symptoms of Sandhoff disease are nearly identical to those of Tay Sachs disease, and include mental disability, muscle deterioration, blindness, deafness, seizures, and cherry red spots in the retina. Currently, Sandhoff disease does not have any treatment or cure. Introduction to HIV 1 Protease Introduction to H uman Immunodeficiency Virus H uman immunodeficiency virus (H IV ), a member of the retrovirus family, is the cause of Acquired Immunodeficiency Syndrome (AIDS), a disease characterized by a suppression of the immune system (Weiss 1993) There are two strains of HIV called
39 HIV 1 and HIV 2, with HIV 1 being far more common (Gilbert 2003) A comparison is these HIV strains in given in Table 1 3 (Reeves and Dorns 2002; Gilbert 2003) Furthermore, HIV 1 is categorized into different groups (or different viral lineages) subtypes (taxonomic groups) and circulating recombinant forms (CRFs) (recombin ant viral forms) (Kantor, Shafer et al. 2005) HIV infection is classified into four stages, the fourth and final called AIDS, which is characterized by CD4 cell count, as well as presence of opportunistic inf ections (OIs ), which are typically normal infections that can become life threatening for the patient with the compromised immune system (Buchbinder, Katz et al. 1994; Lawn 2004) T here are many drugs currently used to slow the progression of the disease, but there is currently no cure (Schneider, Gange et al. 2005; Douek, Roederer et al. 2009; Noble 2009) The life cycle of HIV 1, depicted in Figure 1 17, is dependent upon a human host cell harboring the CD4 cell surface receptor. After a targeted fusion event, the virion are uncoated and the contents unlo aded into the host cell. The viral RNA is reverse transcribed by viral enzyme, and the resultant DNA is imported into the nucleus where it is integrated into the host cell chromosome, transcribed, and translated (Smith and Daniel 2006) These proteins then associate on the outer membrane of the host cell, triggering a budding event that releases immature viral particles. Maturation takes place when HIV 1 protease cleaves the gag and gag pol polyproteins, depicted in Figure 1 18 Genetic Variability HIV 1 has a high level of genetic variability, when generally compared to many other types of viruses. This fact is based upon several governing factors, including its fast replication cycle and its high rate of mutation. It has been determined that 10 9 to
40 10 10 virions can be ge nerated daily, with approximately 3 x 10 mutations per nucleotide base per cycle of replication (Robertson, Hahn et al. 1995; Rambaut, Posada et al. 2004) Because of the wide genetic variability that occurs as a result, HIV is classified into groups. Within the major strain HIV 1, there are three main groups, namely M, N, and O. These classifications are based upon differences in the env region of the genome (Thomson, Perez Alvarez et al. 2002) Group M is by far the most common, and is generally further divided into subtypes. These classifications are based upon differences throughout the entire genome (Carr, Foley et al. 1998) The most common HIV 1 are subtype C, comprising ap proximately 47% of infection worldwide, Subtype A/CRF02_A/G, comprising approximately 26.7%, Subtype B, comprising approximately 12%, Subtype D, comprising approximately 5%, CRF01_A/E, comprising approximately 3%, and 5% comprising all other subtypes and C RFs (Osmanov, Pattou et al. 2002) The majority of HIV research is performed using Subtype B, which is most commonly found in the United States and Europe (Perrin, Kaiser et al. 2003) HIV 1 Protease Human immunodeficiency virus type 1 (HIV 1) protease (HIV 1PR) is a homodimer of two 99 amino acid monomers. The enzyme, an aspartic protease, is required for viral maturation of HIV 1, thus its inhibition is a major target for A IDS antiviral therapy (Ashorn, McQuade et al. 1990) A ccess of the gag and gag pol polyprotein substrates hairpins commonly called the flaps The protease flaps have been shown to undergo a large conformational change during substrate binding and catalysis (Todd, Semo et al. 1998; Ishima, Freedberg et al. 1999; Todd and Freire 1999; Todd, Luque et al. 2000; Velazquez Campoy, Todd et al. 2000; Ishima, Ghirlando et al. 2001; Freedberg, Ishima et al. 2002; Ishima, Torchia et al.
41 2003; Louis, Ishima et a l. 2003) Over 300 structures of HIV 1PR have been reported in the Protein Data Bank (PDB), the first of which was reported in 1989 (Miller, Schneid er et al. 1989) The structure of HIV 1 (top and side views), in both ribbon and space filling form, is shown in Figure 1 19. T his model also clearly demonstrates that the active site is highly inaccessibility to large peptide substrates when the flaps are in the closed conformation. Molecular dynamics simulations have provided some of the most direct evidence of the necessary conformational changes within the flap region to facilitate ligand binding and catalysis. Additionally, m olecular dynamics (MD) simulations performed by Simmerling et al. have pointed to three distinct flap conformations in HI V 1 protease, as illustrated in Figure 1 20 (Hornak, Okur et al. 2006; Hornak, Okur et al. 2006) Conformational Sampling and the Conformational Ensemble of the HIV 1 Protease Flaps The flaps of HIV 1 protease have been shown to undergo a large conformational change upon binding and catalysis, and the sum of the conformations sampled is described as the conformation al ensemble. The preferred conformations of the protease flaps change upon addition of inhibitor or substrate (Galiano, Blackburn et al. 2009) incorporation of naturally occuring or drug pressure selected polymorphisms into the protease (Kear, Blackburn et al. 2009) or t he conditions of the protein solution are changed [Blackburn, unpublished]. Inhibitors of HIV 1 Protease Several points throughout the HIV 1 life cycle are exploited for inhibition, including viral fusion, reverse transcription, integration, and maturati on steps, as shown in Figure 1 21 Ther e are currently 31 antiretroviral drugs (ARVs) approved by the U.S. Food and
42 Drug Administration (FDA) for Highly Active Antiretroviral Therapy (HAART) treatment of HIV infection, a nd these are shown in Table 1 4 T he ARVs are classified into fusion inhibitors (FIs), nucleoside and non nucleoside r everse transcriptase inhibitors (NRTIs and NNRTIs), integrase inhibitors, and protease inhibitors. HAART treatment usually consists of combined treatment with several type s of inhibitors. The research discussed within this dissertation deals specifically with the protease inhibitors. There are nine FDA approved HIV 1 protease inhibitors currently being used in the treatment of HIV 1 infection, namely Ritonavir, Indinavir, Lopinavir, Darunavir, Saquinavir, Tipranavir, Atazanavir, Fosamprenavir ( the pro drug of Amprenavir) and Nelfinavir. Structure of the se protease inhibitors are given in Figure 1 22 Protease inhibitors, first developed in 1989, generally function to tar get the active site of the protease by mimicking the tetrahedryl intermediate of the substrate.
43 Table 1 1. Common monosaccharide abbreviations Name Symbol N acetylgalactosamine GalNAc N acetylglucosamine GlcNac N acetylneuraminic acid NeuAc or Neu5Ac Galactosamine GalN Galactose Gal Glucosamine GlcN Glucose Glc Glucuronic acid GlcA Table 1 2 Representative d iseases associated with abnormal sphingolipid metabolism Disorder Enzyme Deficiency Ac cumulating Molecule Tay Sachs disease HexA GM2 Tay Sachs AB variant GM2AP GM2 Sandhoff disease HexA and HexB GM2, globoside GM1 Gangliosidosis galactosidase 1 GM1 Gaucher disease glucocerebrosidase glucocerebroside Krabbe disease gala ctocerebrosidase galactocerebroside Fucosidosis fucosidase pentahexosylfucoglycolipid Table 1 3 A comparison of HIV 1 and HIV 2 Strain Virulence Infectivity Prevalence Inferred Origin HIV 1 High High Global Chi mpanzee HIV 2 Lower Low Western Africa Sooty Mangabey Table 1 4 FDA approved drugs for the treatment of HIV/AIDS PIs N RTIs NNRTIs EIs IIs Saquinavir Delavirdine Zidovudine Enfuvirtide Raltegravir Ritonavir Efavirenz Emtricitabine Celsentri Indinavir Nevirapine Lamivudine Nelfinavir Stavudine Amprenavir Abacavir Lopinavir Didanosine Atazanavir Tenofovir Tipranavir Fosamprenavir Dar unavir
44 Figure 1 1. Gl ycolipid structure, where Y = lipid Figure 1 2. General structure of a glyceroglycolipid Figure 1 3. General structure of a glycosylphosphotidylinositol
45 Figure 1 4. Structure of sphingosine Figure 1 5. Glycosphingolipid structure Figure 1 6. General structure of a ganglioside
46 Figure 1 7. Ganglioside GM2 structure
47 Figure 1 8. Glycosphingolipid degradation pathway. In green are the names of corresponding lipid storage diseases corresponding with each step in the degradation pathway. L ikewise, in blue are the names of necessary accessory proteins and in red are the names of enzymes involved in the catabolic reactions.
48 Figure 1 9. Crystal structure of Saposin A (PDB ID 2DOB). Image rendered in VMD Figure 1 10. Unit cell c rystal structure of Saposin B (PDB ID 1N69) Figure 1 11. Unit cell crystal structure of Saposin C (PDB ID 2QYP)
49 Figure 1 12. Unit cell c rystal structure of Saposin D (PDB ID 2RB3) Figure 1 13. Representation of ganglioside hydrolysis A) GM2 ganglioside lipid structure, B ) c artoon scheme for hydrol ytic cleavage of GM2 to GM3.
50 Figure 1 14. Cartoon structure of GM2AP showi ng the 4 disulfide bonds in red ( PDB 1G13 ) Figure 1 15. Cartoon representation of GM2AP binding lipid membranes and extracting its specific substrate GM2 ( PDB 1G13 )
51 Figure 1 hexosamin i dase A (PDB ID 2GJX)
52 Figure 1 17. HIV 1 life cycle.
53 Figure 1 18. Cartoon representation of the gag and gag pol polyproteins cleaved by HIV 1 protease during viral maturation. Figure 1 19 Structure of S ubtype B HIV 1 Protease (PDB ID 2BPX) A) Ribbon diagram and B) space filling model of top view of Subtype B HIV 1PR C) Ribbon diagram and D) space filling model of side view of Subtype B HIV 1PR.
54 Figure 1 20. Flap conformations of HIV 1 protease capt ured by molecular dynamics simulations ; A) closed, in presence of inhibitor, B) semi open, and C) wide open.
55 Figure 1 21. Points of inhibition within the HIV 1 viral life cycle.
56 Figure 1 22. Structures of the FDA approved protease inhibitors.
57 CHAPTER 2 INTRODUCTION TO BIOP HYSICAL METHODOLOGIE S Introduction Several different methodologies were utilized in this study, including mass spectrometry, fluorescence spectroscopy, circular dichroism spectroscopy (CD), site directed spin labeling, continu ous wave (CW) and pulsed electron paramagnetic resonance spectroscopy (EPR) General overviews to these methods are given in the following sections. Additionally, each research chapter will have a detailed materials and methods section for all experiment s contained within. Mass Spectrometry Mass spectrometry (MS) is a rich, analytical technique that can be used to determine the exact elemental make up or chemical structure of a molecule or protein (Sparkman 2000; Hoffman and Stroobant 2007) The fundamental principle behind MS is ionization of the molecule followed by identification of the mass to charge ratios. The first ap plication of MS to the study of peptides or proteins came in 1958 (Beckey 1969) but arguable the most significant MS achievement relative to the study of proteins came with the development of matrix assisted laser desorpt ion/ionization (MALDI) by Hillenkamp and Karas in 1985 and the work from several others that immediately followed (Dougherty 1981; Karas, Bachmann et al. 1985; Karas, Bachman et al. 1987; Tanaka, Waki et al. 1988) Two types of approaches are generally used when using proteins are enzymatically digested into peptide fragments then analyzed (Hoffman and Stroobant 2007)
58 A schematic diagram of a simple mass spectrometer is shown in Fig ure 2 1. All MS instruments are composed of three basic parts, the first of which is the ion source (Hoffman and Stroobant 2007) The i on source functions to convert the analyte molecules into ions. Techniques for ionization are variable, and include electron (EI) and chemical ionization (CI) for gases and vapors, electrospray (ESI) and MALDI for liquid and so lid samples, among many others. In CI, the sample is ionized via chemical reactions that occur during collisions between ions and molec ules in the ion source. In EI, electrons interact with molecules to produce ions (Hoffman and Stroobant 2007) After ionization, the i ons are transported via electric or magnetic field to the analyzer, which is the second integral part of a typical MS instrument. The mass analyzer sorts ions by mass to 1) 2) describe the dynamics of charged particles in applied fields under vacuum The Lorentz force is described as the force applied on a charge due to an applied field, where F is the force in newtons, E is the electric field in volts per meter, B is the ma gnetic field in tesla, q is the electric charge of the particle in coulombs, v is the instantaneous velocity in meters per second, and x is the vector (F), mass ( m ), and acceleration (a), and is only valid at ion velocity much lower than the speed of light. Using the expressions given in Equations 2 1 and 2 2, the differential equation shown in Equation 2 3 can be used to describe the motion for charged particles. If th describes the motion of the particle in space and time with respect to mass to charge ( m /Q). Most commonly, this ratio is presented in terms of the dimensionless m/z,
59 where z is the number of elementary charges on the ion (e), as described by Equation 2 4. There exist many common types of mass analyzers including the sector field mass analyzer, the time of flight (TOF), the quadrupole, and the linear quadrupole ion trap. Each t ype of mass analyzer has unique pros and cons, and the choice of analyzer is highly sample dependent (Hoffman and Stroobant 2007) The third and final necessary component of all MS instruments is the detector. The detector provides information regarding the mass to charge ratios and the quantity of each specific ion present (Sparkman 200 0) Standard detectors function in one of two ways; either it records the charge induced or the current produced when the ion passes by or hits a surface (Hoffman and Stroobant 2007) F = m a (2 1) F = Q (E + vxB) (2 2) ( m / Q )a = E + vxB (2 3) z = Q / e (2 4) Fluorescence Spectroscopy Fluorescence spectroscopy, also called fluorometry or spectr ofluorometry is a form of spectroscopy that analyzes fluorescence from a sample. For a thorough review on fluorescence spectroscopy, the reader is referred to the Principles of Fluorescence Spectroscopy textbook, written by J. R. Lakowicz (Lakowicz 2006) This technique utilizes a beam of ultraviolet light to excite electrons via photon absorption within a given system, causing an emission of light. After a species is excited from its ground state to a higher energy state, the species then emits a photon as it returns to the ground electronic state (Lakowicz 2006) Analysis of the frequency and intensity of emission leads to information regarding the structure of the different vibrational levels. A typical
60 experiment results in an emission spectrum, where the various wavelengths of fluorescent light emitted by a sample are moni tored while the excitation wavelength is kept constant (Lakowicz 2006) A typical fluorescence instrument functions by passing an excitation light source through a m onochromator and into the sample cell, thereby exciting molecules within the sample. Typical light sources include lamps, lasers, and photodiodes. The excited molecules fluoresce and that light passes through a second monochromator to the detector. The function of the monochromator is to transmit light at a specific wavelength and intensity. The detector is generally positioned 90 from the incident light to reduce effects from scattering and improve signal to noise (Lakowicz 2006) Circular Dichroism Spectroscopy Circular dichroism (CD) spectroscopy is an optical biophysical technique used to measure differences in absorption of left and right handed circular polarized light by optically active materials (Atkins and Paula 2005) This technique allows for the determination of the average global secondary structure of a peptide or protein. When being used for this application, CD is performed in the far UV regions of 180 260 nm. A CD signal arises when a peptide bond is in a structured environment, and left and right handed circularly polarized light is absorbed to different extents by chiral amino acids (At kins and Paula 2005) The raw data is usually reported in ellipticity with units cm 2 dmol 1 residue 1 ), as given in Equation 2 5, where M r is the protein molecular weight, c is the protein concentration (in mg/mL), l is the cuvette path length and N A is the number of amino acids in the protein (Atkins and Paula 2005)
61 Advantages of CD spectroscopy include that the technique requires very small amounts of protein, is non destructive is highly sensitive toward small changes in secondary structure, and require very l ittle data processing. In addition, even small changes in overall secondary structure can be monitored very accurately. A glaring disadvantage, however, is that the technique is not site specific. In other words, only the overall structure can be determ ined with no indication of what regions contain what elements of secondary structure. In biophysical studies of proteins, CD is often used as positive evidence of proper protein folding, but is not a definitive method. A schematic diagram of a typical c ircular dichroism instrument is shown in Figure 2 3 (Atkins and Paula 2005) (2 5 ) Electron Paramagnetic Resonance Spectroscopy Site D irected Spin L abeling Site directed spin labeling (SDSL) is a technique utilized to incorporate a paramagnetic tag or label, into a protein or other biological molecule at a specific site in order to fa cilitate electron paramagnetic resonance (EPR) studies (Griffith and McConnell 1965; Stone, Buckman et al. 1965) SDSL, first introduced in the mid 1960s by Griffith and McConnell, et al., is an excellent technique for studying the conformations and conformational changes of proteins. Most often, SDSL is carried out by engineering a cysteine residue into the protein of interest via manipulation of the DNA through a technique called site directed mutagenesis The engineered cysteine residue can then be site specifically modified with a spin label.
62 There are many spin labels commonly used in EPR studies today, some of which include (1 Oxyl 2,2,5,5 Tetramethyl Pyrroline 3 Methyl) Methane thiosulfonate (MTSL), 4 Maleimido TEMPO (MSL), 3 (2 Iodoacetamido) PROXYL ( IAP), and 4 (2 Iodoacetamido) TEMPO (IASL). Each of these spin labels are nitroxide based Each one contains a free electron ass ociated with either a five or six membered ring with four methyl groups to help defer collision induced reactions with the radical. Used in this study were three spin labels, MTSL, MSL, and the non native amino acid p acetyl L phenylalanine The followin g sections provide details on each of those labels. Each of the spin labels contain a flexible linker of between four and six bonds th at connect the nitroxide to the CYS residue and a head group of varying bulkiness. Dr. Luis Galiano performed an experi ment to examine the differences in the CW EPR line shapes for HIV 1PR labeled at the K55C position with each of four different spin labels. All spectra were collected at the same temperature with the same concentration of protein; thus t he differences in the line shapes are due to differences in inherent spin label mobility. That data was reported in his 2008 dissertation and is shown in Figure 2 4 (Galiano 2008) (1 Oxyl 2,2,5,5 Tetramethyl Pyrroline 3 Methyl) Methane t hiosulfonate (MTSL) MTSL is the most commonly used spin label (Hubbell, McHaourab et al. 1996; McHaourab 1996; Hubbell, Gross et al. 1998; McHaourab, Kalai et al. 1999; Hubbell, Cafiso et al. 2000; Hubbell 2000) MTSL is incorporated into a protein via thiol based chemistry; a general reaction scheme is shown in Figure 2 5, where the modified cysteine residue is referred to as R1 (Berliner, Grunwald et al. 1982) The aqueous exposed spin labeled side chain can be easily accommodated by most sites within proteins and has been sho wn to be no more perturbing than most other single amino acid substitutions (Hubbell, McHaourab et al. 1996)
63 F igure 2 5 also serves as a g raphical representation of the 4/ 5 model for the intrinsic mobility of the MTSL label, which was developed in the laboratory of Wayne Hubbell at UCLA and is based upon the spin label positioned in a solvent exposed site helix in T4 Lysozyme (Langen, Oh et al. 2000) MTSL has a 5 bond flexible linker but the 4/ 5 model predicts that the carbon interacts sulfur thus restricting the motion of the first three bonds in the linker limiting the rotable degrees of freedom to the 4 and 5 bonds. 4 Maleimido TEMPO (MSL) In comp arison to MTSL, MSL, whose structure is shown in Figure 2 6, has a much bulkie r head group. This slows the overall intrinsic mobility of the label, and lends to the more restricted anisotropic EPR line shape, as shown in Figure 2 4. Another key differenc e between MTSL and MSL is the mechanism of attachment. Again, MTSL has a thiol reactive moiety that leads to an S S bond between the nitroxide and the CYS residue. MSL, on the other hand, bonds to the CYS residue via a C S attachment. This fact makes th e MSL label more stable, with respect to permanence of the attachment. Under certain experimental conditions, this realization becomes very important. Non native amino acid p acetyl L phenylalanine spin label A non native amino acid p acetyl L phenyl alanine and spin label, were generously received from Mark Fleissner and Wayne Hubbell at UCLA, who demonstrated its utility using T4 lysozyme (Fleissner, Brustad et al. 2009) This label differs from traditional nitroxide spin labels based upon i ts unique method of incorporation that does not rely upon the addition of a non native cysteine residue. This is particular useful for proteins that have native cysteines that are either functionally active or are involved in structurally important disul fide bonds. Here, the genetically encoded unnatural amino
64 acid p acetyl L phenylalanine (p AcPhe) is reacted with a hydroxylamine based reagent, resulting in a nitroxide side chain called K1. The structure of K1 and a basic scheme of spin label attachme nt are shown in Figure 2 7. The K1 side chain has a higher degree of freedom than does the R1 side chain, and this fact is indicated by an EPR line shape indicative of a higher degree of mobility and a faster correlation time (Fleissner, Brustad et al. 2009) T he unnatural amino acid which can be site specifically incorporated in using the nonsense amber codon into Escherichia coli (Wang, Zhang et al. 2003) Saccharomyces cerevisiae (Chin, Cropp et al. 2003) and mammalian cells (Liu, Broc k et al. 2007) p AcPhe contains a keto functional group a group that is not present in any of the common amino acids that is reactive in aqueous solution with hydroxylamines (Brustad, Lemke et al. 2008) Continuous Wave Electron Paramagnetic Resonance Spectroscopy (CW EPR) Introduction and h istory Electron paramagnetic resonance (EPR) spectroscopy is the st udy of absorption of microwave radiation by a paramagnetic species in the presence of an external magnetic field. Today, EPR spectroscopy is used in a wide variety of applications for the detection of free radicals and paramagnetic species (Eaton, Eaton et al. 1998) In the study of pro tein systems, EPR is often used in conjunction with site directed spin labeling, which will be discussed in details in the upcoming sections. EPR spectroscopy was discovered and first described in 1944 by Soviet physicist Yevgeny Konstantinovich Zavois ky of Kazan State University in Kazan, Russia, when EPR signals were detected in several salts, including hydrous copper chloride, copper sulfate, and manganese sulfate. He was als o said to have observed nuclear magnetic
65 resonance (NMR) in 1941but dismissed early results on the basis of lack of reproducibility (Eaton, Eaton et al. 1998) Bloch and Purcell first published on NMR in 1946 and subsequently shared the Nobel Prize in p hysics in 1952. CW EPR sample requirements For all work reported within, the CW EPR methodology was performed with loop gap resonator at X band frequencies with a center field of approximately 3480 Gauss. These experiments required nanomole quantities of spin labeled protein. Typical sample sizes were 10 microliters of 50 250 M protein. CW EPR theory The unpaired electron found most often in a free radical or paramagnetic species, experiences a net dipole This dipole is due to a magnetic moment that arises from spin angular momentum of quantum number m = In the simplest case of the free energy of the two spin states m s = is equal. W hen a magnetic fie ld is applied, the electron can align itself either parallel or antiparallel to the field This phenomenon is referred to as the Zeeman Effect. The Zeeman Equation (Eq. 2 6) describes the difference between the energy levels E and E e spectroscopic g factor e is the Bohr magneton, and B is the strength of the applied magnetic field (Weil, Bolton et al. 1972; Poole 1983) The Bohr magneton (9.274x10 24 JT 1 ), is a proportionality constant defined in Equation 2 7 where e is the electric charge, (1.054 10 34 Js), m e is the mass of the electron (9.109 10 31 kg) Energy is absorbed, i.e., resonance occurs, when the applied energy is equal to the difference in
66 energy levels E and E ; this is achieved by maintaining a constant frequency and sweeping the magnetic field. The energy diagram describing this graphically i s given as Figure 2 8 while Figure 2 13B shows a typical derivative absorption spectrum for a free electron in solution (Weil, Bolton et al. 1972; Poole 1983) When EPR is performed in conjunction with SDSL, the free electron (m s = ) associated with the nitroxide interacts with th e nuclear spin from nitrogen (m I = 1) via hyperfine coupling As such, both levels E and E split into three Hyperfine energy levels, afford ing the system three allowed energy transitions. The energy diagram for this type of system, along with the corr esponding derivative absorption spectrum is given in Figure 2 9 (Weil, Bolton et al. 1972; Poole 1983) e B (2 6) e = e m e (2 7 ) Nitroxide spectral line shapes The EPR spectral line shape is sensitive to local secondary structural elements, local dynamics, and conformational changes all of which have an impact on correlation time ( Figure 2 1 0), and because of this, c onformational changes can be easily monitored with SDSL EPR (Altenbach, Flitsch et al. 1989; Farahbakhsh, Altenbach et al. 1992; Altenb ach, Yang et al. 1996; McHaourab, Lietzow et al. 1996; Hubbell, Gross et al. 1998; Altenbach, Cai et al. 1999; Altenbach, Klein Seetharaman et al. 1999; Barnes, Liang et al. 1999; McHaourab, Kalai et al. 1999; Hubbell, Cafiso et al. 2000; Mollaaghababa, St einhoff et al. 2000; Columbus 2001; Columbus, Kalai et al. 2001; Columbus, Kalai et al. 2001; Barnakov, Altenbach et al. 2002; Columbus and Hubbell 2002; Columbus and Hubbell 2002; Columbus and Hubbell 2004; Qin, Feigon et al. 2005; Fanucci and Cafiso 2006 ; Fanucci and Cafiso 2006; Guo, Cascio et al. 2008)
67 The c orrelation time of the paramagnet is described by three modes of motion, namely rotational R ), internal correlation time ( i ), and local dynamics and backbone fluctuations ( B ). Rotational correlation time is the rate at which the protein is tumbling in solution; that rate is clearly affected by the size and tumbling volume of the paramagnetic macromolecule, as well as the viscosity, temperature and presence of solutes in solution. i ) describes the mobility of the spin label, and is affected by the size and structure of the nitroxide spin labe l, intrinsic spin label mobility and the torsional oscillations within the spin label The local dynamics and backbone fluctuation s B ) are characterized by structure (secondary and tertiary), in addition to conformational changes involving the paramagn etic center. The CW EPR spectral line shape is highly sensitive to changes in correlation time and anisotropy. For sites that undergo fast isotropic motion and a low rotational correlation time, the resultant spectra have sharp, narrow peaks (top). As motion becomes more restricted and rotational correlation time increases, resonan ce peaks decrease in sharp ness and line broadening occurs This phenomenon is shown in Figure 2 10, in the inter mediate, slow (middle), and rigid spectra (bottom). CW EPR li ne shapes are ana lyzed in order to extract parameters that describe the motion of the spin label, as described in the previous sections. Perhaps the simplest H pp or the peak to peak width of the central resonanc e line, most commonly reported in units of Gauss. A s shown in Figure 2 10, the line shape resulting from fast, isotropic motion shows sharp, narrow resonance lines that broaden as motion becomes more restricted and anisotropic T he pp there fore, increases with anisotropy and motion restriction. Scaled
68 mobility is essentially a normalized pp and is reported values from data collected on different instruments Scaled mobility, Equation 2 4, is calculated by normali zing experimental values of ( exp ) using spectral line widths of the most mobile ( m ) and immobile ( i ) spectrum (Hubbell, Cafiso et al. 2000) (2 8 ) Pulsed Electron Paramagnetic Resonance Spectroscopy Introduction and history The basic concepts of pulsed EPR are very similar to those of pulsed nuclear magnetic resonance (NMR), wit h only minor perturbations. Pulsed EPR was first when a modulation of the echo amplitude in a Ce3+ doped CaWO4 crystal was observed for the first time In the same year, the Lockheed group in C alifornia observed echo modulations on a single crystal of Ce3+ doped lanthanum magnesium nitrate (Schweiger and Jeschke 2001) It was not until the 1982 that X band pulsed EPR spectrometers became commercially available. Pulsed EPR has since gained popularity as a valuable tool in a wide variety of applications in the fields of chemistr y, physics and biology The specific pulsed technique utilized in this research is called pulsed electron double resonance (PELDOR), or double electron electron resonance (DEER). DEER experiments use two different microwave frequencies to measure the s trength of the coupling between two electron spins; in our case, to measure the distance between two nitroxide spin labels. The following sections will provide details regarding sample
69 requirements, theory of DEER, pulse sequences, data analysis and inter pretation of results, and error analysis. Sample r equirements As with CW EPR, the pulsed EPR methodology utilized here on the ELEXSYS E580 with MD4 or MD5 dielectric resonator requires nanomole quantities of spin labeled protein. Typical sample sizes are a total of 100 L including 70 L of 150 300 M protein and 30 L of d 8 glycerol. Certain samples also contain 3 L protease inhibitor. DEER theory DEER experiments generally utilize two different microwave frequencies in order to extract information regarding the distance between two paramagnets by measuring the strength of the dipolar coupling. In all cases described in this work, that distance refers to the spacing between two nitroxide spin labels at specific sites within either HIV 1 protease or GM2 activator protein. In theory, the DEER technique is capable of extracting distances in the range of 1.5 8 nm, however it is seldom used to reliably extract distances greater than 5 nm. For the purposes of describing the theory behind these experim ents, the spin label at position 1 will be referred to as spin subset A, and the spin label at position 2 will be referred to as spin subset B. The DEER technique can be described using a simplified Hamiltonia n ( Equation 2 6 ). The simplified Hamiltonian is c omprised of three sets of terms, including the Zeeman terms for the spin subsets A and B A S z A and B S z B respectively, and the dipolar coupling term ee Sz A z B The coupling term is proportional to the inverse cube (1/r 3 ) of the distance between the spins where r is said distance.
