Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigating the protein-protein interactions between TAR DNA-binding protein 43 AND p65 subunit of… Lee, Joseph 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_may_lee_joseph.pdf [ 2.63MB ]
Metadata
JSON: 24-1.0224505.json
JSON-LD: 24-1.0224505-ld.json
RDF/XML (Pretty): 24-1.0224505-rdf.xml
RDF/JSON: 24-1.0224505-rdf.json
Turtle: 24-1.0224505-turtle.txt
N-Triples: 24-1.0224505-rdf-ntriples.txt
Original Record: 24-1.0224505-source.json
Full Text
24-1.0224505-fulltext.txt
Citation
24-1.0224505.ris

Full Text

     INVESTIGATING THE PROTEIN-PROTEIN INTERACTIONS BETWEEN TAR DNA-BINDING  PROTEIN 43 AND p65 SUBUNIT OF NF-κB   by  Joseph Lee  M.Sc., Simon Fraser University, 2011 B.Sc., The University of British Columbia, 2006     A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE   in     The Faculty of Graduate and Postdoctoral Studies  (Chemistry)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  February 2016  © Joseph Lee 2016  ii  ABSTRACT TAR DNA-binding Protein 43 (TDP-43), a ubiquitous protein that regulates gene expression, has been found to play a crucial role in the pathogenesis of Amyotrophic Lateral Sclerosis (ALS), in which the disease is characterized by TDP-43 protein inclusion bodies. Relevant literature suggests that the protein either self-aggregates or interacts with various partners to cause this proteinopathy. One of these binding partners was suggested to be the p65 domain of the nuclear factor kappa-B (NF-κB), a transcription protein complex which plays a crucial role in inflammatory and immune responses. It is upon this hypothetical disease pathogenesis that the study of TDP-43 and NF-κB p65 is rationalized. Hence, the first part of the thesis describes the methods that were developed to obtain pure recombinant TDP-43 from an E.coli expression system. Subsequently, the preparation of NF-κB p65 peptides using solid phase peptide synthesis (SPSS) is described in the thesis. Furthermore, the structural conformation of proteins and peptides was explored using molecular dynamics (MD) simulations to predict how they will behave in vivo and also to allow a comparison to in vitro experimentation.   iii  PREFACE My supervisor and I designed the experiments. I carried out the experiments described in this thesis. Molecular dynamics simulations were set up by Dr. Walter Scott and myself. Analysis of the MD simulations was done by Dr. Walter Scott. Finally, I wrote the thesis. My supervisor and Dr. Walter Scott edited it.  iv  TABLE OF CONTENTS ABSTRACT .......................................................................................................... ii PREFACE ........................................................................................................... iii TABLE OF CONTENTS ...................................................................................... iv LIST OF TABLES .............................................................................................. vii LIST OF FIGURES ............................................................................................ viii LIST OF ABBREVIATIONS ................................................................................. x ACKNOWLEDGEMENTS .................................................................................. xii DEDICATION .................................................................................................... xiii CHAPTER 1: INTRODUCTION ............................................................................ 1 1.1 Amyotrophic Lateral Sclerosis: Symptoms, Causes, Previous Research ............................................................................................ 1 1.2  TAR DNA-Binding Protein 43 ........................................................... 3 1.2.1 Normal Physiological Function ........................................................ 3 1.2.2 TDP-43 Domain Function and Arrangement ................................... 5 1.2.3 TAR DNA-Binding Protein 43 Proteinopathy in ALS ....................... 6 1.2.4 Aggregation Model Due to Misfolding of TDP-43 ............................ 8 1.2.5 Dysregulation Model Due to Protein Interaction with p65 of NF-kB ............................................................................................ 11 1.2.6 NF-κB Physiological Role and Domain Function .......................... 13 1.3 Description of Methodology Used ..................................................... 14 1.3.1 Protein Expression ........................................................................ 14 1.3.2 Solid Phase Peptide Synthesis ..................................................... 15 1.3.3 Molecular Dynamics Simulations .................................................. 15 1.4 Overview of Objectives ..................................................................... 16 1.4.1 Long Term Goals .......................................................................... 16 1.4.2 Specific Aims ................................................................................ 17 CHAPTER 2: DEVELOPMENT OF A PURIFICATION PROTOCOL FOR GST-TAGGED TDP-43 ................................................................................ 20 2.1 Introduction ...................................................................................... 20 2.2 Material and Methods ....................................................................... 21 2.2.1 Materials ....................................................................................... 21 2.2.1.1 Plasmids .................................................................................. 21 2.2.1.2 E. coli BL21 (DE3) Cells ........................................................ 22 2.2.2 Methods ........................................................................................ 22 2.2.2.1 Polymerase Chain Reaction ................................................. 22  v  2.2.2.2 Plasmid DNA Preparations ................................................... 23 2.2.2.3 Restriction Digests, Agarose Gels and Plasmid DNA Purification ............................................................................................... 