70 DEER experiments can be carried out using several different pulse sequences. The simplest and first to be developed were the 2 + 1 and 3 pulse DEER, shown in Figures 2 11 and 2 12, respectively. The 2 + 1 sequence is applied at a single frequency. The 3 pulse DEER consists of a two pulse echo forming sequence ( /2 ) separated by a constant delay at a frequency 1 in addition to a pump pulse ( ) at a frequency 2 with a variable time t with respect to the first observer pulse. Both the 2 + 1 and 3 pulse DEER however, experience a dead time of approximately 64 ns during which the system cannot record data. This dead time severely limits accurate experimental measurements (Pannier, Veit et al. 2000) All pulsed EPR work reported in this dissertation made use of the common 4 pulse DEER sequence at X band frequencies. The standard 4 pulse DEER sequence, shown in Figure 2 13 and developed by Pannier et al. negates the dead time problem by introducing a refocusing 180 pulse in the frequency (1) (Pannier, Veit et al. 2000) T he Hahn Echo sequence is used to produce an echo for the spins in resonance with the microwave frequency (1) pulse that flips the magnetization 1 pulse is applie d to invert the magnetization, and after a second 1 an echo is produced along the Y axis. A fter a time 2 pulse at th e microwave frequency 2 flips the spin subset B, which in turn affects the spin subset A. Because the subsets A and B are dipolar coupled, the resultant spin echo oscillates with the dipolar coupling frequency ee (Pannier, Veit et al. 2000) For all work described here, the distance of interest lies between two i dentical spin labels on a single protein macromolecule. Figure 2 14 shows the positioning of both the
71 pump pulse and observer sequence in the field swept echo experiment. Because the paramagnets are identical spin labels, the g values are very similar an d the spectral features of the labels overlap; because of this, selective excitation of a single label is impossible. Due to the larger number of excited spins at the central transition, the 2 is positioned on the central resonance line, prov iding an optimal S/N ratio. The observer pulse is applied to the frequency corresponding to the low field resonance, which, for nitroxide spin labels, is approximately 26G (72 MHz) below the central resonance at X band Given the concentration of protein used for DEER experiments (150 300 M) it is certain that spins will come in inter molecular contact with spins from proteins resulting in a random distribution of spin spin interactions The sum of the se inter molecular interactions giv es rise to a background signal in the form of an exponential decay. Equation 2 10 describes the DEER signal, where V(t) represents the raw experimental dipolar modulation, B(t) represents the background contribution from inter molecular spin spin interactions, and F(t) is a function describing the background subtracted dipolar modulated echo data. After background subtraction the function V(t) can be written as Equation 2 11, where is the kerne l function given in Equation 2 12 In order to ensure that the background subtraction is done to the most accurate extent, a form of self consistent analysis was performed on all data presented in this dissertation. A program called Dee rSim is used to generate a theoretical dipolar evolution that is completely free of any background contributions. This theoretical echo curve is overlain with the background subtracted echo curve. If the two curves overlay
72 exactly, then the background su btraction is accurate and the data is free of contributions from inter protein spin spin distances. Figure 2 15 shows typical dipolar modulated echo data from a system containing two separate nitroxide spin labels, including the raw oscillation, the linea r function B(t) (in red) which describes the contribution from background, and the background subtracted echo data with fit (grey). All DEER data presented in this dissertation was analyzed using the DeerAnalysis software. The DeerAnalysis software uses both shell factorization and Tikhonov regularization in order to extract distance information from the dipolar modulated echo curves. For a comprehensive discussion on DEER data collection and analysis, the reader is referre d to a number of sources (Jeschke 2002; Jeschke 2002; Jeschke 2004; Chiang 2005; Jeschke, Chechik et al. 2006; Jeschke and Polyhach 2007) including the www.epr.ethz.ch Though sever al methods exist by which to analyze DEER data, including the direct Fourier transform and Monte Carlo analyses, this chapter will focus specifically on Tikhonov regularization (TKR) (Tikhonov 1943; Hansen 1998; Chiang, Borbat et al. 2005) TKR analysis is based upon function in Equation 2 13, which is used in order to find the best answer to an ill posed problem by balancing the quality of the fit with the smoothness of the solution by varying the regularization parameter where P is the probability of the spin spin distance, K and L are operators, and S is the experimental data vector. Each individual value is g iven by Equation 2 1 4. ( ) and ( ) are given in Equa tions 2 1 5 and 2 1 6 Following TKR analysis at many values of log ( ) is plotted against log ( ) The aforementioned plot is referred to as the L curve Good quality DEER data will result in an L curve with a sharp point, often referred to an
73 curve is subsequently used to determine the optimal regularization parameter by selecting the value of found at the elbow point within the Figure 2 16 shows a standard L curve, with regions of the L curve designate d at under smoothed, optimal, and over smoothed. Again, selection of the optimal regularization parameter allows for extraction of the most accurate distance profile. Once the correct regularization parameter has been chosen, the optimal distance profil e can be selected. The distance profile can be examined for most probable distance, average distance, and the full width and half maximum intensity (FWHM). T he most probable distance is the distance at which the intensity is the greatest point in the dis tance profile. The FWHM is the breadth, in Angstroms, at half of the maximum intensity and is indicative of flexibility within the system. Figure 2 17 shows two examples of dipolar modulated echo curves with the corresponding distance profiles. The echo curve in black translates to a distance profile with a greater FWHM, most probable distance, and average distance than does the echo curve in grey. This would average, cl oser together and held in a more rigid position than the spin labels in the A more quantitatively descriptive method of analysis can be performed, and this process will be termed the Gaussian reconstruction process. In the Gaussian recons truction process, individual sub populations are identified and used to regenerate the distance distribution profiles in an effort to define the conformational ensemble. A conformational ensemble is the total of all the individual and distinct structural populations. To define a conformational ensemble, one must report the most probable
74 distance, the FWHM, and the relative percent of the total population for each individual Gaussian shaped population. Once the distance profile has been adequately regener ated, a theoretical dipolar modulation is generated and overlain with the background subtracted experimental dipolar modulation. If the two echo curves overlay exactly, the populations used to regenerate the distance profile are used to define the conform ational ensemble. One final and additional step is performed as a method of error analysis. Minor populations are suppressed one at a time, and in linear combinations. These suppressed ensembles are then used to generate a new theoretical echo to overl ay with the background subtracted experimental echo. If the echo curves still overlay perfectly, the population is said to have been successfully suppressed, and that population is discounted from a contributing member of the conformational ensemble. If, on the other hand, the suppression of one or more populations changes the echo such tha t it no longer overlays with the background subtracted dipolar modulation, that population is said to contribute to the total conformational ensemble (2 9) V(t) = F(t) B(t) (2 10 ) (2 11 ) (2 12 ) (2 13 )
75 (2 14 ) (2 15 ) (2 16 ) E580 spectrometer DEER set up This section serves as a detailed operating manual and experimental setup protocol for DEER experiments run on the Bruker EleXsys E580 Spectrometer in the laboratory of Alex Angerhofer. For a more comprehensive manual, readers are referred to the ELDOR manual. General concepts discussed here include sample preparation, equipment and cryogenics setup, cavity tuning, preliminary experiments and DEER experimental setup, closing out an experiment, and shutting down the spectrometer and pumps Two days pr i or to start of DEER experiments, r equest the liquid Helium dewar from the department of physics cryogenic website ( http://www.phys.ufl.edu/ ~cryogenics/ ). Follow the Order Helium link to reach the online liquid helium request form, shown below in Figure 2 x. Fill in each box and press Send in Request. The type of need is Cooldown. Under special needs, select Non magnetic. In the comments section, you should mention how many days you will need to keep the dewar, and also clarify that we are room 303F. Check the level of nitrogen in the dewar in CLB 416. If necessary, order more nitrogen by calling (352) 372 8417 and requesting a low pressure nitrogen dewar, using account number A2504. Be sure to specify if any empty dewars will be ready for pick up.
76 On the day of DEER experiments, p repare purified protein sample for DEER run by buffer exchanging the sample with deuterated buffer using a 5 mL Desalting column. Deuterium ( 2 H) extends time of spin dephasing in DEER experiment. Wash the desalting column with nanopure water (nH 2 O), 1M NaCl, nH 2 O, 0.5 M NaOH, nH 2 O, and NaOAc (pH 5) using 3 to 4 column volumes of each (15 to 20 mL). Equilibrate column with 10 mL deuterated NaOAc (pH 5). Inje ct 1 mL of sample (ideally OD280 = 2.5 to 2.7). Collect 1 mL of flow through in a 1.5 mL tube. Inject 0.5 mL of deuterated NaOAc, collecting the last 0.5 mL of flow through. Continue injecting 1.5 mL of deuterated NaOAc while collecting the buffer exchan ged sample in another tube. Inject 0.5 mL of nH 2 O while collecting the last 0.5 mL of buffer exchanged sample in a third tube. Concentrate the 2 mL sample to approximate OD reading of 3.0. Thoroughly wash and vacuum dry the EPR tube to avoid cross conta deuterat e d glycerol (glassing agent), and any inhibitors or substrate necessary for the experiment. Insert the mixed sample into a clean EPR tube using a skinny piece of tubing connected to a smal l syringe. In order to set up the equipment on the day of the DEER experiment, c heck to be sure correct that the resonator and cryostat are correctly installed. See Alex Angerhofer and refer to ENDOR manual for assistance. Set up temperature reference by f ill ing the reference dewa r with liquid nitrogen and placing it in front of the cavity. Position the blue reference line into the dewar through the cork stopper. Turn on the water supply. Do so by opening four valves, two for water supply and two for wat er return. The values are located overhead on opposite sides of the room, and are currently connected to green water hoses. Pump down the cavity Close off the diffusion pump by closing both the
77 screw valve and the lever value. Open the diffusion pump by pass. Close off the cavity by closing the black valve on the front of the cryostat. Be sure the vacuum line is connected to the cavity. Turn on the rough pump (switch embedded in the black cord) and turn on the vacuum gauge (set to sensor 1). When t he vacuum gauge reads 10 3 Torr, open the black valve in front of the cavity. Let the vacuum return to 10 3 Torr and let it sit for ~ 10 mins. Next open the valves to the diffusion pump, and let the vacuum return to 10 3 Torr and let it sit for ~ 10 mins If diffusion pump is cool, turn it on by plugging it in and flipping the switch on the front of the pump. If diffusion pump is hot/warm, leave the pump off and make sure water supply is all the way open. NEVER TURN ON THE DIFFUSION PUMP IF IT IS HOT T O AVOID OVERHEATING. Switch vacuum gauge to sensor 2. Let vacuum sit at 10 6 10 9 Torr for ~20 mins. Set up cryostat cool down. Retrieve the assigned He dewar from the helium storage room on the third floor of Leigh Hall by disconnecting the computer and the recovery line before moving the dewar. Be sure to close the recovery valve while dewar is disconnected from the wall. Once the dewar is in CLB 303, connect its computer power cord and connect its He recovery line to the central He recovery (trans parent He recovery line from cryostat is always connected to central He recovery). Make sure the recovery valve is open (must remain open throughout the entire DEER experiment). Set up the dewar computer for transfer. Press the far right button on the co mputer. Select transfer user=303F begin transfer. Insert transfer line into He dewar. Be sure the valve on the transfer line is open. Set the transfer line into position at the top of the dewar. Open both the screw valve and the lever valve and slo wly insert the transfer line. It is important that this step is done slowly as the warm transfer line heats the helium as they
78 come into contact and the helium boils off. The faster the transfer line is inserted, the more helium you lose and the more pre ssure builds. Try to keep the pressure below 50 in of water. If pressure builds too fast, you can pull the transfer line back up a little and resume lowering the transfer line once the pressure has dropped. Keep lowering the transfer line until it is al l the way down. Close the screw valve tight around the transfer line. Connect the yellow hose to the top of transfer line (it should be already connected to the bottom valve of He flow network). Insert transfer line into the cryostat. Remove the plug f rom the cryostat, hold the transfer line so that it is straight and aligned with the probe, and slowly insert it into the probe. Once the transfer line is all the way in, tighten the nut halfway to leave the nut loose in order to purge the air out of the probe with He gas. Start cool down. Make sure all the connections in the He system have been made. Open the valve on the board that connects the transfer line to the system (bottom valve). Turn on the He pump (small blue pump behind the magnet). Open the valve on the board that connects the pump to the board (third up from bottom). Open the black handled valve on the wall that leads to the He recovery system. Let system pump for ~10 mins. Tighten the nut to begin liquid He transfer (slightly wiggle the transfer line as you gently tighten the nut, making sure that it is all the way tightened ). Turn on the temperature control unit. Set the parameters of the temperature control device to the following settings : prop = 21 (default is 20). Leave int and der at default values. Set the temperature to 65K. Press the auto button in the block for the heater. Let system equilibrate to reach desired temperature. This can take awhile, if you walk away, be sure to come back in time to slow the gas flow as you reach the set temperature.
7 9 Remember, reference thermocouple of the temperature control unit (blue cord) has to be in the dewar with liquid nitrogen to give you correct temperature reading. While cooling, open the needle valve all the way, in order to get a He flow of approximately 40 psi. Once cooled, start closing the needle valve on the transfer line or on the 1 st bronze round handle in increasingly small increments to find the optimal (and minimal) He flow (usually ~6 10 psi) necessary to maintain 65K. Watch heater bars on the thermo regulator display (3 4 bars is ok, 5 6 bars means that He flow is too fast). If the gas flow is too high and the heater has to work too hard, it can and will burn out. Turn on the magnet and console. Turn on the water c ooler that chills the magnet. Turn on the magnet by hitting a green button, then hit the white reset button. Turn on the console by hitting the green button. Wait for the console to load (about a minute). This is signaled by a green READY button. Set up the computer. Log on to the computer (see Angerhofer to have an account created for you, have a secure password ready). Open Konsole. and t lick OK. Insert sample into probe. Prepare sample tube and adapter. Take a small piece of teflon tape and curl it up into a very small sphere. Drop this into the sample tube. Gently and carefully flick the tube to get the Teflon to the bottom of the tu be. The Teflon ball absorbs the pressure of solution expansion, preventing tube breakage. Find an adapter that fits the tube well. The tube should not fall out once you took it out of the magnet (must fit tightly ). Insert the sample tube and adapter into the sample wand by screwing it. The bottom of the sample tube should be inserted into the threaded end of the adapter. Doing otherwise will damage the adapter. Adjust the sample tube so that the active part (middle) of the
80 sample is 39 mm below the bottom of the adapter if using the MD5 resonator, or 55 mm below the bottom of the adapter if using the MD4 resonator. For any other resonator, be sure to check the manual. Make sure the position is above the teflon ball. Tighten the threads to secure t he tube. Freeze sample first wiping off the sid es of the tube and wand with a K imwipe to remove fingerprints and oil. Fill an open top dewar with liquid N 2 from CLB 416 nitrogen tank. Dip the tube into the liquid nitrogen at the most acute angle possi ble. This helps prevent the sample tube from breaking as a result of water expansion due to freezing. Leave the tube in liquid N 2 until just before you are ready to insert it into the cavity. Insert sample into cavity Turn off the valve that leads to th e He pump (third up from the bottom). It is critical that the vacuum is not on when the cavity is open as you will suck moist air into a cold cavity where it will freeze and form ice crystals. It is also critical that you swap out the wands quickly to min imize the amount of air that the cavity is exposed to and care must obviously be taken to ensure that you do not break your sample tube by hitting on the edge of the nut as you are putting your sample into the cavity. Ice on sample tube will not affect th e signal, but sample will be stuck inside. When the vacuum gauges on the board read zero, loosen the nut around the sample wand. Carefully remove the sample wand (be sure to pull the wand straight up as pulling at an angle will break the sample tube). Quickly but very carefully, insert the other sample wand with your sample tube in it. If you are not placing sample you can just put a wand inside to seal the cold cavity. Lower it gently to the very bottom and re tighten the nut. Make sure this nut is all the way closed, a leak here will introduce moisture into the very cold cavity where it will freeze your sample into place (the only solution to this is to
81 completely warm up the cavity and disassemble it, dry it, and put it back together. Tune cavity by o pen ing the FT Bridge window by clicking on the button that looks like a switch. Under the Bridge Configure tab, select CW mode and stab on. Open the Tuning window and the FT Bridge. In the Tuning window, turn on the dual trace a nd the reference a rm. Set to Tune, then s et the Attenuation to 15dB. If the dip is not seen on the screen, adjust the Frequency slider bar until it is found. Slide the Bias slider bar all the way to the right. Adjust Signal Phase so that the dip has a flat baseline and is on a plateau. Slide the Bias slider bar all the way to the left. Turn the monitor around so that you can stand in front of the spectrometer and see the monitor at the same time. Slowly lift the Q lever to its highest position (use sma ll increments) or until the dip is fully broadened. Adjust the Frequency (use the small black arrows) to center the dip (now broadened). Center the stabilizing frequency (External Stabilizer slider bar) by using the arrows in the Receiver Unit tab of the FT Bridge. Click Operate and close the bridge tuning windows. In the FT Bridge p anel Bridge Configuration tab, switch to Pulse mode. Close window. In the lower right hand corner of the green. Create a folder for a new experiment. My computer Check defense pulse. Create an experiment by opening Experiment window (icon that has EXP written in a comic strip hit Create. Make sure that the amplifier is off (near zero values on display mean that it is on standby) and that the max attenuation power is set at 60dB (in the FT Bridge). Click the
82 Clutch button (icon that resembles a pair of cogwheels ), which will send parameters to the hardware. Set up pulse tables. Open the Parameter panel (button 23), then drag the window to the upper left corner of the screen. Open SpecJet window (will appear on the upper left corner), and the FT Bridge panel. In the FT Bridge wi ndow (Receiver Unit tab), set Video Gain = 66 dB, and Video Band = 25 dB. In SpecJet, set Averages to Position to 200ns and the Length to 32ns. Select Acquisition trigger then from the Channel selection menu, set the Length to 2ns in pulse position 1. Click Start in the deo Ga in if pulses are not visible. Make sure that the 2ns pulse is in the middle of defense pulse!!! Adjust the Phase in the FT bridge if defense pulse is not clearly showing in one of the lines. Click Stop in both the SpecJet and Parameters windows. I f the amplifier is already ON or in make sure the Max Attenuation Power is set to 60 dB and turn on the TWT amplifier (white toggle). Let the amplifier sit for 2 min s to warm up. Test defense pulse once again before switching to Operate (Click Start in Parameters window, and Run in SpecJet window). Click Stop in both the SpecJet and Parameters windows. Push the 4 0 sec for the amplifier to reach 11.8 Check for cavity ringdown. Click Start in the Parameter window and Run in SpecJet. Turn down the High Power Attenuation slowly (this increases power) to check for cavity ringdown (usually to about 6dB). If there is significant ringdown, stop immediately and return the attenuation to 60 dB. Check the Q lever position. If problem
83 persists, seek help from Dr. Angerhofer. Use Video Gain if needed to adjust the size of the signal in the window and adjust signal phas e to see peaks better. Ideally, the intensity should only change within the bounds of the defense pulse. Outside of the defense should just be noise. If there is significant intensity outside the defense pulse, (1) you can damage the detector and (2) it will interfere with your signal. Set Attenuation back to its maximum value (60dB). (Shift+click =10, Ctrl+click =1) Click Stop in both the SpecJet and Parameters windows. The next objectives are to determine the center field of the spectrum, to know th e appropriate attenuation, to determine the values of T2, and roughly, T1. First, we collect a field swept echo spectrum (FS) to determine the pump and observe frequencies. Then, we measure T2 to determine the cor rect value for the D2 parameter. Collect a field swept spectrum Set Up Pulse Table. Open Parameters. protonated matrices) and the respective Lengths to 16 and 32 ns. Pulse 1 is a 90 pulse while pulse 2 is 18 0. We keep the lengths and positions of both pulses constant because we want to maintain a particular bandwidth of excitation to excite 2 particular populations we need. Thus, to calibrate pulses, we need to adjust power instead of the length later. Sele ct Acquisition trigger and set Length to 2ns. Click Start in the Parameters window and Run in SpecJet window. Optimize and phase echo. Lower Attenuation to 6 dB to see the echo. Use Video Gain to adjust the signal intensity (maximum gain ~66 gives the best S/N ratio but a lower value should be used if needed to prevent clipping of the echo). Adjust the Channel offsets (click on Settings in the SpecJet window) so that both the real and imaginary components are near the bottom
84 of the screen in the SpecJe t window. To optimize echo intensity, change the Center Field to maximize the echo (Parameters window Field tab Center Field). Adjust the Phase to make the imaginary channel (bottom) dispersive and symmetric. Set the Center Field position to be the center field transition (usually ~3460 G). Set Sweep Width to 200 G. Decrease high power attenuation (Ctrl+click) until the real echo has the maximum intensity (~0.1 6 dB) and phase again. This is simultaneously increasing power of both pulses (1) and ( 2). The attenuation that generates the largest echo is the correct power to make the first pulse a 90 degree pulse, and the second, 180. Adjust video gain if needed. Re phase if necessary. Determine echo location and width. To acquire only the echo yo u will have to determine the start and end positions of the echo using Acquisition trigger. Adjust the Acquisition trigger Position (will be ~1470 ns) so that the echo starts at 0 ns in SpecJet window. Write this down as n adjust Acquisition trigger Position so that the other end of echo is at 0ns (try not to clip off the echo tail). As much as possible, include more echo rather than less). Calculate the difference (usually 200 300 ns) and write this down as Pulse Length under Acquisition trigger. Return the trigger position to the beginning of echo and length of the pulse width as the length. Click stop in both windows. Set up the remaining parameters. Set the integrator to 4.0 ns. Set shots per point to 50. In the Acquisition tab, set the x axis to magnetic field (x axis size is 1024 points by default). Close all windows. Press play to collect the spectrum. Click on FS (if needed) to show full screen. Write down Center Field (G)/Frequency (GHz) during run. Save the file into folder you created before for this experiment (experiment initials are FS). Export as a DAT file (File Export Change .DTA file extension to .DAT). Here, we collect the
85 echo detected field swept absorbance spectrum of the nitroxide radi cal. From this graph of echo intensity as a function of field, we have to determine and write down the new center field where the echo intensity is maximal. Measure the Tm of the sample, by o pen ing the Parameters window. In the Acquisition tab, select Ti me as the x axis. Under Center Field enter the new center field you determined from the previous experiment. Select Acquisition trigger in the drop down menu and set the Position displacement (pos disp) to 16 ns. In the Patterns tab, and set the Position displacement (pos disp) in the second column to 8 ns. Close the window and press Play to collect the spectrum. You will get exponential decay of your echo intensity as function of time between two pulses. The waves in the beginning a re ESEEM modulations of 2 H in solvent. Write down actual Frequency (GHz). Save file (experiment initials are Tm) and export. Fit the line to button several times until Tau stops changing. Tau = T 2; expected to be within the 2000 16000 ns range. It is hard to get a right fit so you can determine T2 by eye it will be a point on curve a little before the flattening of the curve. T2 must to be greater than or equal to d2, which is always 3000 ns. R ecord the Tau value. Determine the pump and observe frequencies. DEER Set Up: Create Experiment. Attenuation to its maximum (60 dB; tantamount to lowest Clutch button to send parameters to hardware. Set Up Pulse Table. Open the Parameters window, SpecJet and FT Bridge. In the Set up tab: Set the shot repetition time (SRT,
86 SRT correlates with T1 and is twice as long. When SRT can b e doubled and the echo does not change, the ~T1 because the spins have enough time to relax.) to 4000. Set pi/2 equal to 16ns. Set d0 to 0 ns (will be determined shortly). Set d1 to 400 ns. Set d2 to the determined T2 currently 3000 ns is d2 of choice ). 1 This can usually be a little shorter than the actual T2. The trade off is between measuring longer distances (longer d2) and getting better signal to noise (shorter d2). Set pi ELDOR to 32 ns. Set d3 to 100 ns. Drop the High Power Attenuation ( in the FT Bridge) to the previously determined level for the best 90 pulse. Set the Video Gain to 66 dB. In the Field tab change Center field (CF) and make the Sweep Width equal to 160G. Click on Start and Run. Phase Echo. Adjust the channel offset (click on Settings button in Specjet) and use the *2 to see the imaginary component more clearly. In SpecJet window, phase echoes (via Signal Phase) until the imaginary signal (the bottom one) is dispersive. Click Stop in both SpecJet and Parameters wind ows. Collect the spectrum. Select 8 step phase cycling to get only the DEER echo. This type of phase cycling effectively eliminates the two stimulated echoes. Close all windows. Press Play. Save as DEER Setup experiment (experiment initials are DSU) and export to ASCII as .DTA file. Determine Echo position and location. Determine the width of the echo by pointing the cursor to the beginning and end of the echo. First, write down d0 and PG parameter. The latter is essentially the width of the echo. Field swept echo detected spectrum. In the Field Sweep tab, set the pulse gate (or gate; PG) to the echo width. Set d0 to the value that you determined for the left edge of the echo. Set shots per point to 100 and Sweep Width to 160G. Select 2 step p hase cycling. Leave other parameters at their default values Close all windows.