23 2.2.2.4 Ethanol Precipitation of DNA ................................................ 23 2.2.2.5 Phosphatase Treatment of Vector ....................................... 24 2.2.2.6 Ligations................................................................................... 24 2.2.2.7 Competent Cells Transformation ......................................... 24 2.2.2.8 Preparation of pGEX-KG-TDP-43 ....................................... 25 2.2.2.9 Expression of Protein via IPTG Induction .......................... 27 2.2.2.10 GST-tagged Protein Purification .......................................... 27 2.2.2.11 HRV 3C Digestion .................................................................. 28 2.2.2.12 Thrombin Digestion ................................................................ 29 2.2.2.13 Gel Electrophoresis ................................................................ 29 2.3 Results and Discussion .................................................................... 30 2.3.1 Expression of N-terminal GST-tagged TDP-43 RRM-1 ................. 30 2.3.2 Expression and Purification of N-terminal GST-tagged TDP-43 ......................................................................................... 32 2.3.3 Expression and Purification of N-terminal GST-tagged TDP-43 with Glycine Linker Between the GST tag and TDP-43 ......................................................................................... 34 2.4 Conclusion ....................................................................................... 36 CHAPTER 3: DEVELOPMENT OF A SOLID PHASE PEPTIDE SYNTHESIS PROTOCOL FOR THE EIGHT P65 PEPTIDES ..................... 37 3.1 Introduction  ..................................................................................... 37 3.2 Material and Methods ....................................................................... 38 3.2.1 Materials ....................................................................................... 38 3.2.1 Methods ........................................................................................ 39 3.2.1.1 Solid Phase Peptide Synthesis ............................................ 39 3.2.1.2 Peptide Cleavage from Resin and De-protection of Side Chains 39 3.2.1.3 Peptide Precipitation .............................................................. 40 3.2.1.4 High Performance Liquid Chromatography ........................ 40 3.3 Results and Discussion .................................................................... 41 3.3.1 Development of Peptide Synthesis Protocol ................................. 41 3.3.2 Reverse Phase HPLC Purification ................................................ 45 3.4 Conclusion ....................................................................................... 47 CHAPTER 4: MOLECULAR DYNAMIC SIMULATIONS ................................... 48 4.1 Introduction ...................................................................................... 48 4.2 General Methods .............................................................................. 49 4.2.1 Scalar Atomic B-Factors Analysis ................................................. 50 4.2.2 Clustering Analysis ....................................................................... 50 4.2.3 Hydrogen Bonding Analysis .......................................................... 51 4.3 TDP-43 RRM-1 Simulations ............................................................. 52 4.3.1 Methods ........................................................................................ 52  vi  4.3.2 Results and Discussion ................................................................. 54 4.4 NF-Kappa B p65 Peptide Simulations .............................................. 57 4.4.1 Methods ........................................................................................ 58 4.4.2 Results and Discussion ................................................................. 60 4.5 Conclusion ....................................................................................... 67 CHAPTER 5: CONCLUSION AND FUTURE PLANS ........................................ 69 REFERENCES ................................................................................................... 73 APPENDICES .................................................................................................... 81 Appendix A: PCR And Thermocycling Reaction Conditions ............................ 81 Appendix B: Oligonucleotide Primers .............................................................. 82    vii  LIST OF TABLES Table 3-1  Eight peptides from p65 subunit of NF-κB that were found to interact with TDP-43. The full length sequence of human p65 can be found in Uniprot entry Q04206............................................. 38 Table 3-2  List of p65 peptides made using solid phase peptide synthesis ....... 44 Table 3-3  GRAVY indices of the eight original p65 peptides and the NF1-error peptide ............................................................................ 46 Table 4-1  Overview of the four TDP-43-RRM1 variants ................................... 54 Table 4-2  Combined clustering results incorporating conformations from all four TDP-43-RRM1 variants simulations. ................................... 56 Table 4-3   The sequences of the two peptidic fragments simulated, showing their residue numbers in the whole protein. ...................... 59 Table 4-4  Overview of four simulations performed in methanol. ...................... 59 Table 4-5  Combined clustering results incorporating conformations from all four peptide simulations. ............................................................. 62   viii  LIST OF FIGURES  Figure 1-1  Overview of events in the pathogenesis of ALS ............................... 3 Figure 1-2  Schematic and three dimensional structure of TDP-43 ..................... 6 Figure 1-3  Possible pathogenic mechanism of mutant TDP-43 ......................... 7 Figure 1-4  RRM-1 oxidation perturbs its conformation ....................................... 9 Figure 1-5  C173S/C175S mutations in RRM-1 induce TDP-43 aggregation ..................................................................................... 10 Figure 1-6  Proposed mechanism by which TDP-43 aggregates. ..................... 10 Figure 1-7  p65 subunit of nuclear factor κB (NF-κB) ........................................ 12 Figure 1-8  Flow Chart for Expression and Purification off GST-tagged Proteins ........................................................................................... 14 Figure 1-9  Flow Chart for Solid Phase Peptide Synthesis ............................... 15 Figure 2-1  Construction of pGEX-KG-TDP-43 ................................................. 26 Figure 2-2  Expression of GST-RRM-1 ............................................................. 31 Figure 2-3  Expression and Purification of GST-TDP-43 .................................. 33 Figure 2-4  Expression and Purification of GST-gly-TDP-43 ............................. 35 Figure 3-1  MALDI-ToF MS analysis of NF1-error and NF2 crude peptides prepared via SPPS .......................................................................... 42 Figure 3-2  MALDI-ToF MS analysis of NF3 and NF8 crude peptides prepared via SPPS .......................................................................... 43 Figure 3-3  MALDI-ToF MS analysis of NF1-error peptide crude product made by SPPS using threonine preloaded Wang resin .................. 45 Figure 3-4  RP-HPLC purification of SPPS prepared crude NF3 peptide ......... 46 Figure 4-1  Hydrogen bonding analysis ............................................................ 52 Figure 4-2  The initial structures of Wild-Type and single/double mutants of TDP-43-RRM-1 ........................................................................... 53  ix  Figure 4-3  The calculated Cα Isotropic B-factors of the respective TDP-43 RRM-1 variants ............................................................................... 55 Figure 4-4  Representative structure of cluster-0 .............................................. 57 Figure 4-5  Structure of 1IKN ............................................................................ 59 Figure 4-6  The helix and loop starting structures of the four simulations for NFkB p65 peptide ...................................................................... 60 Figure 4-7  The Cα RMSD to the respective starting structures as a function of time ................................................................................ 61 Figure 4-8  Combined clustering of the four p65 peptide simulations ............... 63 Figure 4-9  The representative structure of cluster-0 of the combined cluster analysis shows a loop-like structure stabilized by hydrogen bonds (cyan).................................................................... 64 Figure 4-10 The simulated data points versus the entropic part of the free energy difference between cluster i and cluster-0. .......................... 64 Figure 4-11 The representative structures of cluster-0 of the two NF2 simulations. ..................................................................................... 66   x  LIST OF ABBREVIATIONS  ALS   amyotrophic lateral sclerosis bp:   base pair DCM   dichoromethane DIEA   N,N-Diisopropylethylamine DMF   dimethyformamide EtOH   ethanol FL   full length Fmoc   fluorenylmethyloxycarbonyl HBTU N,N,N′,N′-Tetramethyl-O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate HPLC   high performance liquid chromatography IPTG:   isopropyl β-D-1-thiogalactopyranoside KB    Boltzmann’s constant MD   molecular dynamics NH4Ac  ammonium acetate NF-κB   nuclear factor- kappaB NLS:   nuclear localization sequence NMR:   nuclear magnetic resonance PAG(E):  polyacrylamide gel (electrophoresis) PBS:   phosphate buffered saline PCR:   polymerase chain reaction PMSF:  phenylmethylsulphonylfluoride RRM   RNA recognition motif SAP:   shrimp alkaline phosphatase   xi  SDS   sodium dodecyl sulfate TES   triethysilane T    temperature TDP-43  Transactive response DNA-binding Protein 43 TFA   trifluoroacetic acid Tris:   Tris(hydroxymethyl)methylamine WT:   wild-type      xii  ACKNOWLEDGEMENTS I would like to thank my senior supervisor, Dr. Suzana Straus, for her excellent supervision and suggestions on my thesis work. Your enthusiasm in science and research has motivated me to walk thru the difficulties in life. Thanks also to my examining committee members, Dr. Russ Algar, Dr. Hongbin Li, and  Dr. Alexander Wang. Thanks to all my lab mates, past and present, for being such a great group of people to work with. Special thanks to Dr. Walter Scott for helping with Molecular Dynamics Simulation experiments. Thanks Yurou Sang for showing me all the lab equipment and giving me advice in research. Thanks also to Chantal Mustoe and Prashant Kumar for discussions and ideas about experiments.  Thanks to John Sherman’s lab for letting me use the HPLC located in their laboratory. Special thanks to UBC Biological Services Laboratory for the use of reagents, equipments, and space.  Thank you Dr. Jean-Pierre Julien (Université Laval) for providing the original TDP-43 construct and the sequences of the eight p65 peptides.       xiii  DEDICATION     For my Parents  Thank you for your encouragement  1  CHAPTER 1: INTRODUCTION This thesis describes the purification of TAR DNA-binding Protein 43 (TDP-43), a ubiquitous protein that regulates gene expression, and the synthesis of peptides from the p65 domain of the nuclear factor kappa-B (NF-κB), a transcription protein complex which plays a crucial role in inflammatory and immune responses. TDP-43 has been recently implicated in the pathogenesis of Amyotrophic Lateral Sclerosis (ALS) based on the findings that patients suffering from ALS have TDP-43 protein inclusion bodies. Chapter 1 lays the foundation of the thesis by reviewing the relevant literature, beginning with a general overview of ALS, looking at its symptoms and pathology, followed by the molecular characterization of potential biochemical determinants of the disease. It is upon this hypothetical disease pathogenesis that the study of TDP-43 and NF-κB p65 is rationalized. In this chapter, the methods used are also described. These are: protein expression, solid phase peptide synthesis, and molecular dynamics simulations.  