87 Collect Spectrum. Press Play and write down the current Center Field/Frequency. Save file with an FSD initial. Next, c alculate vELDOR. The pump pulse frequency should b e set to frequency that corresponds to the maximum of the center field transition. The observe frequency should be set to the maximum of the low field transition. Determine the distance between the two transitions ( B) by subtracting the center field fro m the low field. Convert this to MHz ( ) multiplying the field difference (in Gauss) by 2.83. The ELDOR frequency (pump) is obtained by subtracting ( ) from the frequency used to collect the field swept spectrum (because it corresponds to the center fi eld transition). Sample calculation : Frequency = 9.72589GHz Center Field = 3462G Low field = 3435G B = CF LF = 3262 G 3435 G = 26 G (this is the lowest possible value for B) B 2.83 MHz/G = 26 G 2.83 MHz/G = 72 MHz ( should be >65 MHz ) ELDOR = current 9.72589 GHz .072 GHz = 9.71869 GHz ( ELDOR freq.) Next, s et up DEER experiment. Set up puls e parameters. In the 4 P DEER tab, set field position to the low field value you have found. Set the ELDOR frequency to ELDOR you just calculated. Set the Attenuation to 0 dB to get DEER signal. Set shots per point to 100. Set dx = 12 (or 16). dx is time increment for pi pulse in 2 frequency channel (big dx means less steps), dx = Tau (~3000) / # points (255) = ~12 (usually). Select 2 step phase cycling. Set number of scans to 500000 (this will automatically be changed to the maximum possible number of scans, which is 132000). Close all
88 windows. Collect the spectrum. Press Play. Write down experiment start time and date. Stay fo r a couple of scans to check for the DEER signal that looks like oscillations of baseline (baseline should not be flat!) If you see it, pour all the liquid N 2 left into reference thermocouple dewar (enough for 2 3 day experiment). Let your experiment run until you have good signal to noise. An average run takes 2 do not expect substantial signal improvement in the next 24 hours. Recall that S/N is proportional to the square root of the number of scans. Fill both the temperature reference dewar and diffusion pump cooling dewar approximately every 12 hours. This is imperative for the maintaining the temperature of the experiment, as well as the saf ety of the pump. To close out an experiment, w rite down start and stop time of DEER experiment, date of experiment, and number of scans. Hit Stop experiment when enough scans have been collected. Wait for the program to end the phase cycling (try NOT to t ouch ASCII .DAT file. Click on Clutch to stop sending parameters to the spectromete r. Open Parameters window. In the 4 Pulse DEER tab set Attenuation to 30dB, then Switch to Tu Standby (green button) (If you are done with all your DEER experiments, you can also turn off the amplifier with the white toggle switch (put in Standby before turning off).
89 C lose third valve up from the bottom in order to break connection with the Helium pump Loosen nut and take out the sample carefully to avoid breaking the tube. Replace with a sample wand. Tighten nut. Open valve to He pump. Warm up the system. If all DEER experim ents are complete, it is time to warm up the system. There are two ways in which to do this, one that takes approximately 1.5 hours and another than takes approximately 12 15. If there is no time rush, the slow method is simpler. For slow method: Tur n off blue helium pump and close needle valve on the top of the Helium dewar. Reset the temperature control box by turning it off and turning it back on. You have now shut off the helium and set up the system to warm up naturally. After approximately 12 hours, when the system has reached room temperature, it is safe to remove the transfer line from the cavity. If a quick take down is required, a faster method is available which utilizes the system heater. CRUCIAL STEP: If warming up using heater, it is necessary to leave the Helium flowing at a very low rate. DO NOT WARM UP WITH HEATER WITH NO HELIUM FLOW, but rather just decrease flow rate. Now begin to raise the target temperature on the temperature control box. It is important that this be done in small steps (approximately 5 10K at a time), as not to blow the heating element or otherwise cause damage. Either way, once the system has reached room temperature, it is safe to proceed. The needle valve atop the helium dewar should be in the closed position and the helium pump should be turned off. Disconnect the cryogenics. The next step is to remove the transfer line from the room temperature cavity. Close the black valve under where the transfer line meets the cavity, closing off the cavity fro m the vacuum pumps. Unscrew the transfer line and slowly and carefully remove by pulling straight out. Take caution that fragile tip of the
90 transfer line does not get bent when removing. Place the protective cover on the end of the transfer line and scr ew into place. Replace the black cap onto the end of the cavity. Disconnect the yellow hose from the top of the transfer line. Loosen nut on top of dewar. With gloves on, lift the transfer line out of the dewar, then immediately tighten the nut atop th e dewar once the transfer line is clear. Place transfer line in storage place. Close release valve on dewar and wall and disconnect release line End transfer on the dewar computer, then d isconnect computer from dewar. Return dewar to storage room and reconnect to wall. Open release valve on dewar Reconnect dewar computer to computer in dewar storage room Shut down the spectrometer and pumps. Go to the Acquisition window. Disconnect from spectrometer. Exit program. Turn off devices in the follo ffusion pump Leave rough pump and water supply/return until diffusion is cool. Turn off rough pump. Turn off water supply/return.
91 Figu re 2 1. Schematic diagram of a simple mass spectrometer with a sector type mass analyzer. Figure adapted from Wikipedia, and was free for distribution and reprint. Figure 2 2. Schematic diagram of a standard fluorescence spectrometer.
92 Figure 2 3. Schematic diagram of a standard circular dichroism instrumental set up. Image adapted from Wikipedia and was free for distribution and reprint. Figure 2 4. Effect of choice of spin label on the derivative EPR line shape of HIV 1PR Subtype B si wit h A) MTSL, B) MSL, C) IASL, and D) IAP. All spectra were collected at X band frequency with 100 Gauss sweep width.
93 Figure 2 5. Site directed spin labeling scheme showing the reaction of MTSL with a free thiol group of an engineered CYS residue. F igure 2 6. Structure of MSL spin label.
94 Figure 2 7. Spin labeling scheme for the incorporation of the K1 spin label moiety. Figure 2 8. Scheme for electron transition in a magnetic field. A) Energy level diagram for a free electr on in an applied magnetic field, and B) c orresponding derivative of absorption spectrum.
95 A B Figure 2 9. Cartoon r epresentation of EPR. A) Energy diagram for a system with a free electron (m s = ) undergoing hyperfine interaction with the nucleus of nitrogen (m I = 1), and B) r epresentative de rivative EPR spectrum for a nitroxide spin label.
96 Figure 2 10. Dependence of EPR spectral line shape on motion correlation time, and degree of anisotropy Figure 2 11 The 2 + 1 pulse DEER sequence. Figure 2 12 The 3 pulse DEER sequenc e where t ime t is varied and is fixed.
97 Figure 2 13. The 4 (top) with a Hahn echo sequence ( /2 ) followed by a refocusing 2) (bottom). Figure 2 14 Absorption spectra for a nitroxide spin label with positions of the low field transition approximately 26 Gauss (72 MHz) below the central resonance line, and the center field transition mar ked as the
98 Figure 2 15. Sample dipolar evolution curve showing locations of raw dipolar modulation, background subtraction, and background subtracted echo curve with fit. Figure 2 16. Selection of regularization parameter in TKR D EER analysis method. A) Typical L curve showing regions of unde rsmoothed optimal and ov ersmoothed with distance profiles and echo fits for B) undersmoothed, C) over smoothed, and D) optimal regularization parameters.
99 Figure 2 17 Example s of processed DEER data. Examples of A) dipolar modulated echo curves with B) corresponding distance profiles.
100 CHAPTER 3 SURFACE LOOP CONFORM ATIONAL FLEXIBILITY OF THE GM2 ACTIVATOR PROTEIN IS NOT ALTER ED BY LIGAND BINDING BUT IS NECESSARY FOR LIPID LIGAND EXTRACTION FR OM BILAYER SURFACES Introduction Glycosphingolipid catabolism occurs in lysosomal compartments within the cell ( Sandhoff, Kolte r, 1995 ) GSLs are endocytosed, trafficked and sorted through early and late endosomal compartments on the way to the lysosome, where specific enzymes sequentially cleave sugar groups, eventually producing ceramide, which is finally deacylated to sphingos ine ( Griffiths, Hoflack et al, 1996 ) More than ten different enzymes and five accessory proteins are involved in this important process. One of those accessory proteins, the GM2AP, is essential for stimulating th e catabolism of neuronal gangliosides by extracting ganglioside GM2 from intralysosomal vesicular membranes ( Sandhoff, Kolter, 1996 ) As described in detail in Chapter 1, GM2 activator protein (GM2AP) is a general lipid transfer protein that functions in the lysosomal compartment of cells GM2AP extracts ganglioside GM 2 in the late endosome and presents it for hydrolysis by hexosaminidase A to convert ganglioside GM2 to GM3. A ctivator proteins often function by binding to glycosphingolipids and forming aqueous soluble complexes thereby providing an aqueous soluble enz yme access to a lipid generally confined to a hydrophobic environment. Mutations in GM2AP that inactivate the protein lead to the lipid storage disease known as Tay Sachs ( Sandhoff, Kolter, 1995; Gravel, Clarke, 1995 ) The importance of GM2AP in regulating lipid catabolism has been widely p ublished but little is known about the mechanism that allows GM2AP to function. The focus of my research is on elucidating
101 key physical and structural characteristics that allow GM2AP to bind and extract its lipid substrate. In addition to stimulating GM2 degradation, GM2AP has been proposed to act as a general lipid transporter because in vitro GM2AP has been shown to bind and extract GM2 from micelles, transfer glycolipids, phospholipids and fluorescently labeled lipids ( Conzelman, Burg et al 1982; Mahuran, 1998; Ran, Fanucci, 2008 ) bind and inhibit PAF ( Rigat, Wang et al. 1997; Rigat, Yeger et al. 2009 ) and present CD1 molecules to antigens (Kolter, Sanhoff, 2005; Major, Joyce et al. 2006; Zhou, Cantu et al. 2004) In order to carry out its function as a lipid transporter, GM2AP must partition with the lipid bilayer surface. A model of membrane partitioning was recently p roposed based on results of sedimentation assays which showed that only a 15% fraction of GM2AP molecules remains on the bilayer surface, and therefore, implying that GM2AP is undergoing exchange with the bilayer surface ( Ran, Fanucci, 2009 ) From fitting the results of sedimentation assays, we concluded that GM2AP establishes exchange eq uilibria of a minimum of four species: GM2AP in solution, GM2AP on the bilayer surface, GM2AP lipid complex on the bilayer surface and GM2AP lipid complex in solution. GM2AP membrane interactions have also been studied by others using Langmuir monolayers a nd these data suggest that GM2AP is surface associated without deep penetration into the bilayer ( Giehl, Leem et at al, 1999 ) Recently, we also showed with site directed spin labeling (SDSL) and power saturation electron para magnetic resonance (EPR) that the orientation of GM2AP on zwitterionic bilayer surfaces is such that the mobile external loops that line the opening to the lipid binding pocket are positioned at the bilayer surface ( Mathias, Ran et al. 2009 )
102 GM2AP is thought to sample three distinct conformations, although the functional importance of these conformers has not yet been defined (Wri ght, Li et al. 2000) Various X ray structures with different lipid substrates reveal multiple conformations of al conformation change conformational changes of the loop region, site directed spin labeling electron paramagnetic resonance (EPR) spectroscopy was applied to GM2AP (Lundb lad, 2004). In proteins, site directed spin labeling typically proceeds by generating site specific cysteine amino acid substitutions for modification with paramagnetic nitroxide probes, e.g., 4 maleimide TEMPO (Figure 3 1). Reduced cysteine residues act a s nucleophiles capable of reacting with maleimide or thiosulfonate groups to form thioethers or disulfide bonds, respectively (Tipton, Carter 2009 ) Cysteine residues involved in disulfide bonds do not readily react with those functional groups at pH 6.8 7.4 (Lundblad, 2004) To ensure that purified protein contained the proper connectivity of the 8 native cysteine residues and that only the introduced reporter cysteine (L126C) was labeled, a mass spectrometry based assay was developed. Because of the ana lytical challenge associated with verifying multiple disulfide bonds, a multitier assay was needed the full details of this experimen t will be discussed in chapter 4 (Tipton, Carter, 2009) The first aim of my project is to assess the functional importan ce of the residue in each mobile loop is substituted to a CYS. The two mutant CYS moietie s
103 were chemically cross linked and novel fluorescent based assays were used to monitor vesicle sedimentation, ligand extraction, and extraction kinetics of cross linked GM2AP constructs. The results further hypothesize that restricting loop mobility slows the rate of substrate extraction. Materials and Methods Materials T he chemicals, reagents, and supplies were obtained from Fisher Scientific (Pittsburg, Pennsylvania) and used as received, with a few noted exceptions. Methanol and water were purchased fr o m J.T. Baker (Philipsburg, NJ) at HPLC grade. Trypsin, endoproteinase GluC, endoproteinase Asp N, formic acid, and ammonium bicarbonate were purchased from Sigma Aldrich (St. Louis, MO). 1 palmitoyl 2 oleoyl sn glycero 3 phosphocholine (POPC) in chlorof orm, bis(monooleoylglycero)phosphate (BMPdi18:1 or DOBMP) in chloroform was purchased from Avanti Polar Lipids (Alabaster, AL) and used without further purification. N (5 dimethylaminonaphthalene 1 sulfonyl) 1,2 dihexadecanoyl sn glycero 3 hosphoethanolami ne, triethylammonium salt (DDHPE) was purchased from Molecular Probes (Eugene, OR) in the form of powder. Methanethiosulfonate spin label (MTSL) was purchased from Toronto Research Chemicals, Inc. 1,8 bis maleimidotriethlyeneglycol (BM(PEG) 2 ), I 1,11 bis m aleimidotriethyleneglycol (BM(PEG) 3 ), and dithio bismaleimidoethane (DTME) were purchased from Thermo Scientific Rockford IL. Methods GM2AP is sub cloned into pET 16b vector, which is used for expression of histadine tagged proteins. GM2AP is expressed i n inclusion bodies and must be refolded following total denaturation with urea. The following sections describe, in
104 detail, the methods associated with the expression, purification, and spin labeling of the various GM2AP constructs. Expression of GM2AP co nstructs GM2AP constructs are expressed in BL21 (DE3) E. coli as His tagg ed proteins. Transformed cells are grown in 1 L sterile LB media to an OD 600 = 0.8 1.1. The cells are then induced by IPTG for 8 hours at 20 C by adding 500 L of 0.8 M IPTG. Foll owing induction, the cells are pelleted using a Sorvall RC6 centrifuge at 5000 rpm for 20 minutes. GM2AP purification protocol The BL21 (DE3) E. coli. pellets were resuspended in lysis buffer (0.1 M sodium chloride, 0.1 M Sodium phosphate, 1% tween). The c ells were then lysed by two rounds of sonication and French press followed by centrifugation in a Eppendorf 5810 R centrifuge for 20 minutes at 10500 rpm. The isolated inclusion bodie s were then resuspended in 5 M u rea, homogenized, and sonicated once mor e. After centrifuging again in the Eppendorf 5810 R centrifuge for 20 minutes at 10500 rpm, the supernatant was loaded onto a nickel chelating column equilibrated with 5 M urea at pH 8.0. A guanidine hydrochloride buffer (6 M guanidine hydrochloride, 0. 1 M sodium phosphate, 10 mM Tris, and 5% glycerol) was used at pH 8.0, 7.0, and 4.5 to wash and elute the protein off the nickel chelating column. The eluted protein solution was slowly transferred into refolding buffer (50 mM tris, 10% glycerol, and 0.0 5% tween20) for rapid 20 fold dilution using a peristaltic pump. The refolded GM2AP mutant was then dialyzed to a pH of 7.0 and the mutant cysteine was labeled by either maleimide biotin or maleimide 4 tempo.
105 A Q column equilibrated with 25 mM tris, 2. 5% glycerol, and 0.05% tween at a pH of 8.0 was used concentrate to the protein. The GM2AP mutant was then eluted from the Q column by the addition of 150 mM sodium chloride. The concentrated GM2AP eluted from the Q column was loaded onto a S 200 Sephacry l column equilibrated with 25 mM ammonium bicarbonate to remove any tween20 and isolate the misfolded protein from correctly folded protein. Spin labeling The substituted cysteine residue was modified with 4 maleimide TEMPO (4MT) overnight in the dark (4 C) in 25 mM Tris, 2.5% glycerol, and 0.05% Tween 20 buffer (pH 7.0). The spin labeled L126C 4MT GM2AP construct was then dialyzed against 25 mM Tris, 2.5% glycerol, and 0.05% Tween 20 buffer (pH 8.0) and concentrated to 200 L. Concentrated spin labeled protein was loaded onto a size exclus ion S 200 Sephacryl column (Am ersham Biosciences, Piscataway, NJ) and eluted with 25 mM ammonium bicarbonate (pH 8.0) to remove the salts and detergents. Site directed spin labeling is used to determine if there is a c onformational labeling was needed, the purified and refolded protein sample was buffer exchanged into 10 mM Tris HCl, pH 6.9 (further details of spin labeling discussed in next section) after the final purification step. Approximately 1 mg of spin label (IASL, IAP, MTSL or MSL) was dissolved in 100 L ethanol, and added to approximately 40 mL of HIV 1PR in 10 mM Tris HCl, pH 6.9. The spin labeling reaction was carried out in the dark ( via wrapping reaction tube in aluminum foil) at room temperature for approximately 4 6 hours followed by 6 8 hours at 4 C for inactive (D25N) constructs, and at 4 C for
106 approximately 8 12 hours for active (D25) constructs. At this time, the sample solut ion was centrifuged at 12000 rpm for 20 minutes at 4 C to remove solid impurities and aggregated proteins. The sample was then buffer exchanged into 2 mM NaOAc, pH 5, and concentrated to OD 280 =1.25. Crosslinking The substituted double cysteine resid ues (I66C/L126C and I72C/S130C) were modified one of the following cross linking reagents 1,8 bis maleimidotriethlyeneglycol (BM(PEG) 2 ), I 1,11 bis maleimidotriethyleneglycol (BM(PEG) 3 ), and d ithio bismaleimidoethane (DTME). R eactions were allowed to run o vernight (4 C) in 25 mM Tris, 2.5% glycerol, and 0.05% Tween 20 buffer (pH 7.0). Concentrated spin labeled protein was loaded onto a size exclusion S 200 Sephacryl column (Amhersham Biosciences, Piscataway, NJ) and eluted with 25 mM Tris and 2.5% glycero l (pH 8.0) to remove any aggregated protein. Preparation of POPC:Dansyl DHPE liposomes As p rev iously reported by Ran, Fanucci, 2008, POPC in chloroform was used without further purification POPC: d ansyl DHPE ( 4 :1 in molar ratio) vesicles were prepared by m ixing the desired amounts of both lipids (in chloroform), d r ying to a film by a stream of nitrogen and vacuum desiccation for 6 12 hours. POPC:R18 (19:1 in molar ratio) vesicles were prepared by mixing the desired amounts of POPC (in chloroform). For kinet ic measurements, the dried lipid films were hydrated for one hour in an appropriate volume of 50 mM sodium acetate buffer pH 4.8 Large unilamellar vesicles (LUVs) of above lipid samples were prepared by extrusion through 100 nm polycarbonate filters with 55 passes using a hand held mini extruder Final phospholipid
107 concentrations were determined on the basis of total phosphate determination by Malachite Green Phosphate Assay Kit (BioAssay Systems). Results and Discussion The models of GM2AP (PDB ID 1G13) f rom X ray crystallography reveals that the volume of the hydrophobic cavity is roughly six times larger than the volume of a ceramide moiety (500 3 ). In addition, based on X ray structure analysis of five different crystal forms, large differences in the diameter and area of the opening to the lipid binding cavity were detected and attributed to flexibility of loop regions that decorate the rim of the cavity ( Wright Li et al. 2000; Wright, Mi et al. 2004; Wright, Zhao et al. 2003; Wright, Mi et al 2005) These regions (highlighted in Figure 3 2) are of particular interest for this study as their mobility may be related to function. The loops are located on opposi ng sides of the entrance to a prominent hydrophobic cleft whose width varies substantially among the different crystal structures. On one side of has a relatively stable c onformation in all crystal structures examined due to its involvement in crystal lattice contacts (Wright, Zhao et al. 2003) On the opposing side, a reverse turn (S130 T133) containing W131 exhibits two alternative conformations. In structures where the cleft is wide open and bound to lipid (PDBID 1PUB), this loop is flipped out with W131 exposed to solvent (Wright, Zhao et al. 2003) The second conformation of this loop is seen in structures where the cleft is closed, and the loop is tucked in towards the interior of the cleft, burying W131 and st between residues on either side of the cleft (L128, P129, L132 and I72, I66) and PDB ID 1G13 ( Wright, Li et al. 2000; Wright, Mi et al 2005) Given that this mobile loop is preceded by an extended disordered chain
108 segment (V122 P129), mobility of this entire region (V122 T133) can modulate the width of this cleft, and therefore the size of the circumference of the cavity (Figure 3 2 ) which allows for lipid ligand to enter into the binding pocket. The question arises as to whether the features of the protein that are seen in the crystal structure are in conformational exchange in solution and how these various conformations are related to function, as protein mobility and dynamics are often related to function ( Smock, Gierasch, 2009; Clore, 2008 ) For the case of GM2AP, the multiple conformations in the crystal structure, which in effect modulate the size of the entrance to the cavity, may suggest that a conformational change is necessary for extracting li pid ligand from vesicles surfaces, or that the conformat ions will be modulated in the ho lo (lipid bound) or apo protein. Another possible role for the conformational flexibility is in conformational entropy to allow GM2AP to partition with the vesicle surf ace or to interact with hydrolases, such as HexA, in GM2 hydrolysis. To investigate the local structure and mobility of the apolar and mobile loops of GM2AP in solution, SDSL EPR was utilized, which is a powerful spectroscopic tool used to study conformati onal changes in proteins as well as to characterize local backbone motion ( Fanucci, Cadieux et al. 2002; Fanucci, Cafiso, 2006; Hubbel, Cafiso et al. 2000; McHaourad, Lietzow et al. 1996; Columbus, Hubbell, 2004; Columbus, Kalai et al. 2001 ) Thus, ten single CYS variants of GM2AP were previously generated and labeled with MTSL at the sites shown in Figure 3 3 Six of the chosen spin labeled sites are located in the membrane binding loops (A60R1, I66R1, I72R1, L126R1, S130R1, and N136R1) a nd were chosen to investigate the conformational flexibility using EPR
109 spectroscopy. The remaining four CYS variants of GM2AP (V54R1, L87R1, T90R1, and S115R1) were generated to probe sites thought to be either structured ( h elices and sheets) or loop regions of the protein. This conformational exchange process is slow on the EPR time scale such that both spectral components are resolved and the enthalpy and entropy required for the conformational change were found to be 65 kJ/mol and 215 J/mol K, respectively. Figure 3 4 shows the nitroxide EPR line shapes recorded at ambient room temperature for GM2AP in solution under basic pH and acidic pH in the presence of GM2 micelles. The EPR line shapes at each site are consistent with those expected based upon the local structure and dynamics reflected in the x ray structures of GM2AP. This conclusion is draw upon literature reports showing that the R1 line shape correlates with protein structural components and B factors ( Xu, Kim et al. 2008 ) Of the six sites in the flexible loops and disordered strand regions, A60R1, L126R1, and N136R1 have line shapes consistent with fairly mobile and solvent accessible sites on proteins ( McHaourab, Lietzow et al. 1996 ) For example, the spectra from site L126R1 are narrow and intense, reflecting a higher degree of motional averaging. This site is located in a flexible strand with very high crystallographic B factors. Sites I66R1, I72R1 and S130R1 have broadened EPR spectra with structure seen in the high field resonances, which indicate that these spin labeled sites reside in more structured regions of GM2AP. In numerous X ray structures of GM2AP, the side chains of I66 and I72 can be seen to point in towards the hydrophobic cavity, so it is likely that at these sites, the spin label motion is restricted by neighboring amino acid side chains as well as the limi ted space of the cleft leading to the lipid binding pocket. Further, the line shape from site S130R1 is
110 broadened and more reflective of a spin label attached to an alpha helical region of a protein, and some of the crystal structures of GM2AP place this r esidue in a helical structure ( Wright Li et al. ,2000; Wright, Zhao et al. 2003 ) Previous EPR spectra obtained with acidic conditions in the presence of GM2 micel les are very similar to those obtained under basic pH conditions, suggesting that little to no conformational change has occurred in the presence of ligand. In efforts to further characterize possible conformational changes of the loops upon GM2 ligand bin ding, two different double CYS constructs were prepared, I66R1/L126R1 and I72R1/S130R1. These sites allow for distance measurements across hydrophobic cleft between the two loops to be made. Low temperature EPR spectra were collected and analyzed for dista nces across the binding cleft. As can been seen in the spectra in Figure 4A, very little to no differences were detected in the spectra upon addition of ligand. The average distances obtained from analysis of the spectral line shapes using the empirical d 1 /d 0 parameter ( Gross, Columbus et al. 1999 ) are given in Table 3 1, and values are all between 18 22 A slight increase of 2 in the average distance was seen for I66R1/L126R1 upon addition of GM2 micelles. Bec ause the average distances are near 20 which is the upper limit of distances readily detected by this method ( Altennbach, Oh et al. 2001; Rabenstein, Shin, 1995 ) pulsed double electron electron resonance (DEER ) spectroscopy was also performed for these sites. The background subtracted DEER echo curves are shown in Figure 3 5. Unfortunately, the average distance of 20 is also near the lower distance cut off for accurate analysis of DEER distance distribution p rofiles. Nevertheless, the shapes of the echo curves are consistent with an average distance of 20 23 (Data in Table 3 1) and show that no
111 major conformational changes of these loops occur upon binding GM2 ligand. The lack of distinct oscillations in th e DEER echo curve again indicates a high degree of conformational flexibility, which is discussed in more detail below. Because the EPR line shapes and distance data reveal little to no change upon addition of GM2, two different experimental observations t o ensure GM2 was bound were undertaken. In the first of these, samples where GM2 micelles were added, protein:GM2 complexes were purified by size chromatography and the presence of GM2 in the protein fraction was detected via a rescorcinol assay as describ ed previously ( Ran, Fanucci, 2008; Ran, Fanucci, 2009 ) Second, for both A60R1 and L126R1,differential scanning calorimetry showed an increase in T m of 0.9 C of the thermotropic unfolding for the halo protein comp ared to the unliganded protein; thus, further confirming that GM2 was indeed bound to the protein. It is noteworthy that the absolute values of T m differed for the two constructs (Data in Table 3 2) giving an indication of the relative stability of the mo dified proteins compared to WT (Figure 3 10). In addition to the above mentioned six sites in GM2AP, four additional sites were also previously chosen to investigate conformational heterogeneity. The EPR line shapes of all ten R1 labeled GM2AP constructs i n basic solution with the overlaid simulations are shown in Figure 3 11. The EPR spectra were simulated using the MOMD model of Freed and colleagues ( Budil, Lee et al. 1996 ) Using this method, two spectral components are required to adequately regenerate the experimental EPR line shapes for the sites in the flexible loops (A60R1, I66R1, I72R1, L126R1, S130R1, and N136R1). To simplify the fitting, we assumed that one component was mobile with the