1.1   Amyotrophic Lateral Sclerosis: Symptoms, Causes, Previous Research Amyotrophic Lateral Sclerosis, also known as Lou Gehrig’s disease, is a lethal neurodegenerative disorder which affects the brain and spinal cord of approximately 2500-3000 Canadian adults. To date, there is no effective treatment or cure to this devastating disease. ALS is characterized by a subtle  2  onset of focal weakness of the body, typically in the limbs but sometimes in bulbar muscles, and progresses to the paralysis of skeletal muscles. The most common cause of death in ALS is due to respiratory paralysis, which typically occurs within 5 years of diagnosis. Although most cases of ALS are sporadic, approximately 5-10% of the cases are hereditary due to genetic mutations [1]. The first wave of molecular research in ALS began in 1993 with the identification of mutations in the superoxide dismutase 1 (SOD1) gene [2]. However, the major milestone came in 2006 with the identification of TAR DNA-binding protein 43 (TDP-43) inclusions in ALS brains and spinal cords, indicating the importance of this protein in the disease [3], [4]. Later in 2008, the discovery of TDPDBP gene mutations in a small percentage of familial ALS cases confirmed a mechanistic link between TDP-43 and ALS pathogenesis [5], [6]. An overview that encompasses the different ALS disease pathogenic mechanisms is shown in Figure 1-1, identifying the common players such as SOD1, FUS, TDP-43, and Chromosome 9 open reading frame 72 (C9ORF72) [7]. Additional proteins and genes that have also been found to play a more minor role include optineurin (OPTN, Figure 1-1), a causative agent of primary open-angle glaucoma in ALS patients, valosin-containing protein (VCP), link to frontotemporal demential, and finally the UBQLN2 gene, which encodes ubiquitin 2. For SOD1, FUS and TDP-43, a number of mechanisms leading to ALS have been suggested, including protein aggregation, interaction with other binding partners, etc. As TDP-43 is the focus of this thesis, further details on SOD1, FUS, and the minor proteins listed above will not be discussed further. The reader is referred to excellent reviews for more details [7], [8].   3   Figure 1-1  Overview of events in the pathogenesis of ALS Diagram indicating all the major and minor proteins linked to ALS. (adapted from [7] with permission) 1.2  TAR DNA-Binding Protein 43 1.2.1 Normal Physiological Function Proteins perform all kinds of intra- and extra-cellular events. Hence, genetic mutations, deletions, protein misfolding and aggregation lead to gain or loss of protein functions, which are related to various genetic and sporadic diseases. The TAR DNA-binding protein 43 (TDP-43) is a ubiquitously expressed protein whose abnormal aggregation is directly linked to ALS [3], [9], [10].  4  TDP-43 is a highly conserved 43 kDa heterogeneous nuclear ribonucleoprotein (hnRNP) that controls the transcription, splicing, and RNA stability of specific genes [11]. Although it is most abundant in the nucleus, TDP-43 can shuttle between the nuclear and cytoplasmic compartments and be transported along axons [12]. TDP-43 was originally identified as a transcriptional factor, repressing the transcription of the HIV-1 gene and the expression of cyclin-dependent kinase 6 (Cdk6) [13], [14]. TDP-43 is also a splicing factor binding to the intron 8 and exon 9 junction1 of the cystic fibrosis transmembrane conductance regulator (CFTR) gene to inhibit exon splicing [15]–[17]. The protein associates with single-stranded RNA and DNA sequences, and has been shown to exhibit remarkable specificity for UG/TG dinucleotide repeats [17]–[20]. Also, TDP-43 has also been shown to be a mRNA-binding protein in spinal motor neurons, suggesting its involvement in regulating mRNA stability, transport and local translation in neurons [21]. TDP-43 is a key component of dendritic and somatodendritic2 RNA transport granules in neurons [22]. It also plays an important role in neuronal plasticity by regulating local protein synthesis in dendrites [23]. TDP-43 is also involved in the cytoplasmic stress granule response in which protein complexes are formed to sequester mRNAs redundant for survival [24], [25]. This suggests TDP-43 function is linked to conditions of cellular                                                  1 Most eukaryotic genes are composed of coding regions (exons) and noncoding regions (introns). During transcription, the entire gene is copied into a pre-mRNA. During the process of RNA splicing, introns are removed and exons are joined to form a continuous coding sequence, which is ready for translation. 2 Dendrites are neuron projections that have short, narrow, and branched morphology which receives and integrates signal from other neurons or sensory stimuli. It conducts a nerve impulse towards the axon through the cell body. Somatodendrite is the region of neuron that includes the cell body and the dendrite, but excludes the axon.   5  stress. Hence, TDP-43 is normally implicated in gene transcription, exon splicing, translational regulation, and interactions with splicing factors and nuclear bodies.   1.2.2 TDP-43 Domain Function and Arrangement The TARDBP gene is highly conserved in human, mouse, Drosophila melanogaster and Caenorhabditis elegans [26]. Sequence analysis revealed that TDP-43 is a 414- amino acid protein which consists of several functional domains (Figure 1-2a). It includes a bipartite nuclear localizing signal (NLS) in the N-terminal domain (NTD) that is partially structured and likely to be involved in protein oligomerization, followed by two RNA-recognition motifs (RRM-1 (Figure 1-2b) and RRM-2) capable of binding nucleic acids which also includes a nuclear export signal (NES), and ending in an intrinsically disordered C-terminal domain (CTD) implicated in protein interactions. TDP-43 has a similar domain organization to the heterogeneous nuclear ribonucleoproteins (hnRNP) family of proteins, which are RNA binding proteins which appear to influence pre-mRNA processing and other aspects of mRNA metabolism and transport [27]. The RRMs consist of two highly conserved hexameric and octameric segments denoted as ribonucleoprotein 2 (RNP2) and ribonucleoprotein 1 (RNP1) respectively [28]. These RNPs are involved in binding to TAR DNA sequences and RNA sequences that contain UG repeats [13], [29]. The glycine-rich C-terminal tail interacts with hnRNP family proteins to form complexes involved in splicing inhibition [30].    6   Figure 1-2  Schematic and three dimensional structure of TDP-43  (A) Schematic representation of TDP-43 domain structure. TAR DNA-binding protein 43 (TDP-43) protein contains two RNA-recognition motifs (RRM-1 and RRM-2), a carboxy-terminal glycine rich domain, a bipartite nuclear localization signal (NLS) and a nuclear export signal (NES) (adapted from [31] with permission). (B) Three dimensional structure of TDP-43 RRM-1 domain. Residues 101 and 178 are indicated to denote which is the N- and C-terminus respectively. PDB id: 2CQG.pdb. Model generated using CHIMERA [32]  1.2.3 TAR DNA-Binding Protein 43 Proteinopathy in ALS Recent evidence has shown that TDP-43 is a major protein in inclusions from patients suffering from ALS. This has prompted intense investigation with regards to its structure, subcellular localization, interaction partners, and aggregation propensity. In general, neurodegenerative diseases related to TDP-43 deposits are termed “TDP-43 proteinopathies” , which also describes the  7  characteristic histopathological transformation of TDP-43 that occurs in ALS [33], [34]. TDP-43 is predominately localized in the nucleus, but in cases of neurodegenerative proteinopathies, the protein is present as cytoplasmic aggregates presumably with post-translational modifications such as hyper-phosphorylation, ubiquitination and cleavage which results in ~25kDa C-terminal fragments in affected brain regions [35]–[37]. The regional spread of TDP-43 proteinopathy from spinal and corticol motor neurons to other cortical regions can be used to stage ALS progression [38], and this suggests that some or all of the features of transformed TDP-43 protein are linked to pathogenesis (Figure 1-3).   Figure 1-3  Possible pathogenic mechanism of mutant TDP-43 The left panel shows the normal state of TDP43, shuttling between the nucleus and the cytoplasm. In the diseased state, shown on the right, aberrant TDP-43 is formed resulting in aggregation. (adapted from [7] with permission)  8  There are two proposed mechanisms by which neuronal degeneration occurs: 1) either by aggregation of TDP-43 in the cyptoplasm of neurons [39], [40] or 2) by up-regulation of TDP-43 in the nucleus [40]–[43]. In either case, the first RNA recognition motif RRM-1 has been identified as the most likely domain where either aggregation or dysregulation is triggered [43], [44]. Among the unique features of TDP-43 inclusions are that they are not amyloid deposits, and are negative for tau, α-synuclein, and β-amyloid 3, indicating that the inclusions may have a distinct aggregated structure [45]. Interestingly, there is little to no understanding at the molecular level of how RRM-1 interacts with other proteins and how this leads to dysregulation or aggregation. 1.2.4 Aggregation Model Due to Misfolding of TDP-43 As suggested in Figure 1-3, proteinopathy of TDP-43 may be due to its ability to self-assemble, resulting in aggregation. It has been shown using immunofluorescence studies that TDP-43 translocates from the cytoplasm to the nucleus and vice versa [39]. However, inhibition of TDP-43 translocation, overexpression of TDP-43, or mutation of the NLS have been shown to lead to the sequestration of nuclear TDP-43 and the formation of insoluble aggregates in the cytoplasm [39]. Also, it has been found that RRM-1 is prone to misfolding under high pressure conditions and oxidative conditions [44], [46]. In particular, this was demonstrated via solution state NMR studies (two-dimensional HSQC spectra of 15N-labelled RRM-1) to show that the structure of RRM-1 was perturbed when two                                                  3 The abnormal accumulation of tau protein and β-amyloid is the main histopathological feature of Alzheimer’s disease. On the other hand, the neuropathologic feathure of Parkinson’s disease is the presence of Lewy bodies composed of amyloid fibrillar deposits of α-synuclein. For further details, see excellent reviews on Alzheimer’s disease and Parkinson’s disease [74], [75]  9  cysteine residues (Cys173 and Cys175) in the RRM-1 domain are oxidized (Figure 1-4) [46]. To further support this, residues 173 and 175 mutations of cysteine to serine in the RRM-1 demonstrated TDP-43 aggregation in the nucleus and cytosol induced by the mutation [44] (Figure 1-5). The misfolding of RRM-1, creating an enhancement of β-sheet in the region (Figure 1-6), could lead to downstream aggregation events involving the C-terminus of TDP-43, where ALS-associated mutations are located [44].  Figure 1-4  RRM-1 oxidation perturbs its conformation  Two dimensional HSQC spectra of 15N-labelled RRM-1 domain from TDP-43 before (blue) and after (red) oxidation by H2O2 in PBS. Green signals represent a third spectrum where one day after the oxidation reaction, 50mM DTT was added to the sample. Residues labelled in red represent major perturbations (i.e. resonance disappears), while those in black represent minor perturbations (i.e. resonance shifts). Adapted from [46] with permission.   10   Figure 1-5  C173S/C175S mutations in RRM-1 induce TDP-43 aggregation Confocal mages of mammalian HEK293A cells transiently transfected with EGFP-fused full length (FL) TDP-43 (wild type, C173S, C175S, and C173S/C175S) in green. DAPI fluorescent stain was used to bind DNA and locate the position of the nuclei. Single and double C/S mutations resulted in the formation of marked aggregates or inclusions of FL TDP-43, which located in the nucleus and occasionally in the cytoplasm. Adapted from [44] with permission.   