112 value of the order parameter C 20 = 0, as was done previously by o thers ( Columbus, Hubbell, 2004 ) As can be seen in Figure 3 11, the two components have dramatically different line shapes, with one having narrow, isotropic like fea tures, corresponding to a highly mobile site and the other a more broadened line shape corresponding to a site with more restricted motion. We assign the two components to a mobile and immobile spin label conformation, consistent with the alternative confo rmations seen for these flexible regions in the GM2AP X ray structures ( Wright, Li et al. 2000; Wright, Mi et al. 2004 ; Wight, Zhao et al. 2005; Wright, Zhao et al. 2003 ) The relative percentages of each spect ral component are also given in Figure 3 11. To further test the hypothesis that the multiple spectral components reflect protein flexibility, four additional cysteine variants were generated within regions of the protein, which based upon the protein fol d and X ray B factors, would be predicted to have nitroxide line shapes that could be simulated with single component fits. Two of these sites, V54 and S115, are located on the back side of the cup structure in strands. In addition, L87 is a solvent acce ssible site on the short helix, and T90 is located in an unstructured loop. The EPR spectra obtained for these four spin labeled sites are shown in Figure 3 11. In each case, these spectra have line shapes that are consistent with their location in struc tured or unstructured regions in GM2AP, and each could be adequately simulated using a single component fit and the overlay of the experimental and simulated spectra are shown in the top panel in Figure 3 11. Sample homogeneity was assessed by high resolut ion gel filtration and HPLC, which showed single peaks upon elution, verifying that the two components in the EPR data arise from conformational flexibility of a single R1 label and not from mislabeling or misfolding of
113 the disulfide bonds Figure 3 12. To support the claim that the two components in the spectral fits arise from conformational flexibility of the loops and not mislabeling or misfolding of the disulfide bonds GM2AP CYS variants were screened by mass spectrometry methods previously shown ( Tipton, Carter et al. 2009 ) Briefly, the mass spectrometry results verified both the correct location of the spin label and that samples were singly labeled. The isotopic patterns of the two distinct protein forms can be resolved by FTICR MS and thus, can be matched to the calculated isotopic distribution obtained from the molecular formula for the labeled and unlabeled forms of GM2AP. The full experiment for the L126R1 construct has been recently reported a nd will be discussed in detail in chapter 4 ( Tipton, Carter et al. 2009 ) To further investigate the origin of the multiple component spectra for sites located in the flexible loop regions of GM2AP, data were collected as a function of temperature over the range of 10 o C to 30 o C in 5 C increments. A similar approach has been utilized to characterize a conformational change in the transmembrane sequence of the transducer domain of N. pharaonis halobacterial transducer of rho dopsins II (NpHtrII) in lipid membranes ( Bordignon, Klare et al. 2005) For GM2AP, L126R1 in the disordered strand and A60R1 and I66R1 adjacent to the apolar loop were selected for the variable temperature EPR exp eriments. If these regions of the protein are in conformational exchange, the equilibrium populations of the two components should change in a predictable manner with temperature. The EPR line shapes obtained at each temperature were then simulated for I66 R1 and L126R1 using the MOMD model to extract fractions of populations of the two components. The high degree of free spin in the spectra of the A60R1 construct prevented accurate modeling
114 of these spectra. Representative EPR spectra and MOMD simulations of line shapes for L126R1 and I66R1 at select temperatures are shown in Figure 3 13. Changes in the EPR line shapes were seen for L126R1 with increasing temperature; however, no significant changes in the A60R1 or I66R1 line shapes were observed. The perc entages of the mobile and immobile components were obtained from the line shape fittings for L126R1 and I66R1 and are plotted as a function of temperature (Figure 3 13). The plot of component percentages for I66R1 did not show a temperature dependent chang e, which is consistent with the EPR spectra that do not change significantly with temperature. However, it is noted that the MS results indicate under labeling of site I66C, which suggest that spin label incorporation at this site might interfere with WT f lexibility. Keeping this fact in mind, the results indicate that site I66, at the top of the apolar loop is not undergoing conformational exchange and the two components in the spectral fit may represent alternative conformations of the spin label. This in terpretation of the data is also consistent with the crystal structure, which shows little positional change of the backbone of this segment in the three monomers A, B and C (1G13) (Figure 3 2). Therefore, the origin of the two component spectra at this s ite may arise from different rotameric states of the spin label. For L126R1, changes in the EPR line shapes as well as in the percentages of mobile and immobile components were observed as the temperature was varied. At low temperatures, the percent mobil e and immobile components were about 10% and 90%, respectively. With increasing temperature, the percentage of mobile component increased and the percentage of the immobile component decreased until almost equal populations of the two components were seen at 25 C. The natural logarithm of the
115 ratio of the fractions of the two components exhibited a linear dependence with inverse temperature (Figure 3 14). Here, f 1 refers to the fraction of the mobile component and f 2 refers to the fraction of the immobile component. The data shown in Figure 3 14 were the result of two independent experiments performed on two separate samples prepared from different protein expression, refolding, spin labeling and purification preparations. The simulations were also performe d independently by different researchers. Results for the relative percentages of each component are strikingly similar. Both data sets and simulations were plotted together and the slopes of the lines were reproducible within error. The solid line shown is a linear fit of all the data taken together with values of 7.8 0.8 for the slope and 25.8 2.8 for the intercept, from which, the enthalpy and entropy of the conformational change were calculated to be 65 kJ/mol and 215 J/mol K, respectively. From t he values of f 1 and f 2 determined from the spectral fits, the value of the equilibrium constant, K at 298.15 K was 0.7, indicating that the activation energy for this conformational change is on the order of a few kT and that both conformations are easily accessible at physiological temperatures. L126R1 is located in the conformationally disordered strand of GM2AP, and its position is poorly defined in all 3 monomers of the X ray structure (1G13). In monomer A, the conformation of the loop is such that the side chain of L126 appears to point towards the protein interior while it extends out into solution in monomer C. The two spectral components determined from the MOMD simulation are consistent with these different environments of the spin label at this si te. The immobile component may arise from a fraction of protein conformers having L126R1 pointing in towards the lipid
116 binding cavity, where the motion of the spin label can be restricted by neighboring amino acid side chains as well as by the limited spa ce in the cleft of the cavity entrance. The mobile spectral component is consistent with the strand conformation seen in monomer C, where the spin label would extend out to solution and have a higher degree of rotational freedom. In addition to binding it s specific ligand, GM2AP has been shown in vitro to bind to other fatty acids (Wright, Li et al. 2005). Taking advantage of this result a dansyl fluorescence based assay for monitoring the kinetics of lipid extraction was developed and prev iously reporte d by Ran, Fanucci 2008 The assay is uses the fluorescent properties of phosphatidylethanolamine with a dansyl labeled headgroup (dansyl DHPE) in which the wavelength of maximum emission and quantum yield is sensitive to the polarity of the local environm ent. Upon sequestration of the dansyl head group into the hydrophobic cavity of GM2AP there i s an observable blue shift to 518 nm from 484 nm corresponding to the dansyl residing in large unilamellar vesicles of POPC:dansyl DHPE(4:1) to being bound in the hydrophobic pocket of GM2AP A representation of this assay can be viewed in Figure 3 15. To further investigate whether loop plasticity is of functional importance the apolar and disordered loops from previously generated double CYS constructs I66C/L126 C and I72C/S130C were physically tethered using three different cross linking reagents : 1,8 bis maleimidotriethlyeneglycol (BM(PEG) 2 ), 1,11 bis maleimidotriethyleneglycol (BM(PEG) 3 ), and dithio bismaleimidoethane (DTME). The results (shown in Figure 3 16) dans yl DHPE extraction assays show that merely mutating the native residues to cysteine slow the rate of ligand extraction by GM2AP. The addition the tethering
117 reagent further diminishes the rate of ligand extraction leading to the hypothesis the conform ational plasticity is an important component in the ligand extraction mechanism of GM2AP. To confirm that the physical tethering of loops also per turbed the extraction of the preferential biological ligand a GM2 extraction assay was developed The assay was used to monitor the efficiency of GM2AP at extracting GM2 from unilamellar vesicles containing PC/GM2/cholesterol/BMP (50:10:20:20 molar ratio) (Assay schematic shown in Figure 3 17) (Ran, Fanucci, 2009) W e define extraction efficiency in this assa y as the number of moles of GM2 divide by the number of moles of GM2AP that elute in tandem fraction s from an S200 size column. The results again indicated that merely mutating the native residues in the apolar and disordered loops had an effect on the mec hanism of extraction. While the physical tethering of the loops further reduced the number molecules of GM2 that GM2AP was able to extract (Data in Table 3 3). Site directed spin labeling was utilized to investigate the flexible loop regions of GM2AP an accessory protein with eight native cysteines that form four disulfide bonds essential for the stability of exposed loop regions. We were able to engineer a ninth and tenth cysteine in th e protein as well as express and purify, then isolate homogeneous sa mples of spin labeled GM2AP. Analysis of the EPR spectral line shapes and MOMD simulations for spin labels in the disordered chain segment and both loops are consistent with the various conformations seen in X ray structures. Multiple conformations of the mobile loop in solution were detected by EPR, indicating that the conformations seen in the crystal structure are present in solution and may
118 play a functional role in ligand binding or interactions with lipid vesicles. Although, distance measurement s fro m EPR were not able to detect local conformational changes in the loop regions as a result of ligand binding. The ho lo state of R1 mutants was confirmed by differential scanning calorimetry, which observed a 0.9 C increase in the thermotropic unfoldi ng o f GM2AP constructs with GM2 bound Furthermore, functional assays showed that immobilization of the loops inhibit ed the ability of GM2AP to bind its ligand, i ndicating that although GM2AP may not undergo a global conformational change as a result of liga nd binding loop plasticity is essential to its role as a general lipid transfer protein.
119 Table 3 1. Summary of Double CYS R1 GM2AP Construct Distance Experiments Construct d1/d0 CW Distance DEER Distance Distance FWHM Crystal Structure Theoretical I66R1/L126R1 0.39 182 182 12 18 18 30 I66R1/L126R1 +GM2 0.37 202 182 10 18 18 30 I72R1/S130R1 0.40 182 182 15 19 18 30 I72R1/S130R1 +GM2 0.38 182 182 12 19 18 30 Table 3 2. Differential scanning calorimetry values for GM2AP constructs Construct T m A60R1 75.0 C A60R1+GM2 75.9 C L126R1 72.7 C L126R1+GM2 73.8 C Table 3 3 GM2AP cross linking ligand extraction data Construct Extraction Efficiency WT 0.69 I66CL12 6C 0.40 I66CL126C BM(PEG) 3 0.13 I72CS130C 0.51 I72CS130C BM(PEG) 3 0.25
120 Figure 3 1 Structures of GM2AP and MTSL. A ) Structure, molecular weight, and elemental composition of 4 maleimide TEMPO. B : GM2AP crystal structure (PDB ID 1G13) with MTSL represented in blue at position L126. Figure 3 2 The three unique monomers (A, B, C) of apo GM2AP (PDB ID 1G13) showing the different conformations of the loop regions at the cleft entrance. The four disulfide bonds are shown in orange, the apolar lo op (V54 W6 3) is shown in green, the reverse turn (S130 T133) is shown in red and disordered strand (V122 P129) is shown in blue. T aken together, the reverse turn and disordered strand are collectively referred to within the text as a mobile loop.
121 Figure 3 3 The red spheres in monomer A of GM2AP (PDB 1G13) represent the C positions of the reporter sites chosen for SDSL EPR experiments. Figure 3 4 Stack plot of EPR line shapes from six sites in the flexible loops of GM2AP. Spectra were collected at ambien t room temperature without temperature regulation. The black traces are for samples prepared in basic pH (8.0) whereas those in gray were colle c ted at pH 4.8 in the presence of 4x molar excess GM2 micelles. All spectra have 100 G sweep widths. Note the spe ctra obtained for I72R1 in the presence of GM2 showed the most change, but this effect results from protein instability that lead to protein precipitation over time, with increased broadening of the spectral line shape.
122 Figure 3 5 DEER data from GM2AP double mutants. A) Absorption and der ivative CW EPR spectra of double CYS constructs I66R1/L126R1 and I72R1/S130R1 collected in the frozen state at 130 C. B) Dipolar evolution evolution spectra for double CYS constructs I66R1/L126R1 and I72R1/S130R1 collected in the frozen state at 208 C
123 Figure 3 6 GM2AP construct I72C/ S130C. A) Long pass filtered and background subtracted dipolar echo curve (black) overlaid with the TKR regenerated echo curve from DeerAnalysis (red). B) L curve utilized to choose the appropriate re analysis. D) Frequency domain spectrum. A B C D
124 Figure 3 7 GM2AP construct I72C/ S130C bound to GM2 A) Long pass filtered and background subtracted dipolar echo curve (black) overla id with the TKR regenerated echo curve from DeerAnalysis (red). B) L curve utilized to from TKR analysis. D) Frequency domain spectrum. A B C D
125 Figure 3 8 GM2AP construct I66C/S126 C. A) Long pass filtered and background subtracted dipolar echo curve (black) overlaid with the TKR regenerated echo curve from DeerAnalysis (red). B) L curve utilized to choose the ile from TKR analysis. D) Frequency domain spectrum. A B C D
126 Figure 3 9 GM2AP construct I66C/L126 C bound to GM2 A) Long pass filtered and background subtracted dipolar echo curve (black) overlaid with the TKR regenerated echo curve from DeerAnaly sis (red). B) L curve utilized to from TKR analysis. D) Frequency domain spectrum. A B C D
127 Figure 3 10 Differential scanning calorimetry for GM2AP constructs A60R1 and L126R1. A) D ifferential scanning calorimetry show the thermal of unfolding of A60R1 B) Diffe rential scanning calorimetry show the thermal of unfolding of A60R1 with GM2 bound. C) Diffe rential scanning calorimetry show the thermal of unfolding of L126R1. D) Diffe rential scanning calorimetry show the thermal of unfolding of AL126R1 with G m2 bound.
128 Figure 3 11 Overlay of area normalized EPR spectra (gray) and MOMD simulated spectra (black) of the ten spin labeled GM2AP constructs. All spectra have 100 G sweep widths and were collected under temperature control at 25C. Adequate agreement between the experimental and simulated line shapes were obtained for sites V54R1, L87R1, T90R1, and S115R1 (top) using only a single spectral component. However, the spectra of the remaining sites (bottom) required two components for adequate agr eement and the corresponding two spectral components are shown with relative percentages of mobile (gray) and immobile (black) components given.
129 Figure 3 12 GM2AP chromatograms from pro tein purification. A) FPLC chromatogram for GM2AP L126R1 elution profile from S 200 sephacryl column dotted red li nes represent fractions collected B) FPLC chromatogram for GM2AP L126R1 elution profile from tricorn superose column. C) HPLC chromatogram for GM2AP L126R1 elution profile from C18 column. D) 16.5% Tris HCL Biorad precast gel of broad range molecular weight marker in lane 1 and GM2AP L126R1.
130 Figure 3 13 Representative EPR spectra of GM2AP L126R1 and I66R1 as a function of temperature (C). Overlain experimental spectra (black) and MOMD simulations (red) are shown. Each simulation required two spe ctral components for an adequate fit to the experimental data. B. Plots of the relative percentages of mobile and immobile components obtained from the spectral fits as a function of temperature (K).
131 Figure 3 14 Linear regression of the ratio of the fractions of mobile and immobile L126R1 populations plotted as a function of inverse temperature. Two separate EPR data sets were collected from different protein sample preparations, and the simulations were performed independently for each set. The two data sets are indicated by triangles and circles. The solid line is the linear regression from both data sets taken together. From the slope of the line ( 7.8 0.8) and the intercept (25.8 2.8), the enthalpy and entropy of loop motion was found to be 65 kJ/mol and 215 J/mol K, respectively. The value of the equilibrium constant K at 298.15 K was calculated to be 0.7.
132 Figure 3 15 Representative diagram of the dansy l DHPE fluorescent lipid extraction assay. For the protein:lipid complex, the maximal emission fluorescence wavelength is near 484 nm, which is similar to dansyl DHPE dissolved in benzene. A red shift to 518 nm is obtained for extruded 100 nm large unilamellar vesicles of POPC:dansyl DHPE(4:1), and is similar the emission wavelength when d ansyl DHPE is dissolved in methanol 100 nmole s of GM2AP was prepared i n 50 mM sodium acetate buffer at pH 4.8 was injected into 4 00 nmole s d ansyl DHPE (Ran, Fanucci, 2008)
133 Figure 3 16 Tethering results from double mutant GM2AP constructs. A) Dansyl DHPE extraction results for tethering the loops in I66C /L126C. B) Dansyl DHPE extraction result s for tethering the loops in I 72 C/ S 1 30 C. Figure 3 17 Representative diagram of the gel filtration GM2 lipid extraction assay Gel filtration assay for GM2 extraction. The elution profile of the mixture of 7.5 nmol GM2AP with 200 nmol POPC: GM2:CHOL:BMP (50:10:20:20) vesicles in 50 mM NaOAc pH 4.8 buffer with total volume of 100 L The average fraction size was ~50 L The blue columns show the concentration of GM2 (determined from resorcinol assay) in each fraction. The black columns show t he GM2AP concentration (determined from UV VIS OD220) in each fraction. Fractions that contained vesicles were determined by light scattering at 550 nm.
134 CHAPTER 4 MASS SPECTROMETRIC ANALYSIS AND CONFIRMATION OF DISULFIDE BONDING AND REPORTER SITE LABELIN G EFFICIENCY IN GM2AP Introduction Much of the work described in this chapter was published in an Analytical Chemistry Field FTICR MS To Determine Disulfide Connectivity and 4 Maleimide TEMPO Spin Label Loc ation in Mathias, Mark R. Emmett, Gail E. Fanucci, and Alan G. Marshall (Tipton, Carter et al 2009) Jeremiah Tipton was a post do ctoral fellow at the National High Magnetic Field Laboratory (NHMFL), and Mark Emmett and Alan Marshall are at the Department of Chemistry and Biochemistry at Florida State University in Tallahassee. The work was supported by NIH (GM 78359), NIH (R01GM0772 32), NSF Division of Materials Research through DMR 0654118, and the State of Florida (Tipton, Carter et al. 2009) To validate the EPR results that conformational heterogen e ity observed in EPR spectra arose from m ultiple conformations of GMAP not mis labeling a mass spectrometry protocol was developed to ensure that spin label reactions only modified the mutant reporter cysteine (Cys ) Specifically, this assay will be used to demonstrate that the spin label has a ttached to only the engineered reporter Cys that was introduced into the primary sequence via mutagenesis. Electrospray Ionization combined with Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (ESI FT ICR MS) with a 14.5 T superconducting magne t was chosen to ensure proper labeling and disulfide bond formation in GM2AP.
135 The goal was to develop a novel mass spectro metry (MS) method to isolate Cys residues in order to determine whether the Cys moieties were in disulfide bonds or accessible to mod ification (i.e. biologically active). This aim arose as a product of many difficulties experienced in obtaining preliminary mass spectra due to poor ionization efficiency of particular peptides of interest. For a series of GM2AP single Cys mutants a ninth CYS introduced into various sites of the protein to serve as a Cys reporter site for modification. Through a collaboration with Alan Marshall at the National High Field Magnetic Laboratory, (Tallahassee, FL) we were able to determine that the ninth report er Cys residue was the only Cys residue modified by our labeling reagents and co nfirmed that the other eight Cys maintained the naturally occurring disulfide bonds. Methods Direct Infusion and Reversed Phase Capillary Liquid Chromatography Direct infusio n of the intact proteins was performed with an Advion BioSystems Nanomate (Ithaca, NY). The intact proteins were diluted to 1 5 pmol/ L in H2O/methanol/formic acid, 43/50/2 (v/v/v). Digests were separated by reversed phase nanoliquid chromatography (10 L sample) with a C18 capillary column (New Objective, Woburn, MA] 5 cm 0. 75 m, 15 m i.d spray tip). An Eksigent NanoLC (Dublin, CA) w as used to deliver a 35 min gradient (5 to 95% B) at 400 nL/min with solution A as 0.5% formic acid (v/v) in 5% aqueous methanol and solution B as 0.5% formic acid (v/v) in 95% aqueous methanol. Mass Spectrometry Mass spectrometry was performed with either a modified hybrid linear quadrupole ion trap FTICR mass spectrometer (LTQ FT, Thermo Fisher Corp., Bremen, Germany) equipped with an actively shielded 14.5 T superconducting magnet (Magnex, Oxford,
136 U.K.) or a custom built 9.4 T FTICR MS (Senko, Hendrickso n et al. 1996; Hakansson, Chalmers et al. 2003; Marshall, Guan, 1996). The protease treated samples were screened by online LC/MS with the 14.5 T LTQ FTICR instrument operated in top 5 data dependent mode (high resolution FTICR, low resolution LTQ CID ( collision induced dissociation) MS/MS. For each precursor ion measurement, 1 million charges were accumulated in the LTQ prior to transfer (1 ms transfer period) through three octopole ions guides (2.2 MHz, 250 Vp p) to a capacitively coupled open cylindr ical ICR cell for analysis (Beu, Laude et al. 1992). The robust automatic gain control for transferring the same number of ions for each FTICR MS scan yielded less than 0.500 ppm rms mass error with external calibration (Schwartz, Senko et al. 2002). Ea ch of the five most abundant ions was collisionally dissociated in the LTQ for low resolution MS/MS (3 microscans, 10,000 target ions, 2.0 Da isolation width, 35% relative collision energy, 0.250 activation q, 30 ms activation period, and dynamic exclusion list size of 60 for 1 min). Data were collected with Xcalibur 2.0 software (Thermo Fisher). Following the identification of disulfide containing peptides from the first pass low resolution LTQ MS/MS analysis (as described in the Data Analysis and Informat ics section), the samples were reanalyzed by nano LC, selected ion high resolution MS/MS to confirm charge state and fragment ion assignments. The second experiment is needed for verification of disulfide connectivity. The scan type for the high resolution MS/MS was set at 7 to 10 scans per segment. The first scan was the same as the precursor ion scan found in the top 5 data dependent experiment. The next 2 to 9 scans were set to select relevant ions (disulfide connected peptide fragments) for high resolut ion MS/MS. The 2 or 3 most abundant charge states
137 (for relevant disulfide connected proteolytic fragment targets) were selected for high resolution MS/MS analysis because different charge states are known to fragment differently with CID (Tangm Thibault et al. 1993; Griffiths, Jonsson, 2001; Tsapraili, Somogyi et al. 1999). A selected ion high resolution MS/MS scan consisted of a target ion accumulation target of 50,000 ions with a 5 Da mass selection window, CID fragmentation in the LTQ (35.0 scaled co llision energy, 0.250 activation q, and 30 ms activation period), and transfer to the cell for measurement. Low resolution MS/MS spectra exhibiting ambiguous fragment assignments were easily assigned by high resolution MS/MS based on charge state validatio n and mass errors less than 0.500 ppm. High resolution MS/MS data were collected with both Xcalibur 2.0 and MIDAS software. Electron capture dissociation (ECD) was performed with a custom built 9.4 T FTICR and is known to specifically break disulfide bond s more readily than the polypeptide backbone (Zubarec, Krunger et al. 1999; Zubarev, Krunger et al. 2004; Uggerud, 2004).. Thus, disulfide bonds not verified by high resolution CID MS/MS were analyzed by ECD FTICR MS. Ions were selected with a quadrupol e mass filter and accumulated in a second octopole for 1 s (Senko, Hendrickson et al. 1997). The ions were then transferred through multiple ion guides and trapped in an open ended cylindrical Penning trap by gated trapping for ECD analysis (Tsybin, Hendr ickson et al. 2006) Accurate mass and time tag knowledge of disulfide peptides from the top 5 data the Eksigent nano LC (Pasa Tolic, Masselon, 2004). Namely, at the elution time for the peptide of interest, the flow rate was reduced from 400 nL/min to 50 nL/min so that 25 ECD spectra could be signal averaged. Data acquisition was performed by use of a
138 Predator data station and data analysis by MIDAS 3.4 software (Blak ney, Robinson et al. 1990). Sequential Digestion with Trypsin, GluC, and AspN The solvent exchanged sample was diluted to 50 pmol/ L (i.e., 50 M), with a final volume of 160 L. To that solution, 100 ng of trypsin (final substrate:trypsin ratio 1000:1) was added and incubated overnight at 36 C. Next, a 100 L aliquot of the trypsin digest was incubated overnight with 200 ng of endo proteinase GluC (substrate:GluC, 400:1) at 36 C. Finally, 50 L of the trypsin/Glu C digested solution was incubated with 100 ng of endoproteinase AspN (substrate:AspN, 400:1). Each sample was diluted to 1 pmol/ L prior to LC MS analysis. Data Analysis an d Informatics Peaks from the Xcaliber files were extracted with a customized peak picking algorithm, thresholded at 10% of the maximum peak magnitude for low resolution MS/MS and 1% for high resolution MS. The resulting files were searched with MASCOT (Ma trix Science, Cambridge, UK) against a custom created database containing the wild type and L126C mutant sequences. To aid in identification of the MSL labeled Cys the mass of the MSL label (252.14739) was added to the MASCOT modification database. Result s from MASCOT were visualized and validated with Scaffold (Proteomics Software, Portland, OR) software. After identification of the MSL labeled peptide, the remaining disulfide peptides from each digest were identified by comparing high resolution MS prec ursor ion measurements to values calculated from the elemental composition of the assigned peptides. High resolution MS/MS and ECD fragment ion assignments were performed by hand.
139 Results and Discussion Mass spectrometry enables direct identification of po sttranslational modifications, including disulfide bonds ( Tipton, Carte r et al. 2009; Bean, Carr et al. 1992; Badock,Raida et al. 1998; Gorman,Wallis, 2002; Xu, Zhang et al. 2008; Yen, Joshi et al. 2002; Nair, Nilson et al. ,2006; Popolo, Ragni et al. 2 008 ). Because of the analytical challenge associated with verifying multiple disulfide bonds, a multitier assay was developed (Tipton, Carter 2009) First, the expressed and purified GM2AP constructs (Protein gel shown in Figure 4 1) were a nalyzed to deter mine the mass of the intact protein by direct infusion ESI 14.5 T LTQ FTICR MS ( Figure 4 2). A 240 Da mass sh ift was observed between the wild type and spin labeled constructs, correspo nding to the addition of one MSL label at the engineered reporter Cys (L126 MSL ). Also, a 480 D a shi f t was observed in the double mutant I66 MSL/L126 MSL Although, initial intact protein experiments observed that the I66C construct provided a poor label i ng reporter site which is most likely due to the Cys side chain residi ng in the hydrophobic cleft. The spectra also show several salt and oxidation adducts, both common for intact protein analysis. A second criterion for identification of the correct construct is presented in Figure 4 3, top. If there are less than four dis ulfide bonds, the 16+ charge state isotopic distribution, calculated from the mol ecular formula for the L126C MSL GM2AP construct, will decrease by 0.12598 Da (2)(1.007825)/16) for each addition of 2 hydrogens. Thus, a positive match between the calculated and experimental isotopic distributions is first checked before enzymatically digested samples are analyzed For this case, Figure 4 3 bottom illustrates an excellent agreement between the measured and experimental isotopic distributions for the 16+ char ge state isoto pic distribution of the L126C MSL GM2AP construct. Thus, m ass analysis of intact L126C MSL GM2AP
140 unambiguously confirms the presence of four dis ulfide bonds and a single MSL modification but not the disulfide bonding connectivity Following the identification of the intact protein a systematic approach was used to verify proper folding and labeling of a recombinantly overexpressed and MSL labeled L126C GM2AP construct. A sequential proteolytic digestion scheme was used, and include the enzym es, trypsin, endoproteinase GluC, and endoproteinase AspN. Next, a combination of standard proteomic techniques and multiple scan functions with high resolution MS and/or low/high resolution MS/MS were preformed on a modified 14.5 T LTQ FTICR MS. FTICR MS provides high mass accuracy and allows for quick peptide and fragment ion assignments with rms mas s errors of less than 0.5 ppm (Schaub, Hendrickson et al. 2008). Tryptic digestions were analyzed with LC high resolution MS and low resolution MS/MS. The lo w resolution MS/MS was chosen at first because of the slower duty cycle associated with high resolution MS/MS. Figure 4 4 (top) presents the sequence of L126C GM2AP, expected disulfide connectivity, location of the reporter Cys and sites of proteolytic cl eavage for different proteases. MASCOT/Scaffold analysis identifies the tryptic digest fragments shown in red in Figure 4 4, bottom but cannot identify the two overlapping disulfide containing peptides ( Figure 4 4, bottom). MASCOT is typically used to anal yze proteomic data sets in which the protein sample from cell lysate (including proteins containing disulfide bonds) have been reduced with dithiothreitol and alkylated with iodoacetamide. Thus, in the previous mentioned case peptides that contain Cys will have reacted with iodoacetimide making the disulfide connectivity pattern ambiguous. This experiment
141 was designed to retain the disulfide connectivity. T hus, t he current version of MASCOT is not useful to identify peptides that are linked by disulfide bon ds. However, based on the known primary sequence, accurate mass measurement of the tryptic digest peptide precursor ions, and sequence tags from subsequent MS/MS, manual analysis revealed all of peptides containing one or more disulfide linkages, thus prov iding 100% sequence coverage. In addition, from Figure 4 4 (top), the connectivity for tryptic peptide Cys 1 (linking C8 to C152), containing a single disulfide bond (see Figure 4 5) could be uniquely determined Thus, from the predicted sequence and the r esults of the tryptic digest, the expected tryptic disulfide peptide sequences and masses could be generated. To accurately verify the disulfide bound peptide fragments gen erated the need for the sequential digestion platform. Furthermore, high mass accura cy reduces the number of possible matches for a measured peptide to the possibilities found in a database (or protein) of known sequences. For example, if the tryptic peptide IESVLSSSGK, mass 1005.5342 Da, is searched against the possible peptide fragments from the sequence of GM2AP at 1, 20, and 2000 ppm mass error, the numbers of matched peptide sequences are 1, 4, and 7, respectively. Thus, high mass accuracy yields higher confidence in assignments by reducing the number of possible peptide matches. The peptide containing the MSL spin label was observed and confirmed with a mass error of 0.340 ppm, F igure 4 5 shows the observed b and y ion series for the tryptic peptide containing the MSL reporter Cys (in green). In contrast to other labile post translat ional modifications, the data illustrate s that the MSL thioe t h er bond does not break during CID fragmentation, because the b9 15 and y13 14 ion series retain
142 the mass associated with MSL addition (Cooper, Hakansson et al. 2005; Syka, Coon et al. 2004; Be ausoleil, Jedrychowski, 2004) Furthermore, if a population of the expressed protein were not properly folded, other non reporter Cys resid ues would be modified by the MSL label. This is not the case because other MSL modified non reporter Cys were not obs erved. Furthermore, verification of disulfide connectivity was preformed from sequential trypsin, Glu C, and Asp N enzymatic digestions. From the tryptic peptide assignments and the predicted subsequent Glu C and AspN proteolytic cleavage sites, the remai ning disulfide connected peptides were generated and denoted as Cys 2a, Cys 2b, and Cys 2c (see Figure 4 6). Charge state m / z values are also calculated for nano LC selected ion high resolution MS/MS. Figure 4 7 presents the fragment ions from the disulfi de containing peptides identified by high mass accuracy MS and CID MS/MS. From the trypsin digest, Cys 1 ( Figure 4 7, top; Table 4 1) resulted in b and y ions that retained the mass increase associated with the second, shorter LGCIK sequence. High resolut ion MS/MS provided unambiguous assignment of the fragment ions for Cys 1; i.e., at low resolution/mass accuracy in the LTQ many of the fragments had assignment ambiguities. From the trypsin/Glu C digestion, Cys 2a was verified ( Figure 4 7, bottom; Table 4 2) with a tripeptide sequence tag. The Cys precursor ion mass measured to within 0.340 ppm mass error (Table 4 3) includes the loss of 2 hydrogens for the disulfide bond formed between C68 and C75. As for Cys 2 and Cys 2b, high resolution CID MS/MS provid ed a sequence tag for identification but did not reveal the disulfide bond connectivity. Cys 2 and Cys expected to fragment poorly by low energy CID ( Bean Carr et al. 1992 ; Badcock,
143 Raida et al. 1998 ) Therefore, those peptides were fragmented by ECD with a custom built 9.4 T FTICR MS. Both peptides resulted in only a charge reduced species, not fragmentation. The trypsin/Glu C sample was therefore digested with Asp N to release Cys 2c ( Figure 4 6). Once again, high resolution MS/MS sequencing could not verify the final two disulfide bonds; however, peptide identity was confirmed. Figure 4 peak parking is initiated, i.e., flow rate slowed fro m 200 nL/min to 50 nL/min, the separation will be less than optimal for the remainder of the analysis. However, the ECD experiment was designed for analysis of only Cys 2c, thus the loss in separation efficiency for the remainder of the gradient was irrele vant. ECD is known to specificall y fragment disulfide bonds (Zubarec, Krunger et al. 1999; Zubarev, Krunger et al. 2004; Uggerud, 2004) In this case, ECD specifically broke the C94 C105 disulfide bond to generate Frag2 ( Figure 4 8) and EPC 94 PEPLR. ECD a nalysis thus provided the final piece of evidence to verify the last two disulfide bonds. Finally, Table 3 summarizes the tryptic peptides identified by MASCOT, along with the manually identified disulfide bonds. The data acquired from all three sequential digests by accurate mass measurement MS and MS/MS sequence tags resulted in quick identification of all four disulfide peptides as well as the sequence location of the MSL C ys The result is a high confidence MS assay that confirms the 4 native disulfide bonds are intact in GM2AP MSL labeled constructs and that only the reporter Cys is labeled with the paramagnetic spin label.
144 Table 4 1 Peak list and mass errors from high resolution CID MS/MS of Cys 1 Ion Charge Experimental (m/z) Experimental Mass (Da) Calculated Mass Error (ppm) b 8 b 11 b 11 H 2 O b 12 b 14 1 2 2 2 2 978.3773 920.8761 911.8707 985.3975 1077.9551 977.3700 1839.7377 1821.7269 1968.7805 2153.8957 977.3996 1839.7383 1821.7277 1968.7809 2153.8973 0.466 0.348 0.478 0.223 0.761 b 15 b 15 b 15 H 2 O 2 3 3 1135.4694 757.3150 751.3115 2268.9243 2268.9232 2250.9127 2268.9242 3368.9242 2250.9136 0.026 0.447 0.420 y 5 y 9 1 2 555.3612 492.7772 544.3539 983.5399 554.3535 983.5395 0.821 0.386 y 9 H 2 O y 12 2 2 483.7719 923.9610 965.5295 1845.9081 965.5395 1845.9082 0.212 0.347 y 12 y 13 y 14 y 15 y 17 y 18 3 2 2 2 2 3 616.3098 981.4744 1074.5148 1118.0305 1235.0806 852.7333 1845.9081 1960.9351 2147.0151 2234.0465 2468.1467 2555.1781 1845.9082 1960.9351 2147.0144 2234.0464 2468.1469 2555.1789 0.279 0.428 0.307 0.027 0.097 0.319 Y 19 3 896.4133 2686.2181 2686.2194 0.490 Table 4 2 Peak list and mass errors from high resolution CID MS/MS of Cys 2a Ion Charge Experimental (m/z) Experimental Mass (Da) Calculated Mass Error (ppm) M 2H 2 O b 10 b 10 H 2 O b 11 b 11 H 2 O 1 1 1 1 1 705.7891 1051.4226 1033.4120 1152.4700 1134.4594 1409.5637 10 50.4153 1032.4047 1151.4627 1133.4521 1409.5631 1050.4150 1032.4044 1151.4627 1133.4521 0.390 0.271 0.286 0.004 0.013 b 12 b 12 H 2 O 1 1 1299.5388 1281.5282 1298.5316 1280.5209 1298.5311 1280.5205 0.358 0.308
145 Table 4 3 Peak list and mass errors for relevant peptides in GM2AP L126C MSL Experimental (m/z) Experimental Mass (Da) Calculated Mass Error (ppm) Peptide 568.2611 670.3883 1601.8937 408.2472 458.2685 447.7318 855.7492 2269.0151 669.3810 3201.7728 814.4799 914.5224 893.4490 2564.2256 2269.0148 669.3810 3201.7747 814.4800 914.5225 893.4494 2564.2257 0.138 0.021 0.596 0.110 0.084 0.456 0.014 GHHHHHHHHHHSSGHIEGR DPAVIR SLTLEPDPIVVPGNVTLSVVGS TSVPLSSPK DLVLEK EVAGLWIK EGTYSLPK SEFVVPDC* ELPSWLTTGNYR 503.7744 706.82 67 845.8801 723.7993 911.4231 869.3866 1005.5342 2823.2777 5069.2368 2562.2109 3641.6632 2605.1378 1005.5342 2923.2777 5069.2374 2562.2100 3641.6635 2605.1378 0.015 0.000 0.120 0.340 0.126 0.002 IESVLSSSGK Cys 1 Cys 2 Cys 2a Cys 2b Cys 2
146 Figur e 4 1 16.5% Tris HCL Biorad pre cast gel of GM2AP constructs purified for MS analysis Lanes 1 8 and 11 12 show GM2AP constructs after purification by sephacryl S 200 column. Lanes 9 10 show pre sephacryl protein sample after concentration by an anion e xchange Q column.
147 Figure 4 2 Broadband 14.5 T electrospray positive i on of intact wild type, L126C MSL, I66C MSL, N136C MSL, I66C/L126C MSL respectively The recombinant CYS GM2AP constructs were labeled with 4 maleimide TEMPO. The shift in the m/z f or the observed charge states indicates labeling of the protein. Adapted from Tipton, Carte r et al. 2009
148 Figure 4 3 Parent ion FTICR data for GM2AP MSL. Top : Calculated isoptopic distributions corresponding to 0, 1, 2, 3, and 4 disulfide bonds for L126C MSL GM2AP [M+16H] 16+ The calculated m/z distribution decreases by (2)(1.007825)/16= 0.1260 for each additi onal disulfide linkage. Bottom : Calculated and experimental distribution confirms that the protein has the correct amino acid sequence, with four di sulfide bonds, and a singl e MSL label on a cysteine residue. The five highest magnitude in the isoptopic distribution were chosen for comparison to calculated masses (red dots). Adapted from Tipton, Carte r et al. 2009
149 Figure 4 4 Protease digestion scheme for GM2AP L126 MSL. Top : L126C GM2AP amino acid sequence w ith the location of reported disulfide bonds and predicted cleavages for trypsin (blue :K,R), Glu C(dark green :E), and AspN (orange :D). The location of the L126C amino acid substitution is shown in red. Bottom : Se quence coverage for the L126C MSL Gm2AP construct after trypsin digestion and nano LC analysis. The green sequence was identified by MASCOT ; the blue sequence was identified by hand and corresponds to the peptides still conatining disulfide bonds. Adapted from Tipton, Carte r et al. 2009
150 Figure 4 5 FTICR results for GM2AP L126 MSL peptide. Top left: identified peptide containing MSL label from L126C MSL trypti c digests for CID MS/MS. Bottom : Identified fragments by low resolution CID MS/MS fr agment s in green contain the MSL label. Adapted from Tipton, Carte r et al. 2009
151 Figure 4 6 Scheme presenting the theoretical disulfide bound peptides after sequential proteolytic digestion. Various m/z values were calculated for charge states to aid analysos and the selected ion monitoring MS/MS experiment. Adapted from Tipton, Carte r et al. 2009
152 Figure 4 7 CID MS/MS disulfide identification. Top : Disulfide bonds that were verified through CID MS/MS analysis. Top : high resolution MS/MS of Cys 1. Several large peptide fragment masses veri fy the peptide sequence. Bottom : high resolution MS/MS of Cys 2a. The calculated mass for the intact peptide mathes the experimental mass corresponding to one disulfide bond (Table 1). A sequence tag of three amino acids verifies the identity of the disulfide containing peptide. As expected, the cyclic peptide do es not yeild fragments. Adapted from Tipton, Carte r et al. 2009
153 Figure 4 8 ECD MS/MS disulfide identification. Peak parked nano LC high resolution ECD MS/MS (25 scans, 3 min). ECD typically breaks disulfide bonds prior to amino acid backbone fragmentation. Two fragments corresp onding to the specific fragmentation of the C 94 C 105 disulfide bond were observed. Adapted from Tipton, Carte r et al. 2009
154 CHAPTER 5 DOUBLE ELECTRON ELECTRON RESONANCE S TUDIES OF HIV 1 PROTEASE Introduction and Pr evious Work HIV 1 pro tease (HIV 1PR) is essential f or the maturation of the retrovirus that continues HIV replication, infection, and progression to AIDS. The aspartic protease homodimer cleaves polyproteins gag and gag pol into functional proteins. The active site pocket hous es catalytic D25 residues, and is physically mediated by the conformational changes of the hairpin flap region HIV 1PR is a target of antiretroviral drugs (ARVs) that slow the rate of viral mat uration (Ashorn, McQuade et al. 1990 ) The ribbon diagram in Figure 5 1 details critical regions of HIV 1PR (PDB ID 2PBX), including the a ctive site pocket, the flap and hinge regions, and the sites L63 and A71. The double electron electron resonance (DEER) technique in combination with site directed spin labeling (SDSL), is an e xcellent tool for examining changes in the conformational ens emble of the HIV 1 protease flap region ( Galiano, Bonora et al. 2007; Galiano, Ding et al. 2009, Kear, Blackburn et al. 2009; Torbeev, Raghuraman et al. 2009; Pannier, Velt et al. 2000 ). DEER measures the strength of the dipolar coupling between two e lectron spins (here, spin labels) spaced at distances between 20 60 where the coupling is proportional to r 3 with r representing the distance between the spins ( Jeschke, Chechik et al. 2006; Ding, Layen et al. 2008 ) Conformers design ated as wide open semi open closed and tucked/curled have been identified with SDSL DEER and X ray crystallography techniques and modeled using molecular dynamic simulatio ns ( Kear, Blackburn et al. 2009; Torbeev, Raghuraman e t al. 2009; Freedberg, Ishima et al. 2002; Hornak, Okur et al. 2006
155 Ishima, Freedberg, 1999; Todd, and Freire, 1999; Todd, Luque et al. 2000; Todd, Semo et al. 1998; Velazquez Campoy, Todd et al. 2000; Goodenow, Bloom et al. 2002 ) Initial work on DEER measurements of HIV 1PR was performed by former Fanucci group member Dr. Luis Galiano (Ph.D. 2008). At that time, we did not have experiments were performed in the lab of Peter Fajer at the National High Magnetic Field Lab (NHMFL) with the help of post doctoral research fellow Marco Bonora. This original data was collected at X band on a Brker EleXsys E580/E680 equipped with the ER 4118X MD5 Dielectric Ring success of the DEER methodology when applied to the study of flap conformations in Subtype B si HIV 1PR. These results showed clear differences in the most probable distance between the flaps and the flexibi lity of the flaps in the presence and absence of the protease inhibitor Ritonavir (Galiano, Bonora et al 2007) The dipolar modulated echo dat a for the apo construct (blue ) is noticeably different than that of the protease in the presence of protease inhibitor Ritonavi r ( red ) as shown in Figure 5 2A The resulting distance distribution profiles ( Figure 5 2B) report very different flap conformations a nd flexibility. T he average distance between the labeled sites on each flap shift ed by about 3 (from 36 to 33 ) fro m open to c losed upon binding Ritonavir T he breadth of the distance distribution profile decreased dramati cally upon addition of the inhibitor indicating that the flexibility of the flaps decreased as well, demonstrating a structural mechanism of inhibit or resistance.