Figure 1-6  Proposed mechanism by which TDP-43 aggregates.   Exposure of the β hairpin in region C may lead to RRM-1 as well as C-terminus aggregation, resulting in the formation of inclusions found in pathological TDP-43 proteinopathies. The region C was identified through high pressure solution state NMR experiments. Adapted from [44] with permission.    11  One of the aims of this thesis is to look at the structure of RRM-1 and its mutant variants and how misfolding would lead to proteinopathy. It will be important to look at the RRM-1 domain on its own and also its behaviour when embedded in the whole TDP-43 protein.  1.2.5 Dysregulation Model Due to Protein Interaction with p65 of NF-kB Expression of TDP-43 is often mediated by a host of interacting proteins [47]. Therefore, if up-regulation of TDP-43 is an important factor in its proteinopathy then there should be specific binding partners which interact with it. Recent work has reported that the protein expression of TDP-43 and nuclear factor kappaB (NF-κB, described in detail in section 1.2.6) p65 subunit (an inflammation-regulating protein) is higher in spinal cords of ALS patients than healthy individuals [43]. More importantly, TDP-43 has been found to interact and colocalize with p65 in neuronal cells from ALS patients, but not in cells from healthy individuals. The interaction was probed using immunoprecipitation and was localized to the N-terminus and RRM-1 domain of TDP-43 [43]. The nature of the interaction between TDP-43 and p65 was found to be a protein-protein one, as demonstrated by the fact that the presence of proteinase K (a protease that cleaves peptide bonds at the carboxylic sides of aliphatic, aromatic, or hydrophobic amino acids) would abolish the interaction, whereas RNase A and DNase 1 had no effect [43]. It is therefore suggested that TDP-43 is associated with NF-κB activation, leading to exaggerated immune response resulting in neuro-inflammation and motor neuron destruction. More recently, researchers have cut the full length (FL) NF-κB p65 subunit seqeunce into 10-20 residue  12  overlapping peptides and attached them onto a cellulose surface to look at possible binding interaction between TDP-43 and p65 (Jie et al., unpublished). The preliminary study has identified 8 possible binding regions on the p65 subunit that could interact with TDP-43 (Figure 1-7a). Hence another aim of this thesis is to investigate the interactions between TDP-43 RRM-1 and the p65 subunit peptides.     Figure 1-7  p65 subunit of nuclear factor κB (NF-κB) (A) Amino acid sequence of p65 from NF-κB. Residues in green/purple represent residues 20-291, for which there is a three-dimensional crystal structure as shown in B. The purple/red segments represent 8 peptides found to interact with TDP-43 (Jie et al., unpublished). (B) Three dimensional structure of part of p65, PDB id: 2O61.pdb. A.               B.   13  1.2.6 NF-κB Physiological Role and Domain Function Before proceeding into further discussion about TDP-43, a brief description of NF-κB will be given here. The NF-κB is a nuclear factor that binds to immunoglobulins in the nucleus of B cells [48]. NF-κB is usually present in a latent form within the cytoplasm, and when activated, it translocates into the nucleus to bind DNA. Upon binding to DNA, NF-κB regulates various gene expression steps that in turn relate to the control of the immune system, growth, and inflammation. Since NF-κB plays an essential role in normal physiology, dysfunction of NF-κB is associated with many disease and disorders such as diabetes, cancers, chronic inflammation, as well as cardiovascular and central nervous system diseases [49].  The NF-κB family is comprised of five polypeptides/proteins – p50, p52, p65 (RelA), c-Rel, and RelB [50]. These five proteins share a homologous N-terminal rel homology domain (RHD), which comprises a DNA-binding domain, a dimerization domain, and a nuclear localization signal (NLS). NF-κB exists as a homo- or hetero-dimer by dimerization of the RHD domain with different affinities and specificities. Each NF-κB dimer has distinct but often overlapping properties such as their regulation by inhibitor κB proteins, nuclear translocation, and DNA recognition and binding. The DNA-binding RHD of NF-κB proteins is bipartite and folds into two immunoglobulin-like (Ig-like) domains linked by a short flexible linker (Figure 1-7b). The N-terminal Ig-like domain is responsible for specific sequence DNA binding,  14  whereas the C-terminal Ig-like domain is responsible for subunit dimerization and non-specific DNA binding [51]. 1.3 Description of Methodology Used 1.3.1 Protein Expression In order to study a protein using biophysical techniques, it is often necessary to produce it in large quantities. Protein expression has become a common tool to achieve these aims, in particular using E.coli as an expression system. Many reviews [52], [53] describe the methodology. In Figure 1-8, an overview of the general steps involved are given.  Figure 1-8  Flow Chart for Expression and Purification off GST-tagged Proteins A standard protein expression and purification overview is shown. Please refer to section 2.2.2.9 – 2.2.2.12 for detailed protocol.   15  1.3.2 Solid Phase Peptide Synthesis  Similarly to protein expression, solid phase peptide synthesis is an important methodology used to produce large quantities of peptides. A general overview of the steps involved is shown in Figure 1-9. Further details can be found by consulting references [54], [55].   Figure 1-9  Flow Chart for Solid Phase Peptide Synthesis A standard Fmoc chemistry peptide synthesis is shown. Please refer to section 3.2.1 for detailed protocol. HBTU, N,N,N′,N′-Tetramethyl-O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate; DMF, Dimethylformamide; DCM, Dichloromethane.  1.3.3 Molecular Dynamics Simulations Molecular dynamic simulations play an important role in recent studies of biological macromolecules as they provide details related to motions of individual atoms as a function of time. This tool gives users an atomistic picture of interactions and provides important insights toward structural conformations,  16  inter/intra-molecular interactions, and reaction pathways. The use of MD to simulate properties of biomolecules allows researchers to investigate specific domains, motifs, or even amino acids under different conditions. However, it should be emphasized that MD simulation results should always be validated by actual experimental data in order to compare simulation calculated data and experimental data to determine the accuracy of the results and also to help improve MD simulation methodology.  In this thesis, MD simulations will be used to look at two different aspects. First, MD will be used to determine if there are any structural and conformational differences between wild-type (WT) TDP-43 RRM-1 and mutant variants where the two cysteine residues (Cys173 and Cys175) in RRM-1 were substituted into serine residues. In a second instance, MD simulation will be used to examine the similarities and differences between the NF-κB p65 peptides identified in section 1.2.5. A particular emphasis was placed on the NF1 peptide (AGSIPGERSTDTTK; residues 43-56) and NF2 peptide (KRDLEQAISQRIQT; residues 123-136) (Figure 1-7a) since their secondary structure is known (Figure 1-7b). The results of these simulations will serve as the stepping stone for future experimental considerations. 1.4 Overview of Objectives 1.4.1 Long Term Goals Recent research has identified TDP-43 as a causative agent of ALS. More interestingly, RRM-1 has been implicated in both the dysregulation and  17  aggregation scenario. However, only a few studies at the molecular level clearly demonstrate how RRM-1 binds to other proteins or itself. Hence, the goal of this project is to identify binding interactions between RRM-1 and the p65 subunit of NF-κB, and potentially between RRM-1 and itself.   1.4.2 Specific Aims The thesis aims to report on some key experiments to elucidate the function of the RRM-1 domain in the proteinopathy of TDP-43. Therefore the specific aims of this project are as follows:  Aim 1: Exploring the aggregation hypothesis by comparing aggregation propensities in wild type TDP-43 versus a Cys173 and Cys175 mutant  As mentioned in section 1.2.4, the structure of RRM-1 was perturbed with just having the two cysteine residues (Cys173 and Cys175) being oxidized. It would be interesting to find out what would happen when the residues are substituted with serine residues, as such mutation is a common cause of pathogenesis. By preparing the wild-type RRM-1 and RRM-1 cysteine mutants, we may be able to identify if there are differences in the structure of the aggregates among the variants. This work will be done via nuclear magnetic resonance (NMR) spectroscopy which will enable us to look at structural perturbation at a molecular level. In order to examine this further, MD simulations of RRM-1 mutants will be performed to look at their different protein conformation and dynamics which may alter the oligomerization propensity.    18  Aim 2: Synthesis of the 8 peptides derived from p65 subunit of NF-κB found to interact with TDP-43.  Frist, the 8 peptides shown in Figure 1-7a will be synthesized in-house via a solid phase peptide synthesizer utilizing Fmoc chemistry. Crude peptides will be purified using reversed phase High Performance Liquid Chromatography (RP-HPLC) to eliminate any fragmented variants. The purified product will be confirmed via Mass Spectrometry to ensure the right molecular weight and is free from fragmented by-products.   Aim 3: Over-expression of TDP-43 for structural and binding studies In preparation for the interaction studies listed in Aim 4 below, expression and isolation of TDP-43 RRM-1 will be carried out in parallel to obtain purified protein.  Aim 4: Mapping where the 8 peptides from p65 bind to RRM-1 using various biochemical and biophysical methods. We wish to explore if there is a specific interaction that exists between one or more of the 8 peptides and RRM-1. In order to do so, we will employ different biochemical and biophysical methods such as protein pull-down assays and isothermal titration calorimetry (ITC). These experiments will be necessary to confirm that the interactions exist and will enable us to determine a dissociation constant, KD, for each of the peptides with RRM-1. Moreover, NMR spectroscopy  19  will be employed to look at the binding at a molecular level. Hence, we will express and purify 15N-labelled RRM-1 and titrate in each peptide in turn to detect binding. Any binding that occurs will cause a shift in the chemical shift of a given residue in the HSQC spectrum.  Overall, these studies will ensure a better understanding of how TDP-43 may contribute to ALS symptoms.  20  CHAPTER 2: DEVELOPMENT OF A PURIFICATION PROTOCOL FOR GST-TAGGED TDP-43 2.1 Introduction Development of the protocol to produce large quantity of TDP-43 is needed to study the aggregation model and also for the use of the interaction studies. Hence, the first goal to achieve was to develop protocols for efficient expression and purification of soluble protein. The hallmarks of an efficient protocol would be the acquisition of pure protein with high yield and with the fewest steps. The potential challenges when working with proteins that tend to aggregate is their intrinsic low-water solubility and consequent requirement for detergents. The workup to isolate pure TDP-43, and furthermore TDP-43 RRM-1, will serve as the foundation which will be necessary for many future experiments such as the binding studies and the mutational studies mentioned in Chapter 1.  Since the expression of large quantities of TDP-43 has never been reported in the literature, it was anticipated that it might be a difficult task which would involve screening of different conditions. Since this protein is prone to aggregation, it was anticipated that the protein may be sequestered into inclusion bodies once expressed in the cytoplasm. These inclusion bodies can affect the protein expression in E. coli and can also affect the solubility of the protein of interest. However, once the protocol is fully optimized, heterologous expression in  21  E. coli will be a cost effective way to isolate large quantities of protein. The account that follows describes how the protocol was optimized.  