156 Steve Kent and co workers obtained active and inactive (D25N) K55MTSL labeled HIV 1 protease constructs via total chemical synthesis and using DEER, reported interflap distances in the presence of three pepti domimetic inhibitors. Each of T he inhibitors, called MVT 101, KVS 1, and JG 365, capture different stages of the p eptide bond hydrolysis reaction, where MVT 101 locks the protease in an early transition like state KVS 1 mimics the tetrahedral intermediate in the reaction, and JG 365 mi mics a later transition like state (Baca and Kent 1993; Torbeev, Mandal et al. 2008) Th e respective distance distribution profiles sugges ted that the flaps adopted different catalytic properties throughout the cou rse of the catalytic reaction, with the flaps predominantly closed and with minimal flexibility during the e arly stages of the rea ction, with a more open conformation with increased flexibility aiding in product release as the reaction proceeds (Torbeev, Raghuraman et al. 2008) Dr. Luis Galiano and Dr. Ralph Weber, Senior Applications Scien tist for Brker's EPR division, later focused on understanding the role of secondary polymorphisms on acquired drug resistance by t he protease. V6 and MDR769 are patient isolates where V6 is from a pediatric patient treated with RTV and MDR769 was isolated from a patient previously treated with IDV, NFV, SQV, and APV. Findings showed t hat mutations that arise in response to PI treatment alter the flap conformations of the apoenzyme, thus affecting the conformational ensemble of the protease. In order to p rovide structural insight into the experimental data, MD simulations were performed in the lab of Carlos Simmerling, and the structural data obtained was in excellent agreement with the previous experimental data. The Subtype B flap region adapts an aver age conformation described as semi
157 open conformation by X ray crystallographic studies. MDR769 adopts a more open average conformation and V6 a more closed average conformation with respect to Subtype B. These results featured in an issue of Chemical and Engineering News (Drahl 2009) were important because they demonstrated t hat drug induced polymorphisms affect flap conformations and flexibility, a likely contributor to drug resistance. Dr. Mandy E. Blackburn compared the distance distribution profiles of HIV 1PR in the apo form with those in the presence of nine separate FDA appro ved protease in hibitors and a substrate mimic (Galiano, Blackburn et al. 2009) le ading to a publication found among the top ten most accessed Biochemistry rapid reports in July September 2009. HIV 1PR distance profiles were split into two groups by inhibitor and those that revealed a Inhibitors defined as having strong interactions must show at least 70% of the conformational ensemble in the closed flap conformation and include all but IDV, NFV, and ATV which y because the flaps are predominantly found in the semi open conformation. M oderate affects were seen for ATV, where approximately 40% of the conformational ensemble is in the closed conformation, while IDV and NFV had less than 20% of the conformational ensemble in the closed conformation (Galiano, Blackburn e t al. 2009) No strong correlation seems to exist between the relative percent closed conformation and the K i or K D values. T he DEER data does however, seem to be in agreement with the number of non water mediated hydrogen bonds between the inhibitor an d the prot ease
158 Dr. Jamie L. Kear focused on examining conformational sampling in the apoenzyme of various subtypes and patient isolates to determine how naturally evolved and drug induced polymorphisms alter the effect of inhibitors on the conformational ensemble and flap flexibility of non B subtypes and drug resistant patient isolates from protea se inhibitor exposed patients and found that the flap conformational ensembles differed greatly among the apo protease constructs. These results demonstrated th at natural and drug induced polymorphisms in the amino acid sequence of various subtypes and patient isolates, whether because of naturally occurring amino acid substitutions or drug pressure selected mutations alter the average flap conformations and flex ibility of the flaps. These resul ts were featured in an issue of AIDS Weekly (30 Nov. 2009) Here SDSL DEER has been used to further the examination of HIV 1PR flap conformational ensembles DEE R data are generally collected at cryogenic temperatures due to a phase memory time that is too short for detection of the spin echo at room temperature. The question is often raised on whether the conformations trapped at 65K accurately represent the ther mo dynamic conformations sampled. Distances between spin labeled sites in the flap region of an apo, and CA p2, Indinavir (IDV) and Tipranavir (TPV) bound HIV 1PR Subtype C construct were measured following four different freezing conditions : a slow 20 C freeze, an intermediate liquid nitrogen freezing freeze, and rapid freezes from both 25 C and 37 C using isopentane in c ombination with liquid nitrogen ( Figure 5 3 ) Additionally studies were done to determine the conformational ensembles for four protea se constructs, namely PRE and POST, and L63P and A71V secondary
159 mutants. PRE and POST were derived from a pediatric subject infected via maternal transmission with HIV ( Ho, Coman et al. 2008 ) before and after PI treatment; respectively. T he gag pol allel es were isolated from serial blood samples obtained over 7 years, starting before therapy initiation (PRE) and after the development of multiple drug resistance following 77 weeks of PI therapy with Ritona vi r (RTV) and an additional 16 weeks with Indinavir (IDV) (POST) (Barrie, Perez et al. 1996). PRE contains polymorphisms caused by the norma l process of genetic drift. Results indicate that PRE adopts a predominately closed flap conformation, but that the emergence of drug pressure selected mutations afte r (POST) causes the conformational sampling to favor a semi open population. Also investigated were the effects of two second ary mutations (away from the active site cavity, Figure 5 4), L63P and A71V on the conformational ensemble L63P is hypothesized to alter the conformational sampling resulting in a more closed state in the PRE construct. A71V a common secondary mutation seen in multi drug resistant constructs (Clemente, Hemraini et al. 2003), is thought to induce a more closed conformation. A mini mal amount of early work was done to incorporate and site directed spin lab el an un natural amino acid into HIV 1PR at position 55 of the flaps. This method differs from the traditional approach to site directed spin labeling, in which a thiol reactive spin label is attached to a native or non native CYS residue. Previous work from the lab of Wayne Hubbel reported an orthogonal labeling strategy that does not label any functional group found within the 20 naturally occurring amino acids (Fleissner, Brustad et al. 2009). In this method, Fleissner utilized the genetically encoded unnatural amino acid p acetyl L phenylalanine (p AcPhe) ( Figure 5 5) and
160 reacted it with a hydroxylamine reagent to yield a nitroxide side chain termed K1 ( Figure 5 6). They were ab le to demonstrate success using seven different mutants of T4 lysozyme, each containing a single p AcPhe at a solvent exposed helix site. It was shown that the EPR spectra of the K1 mutants had higher nitroxide mobiliti es than the spectra of mutants contai ning the R1 side chain Here we were successful in expressing HIV 1 protease K55 p acetyl L phenylalanine in large scale inclusion bodies, purifying, refolding (with mediocre success) in acidic solution, and spin labeling. A low quality DEER dataset was o btained, and the results will be discussed here. Materials and Methods Materials The chemicals, reagents, and supplies were obtained from Fisher Scientific (Pittsburg, Pennsylvania) and used as received, with a few noted exceptions. pET23 plasmid DNA wa s purchased from Novagen (Gibbstown, New Jersey). HiTrap Q HP Anion Exchange column, HiPrep 16/60 Sephacryl S 200 High Resolution Size Exclusion column was purchased from GE Biosciences (formerly Amersham, Pittsburg, Pennsylvania). 4 Maleimido 2,2,6,6 tet ramethyl 1 piperidinyloxy (4 Maleimido TEMPO, MSL) was purchased from Sigma Aldrich (St. Louis, MO). (1 Oxyl 2,2,5,5 tetr amethyl pyrroline 3 methyl) M ethanethiosulfonate spin label (MTSL) was purchased from Toronto Research Chemicals, Inc. (North York, Ontario, Canada). 0.60 I.D. x 0.84 O.D. capillary tubes were purchased from Fiber Optic Center (New Bedford, Massachusett s). BL21*(DE3) pLysS E. coli cells were purchased from Invitrogen (Carlsbad, California). NaOAc (C 2 D 3 O 2 ), D 2 O, d 8 glycerol were purchased from Cambridge Isotope Labs (Andover, MA). Ritonavir, Indinavir, Tipranavir, Darunavir,
161 Amprenavir, Atazanavir, Nelf inivir, Saquinavir, and Lopinavir were generously received from the AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH ( Bethesda, Maryland) (NIH) The non hydrolysable substrate mimic, CA p2 (H Arg Val Leu r Phe Glu Ala Nle NH2 (R V L r F E A Nle NH2, r = reduced) was synthesized and purchased from the University of Florida Protein Chemistry Core Facilit y (Gainesville, Florida). Methods Expression of HIV 1 protease pET 23a vectors with HIV 1PR genes were transformed separately into E. coli strain BL21 (DE3) pLysS via standard heat shock methodology. The transformed cells were inoculated in 5 mL sterile Luria Bertani (LB) media ( Table 3 9 ) and grown at 37 C with shaking at 250 rpm to an optical density (OD 600 ) of approximately 0.60 then transferred to 1 L sterile LB media and grown to an OD 600 of approximately 1.0 with shaking at approximately 200 RPM. Cells were then induced using 1 mM isopropyl beta D thiogalactopyranoside (IPTG ) The cultur e was then incubated with shaking at 250 rpm for 5 6 hours at 37C Cells were harvested by centrifugation for 15 minutes at 8500 g using a Sorvall RC6 floor mo del centrifuge with SLA 3000 rotor at 4 C and s upernatant was disc arded. Purification of HIV 1 protease Prior to protein purification, 250 mL of 9 M and 150 mL of 1 M urea buffers were freshly prepared in separate containers and 1 2 g AG 501 X8 (D) resi n ( 20 50 mesh ) was added to each of the urea buffers. These solutions were placed on a heated (30 C) stir plate and mixed by stir bar for approximately 2 hours to dissolve the urea and the resin was removed by filtration. All buffer exchange steps descr ibed in the
162 following sections were carried out using a 5 mL HiTrap Desalting column from GE Healthcare (packed with Sephadex G25), which is first washed successively with 3 4 column volumes (15 20 mL) of nanopure water (nH 2 O), 1 M NaCl, nH 2 O, 0.5 M Na OH, nH 2 O, and then equilibrated in the desired buffer. Pelleted cells from 1 L growth were resuspended in 30 mL resuspension buffer Mercaptoethanol ( B ME) was added to the reaction and mixed well by swirling. The approximate weight of t he wet pellet was generally 5 6 grams, but varied from purification to purification. In order to lyse the cells and release the cellular contents, t he sample was sonic ated for 2 minutes using a Fisher b rand tip sonicator at approximately 25 watts output p ower Sonication was always performed in cycles of 5 sec onds on followed by 5 sec onds off to avoid shearing the proteins present in the lysate. The sample was then passed 3 times through a 35 mL French pressure cell (Thermo Scientific, Waltham, Massach usetts) operating at approximately 1200 pounds per square inch ( psi ) Next, the lysed cells were centrifuged for 30 minutes at 18500 x g and 4 C using the Eppendorf 5810R centrifuge with F34 6 38 rotor t o collect cell debris and protease containing inclu sion bodies, and the supernatant was discarded. The inclusion body containing pelle t was resuspended, homogenized (using a 50 mL Dounce Tissue Homogenizer) and sonicated in 40 mL fresh wash buffer #1 then centrifuged for 30 minutes at 18500 x g and 4 C, and the supernatant was discarded. This process was repeating with 40 mL wash buffer #2 and 40 mL wash b uffer #3 Each of these steps functioned to isolate and wash the inclusion bodies, removing non target proteins and remaining cellular components. I n order to solubilize the inclusion bodies, t he pellet
163 was then resuspended, homog enized and sonicated in 30 mL inclusion body resuspension buffer containing 9 M urea, then centrifuged at 18500 x g and 4 C for 30 minutes. The supernatant, which contained solubilized target proteins in unfolded, monomeric form was collected. Because purification proceeds by anion exchange chromatography, the pH of the inclusion body resuspension buffer needs must be adjusted according to the specific isoelectric point (p I) of the protein being purified. A 5 mL HiTrap Q HP Anion Exchange column was equilibrated with inclusion body resuspension buffer on an Akta Prime liquid chromatography system, and the supernatant containing solubilized HIV 1PR monomers was ap plied to th e column at a rate of 5 mL/min. Fraction collection was started immediately and flow through was collected in 4 mL fractions, (specific fraction numbers depend on the results of the chromatogram ). Fractions containing HIV 1PR were collected and pooled int o a clean 50 mL Eppendorf tube (approximately 32 mL from 8 separate 4 mL fractions), then approximately 12 hours at 4 C to allow some contaminants to precipitate, after which time the protease sample was decanted into a 50 mL polypropylene Eppendorf centrifuge tube and centrifuged for 30 minutes at 12000 rpm and 6 C to separate precipitated contaminants. 300 mL 10 mM formic acid solution was prepared in a clean beaker and coo led on ice to approximately 0 C. HIV 1PR was refolded by doing a 10 fold stepwise dilution on ice using a peristaltic pump for approximately 2 hours, and the pH was subsequently adjusted to approximately 3.8 (typically by adding about 1 mL 2.5 M sodium acetate). The solution temperature was brought to approximately 3 0 C and the pH was adjusted to 5 (typically by adding about 3 mL 2.5 M sodium
164 acetate). After approximately 20 minutes of wait time the solution was moved to balanced centrifuge tubes and centrifuged for 20 minutes at 18500 x g and 23 C to remove contam inants that precipitated during refolding. The sample was concentrated to OD 280 = 0.5 using an Amicon 100 mL equipped with a Millipore 10,000 Da MW cut off polyethersulfone membrane. If further purification was required, the protease sample was buffer ex changed into 50 mM NaOAc, pH 5, and 5 mL of the concentrated protein sample was loaded on to a HiPrep 16/60 Sephacryl S 200 equilibrated with the same buffer and run at 0.5 mL/min and 2 mL fractions containing HIV 1PR were collected (fraction numbers vary according to chromatogram). The result is >95% pure HIV 1 Protease. N o protease inhibitors were added to the lysate at any time because the target protein is a protease. Incorporation of the unnatural amino acid The plasmid pSUPAR3 (Figure 5 7) was co tra nsformed into BL21(DE3) with the expression plasmid containing the HIV 1PR gene and subsequently plate d onto media containing ampicillin and chloramphenicol The transformed cells were inoculated in 5 mL sterile Luria Bertani (LB) media and grown at 37 C with shaking at 250 rpm to an optical density (OD 600 ) of approximately 0.60, then transferred to 1 L sterile LB media and grown to an OD 600 of approximately 1.0 with shaking at approximately 200 RPM. Cells were then induced using IPTG and arabinose. The cultur e was then incubated with shaking at 250 rpm for 5 6 hours at 37C Cells were harvested by centrifugation for 15 minutes at 8500 g using a Sorvall RC6 floor model centrifuge with SLA 3000 rotor at 4 C and s upernatant was discarded. Purification was carried out as described above. Protein containing p acetylphenylalanine was subsequently reacted with a ketone specific reagent (HO 4120) to produce ketoxime linked spin label side chain
165 DEER experiments P ulsed EPR data were collected on a Brker E leXsys E580 EPR spectrometer (Billerica, MA) equipped with the ER 4118X MD 5 dielectric ring resonator at a temperature of 65 K (cooled via liquid helium) and 4 pulse DEER sequence described in Chapter 2. DEER data collection was typically preceded by a series of preliminary experiments also described in Chapter 2, designed to accurately determine the center field, T m the d 0 (time in ns at which the echo begins) and pulse gate (the breadth of the echo in ns), and appropriate positions for the observer s equence and the pump pulse. below the central resonance DEER data analysis Data analysis was carried out as described previously by Blackburn et al. (Galiano, Blackburn et al. 2009) The raw experimental data was processed via Tikho nov Regularization using DeerAnalysis2008 software, available at http://www.epr.ethz.ch/software/index and then the distance profiles were reconstructed with a series of Gaussian functions via an in ho use Matlab based program called DeerSim. Population validation Population validation was performed as discussed previously by Blackburn et al. (Galiano, Blackburn et al. 2009) The validity of each population with a relative percentage of 10% or below was tested via population suppression (discussed in detail in Chapter 2) using Deer Sim and DeerAnalysis2008.
166 Results and Discussion E ffect of Sample Freezing Conditions on the Conformational E nsemble of HIV 1PR D ouble electron electron resonance (DEER) is a pulsed electron paramagnetic resonance (EPR) spectroscopy technique also referred to as pulsed ele ctron double resonance (PELDOR). Used in conjunction with site directed spin labeling (SDSL) DEER has been utilized to examine the conformational ensembles of the flaps in HIV 1PR under various conditions (Galiano, Bonora et al. 2007; Blackburn, Veloro et al. 2009; Galiano, Blackburn et al. 2009; Galiano, Ding et al. 2009; Kear, Blac d kburn et al. 2009) Distance measurements by SDSL DEER are based on the strength of the dipolar coupling of the unpa ired spins (Pannier, Veit et al. 2000; Jeschke, Chechik et al. 2006) and can yield distances of 20 60 between spin labels Results have provided detailed information regarding flap conformations sampled by the p rotease, with conformers designated as wide open, semi open, closed, and tucked or curled. These states were previously modeled using molecular dynamic simulations (Ding, Layten et al. 2008) DEER experiments ar e typically run at cryogenic temperatures; particularly, ours are collected at 65 K. Prior to inserting the sample into the temperature controlled cryostat, the sample must be frozen to a glass state. Most commonly, samples are frozen by submerging the sam ple tube in liquid nitrogen (LN 2 ) until a glass state is achieved. Work here was designed to determine what, if any, effects are created by the method of freezing and speed at which sample freezing takes place. Four different freezing techniques were anal yzed, all four using LN 2 as the cooling agent. A room temperature sample was frozen by direct immersion for approximately 30 seconds in
167 LN 2 A room temperature sample was first frozen at 20 C, followed by immersion in LN 2 A more rapid freezing method t han those previously described can be achieved by using isopentane cooled by immersion into liquid nitrogen as the medium for freezing. Isopentane is desirable for this purpose because it is a liquid at room temperature, it is less volatile with a higher b oiling point than LN 2 and it conducts heat very well. Accordingly, both a room temperature sample and a sample equilibrated to biological temperature (37 C) were immersed in isopentane suspended in LN 2 as the method of free zing the sample prior to data c ollection. For all samples, experimental dipolar modulated echo curves were analyzed via Tikhonov regularization (TKR) with DeerAnalysis2008, (Jeschke, Chechik et al. 2006) and populations assigned to flap conformations of curled/tucked, closed, semi open and wide open were obtained. In this case, we used a subtype C construct modified with two EPR active spin (Figure 5 1). The details of the subtype C construct are given in Figure 5 8 The conformational ensembles were determined for the apo protease and the protease bound to indina vir, tipranavir, and the non hydrolyzable substrate mimic called CA p2. The conformational ensemble for apo subtype C HIV 1PR is reported in the literature, with the average distance and relative percentage for each population as 29.7 and 13%, 33.3 and 8%, 36.7 and 52%, and 40. 2 and 2 7% and assigned to conformations where the flaps are tucked/curled closed, semi open, and wide open, respectively (Kear, Blackburn et al. 2009) ; to date, no conformational ensemble is reported for subtype C with CA p2 IDV, or TPV
168 The distance parameters for HIV 1PR subtype C (apo) under various freezing conditions are shown in Table 5 1 and Figures 5 9 5 1 3; subtype C (IDV) are shown in Table 5 2 and Figures 5 14 5 18; subtype C (TPV) are shown in Table 5 3 and Figures 5 19 5 23; subtype C (CaP2) are shown in Table 5 4 and Figures 5 24 5 28. None of the Subtype C constructs (apo, IDV, TPV, or CaP2) s howed drastic changes in the determined conformational ensemble when different freezing methods were employed. The most variation, though still not dramatic, was found to occur when the protease was bound to IDV. This result is rational because the bindi ng of IDV is known to be an entropy driven process. Likely, the fastest freezing method is the most a conformational ensemble that is an inaccurate description of the t rue conformational ensemble. Thi s work is signific ant because it address es a concern previously expressed about the DEER method. The pulsed EPR data are collected at cryogenic temperatures because at room temperature, the spin memory time is too short f or detection of the spin echo. Often, the question of whether the conformations trapped at 40 80 K accurately represent the the rmodynamic conformations sampled. Single Point Drug Induced Mutations Confer Alterations in the Conformational E nsemble of HIV 1 Protease Figure 5 29 shows the background subtracted dipolar modulated echo curves (Fig 5 29 A) and corresponding distance profiles ( Figure 5 29 B) for the four constructs. Data were analyzed via Tikhonov regularization with DeerAnalysis2008 (Jeschke, Timme l et al. 2006 ). The dotted lines in Figure 5 2 9 B indicate the distances consistent
169 with the closed (33 ) and semi open (36 ) populations (Blackburn, Veloro, 2009; Kear, Blackburn et al. 2009 ). It is c lear that for these data, PRE, L 63P and A71V have a most probable distance of about 33 which corresponds with the closed conformation, whereas POST has a m ost probable distance of 37 similar to Subtype B(si), which is consistent with a predominant semi open conformation al Table 5 5 summarizes the res ults of Figure 5 2 9 B. We have shown that a detailed analysis of the TKR distance profiles is possible via a Gaussian reconstruction analysis, leading to population ensembles for flap conformations ( Blackburn, Veloro, 2009; Kear, Blackburn et al. 2009 ). F igure 5 2 9 C shows the results of the population analysis of L63P HIV 1PR conformational sampling based upon a Gaussian reconstruction of the data in Fig 5 2 9 B. These analyses provide details of the relative populations in the flap co nformational ensem ble c omprised of states as signed to flap conformations of the tucked/curled, closed, semi open, and wide open states. Within the signal to noise of our experimental data, populations as low as 5% may be extracted with low error ( Blackburn, Veloro, 2009; Kear, B lackburn et al. 2009 ) Based upon ch aracterization of subtype B HIV 1PR ( Galiano, Ding, 2009; Blackburn, Veloro, 2009; Kear, Blackburn et al. 2009; Torbeev, Raghuraman, 2009;Ding, and Simmerling, 2008; Hornak, Rizzo et al. 2006; Scott, and Schiffer, 200 0; Heaslet, Rosenfeld et al. 2007; Foulk es Murzycki, Scott et al. 2007) the average distances for the preceding populations are 24 28, 33 36 and 42 respectively Shown in Figure 5 3 0 A are the relative populations of each distinct flap conformation that, in sum, make up the conformational ensembles of each of the HIV 1 PR apo constructs, which correspond to the structures shown in Figure 5 3 0 B. It is
170 noteworthy that although the distance between the MTSL spin labels in the tucked/curled conformation are shorter than those in the closed state, the experimentally obtained distances are consistent with structures seen in previous MD work, where the flaps are turned in towards the active site cavity and are described as a trigger for the wide open states and where an opening to the active site poc ket can be seen Individual complete datasets from each of the four samples are shown in Figures 5 31 5 34. Figure 3 35 summarizes and compares the relative populations (%) for closed, semi open, tucked/curled and wide open conformations of PRE apo (blue) and POST apo (red). Figures 3 36 3 39 show the mass spectra of spin labeled constructs showing major peaks with charge states indicated. Figure 3 40 summarizes the analysis of the mass spectrometry data i ncluding theoretical and observed molecular weights. The conformational ensemble of PRE, giving a predominantly closed conformation, is unexpected because, up to now, a ligand free construct with a most probable distance corresponding with the closed f lap conformation has not been reported. X ray crystal structures of ligand free HIV 1PR primarily exhibit the semi open conformation ( Ishima, Freedberg et al. 1999; Freedberg, Ishima et al. 2002; Todd, Semo et al. 2002;Todd, Freire et al. 1999; Todd, L uque et al. 2000; Hornak, Okur et al. ,2006) and previous DEER studies have shown subtypes B, C, CRF01_A/E, F and multi drug resistant constructs V6 and MDR769 exhibit pred ominantly semi open flap state (Kear, Blackburn et al. 2009 ). Inhibitor bound HIV 1PR has been shown to primarily adopt the closed conformation, caused at least in part by interaction between via a conserved H 2 O ( Galiano, Bonora et al.,
17 1 2007; Galiano, Ding et al. 2009; Blackburn, Veloro, 2009; Kear, Blackburn et al. 2009; Ding, Layten et al. 2008; Rose, Craik, 1998) Intraflap hydrogen bonding, which may stabilize the closed flap conformation, has been observed in MD simulations while also demonstrating that the average flap distance in the closed state is similar for ligand bound and unbound protease ( Ding, Layten et al. 2008 ), indicating that despite having no flap inhibitor interactions, ligand free protease may adopt a closed conformation. The conformational ensemble for Subtype B, reported in a previous work by Kear et al. consists of a most probable distance of 35.2 that corresponds with the semi open conformation. The addition of the single mutation A71V, and to a slightly lesser extent L63P, shifts the conformational ensemble to favor the closed conformation with a most probable distance to 33 This result is significant because it provides physical evidence that a single amino acid mutation is sufficient to cause a substantial shift in the flap conformations of HIV 1PR, and also because it suggests that it s not the lack of stabilizing mutations causing the staggering difference between the PRE and Subtype Bsi ( subscript s indicates the presence of three stabilizing mutations, namely Q7K, L33I, and L63I; subscript indicates that th e protease contains the inactivating D25N mutation) ensembles. The PRE construct differs from Subtype B LAI sequence by four polymorphisms, likely the result of natural genetic drift in the mother of the infected child from which the construct was derive d. The POST construct differs from the PRE construct through an additional 8 drug selected polymorphisms, including L10I, I15V, E34Q (negative to uncharged), M36I, T37N, I54A, R57E (positive to negative), and V82A. The combination of these mutations induce s a substantial shift of the conformational
172 ensemble of the protease, from a primary conformation of closed to semi open. M36I, located in the hydrophobic core, is thought to contribute to drug resistance and likely plays an important role in the conformat ional switch via the hydrophobic sliding mechanism ( Foulkes Murzycki, Scott et al. 2007 ). The DEER results rep orted in this work show that dru g induced mutations, even outsid e of the active site pocket, can significantly affect the conformational ensembl e of HIV 1PR, which likely play an important role in drug resistance and efficacy It is worthwhile to note here that the first generation of PI s used in treatment of HIV 1, including Saquinavir, Indivinavir, and Ritonavir, were designed with respect to su btype B but that subtype B virus constitutes only a small percentage of HIV worldwide. A second generation of PIs addressed the occurrence of resistant variants. Darunavir is the latest drug approved and is most effective against the highly drug resistant forms. The data reported here illustrates the importance of these later generation PIs, and also supports the need for ongoing drug development focused on efficacy against common polymorphisms and prominent subtypes. Electron Paramagnetic Reson ance Studi es of HIV 1 Protease V ia Incorporation of the Unnatural Amino Acid Acetyl Phenylalanine We have shown previously, via traditional SDSL EPR approaches, that the FDA approved HIV 1PR inhibitors induce changes in the conformational s ampling of the flap region s of the protease. However, traditional approaches require removal of the native CYS residues. A non native amino acid spin label, p acetyl L phenylalanine, was received from Mark Fleissner and Wayne Hubbel at UCLA, who demonstrated its utility using T4 l ysozyme (Jeschke, Polyhach et al. 2000) By incorporating the p acetyl L phenylalanine, we can compare the distance profiles obtained for protease with our
173 original DEER results (with MTSL label). As mentioned in the introduction section, HIV 1PR was s uc cessfully expressed with p acetyl L phenylalanine and subsequently spin labeled to yeast HIV 1PR K55K1. P reliminary DEER experiments (Figure 5 41) achieved low levels of SNR, and parameter s are currently require optimization.
174 Table 5 1. Distanc e paramete rs for HIV 1PR C (apo) and various freezing conditions. Freezing method Tucked/curled Closed Semi open Wide open Isopentane/liquid N 2 from 37 C R( ) 29.7 33.3 36.7 40.2 FWHM 3.1 2.8 4.0 3.3 % Conformation 6 12 55 27 Isopentane/liquid N 2 from 25 C R( ) 29.7 33.3 36.7 40.2 FWHM 3.1 2.8 4.0 3.3 % Conformation 6 13 54 27 Liquid N2 from 25 C R( ) 29.7 33.3 36.7 40.2 FWHM 3.1 2.8 4.0 3.3 % Conformation 11 8 54 27 20 C from 25 C R( ) 27.2 33.8 37.1 40.0 FWHM 4.4 5.0 4.9 3.7 % Conformation 11 8 54 27 Table 5 2. Distanc e parameters for HIV 1PR C (IDV) and various freezing conditions. Freezing method Tucked/curled Closed Semi open Wide open Isopentane/liquid N 2 from 37 C R( ) 32.5 37.9 40.0 FWHM 5.0 4.9 4.0 % Conformation 30 50 20 Isopentane/liquid N 2 from 25 C R( ) 26.2 33.1 36.3 39.3 FWHM 4.8 4.8 3.7 3.7 % Conformation 6 27 5 0 17 Liquid N2 from 25 C R( ) 33.2 36.9 40.0 FWHM 4.6 4.8 4.0 % Conformation 30 50 20 20 C from 25 C R( ) 27.2 33.8 37.1 40.0 FWHM 4.4 5.0 4.9 3.7 % Conformation 6 27 50 17
175 Table 5 3. Distanc e parameters for HIV 1PR (TPV) and various freezing conditions. Freezing method Tucked/curled Closed Semi open Wide open Isopentane/liquid N 2 from 37 C R( ) 33.0 37.5 41.0 FWHM 2.8 3.3 1.4 % Conformation 86 12 2 Isopentane/liquid N 2 from 25 C R( ) 33.0 36.8 41.5 FWHM 2.7 2.6 1.7 % Conformation 84 12 4 Liquid N2 from 25 C R( ) 32.7 34.9 39.0 FWHM 3.1 3.3 3.0 % Conformation 83 12 5 20 C from 25 C R( ) 32.9 37.1 40.0 FWHM 2.8 4.9 3.7 % Conformation 82 12 6 Table 5 4. Distance parameters for HIV 1PR ( CaP2 ) and various freezing conditions. Freezing method Tucked/curled Closed Semi open Wide open Isopentane/liquid N 2 from 37 C R( ) 32.3 34.1 FWHM 2.7 4.0 % Conformation 10 90 Isopentane/liquid N 2 from 25 C R( ) 31 .9 33.9 FWHM 1.9 3.1 % Conformation 12 88 Liquid N2 from 25 C R( ) 33.2 36.9 40.0 FWHM 4.0 4.8 4.0 % Conformation 30 50 20 20 C from 25 C R( ) 27.2 33.9 37.1 40.0 FWHM 4.3 5.0 4.9 3.7 % Conformation 6 27 50 17
176 Table 5 5. Summary of distance parameters obtained from DEER EPR of HIV 1PR. Construct Range (span) Most probable distance Average distance Subtype B* 24 45 (21) 35.2 35.2 PRE 25 45 (20) 33.2 32.8 POST 29 41 (12) 37.0 36.7 L63P 31 43 (12) 33.2 34.4 A71V 31 43 (12) 33.0 33.2 *Data from Kear, et al. Figure 5 1. H IV crystal structure. A) Front view ribbon diagram of HIV 1PR (PDB ID ter sites shown in tan, active site cavity and catalytic D25 residues in blue, the hinge regions in red, and L63 and A71 residues as green and pink, respectively. (B) Top View. (C) Side view.
177 Fi gure 5 2. DEER results of subtype B HIV 1Pr with (blue) a nd without (red) the FDA approved inhibitor Ritonavir. (A) Dipolar modulated echo data and (B) resulting distance distribution profile. Figure 5 3 DEER experiments are typically run at cryogenic temperatures, ours at 65 K. Prior to inserting the samp le into the temperature controlled cryostat, the sample must be frozen in a glass state. This work was designed to determine what, if any, effects are created by the method of freezing and speed at which sample freezing takes place. Four freezi ng technique s were considered: 1. placing the 25 C sample tube at 20 C freezer, 2. submerging the 25 C sample in liquid N 2 3. dipping the 25 C sample tube in isopentane suspended in liquid N 2 and 4. dipping the 37 C sample tube in isopentane suspended in liqu id N 2
178 Figure 5 4 Locations of polymorphisms in PRE and POST, with respect to Subtype B (LAI). Figure 5 5. Unnatural amino acid p acetylphenylalanine and structurally similar amino acid tyrosine. Figure 5 6. Chemical reaction for Incorporation o f unnatural amino acid to form spin labeled side chain K1
179 Figure 5 7. pSUPAR3 plasmid map used for co expression of p acetylphenylalanine, which is subsequently reacted with a ketone specific reagent (HO 4120) to produce ketoxime linked spin label si de chain Construc t Sequence Position 10 20 30 40 50 LAI PQITLWQRPL VTIKIGGQLK EALLDTGADD TVLEEMSLPG RWKPKMIGGI B -----K --------------N -----I ---------------PRE ---------------------N ---------T T ---------L63P -----K --------------N -----I ---------------A71V -----K --------------N -----I ---------------POST Csi --------I ---V -------N ------Q IN T -------------K -S -----I ---N -----I -IA -----------Position 60 70 80 90 99 LAI GGFIKVRQYD QILIEICGHK AIGTVLVGPT PVNIIGRNLL TQIGCTLNF B ---C -----I --A -----------------------A ---PRE ---C ----VP --A -----------------------A ---L63P ---C -----P --A -----------------------A ---A71V ---C -----I --A --V --------------------A ---POST Csi --A C E --VP --A -----------A ----------A ------C -----I --A K ----------------M -L A ---Figure 5 8. Sequence alignment for HIV 1 protease LAI with constructs B si C si PRE i I63P si A71V si and POST i to inactive protease that contains the D25N mutation.
180 Csi apo 20 C R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 29.7 3.1 11 Closed 33. 3 2.8 8 Semi open 3 6.7 4.0 54 Wide open 40 2 3. 3 27 Figure 5 9 HIV 1PR Subtype C apo frozen at 20. B ackground subtracted dipolar echo curve (black) over laid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (blac k) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the distance profile generated from TK R; Frequency domain spectrum; Individual populations necessary for Gaussian reconst ruction of the distance profile generated from TKR.
181 Figure 5 10 HIV 1PR Subtype C apo frozen in liquid nitrogen. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussia n populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectru m Csi apo N 2 R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 29.7 3.1 11 Closed 33. 3 2.8 8 Semi open 3 6.7 4.0 54 Wide open 40 2 3. 3 27
182 Figure 5 11 HIV 1PR Subtype C apo frozen in isopentane. B ackground subtracted dipolar echo c urve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from T KR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Csi apo IP R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 29.7 3.1 6 Closed 33. 3 2.8 13 Semi open 3 6.7 4.0 54 Wide open 40 2 3. 3 27
183 Figure 5 12 HIV 1PR Subtype C apo frozen in isopentane from 37C B ackground subtracted dipolar echo curve (black) overlaid with the T KR regenerated e cho curve (blue ) and reconstruct ed echo curve (gray ) corresponding to the sum of distance profiles ; L curve; Distance profile from TKR ana lysis (black) overlain with summed Gaussian population s (gray dashed); I ndividual populations for Gaus sian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum; Individual populations necessary for Gaussian reconstruction of the distance profile generated from TKR. Csi apo 37 C IP R() 0. 3 FWH M () 0. 5 % ensemble 5% Tucked 29.7 3.1 6 Closed 33. 3 2.8 12 Semi open 3 6.7 4.0 55 Wide open 40 2 3. 3 27
184 Figure 5 13 Stacked distance distribution profiles of H IV 1PR Subtype C, K55R1, apo after various freezing conditions, including 20 C, liquid nitrogen, isopentane/liquid nitrogen from room temperature, and isopentane/liquid nitrogen after heating to 37 C.
185 Figure 5 14 HIV 1PR Subtype C TPV bound frozen at 20 C. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gr ay dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Csi TP V 20 C R() 0. 3 FWHM () 0. 5 % ensemble 5% Closed 32.9 2.8 82 Semi open 3 6.8 2.8 1 2 Wide open 41.1 2.0 6
186 Figure 5 15 HIV 1PR Subtype C TPV bound frozen in liquid nitrogen. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from Deer Sim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gauss ian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Csi TPV N 2 R() 0. 3 FWHM () 0. 5 % ensemble 5% Closed 3 2.7 3. 1 8 3 Semi open 3 4 9 3.3 1 2 Wide open 39 .0 3 0 5
187 Figure 5 16 HIV 1PR Subtype C TP V bound frozen in isopentane. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve (blue ) and with the reconstruct ed echo curve (gray ) corresponding to the sum of distance profiles ; L curve; Distance profile from TK R analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum; Individual populations in Gaussian re construction of the TKR dis tance profile Csi TPV IP R() 0. 3 FWHM () 0. 5 % ensemble 5% Closed 3 3.0 2. 7 8 4 Semi open 3 6.8 2.6 1 2 Wide open 41. 5 1.7 4
188 Figure 5 17 HIV 1PR Subtype C TPV bound apo heated to 37 C then frozen in isopentane. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga uss ian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction ; Frequency domain spectrum Csi TPV 37 C IP R() 0. 3 FWHM () 0. 5 % ensemble 5% Closed 33.0 2.8 8 6 Semi open 3 7.5 3. 3 1 2 Wide open 41.0 1.4 2
189 Figure 5 18. Stacked distance distribution profiles of HIV 1PR Subtype C K55R1 bound to TPV, after various freezing conditions, including 20 C, liquid nitrogen, isopentane/liquid nitrogen from room temperature, and isopentane/liquid nitrogen after heating to 37 C.
190 Figure 5 19 HIV 1PR Subtype C IDV bound frozen at 20. B ackground subtracted dipolar echo curve (black) overlaid wi th the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) over lain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile; Frequency domain spectrum. Csi IDV 20 C R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 27.2 4.3 6 Closed 33. 9 5.0 27 Semi open 3 7.1 4.9 50 Wide open 40 0 3. 7 17
191 Figure 5 20 HIV 1PR Subtype C IDV bound frozen in liquid nitrogen. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individu al populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Csi IDV N 2 R() 0. 3 FWHM () 0. 5 % ensemble 5% Closed 33. 2 4.6 30 Semi open 3 6.9 4.8 50 Wide open 40 0 4.0 20
192 Figure 5 21 HIV 1PR Subtype C IDV bound frozen in isopentane. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (g ray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian re construction of the dist ance profile generated from TKR; Frequency domain spectrum Csi IDV IP R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 26.2 4.8 6 Closed 33. 1 4.8 27 Semi open 3 6.3 3.7 50 Wide o pen 3.93 3. 7 17
193 Figure 5 22 HIV 1PR Subtype C IDV b ound apo heated to 37 C then frozen in isopentane. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the dist ance pro file generated from TKR; Frequency domain spectrum Csi IDV 37 C IP R() 0. 3 FWHM () 0. 5 % ensemble 5% Closed 32 5 5.0 30 Semi open 3 7.9 4.9 50 Wide open 40 0 4.0 20
194 Figure 5 23. Stacked distance distribution profiles of HIV 1PR Subtype C, K55R1, bound with IDV after various freezing conditions, including 20 C, liquid nitrogen, isopentane/liquid nitrogen from room t emperature, and isopentane/liquid nitrogen after heating to 37 C.
195 Figure 5 24 HIV 1PR Subtype C CA p2 bound frozen at 20 C. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the G a ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Csi CA p2 20 C R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 31.8 1.9 13 Closed 33. 7 3.4 80 Semi open 3 8.1 3.4 7
196 Figure 5 25 HIV 1PR Subtype C CA p2 bound frozen in liquid nitrogen. B ackground subtracted dipolar echo curve (black) overlaid w ith the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) ove rlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Csi CA p2 N 2 R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 32.7 3.7 17 Closed 33. 8 3.9 83
197 Figure 5 26 HIV 1PR Subtype C CA p2 bound frozen in isopentane. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstru ct ed echo curve from DeerSim (gray ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual popula tions necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Csi CA p2 IP R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 31.9 1.9 12 Closed 33. 9 3.1 88
198 Figure 5 27 HIV 1PR Subty pe C CA p2 bound apo heated to 37 C then frozen in isopentane. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (blue ) and with the reconstruct ed echo curve from DeerSim (gray ) corresponding t o the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (gray dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Csi CA p2 37 C IP R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 32.3 2.7 10 Closed 3 4.1 4.0 90
199 Figure 5 28. Stacked distance distribution profiles of HIV 1PR Subtype C K55R1, bound with CA p2 after various freezing conditions, including 20 C, liquid nitrogen, isopentane/liquid nitro gen from room temperature, and isopentane/liquid nitrogen after heating to 37 C. Figure 5 29. DEER results for drug pressure mutations. (A) Background subtracted DEER echo curves for PRE (grey), L63P (red), A71V (blue), and POST (orange) constructs, (B) distance profiles from TKR analysis, and (C) Gaussian populations used to regenerate the TKR distance profile of L63P, showing contributions from individual conformations described as tucked/curled (black), closed (red), semi open (blue), and wide open (green).
200 Figure 5 30. Population analysis and molecular dynamics simulations for HIV 1 PR. (A) Conformationa l ensembles consisting of tucked/curled, closed, semi open and wide open conformations determined via Gaussian regeneration of the DEER distance profiles for each construct. Error was approximated at (5%). Populations are summed to define the conformationa l ensemble of each construct. (B) Structures of HIV 1 PR in the closed, semi open, tucked/curled and wide open flap states. In each structure, the MTSL spin label at position 55 is shown in blue.
201 Figure 5 31 HIV 1PR PRE apo. A B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho curve from DeerAnalysis (re d ) and with the reconstruct ed echo curve from DeerSim (blue ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (red da shed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum PRE apo R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 22.0 2.8 4 Tucked 25.1 3.4 18 Closed 33.2 3.3 56 Semi open 38.3 3.7 17 Wide open 45.1 3.2 5
202 Figure 5 32 HI V 1PR POST a po. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e c ho curve from DeerAnalysis (red ) and with the reconstruct ed echo curve from DeerSim (blue ) corresponding to t he sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (red dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Post apo R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 28.9 4.2 8 Closed 33.4 4.2 18 Semi open 36.9 5. 1 45 Wide open 40.6 5.5 29
203 Figure 5 33 HIV 1PR Subtype B L63P a p o. B ackground subtracted dipolar echo curve (black) overlaid with the TKR regenerated e c ho curve from DeerAnalysis (red ) and with the reconstruct ed echo curve from DeerSim (blue ) corresponding to the sum of distance profiles comprising the Ga ussian populat ions; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (red dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Subtype B L63P apo R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 30.5 3.7 10 Closed 33.2 3.5 57 Semi open 36.6 3.5 27 Wide open 42.8 3.8 6
204 Figure 5 34 HIV 1PR Subtype B A71V Apo. B ackground subtracted dipolar echo curve (black) overla id with the TKR regenerated e c ho curve from DeerAnalysis (red ) and with the reconstruct ed echo curve from DeerSim (blue ) corresponding to the sum of distance profiles comprising the Ga ussian populations; L curve; Distance profile from TKR analysis (black) overlain with the summed Gaussian population profile (red dashed); Individual populations necessary for Gaussian reconstruction of the dist ance profile generated from TKR; Frequency domain spectrum Subtype B A71V apo R() 0. 3 FWHM () 0. 5 % ensemble 5% Tucked 30.6 2.6 17 Closed 33.0 3.0 67 Semi open 36.0 2.8 7 Wide open 42.8 3.0 9
205 Figure 5 35 The relative populations (%) for closed semi open, tucked/curled and wide open co nformations of PRE apo (blue) and POST apo (red). Figure 5 36 Mass spectra of spin labeled PRE. The spectrum shows major peaks with charged states indicated.
206 Figure 5 37 Mass spectra of spin labeled POST. The spectrum shows major peaks with charged states indicated. Figure 5 38 Mass spectra of spin labeled L63P. The spectrum shows major peaks with charged states indicated.
207 Figure 5 39 Mass spectra of spin labeled A71V. The spectrum s hows major peak s with charged states indicated. HIV Protease Construct Theoretical MW(Da) (without MTSL) Theoretical MW(Da) (with MTSL) Observed MW(Da) (with MTSL) Pre 10690.6 10875.0 10874.9 0.2 Post 10 601.4 10785.8 10785.6 0 .2 Bsi L63P 10686.6 10871.0 10870.9 0.1 Bsi A71V 10730.8 10915.2 10915.0 0.2 Figure 5 40. Analysis of mass spectra for HIV 1 protease constructs.
208 Figure 5 41 Bsi Amber. Long pass filtered and background subtracted dipolar echo curve (black) overlaid with the TKR regenerated e cho c urve from DeerAnalysis (red); L curv e utilized to choose the appropriate regu larization Dist ance profile from TKR analysis ; Frequency domain spectrum; Protein gel after Q column isolation of Bsi Amber. Subtype B Amber apo R() 0. 3 FWHM () 0. 5 % ensemble 5% Tuc ked 24. 7 5. 2 2 3 Closed 33.4 4.0 29 Semi open 35.3 4.4 27 Wide open 52.5 4.6 21
209 CHAPTER 6 FUTURE DIRECTIONS The results from SDSL EPR, DSC, and fluorescence experiments of GM2AP CYS constr ucts indicate that the flexible loops do not undergo conformational exchange as a result of ligand binding. But, loop flexibility is an important attribute in the ability of GM2AP to extract its multiple ligands. T o further investigate how ligand binding alters the average conformation of GM2AP, uniformly 15 N labeled protein was generated for solution nuclear magnetic resonance ( NMR ) heteronuclear quantum coherence ( HSQC ) measurements were preformed by our post doctoral fellow Yong Ran Figure 6 1 shows H SQC spectra of GM2AP and with GM2 micelles (red) at pH 4.8 (blue) The bind ing of GM2 also does not induce a shift i n most of the resonances, only a few peaks are seen to shift or disappear. Indicating that GM2AP does not undergo a global conformational s hi f t as a result of ligand binding. NMR a ssignments for GM2AP are lacking, and future NMR studies are aimed at mapping out the specific regions in GM2AP that a re altered during interactions with ligands. In attempt to overcome the overlapping NMR resonance peaks from the fast conformational exchange sub s timescale, crystallizations trials (Figure 6 2) have begun in attempted to gain amino acid specific assignments from solid state NMR.
210 Figure 6 1 HSQC NMR spectra of uniformly labeled 15N GM2AP upo n addition of 4x molar excess GM2 micelles (red) at acidic pH 4.8 (blu e). Sample buffers were exchanged via gel filtration chromatography and not as titrations. The final sample contains 0.15 mM GM2AP and 20mM NaOAc. For the GM2 binding experiments, the sp ectra were collected before and after incubate with 0.6mM GM2 micelles at room temperature. 2D 1 H, 15 N HSQC spectra were collected on Bruker 600 Mhz spectrometer with 1mm cryoprobe at 20 C. The spectra were processed with NMRPipe and Sparky.
211 Figure 6 2 Crystallization trials of wild type GM2AP. A) WT GM2AP hanging drop crystal trial in Hepes buffered solution (0.1 M, pH 7.5) containing 30% (w/v) polyethyleneglycol (MM 4000), 10% (v/v) propanol B) WT GM2AP hanging drop crystal trial in Hepes buffered solution (0.1 M, pH 7.5) containing 25% (w/v) polyethyleneglycol (MM 4000), 10% (v/v) propanol. C) WT GM2AP hanging drop crystal trial in Hepes buffered solution (0.1 M, pH 7.5) containing 20% (w/v) polyethyleneglycol (MM 4000), 10% (v/v) propanol.
212 APPENDIX CW AND PULSED EPR PA RAMETERS Table A 1. Typical CW EPR parameters used for work reported within Parameter Value Number of points 1024 Center field ~3450 G Number of scans 5 25 Sweep width 20 100 G Acquisition time 40.63 sec Frequency ~9.6 GHz Power 20 dB 20 mW Receiver Gain 1x10 3 5x10 5 Modulation Amplitude ~1 G Time Constant 0.082 0.164 sec Receiver Phase 100 Table A 2. Typical pulsed EPR parameters used for DEER report ed within Parameter Value Shot repetition time 4000 Sweep width 160 200 G Number of scans variable Shots/point 100 Center Field ~3460 G Low Field ~3432 G Frequency ~9.6 GHz Pulsed Attenuation 0.1 Video Bandw idth 25 MHz 5 Modulation Amplitude ~1 G Time Constant 0.082 0.164 sec Receiver Phase 100 Table A 3. 4 pulse table used for DEER experiments performed in deuterated matrices. +x Pulse Acquisition Trigger Position 200 600 Position Pulse Length 16 32 Pulse Length 16 Pos. Display 8 Pos. Display
213 LIST OF REFERENCES Ahn, V. E., K. F. Faull, et al. (2003) Crystal structure of saposin B reveals a di meric shell for lipid binding. PNAS 1 00 : 38 43. Ahn, V E., P. Leyko, et al. (2006). Crystal structures of saposins A and C Protein Sci 15 : 1849 57. Altenbach, C., K. Cai, et al. (1999). Structural features and light dependent changes in the sequence 306 322 extending from helix VII to the p almitoylation sites in rhodopsin: a site directed spin labeling study Biochemistry 38 : 7931 7. Altenbach, C., S. L. Flitsch, et al. (1989). Structural studies on transmembrane proteins. 2. Spin labeling of bacteriorhodopsin mutants at unique cysteines. Bi ochemistry 28 : 7806 12. Altenbach, C., J. Klein Seetharaman, et al. (1999). Structural features and light dependent changes in the sequence 59 75 connecting helices I and II in rhodopsin: a site directed spin labeling study. Biochemistry 38 : 7945 9. Altenb ach, C., Oh, K. J., Trabanino, R. J., Hideg, K., and Hubbell, W. L. (2001) Estimation of inter residue distances in spin labeled proteins at physiological temperatures: experimental strat egies and practical limitations Biochemistry 40 : 15471 15482. Altenb ach, C., K. Yang, et al. (1996). Structural features and light dependent changes in the cytoplasmic interhelical E F loop region of rhodopsin: a site directed spin labeling study. Biochemistry 35 : 12470 8. Ashorn, P., T. J. McQuade, et al. (1990). An inhib itor of the protease blocks maturation of human and simian immunodeficiency viruses and spread of infection. PNAS 87 : 7472 6. Atkins, P. and J. d. Paula, Eds. (2005). Elements of physical chemistry. Physical Chemistry. New York, W.H. Freeman and Company. B aca, M. and S. B. Kent (1993). Catalytic contribution of flap substrate hydrogen bonds in HIV 1 protease explored by chemical synthesis. PNAS 90 : 11638 11642. Badock, V, M. Raida, et al. (1998). Distinction between the three disulfide isomers of guanylin 9 9 115 by low energy collision induced dissociation Rapid Commun. Mass Spectrom 12 : 1952 1956. Barnakov, A., C. Altenbach, et al. (2002). Site directed spin labeling of a bacterial chemoreceptor reveals a dynamic, loosely packed transmembrane domain. Prote in Sci 11 : 1472 81.
214 Barnes, J. P., Z. Liang, et al. (1999). A multifrequency electron spin resonance study of T4 lysozyme dynamics. Biophys J 76 : 3298 306. Bean, M. F., S. Carr, et al. (1992). Characterization of disulfide bond position in proteins and seq uence analysis of cystine bridged peptides by tandem mass spectrometry. Anal Biochem 201 : 216 226. Beasusoleil, S. A., M. Jedrychowski, et al. (2004). Large scale characterization of HeLa cell nuclear phosphoproteins. PNAS 101 : 12130 12135. Beckey, H. D. (1969). Field ionizaton mass spectrometry. Research/Development 20 : 26. Berliner, L. J., J. Grunwald, et al. (1982). A novel reversible thiol specific spin label: papain active site labeling and inhibition. Anal Biochem 119 : 450 5. Beu,S.C, D. Laude, et al (1992). Elimination of axial ejection during excitation with a capacitively coupled open trapped ion cell for Fourier transform ion cyclotron resonance mass spectrometry Anal Chem 64 : 177 180. Blackburn, M. E., A. M. Veloro, et al. (2009). Monitoring i nhibitor induced conformational population shifts in HIV 1 protease by pulsed EPR spectroscopy. Biochemistry 48 : 8765 8767. Blakney, G. T., D. Robinson, et al. (2005). 53rd Amer. Soc. Mass Spectrom. Ann. Conf. on Mass Spectrom and Allied Topics San Antoni o, TX Bordignon, E., Klare, J. P., Doebber, M., Wegener, A. A., Martell, S., Engelhard, M., and Steinhoff, H. J. (2005) Structural analysis of a HAMP domain: the linker region of the phototransducer in co mplex with sensory rhodopsin II J Biol Chem 280 : 3 8767 38775. Brustad, E. M., E. A. Lemke, et al. (2008). A general and efficient method for the site specific dual labeling of proteins for single molecule fluorescence resonance energy transfer. J Am Chem Soc 130 : 17664 5. Buchbinder, S. P., M. H. Katz, et al. (1994). Long term HIV 1 infection without imnunological progression. AIDS 8 : 1123 1128. Budil, D. E., Lee, S., Saxena, S., Freed, J. H. (1996) Nonlinear Least Squares Analysis of Slow Motion EPR Spectra in One and Two Dimensions Using a Modified Levenberg Marquardt Algorithm J Magn Reson 120 : 155 189. Carr, J. K., B. T. Foley, et al. (1998). Reference sequences representing the principal genetic diversity of HIV 1 in the pandemic. HIV Sequence Compendium : 10 19.
215 Chiang Y. W., P. P. Borbat, et al. (2005). Maximum entropy: A complement to Tikhonov regularization for determination of pair distance distributions by pulsed ESR J Magn Reson 177 : 184 196. Chiang, Y. W. B., Peter P.; Freed, Jack H. (2005). Maximum entropy: A complement to Tikhonov regularization for determination of pair distance distributions by pulsed ESR. J Magn Reson 177 : 184 196. Chin, J. W., T. A. Cropp, et al. (2003). Progress toward an expanded eukaryotic genetic code. Chem Biol 10 : 511 9. Ciaffoni, F. R. Salvioli, et al. (2001). Saposin D solubilizes anionic phospholipid containing membranes. J Biol Chem 276 : 31583 9. Ciaffoni, F., M. Tatti, et al. (2006). Saposin B binds and transfers phospholipids. J Lipid Res 47 : 1045 53. Clemente, J., R. Hemrajani et al., (2003). Secondary mutations M36I and A71V in the human immunodeficiency virus type 1 protease can provide an advantage for the emergence of the primary mutation D30N. Biochemistry 42 : 15029 35. Clore, G. M. (2008) Visualizing lowly populated regio ns of the free energy landscape of macromolecular complexes by para magnetic relaxation enhancement Mol Biosyst 4 : 1058 1069. Columbus, L. and W. L. Hubbell (2002). A new spin on protein dynamics. Trends Biochem Sci 27 : 288 295. Columbus, L. and W. L. Hubb ell (2004). Mapping backbone dynamics in solution with site directed spin labeling: GCN4 58 bZip free and bound to DNA. Biochemistry 43 : 7273 87. Columbus, L., T. Kalai, et al. (2001). Molecular motion of spin labeled side chains in alpha helices: analysis by variation of side chain structure. Biochemistry 40 : 3828 3846.. Conzelmann, E., Burg, J., Stephan, G., and Sandhoff, K. (1982) Complexing of glycolipids and their transfer between membranes by the activator protein for degradation of lysosomal ganglios ide GM2 Eur J Biochem 123 : 455 464. Conzelmann, E. and K. Sandhoff (1979). Purification and characterization of an activator protein for the degradation of glycolipids GM2 and GA2 by hexosaminidase A. Hoppe Seylers Z Physiol Chem 360 : 1837 49. Cooper, H. J., C. Hakansson, et al. (2005). The role of electron capture dissociation in biomolecular analysis. Mass Spectrom Rev 2 4 : 201 222.
216 Darmoise, A., P. Maschmeyer, et al. (2010). The immunological functions of saposins. Adv Immunol 105 : 25 62. Ding, F., M. L ayten, et al. (2008). Solution structure of HIV 1 protease flaps probed by comparison of molecular dynamics simulation ensembles and EPR experiments. J Am Chem Soc 130 : 7184 5. Douek, D. C., M. Roederer, et al. (2009). Emerging concepts in the immunopathog enesis of AIDS. Annu Rev Med 60 : 471 484. Dougherty, R. C. (1981). Negative chemical ionization mass spectrometry: applications in environmental analytical chemistry. Biomed Mass Spectrom 8 : 283 92. Drahl, C. e. (2009). Protease Flaps Give HIV Drugs the Slip. Chemical and Engineering News 87: 1. Eaton, G. R., S. S. Eaton, et al. (1998). Foundations of Modern EPR River Edge, World Scientific Publishing Co. Pte. Ltd. Fanucci, G. E., Cadieux, N., Piedmont, C. A., Kadner, R. J., and Cafiso, D. S. (2002) Str ucture and dynamics of the beta barrel of the membrane transporter BtuB by site directed spin labeling Biochemistry 41 : 11543 11551. Fanucci, G. E. and D. S. Cafiso (2006). Recent advances and applications of site directed spin labeling. Curr Opin Struct Biol 16 : 644 653. Farahbakhsh, Z. T., C. Altenbach, et al. (1992). Spin labeled cysteines as sensors for protein lipid interaction and conformation in rhodopsin. Photochem Photobiol 56 : 1019 33. Ferguson, M. A. and A. F. Williams (1988). Cell surface ancho ring of proteins via glycosyl phosphatidylinositol structures. Annu Rev Biochem 57 : 285 320. Fleissner, M. R., E. M. Brustad, et al. (2009). Site directed spin labeling of a genetically encoded unnatural amino acid. PNAS 106 : 21637 42. Foulkes Murzycki, J. E. W. Scott, et al. (2007) Hydrophobic sliding: a possible mechanism for drug resistance in human immunodeficiency virus type 1 protease. Structure 15 : 225 33. Freedberg, D. I., R. Ishima, et al. (2002). Rapid structural fluctuations of the free HIV prot ease flaps in solution: relationship to crystal structures and comparison with predictions of dynamics calculations. Protein Sci 11 : 221 232. Furst, W., W. Machleidt, et al. (1988). The precursor of sulfatide activator protein is processed to three differe nt proteins. Biol Chem Hoppe Seyler 369 : 317 28.
217 Furst, W. and K. Sandhoff (1992). Activator proteins and topology of lysosomal sphingolipid catabolism. Biochim Biophys Acta 1126 : 1 16. Galiano, L. (2008). Site directed spin labeling EPR studies of conform ational dynamics in the flap region of HIV 1 protease Department of Chemistry Gainesville, University of Florida. Ph.D .: 229 Galiano, L., M. E. Blackburn, et al. (2009). Solute effects on spin labels at an aqueous exposed site in the flap region of HIV 1 protease. J Phys Chem B 113 : 1673 80. Galiano, L., M. Bonora, et al. (2007). Inter flap distances in HIV 1 Protease determined by pulsed EPR measurements. J Am Chem Soc 129 : 11004 5. Galiano, L., F. Ding, et al. (2009). Drug pressure selected mutations in HIV 1 protease alter flap conformations. J Am Chem Soc 131 : 430 1. Giehl, A., Lemm, T., Bartelsen, O., Sandhoff, K., and Blume, A. (1999) Interaction of the GM2 activator protein with phospholipid ganglioside bilayer membranes and with monolayers at the a ir water interface Eur J Biochem 261 : 650 658. Gilbert, P. B. (2003). Comparison of HIV 1 and HIV 2 infectivity from a prospective cohort study in Senegal. Statistics in Medicine 22 : 573 593. Gorman, J. J., T. Wallis, et al. (2002). Protein disulfide bond det ermination by mass spectrometry. J. Mass Spectrom. Rev 2 1 : 183 216. Gravel, R., Clarke, J., Kaback, M., Mahuran, D., Sandhoff, K., and Suzuki, K (1995) The GM2 Gangliosidoses 7 ed ., McGraw Hill Inc., New York. Griffith, H. and H. M. McConnell (1965). A nitroxide maleimide spin label. PNAS 55 : 8 11. Griffiths, G., B. Hoflack, et al. (1988). The mannose 6 phosphate receptor and the biogenesis of lysosomes. Cell 52 : 329 41. Griffths, W.J., P. Jonsson, et al. (2001). Gas phase conformation can have an inf luence on peptide fragmentation. Proteomics 1 : 934 945. Goodenow, M.M., G. Bloom, et al. (2002). Naturally occurring amino acid polymorphisms in human immunodeficiency virus type 1 (HIV 1) Gag NC p7 and the C cleavage site impact GagPol processing by HIV 1 protease. Virology 292 : 137 149. Gross, A., Columbus, L., Hideg, K., Altenbach, C., and Hubbell, W. L. (1999) Structure of the KcsA potassium channel from Streptomyces lividans: a site directed spin labeling study of t he second transmembrane segment Bio chemistry 38 : 10324 10335.
218 Guo, Z., D. Cascio, et al. (2008). Structural determinants of nitroxide motion in spin labeled proteins: solvent exposed sites in helix B of T4 lysozyme. Protein Sci 17 : 228 39. Hakansson, K., M. Chalmers, et al. (2003). Combined electron capture and infrared multiphoton dissociation for multistage MS/MS in a Fourier transform ion cyclotron resonance mass spectrometer. Anal. Chem 75 : 3256 3262. Hansen, P. C. (1998). Rank Deficient and Discrete Ill Posed Problems: Numerical Aspects of Linear Inversion. Philadelphia, SIAM. Heaslet, H., R. Ros enfeld, et al. (2007). Conformational flexibility in the flap doma ins of ligand free HIV protease Acta Crystallogr D Biol Crystallogr 63 : 866 75. Hendrickson, C.L., J. Drader, et al. (1996). Fourier transform ion cyclotron resonance mass spectrometry in a 20 T resistive magnet S Rapid Commun. Mass Spectrom. 10 : 1829 1832. Ho, M. W. and J. S. O'Brien (1971). Gaucher's disease: deficiency of 'acid' glucosidase and reconstitution of enzyme ac tivity in vitro. PNAS 68 : 2810 3. Ho, S., R. Coman, et al. (2008). Drug associated changes in amino acid residues in Gag p2, p7(NC), and p6(Gag)/p6(Pol) in human immunodeficiency virus type 1 (HIV 1) display a dominant effect on replicative fitness and dru g response. Viro logy 378 : 272 281. Hoffman, E. d. and V. Stroobant (2007). Mass spectrometry: principles and applications. West Sussex, Jon Wiley and Sons Ltd. Hornak, V., A. Okur, et al. (2006). HIV 1 protease flaps spontaneously close to the correct str ucture in simulations following manual placement of an inhibitor into the open state. J Am Chem Soc 128 : 2812 2813. Hornak, V., A. Okur, et al. (2006). HIV 1 protease flaps spontaneously open and reclose in molecular dynamics simulations. PNAS 103 : 915 20. Hubbell, W. L., D. S. Cafiso, et al. (2000). Identifying conformational changes with site directed spin labeling. Nat Struct Biol 7 : 735 739. Hubbell, W. L., A. Gross, et al. (1998). Recent advances in site directed spin labeling of proteins Curr Opin St ruct Biol 8 : 649 656. Hubbell, W. L., H. S. McHaourab, et al. (1996). Watching proteins move using site directed spin labeling. Structure 4 : 779 83. Hubbell, W. L. C., D. S.; Altenbach, C. (2000). Identifying conformational changes wi th site directed spin labeling. Nat Struct Biol 7 : 735 739.
219 Ishima, R., D. I. Freedberg, et al. (1999). Flap opening and dimer interface flexibility in the free and inhibitor bound HIV protease, and their implications for function. Structure 7 : 1047 1055. Ishima, R., R. Ghirlan do, et al. (2001). Folded monomer of HIV 1 protease. J Biol Chem 276 : 49110 49116. Ishima, R., D. A. Torchia, et al. (2003). Solution structure of the mature HIV 1 protease monomer: insight into the tertiary fold and stability of a precursor. J Biol Chem 2 78 : 43311 43319. Jeschke, G. (2002). Distance Measurements in the Nanometer Range by Pulse EPR. Chem Phys Chem 3 : 927 932. Jeschke, G., V. Chechik, et al. (2006). DeerAnalysis2006 A Comprehensive Software Package for Analyzing Pulsed ELDOR Data. Appl Mag Reson 30 : 473 498. Jeschke, G. and Y. Polyhach (2007). Distance measurements on spin labelled biomacromolecules by pulsed electron paramagnetic resonance. Physical Chemistry Chemical Physics 9 (16): 1895 1910. Jeschke, G. K., A.; Jonas, U.; Godt, A. (2002) Direct Conversion of EPR Dipolar Time Evolution Data to Distance Distributions. J Magn Reson 155 : 72 82. Jeschke, G. P., H.; Zimmerman, H.; Godt, A. (2004). Sensitivity Enhancement in Pulse EPR Distance Measurements. J Magn Reson 169 : 1 12. Kantor, R., R W. Shafer, et al. (2005). The HIV 1 Non subtype B Workgroup: an international collaboration for the collection and analysis of HIV 1 non subtype B data. MedGenMed 7 : 71. Karas, M., D. Bachman, et al. (1987). Matrix Assisted Ultraviolet Laser Desorption o f Non Volatile Compounds. Int J Mass Spectrom Ion Proc 78 : 53 68. Karas, M., D. Bachmann, et al. (1985). Influence of the Wavelength in High Irradiance Ultraviolet Laser Desorption Mass Spectrometry of Organic Molecules Anal Chem 57 : 2935 2939. Kear, J. L ., M. E. Blackburn, et al. (2009). Subtype polymorphisms among HIV 1 protease variants confer altered flap conformations and flexibility. J Am Chem Soc 131 : 14650 14651. Kolter, T., and Sandhoff, K. (2005) Principles of lysosomal membrane digestion: stimul ation of sphingolipid degradation by sphingolipid activator protei ns and anionic lysosomal lipids Annu Rev Cell Dev Biol 21 : 81 103. Lakowicz, J. R. (2006). Principles of Fluorescence Spectroscopy New York, Springer Science and Business Media.
220 Langen, R. K. J. Oh, et al. (2000). Crystal structures of spin labeled T4 lysozyme mutants: implications for the interpretation of EPR spectra in terms of structure. Biochemistry 39 : 8396 8405. Lawn, S. D. (2004). AIDS in Africa: the impact of coinfectios on the pa thogenesis of HIV 1 infection. J Infect Dis 48 : 1 12. Lemieux, M. J., B. L. Mark, et al. (2006). Crystallographic structure of human beta hexosaminidase A: interpretation of Tay Sachs mutations and loss of GM2 ganglioside hydrolysis. J Mol Biol 359 : 913 29 Li, S. C., H. Kihara, et al. (1985). Activator protein required for the enzymatic hydrolysis of cerebroside sulfate. Deficiency in urine of patients affected with cerebroside sulfatase activator deficiency and identity with activators for the enzymatic h ydrolysis of GM1 ganglioside and globotriaosylceramide. J Biol Chem 260 : 1867 71. Liu, W., A. Brock, et al. (2007). Genetic incorporation of unnatural amino acids into proteins in mammalian cells. Nat Methods 4 : 239 44. Locatelli Hoops, S., N. Remmel, et a l. (2006). Saposin A mobilizes lipids from low cholesterol and high bis(monoacylglycerol)phosphate containing membranes: patient variant Saposin A lacks lipid extraction capacity. J Biol Chem 281 : 32451 60. Louis, J. M., R. Ishima, et al. (2003). Revisitin g monomeric HIV 1 protease. Characterization and redesign for improved properties. J Biol Chem 278 : 6085 6092. Low, M. G. and A. R. Saltiel (1988). Structural and functional roles of glycosyl phosphatidylinositol in membranes. Science 239 : 268 75. Lundblad R. L. (2004). Chemical Reagents for Protein Modification 3rd ed.; CRC Press. Major, A. S., Joyce, S., and Van Kaer, L. (2006) Lipid metabolism, atherogenesis and CD1 restricted antigen presentation Trends Mol Med 12 : 270 278. Mathias, J. D., Ran, Y., Ca rter, J. D., and Fanucci, G. E. (2009) Interactions of the GM2 Activator Protein with Phosphatidylcholine Bilayers: A Site Directed Spin Labeling Power Saturation Study Biophysical Journal 97 : 1436 1444 Matsuda, J., M. Kido, et al. (2004). Mutation in sa posin D domain of sphingolipid activator protein gene causes urinary system defects and cerebellar Purkinje cell degeneration with accumulation of hydroxy fatty acid containing ceramide in mouse. Hum Mol Genet 13 : 2709 23.
221 Matsuda, J., M. T. Vanier, et al. (2001). A mutation in the saposin A domain of the sphingolipid activator protein (prosaposin) gene results in a late onset, chronic form of globoid cell leukodystrophy in the mouse Hum Mol Genet 10 : 1191 9. Mahuran, D. J. (1998) The GM2 activator protein its roles as a co factor in GM2 hydrolysis and as a gener al glycolipid transport protein. Biochim Biophys Acta 1393 : 1 18. McHaourab, H. S., T. Kalai, et al. (1999). Motion of spin labeled side chains in T4 lysozyme: effect of side chain structure Bioch emistry 38 : 2947 55. McHaourab, H. S., M. A. Lietzow, et al. (1996). Motion of spin labeled side chains in T4 lysozyme. Correlation with protein structure and dynamics. Biochemistry 35 : 7692 7704. McHaourab, H. S., Lietzow, M. A., Hideg, K. & Hubbell, W. L. (1996). Motion of spin labeled side chains in T4 lysozyme. Correlation with protein structure and dynamics. Biochemistry 35 : 7692 7704. Mehl, E. and H. Jatzkewitz (1964). [A cerebrosidesulfatase from swine kidney]. Hoppe Seylers Z Physiol Chem 339 : 260 76. Miller, M., J. Schneider, et al. (1989). Structure of complex of synthetic HIV 1 protease with a substrate based inhibitor at 2.3 A resolution. Science 246 : 1149 1152. Mollaaghababa, R., H. J. Steinhoff, et al. (2000). Time resolved site direct ed spin labeling studies of bacteriorhodopsin: loop specific conformational changes in M. Biochemistry 39 : 1120 7. Morimoto, S., B. M. Martin, et al. (1989). Saposin A: second cerebrosidase activator protein. P NAS 86 : 3389 93. Nair, S.S., C. Nilsson, et al. (2006). De novo sequencing and disulfide mapping of a bromotryptophan containing conotoxin by Fourier transform ion cyclotron resonance mass spectrometry. Anal Chem 78 : 8082 8088. Noble, R. (2009). HIV types, subtypes, groups, and strains. 201 0, from http://www.avert.org/hiv types.htm O'Brien, J. S. and Y. Kishimoto (1991). Saposin proteins: structure, function, and role in human lysosomal storage disorders. FASEB J 5 : 301 8. O'Brien, J. S. K. A. Kretz, et al. (1988). Coding of two sphingolipid activator proteins (SAP 1 and SAP 2) by same genetic locus. Science 241 : 1098 101. Osmanov, S., C. Pattou, et al. (2002). Estimated global distribution and regional spread of HiV 1 genetic subt ypes in the year 2000. Acquir Immune Defic Syndr 29 : 184 190.
222 Pannier, M., S. Veit, et al. (2000). Dead time free measurement of dipole dipole interactions between electron spins. J Magn Reson 142 : 331 40. Pasa Tolic, Y.L., C. Masselon, et al. (2004) Proteomic analyses using an accurate mass and time tag strategy. BioTechniques 37 : 621 636. Perrin, L., L. Kaiser, et al. (2003). Travel and the spread of HIV 1 genetic variants. Lancet Infect Dis 3 : 22 7. Poole, C. P., Jr. (1983). Electron Spin Resonance. A Comprehensive Treatise on Experimental Techniques, Second Edition Mineola, Dover Publications, Inc. Popolo, L., E. Ragni, et al. (2008). Disulfide bond structure and domain organization of yeast beta(1,3) glucanosyltransferases involved in c ell wall biogenesis. J Biol Chem 283 : 18553 18565. Qin, P. Z., J. Feigon, et al. (2005). Site directed spin labeling studies reveal solution conformational changes in a GAAA tetraloop receptor upon Mg(2+) dependent docking of a GAAA tetraloop. J Mol Bi ol 351 : 1 8. Rabenstein, M. D., and Shin, Y. K. (1995) Determination of the distance between two spin lab els attached to a macromolecule P NAS 92 : 8239 8243. Rambaut, A., D. Posada, et al. (2004). The causes and consequences of HIV e volution. Nat Rev Genet 5 : 52 61. Ran, Y., and Fanucci, G. E. (2008) A dansyl fluorescence based assay for monitoring kinetics o f lipid extraction and transfer Anal Biochem 382 : 132 134. Ran, Y., Fanucci, G.E. (2009) Ligand Extraction Properties the GM 2 Activator Protein and its I nteractions with Lipid Vesicles Biophysical Journal 97 : 257 266. Reeves, J. D. and R. W. Dorns (2002). Human immunodeficiency virus type 2. J Gen Virol 83 : 1253 1265 Rigat, B., Wang, W., Leung, A., and Mahuran, D. J. (1 997) Two mechanisms for the recapture of extracellular GM2 activator protein: evidence for a majo r secretory form of the protein Biochemistry 36 : 8325 8331 Rigat, B., Yeger, H., Shehnaz, D., and Mahuran, D. (2009) GM2 activator protein inhibits platelet acti vating factor signaling in rats Biochem Biophys Res Commun 385 : 576 580. Robertson, D. L., B. H. Hahn, et al. (1995). Recombination in AIDS viruses. J Mol Evol 40 : 249 259.
223 Rose, R.B., C. Craik, et al. (1998). Stroud, R.M. Domain flexibility in ret roviral proteases: structural implications for drug resistant mutations Biochemistry 37 : 2607 2621. Rossmann, M., R. Schultz Heienbrok, et al. (2008). Crystal structures of human saposins C andD: implications for lipid recognition and membrane interac tions. Structure 16 : 809 17. Sandhoff, K. and T. Kolter (1995). [Glycolipids of the cell surface -biochemistry of their decomposition]. Naturwissenschaften 82 : 403 13. Sandhoff, K. and T. Kolter (1996). Topology of glycosphingolipid degradation. Tren ds Cell Biol 6 : 98 103. Schaub, T. M., C. Hendrickson, et al. (2008). High performance mass spectrometry: Fourier transform ion cyclotron resonance at 14.5 Tesla. Anal Chem 80 : 3985 3990. Schlote, W., K. Harzer, et al. (1991). Sphingolipid activator protein 1 deficiency in metachromatic leucodystrophy with normal arylsulphatase A activity. A clinical, morphological, biochemical, and immunological study. Eur J Pediatr 150 : 584 91. Schneider, M. F., S. J. Gange, et al. (2005). Patterns of the hazard of death after AIDS through the evolution of antiretroviral therapy. AIDS 19 : 2009 2018. Schwartz, J.C., M. Senko, (2002). A two dimensional quadrupole ion trap mass spectrometer. J Am Soc Mass Spectrom 13 : 659 669. Schweiger, A. and G. Jesc hke (2001). Principles of pulse electron paramagnetic resonance New York, Oxford University Press, Inc. Scott, W. and C. Schiffer, (2000) Curling of flap tips in HIV 1 protease as a mechanism for substrate entry and tolerance of drug resistance. Structu re 8 : 1259 65. Senko, M.W., C. Hendrickson, et al. (1996). A two dimensional quadrupole ion trap mass spectrometer. J Am Soc Mass Spectrom 13 : 659 669. Senko, M.W., C. Hendrickson, et al. (1997). External accumulation of ions for enhanced electros pray ionization Fourier transform ion cyclotron resonance mass spectrometry J Am Soc Mass Spectrom 8 : 970 976. Smith, J. A. and R. Daniel (2006). Following the path of the virus: the exploitation of host DNA repair mechanisms by retroviruses ACS Chem Bio l 1 : 217 226. Smock, R. G., and Gierasch, L. M. (20 09) Sending signals dynamically Science 324 : 198 203.
224 Sparkman, O. D. (2000). Mass spectrometry desk reference Pittsburgh, Global View Publishing. Spiegel, R., G. Bach, et al. (2005). A mutation in the saposin A coding region of the prosaposin gene in an infant presenting as Krabbe disease: first report of saposin A deficiency in humans. Mol Genet Metab 84 : 160 6. Stone, T. L., T. Buckman, et al. (1965). Spin labeled biomolecules. PNAS 54 : 1010 1017. Stults, C. L., C. C. Sweeley, et al. (1989). Glycosphingolipids: structure, biological source, and properties. Methods Enzymol 179 : 167 214. Sun, Y., D. P. Witte, et al. (2008). Neurological deficits and glycosphingolipid accumulation in saposin B d eficient mice. Hum Mol Genet 17 : 2345 56. Syka, J. E., J. Coon, et al. (2004) Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. PNAS 101 : 9528 9533. Tanaka, K., H. Waki, et al. (1988). Protein and Polymer An alyses up to m/z 100 000 by Laser Ionization Time of flight Mass Spectrometry. Rapid Commun Mass Spectrom 2 : 151 153. Tang, X.J., P. Thibault, et al. (1993). Fragmentation reactions of multiply protonated peptides and implications for sequencing by tan dem mass spectrometry with low energy collision induced dissociation. Anal Chem 65 : 2824 2834. Thomson, J. J. (1913). Rays of positive electricity. Proceedings of the Royal Society A89 : 1 20. Thomson, M. M., L. Perez Alvarez, et al. (2002). Molecular e pidemiology of HIV 1 genetic forms and its significace for vaccine development and therapy. Lancet Infect Dis 2 : 461 471. Tikhonov, A. N. (1943). On the stability of inverse problems. Dokl. Akad. Nauk USSR 39 : 195 198. Tipton, J. D., J. D. Carter, et al. (2009). Sequential proteolysis and high field FTICR MS to determine disulfide connectivity and 4 maleimide TEMPO spin label location in L126C GM2 activator protein. Anal Chem 8 1 : 7611 7. Todd, M. J. and E. Freire (1999). The effect of inhibitor bi nding on the structural stability and cooperativity of the HIV 1 protease. Proteins 36 : 147 156. Todd, M. J., I. Luque, et al. (2000). Thermodynamic basis of resistance to HIV 1 protease inhibition: calorimetric analysis of the V82F/I84V active site res istant mutant. Biochemistry 39 : 11876 11883.
225 Todd, M. J., N. Semo, et al. (1998). The structural stability of the HIV 1 protease. J Mol Biol 283 : 475 488. Torbeev, V. Y., K. Mandal, et al. (2008). Crystal structure of chemically synthesized HIV 1 pr otease and a ketomethylene isostere inhibitor based on the p2/NC cleavage site. Bioorg Med Che m Lett 18 : 4554 4557. Torbeev, V. Y., H. Raghuraman, et al. (2008). Dynamics of Flap Structures in Three HIV 1 Protease/Inhibitor Complexes Probes by Tota l Chemical Synthesis and Pulse EPR Spectroscopy. J Am Chem Soc 131 : 884 885. Tsapraili, G., H. Hair, et al. (1999). Influence of secondary structure on the fragmentation of protonated peptides J Am Chem Soc 121 : 5142 5154. Tsybin, Y.L., C. Hendrockson, et al. (2206). Impac t of Ion Magnetron Motion on Electron Capture Dissociation Fourier Transform Ion Cyclotron Resonance Mass Spectrometry Int J Mass Spectrom 126 : 144 149. Uggerud, E. (2004). Electron capture dissociation of the disulfide bond a quantum chemical model study Int J Mass Spectrom 234 : 45 50. Velazquez Campoy, A., M. J. Todd, et al. (2000). HIV 1 protease inhibitors: enthalpic versus entropic optimization of the binding affinity. Biochemistry 39 : 2201 2207. Wang, L., Z. Zhang, et al. (2003). Addition of the keto functional group to the genetic code of Escherichia coli. PNAS 100 : 56 61. Wendeler, M., Werth, N., Maier, T., Schwarzmann, G., Kolter, T., Schoeniger, M., Hoffmann, D., Lemm, T., Saenger, W., and Sandhoff, K. (2006) The enzyme binding region of hu man GM2 activator protein FEBS J 273 : 982 991. Weil, J. A., J. R. Bolton, et al. (1972). Electron Spin Resonance: Elementary Theory and Practical Applications John Wiley and Sons, Inc. Weiss, R. A. (1993). How does HIV cause AIDS? Science 260 : 1273 1279. Wenger, D. A., G. DeGala, et al. (1989). Clinical, pathological, and biochemical studies on an infantile case of sulfatide/GM1 activator protein deficiency. Am J Med Genet 33 : 255 65. Wilkening, G., T. Linke, et al. (2000). Degradation of membr ane bound ganglioside GM1. Stimulation by bis(monoacylglycero)phosphate and the activator proteins SAP B and GM2 AP. J Biol Chem 275 : 35814 9. Wright, C. S., S. C. Li, et al. (2000). Crystal structure of human GM2 activator protein with a novel beta cu p topology. J Mol Biol 304 : 411 22.
226 Wright, C. S., Mi, L. Z., and Rastinejad, F. (2004) Evidence for lipid packaging in the crystal structure of the GM2 activator complex with platelet activating factor J Mol Biol 342 : 585 592. Wright, C. S., Mi, L. Z. Lee, S., and Rastinejad, F. (2005) Crystal structure analysis of phosphatidylcholine GM2 activator product complexes: evidence for hydrolase activity Biochemistry 44 : 13510 13521. Wright, C. S., Zhao, Q., and Rastinejad, F. (2003) Structural analysis of lipid com plexes of GM2 activator protein J Mol Biol 331 : 951 964. Xu, H., M. Freitas, et al. (2008). Identification and characterization of disulfide bonds in proteins and peptides from tandem MS data by use of the MassMatrix MS/MS search engine. J Pro teome Res 7 : 138 144. Xu, Q., Kim, M., Ho, K. W., Lachowicz, P., Fanucci, G. E., and Cafiso, D. S. (2008) Membrane hydrocarbon thickness modulates the dynamics of a membrane transport p rotein Biophys J 95 : 2849 2858. Yen, T.Y., H. Yan, et al. (2002). Characterizing closely spaced, complex disulfide bond patterns in peptides and proteins by liquid chromatography/electrospray ionization tandem mass spectrometry. J Mass Spectrom 37 : 15 30. Zhang,S., C. Van Pelt, et al (2003). Automated chip based nan oelectrospray mass spectrometry for rapid identification of proteins separated by two dimensional gel electrophoresis. Electrophoresis 24 : 3620 3632. Zhou, D., Cantu, C., 3rd, Sagiv, Y., Schrantz, N., Kulkarni, A. B., Qi, X., Mahuran, D. J., Morales, C R., Grabowski, G. A., Benlagha, K., Savage, P., Bendelac, A., and Teyton, L. (2004) Editing of CD1d bound lipid antigens by en dosomal lipid transfer proteins Science 303 : 523 527. Zubarez, R. A. (2004). Electron capture dissociation tandem mass spectrom etry. Curr Opini Biotechnol 15 : 12 16. Zubarev, R. A., N. Kruger, (1999). Electron capture dissciociation of gases multiply charged proteins is favored at disulfide bonds and other sites of hydrogen atom affinity J Am Chem Soc 121 : 250 258.
227 BIOGRAPHICAL S KETCH Jeffrey David Carter was born in 1983 and graduated from high school in 2002. Then pursued as biochemistry degree at Virginia Tech and graduated with a Bachelor of Science in 2006 He was admitted to the D epartm ent of C hemistry graduate program at t he University of Florida in 2006, where he joined the research group of Dr. Gail E. Fanucci. H e obtained his Ph.D. in c hemistry in 2012.