2.2 Material and Methods 2.2.1 Materials  Isopropylthio-β-galactoside (IPTG) was purchased from Invitrogen. Oligonucleotide primers were obtained from NAPS-Integrated DNA Technologies (IDT). dNTP mix, Phusion High-Fidelity DNA polymerase and buffers were brought from New England Biolabs. Thrombin protease, Glutathione Sepharose 4B, and dNTPs were from GE Healthcare. E. Coli XL1-Blue® Supercompetent Cells were from Stratagene. DNase I, RNase A, Ampicillin, and Protease inhibitor cocktail were from Sigma. B21(DE3) Competent Cells and pGEX-6P-1 expression vector were from Novagen. GeneRuler 1kb DNA ladder standards, Qiagen Plasmid Mini DNA extraction kits, QIAquick gel extraction kits and QIAquick PCR purification kits were ordered from Qiagen. Restriction enzymes and accompanying buffers, Calf Intestinal Alkaline Phosphatase and T4 ligase were from Thermo Fisher. Syringe filter units (0.22 µm and 0.45 µm) were from Millipore. Sequencing was done by UBC NAPS.  2.2.1.1 Plasmids The pGEX-6P-1 vector is a bacterial expression plasmid that uses an IPTG-inducible tac promoter. It has a N-terminal GST-tag sequence followed by a HRV 3C protease site and cloning sites. pGEX-6P-1 also has a pBR322 origin and an ampicillin resistant gene. pGEX-6P-1 was propagated in either: 1)  22  XL1-Blue subcloning strain of E. coli in LB (10 g NaCl, 10 g bacto-tryptone, 5 g yeast extract/L, pH 7.0) media supplemented with 100 µg/ml ampicillin, or 2) BL21 (DE3) expression strain of E. coli in LB media supplemented with 100 µg/ml ampicillin The pGEX-KG vector is a derivative of the pGEX vector that is used as a bacterial expression plasmid similar to the pGEX-6P-1. However, the N-terminal GST-tag is followed by a thrombin protease cleavage site and a poly-glycine linker before the cloning sites. This is suggested to aid in the protease digestion after purification of the GST-tagged protein.  2.2.1.2 E. coli BL21 (DE3) Cells BL21 (DE3) cells are used for high-level expression of bacterial proteins with expression plasmid containing the T7 promoter. The chemically competent cells have a transformation efficiency of >1 x 107 CFU/µg which offers a superior value for everyday protein expression work. Hence, BL21 (DE3) cells are commonly used for high-level expression of soluble bacterial proteins. 2.2.2 Methods 2.2.2.1 Polymerase Chain Reaction The standard molecular cloning techniques, namely polymerase chain reaction (PCR), sequencing, and plasmid construction, were used. PCR fragments were generated with Phusion High-Fidelity DNA polymerase, following NEB’s recommendations for reaction conditions, using a Bio-Rad T100TM Thermal cycler. (See Appendix A for details on PCR reactions).  23  2.2.2.2 Plasmid DNA Preparations Plasmid isolations from bacterial cells were carried out using QIAGEN plasmid extraction kit, employing the alkaline lysis method, following QIAGEN’s specific instructions. 2.2.2.3 Restriction Digests, Agarose Gels and Plasmid DNA Purification  Plasmid DNA was mixed with a variety of nucleases in the corresponding buffer (e.g. HindIII and XbaI nucleases in FastDigest buffer) and incubated at 37˚C for 2-3 h. Nuclease-digested plasmid DNA was run on 1% agarose gel (0.5 g agarose in 50 ml Tris-acetate buffer (40 mM Tris, 0.1% (v/v) glacial acetic acid, 1 mM EDTA, pH 8.0) 0.5 µg/ml ethidium bromide) at 100 volts on a Bio-Rad PowerPacTM Universal power supply for approximately 60 min (or until dye front reached the end of the gel for maximal band separation), to separate fragments of different sizes. Gels were visualized on a 360 nm long wave UV box or on a Bio-Rad Gel DocTM gel analysis instrument. Fragments of digested plasmid DNA were excised from agarose gels and purified according to instructions provided in the Invitrogen PureLink™ Quick Gel Extraction Kit. 2.2.2.4 Ethanol Precipitation of DNA To a sample size of 100µl, 2µl of linear polyacrylamide carrier (5 mg/mL), 12.5 µl of 8 M NH4Ac, and 200 µl of 95% EtOH on ice was added, and then quickly vortexed and incubated at -20°C for 60 min. Samples were centrifuged at 12470 x g, at 4°C for 30 min. The supernatant was carefully aspirated from the  24  sample and discarded. The DNA pellet was washed with ice cold 75% EtOH to remove the salt in the sample. After 10 min on ice, the sample was centrifuged for 5 min as above. The supernatant was discarded and the pellet washed with ice cold anhydrous EtOH before centrifuging for 5 min. The supernatant was discarded and the pellet was air dried at room temperature until all traces of EtOH had evaporated. The DNA pellet was resuspended in 10 mM Tris pH 7.4 or ddH2O. 2.2.2.5 Phosphatase Treatment of Vector  The digested vector was mixed with one unit of calf intestinal alkaline phosphatase (CIAP) per picomole of 5’ ends in dephosphorylation buffer (50 mM Tris-HCl, 1 mM EDTA, pH 8.5) in a final volume of 10 µl. The reaction was carried out for 1h at 37°C and then deactivated by incubation at 65°C for 15 min. 2.2.2.6 Ligations Ligation reactions were performed with a 3:1 molar ratio of insert DNA to Sap prepared vector in Thermo Fisher 10X T4 DNA Ligase Buffer (final 1X: 40 mM Tris-HCl pH 7.8, 10 mM MgCl2, 0.5mM ATP, 10 mM DTT) with 5 units of T4 DNA ligase in a final volume of 20 µL. The Thermo Fisher Ligation Kit was used as per the manufacturer instructions, with a 1h incubation period at room temperature (RT). 2.2.2.7 Competent Cells Transformation Competent cells (50 µl) were thawed on ice for 10 min prior to the gentle addition of 1.0 µg of plasmid DNA or 5 µL ligation product, and were then  25  incubated on ice for 20 min. The cells were heat-shocked in a 42°C water bath for 45 sec and transferred to ice for 2 min. 0.5 ml of pre-warmed LB media was added and the culture was shaken at 225 rpm for 1 h at 37°C. 100 and 400 µls of transfected cells were plated on LB / agar plates containing the appropriate antibiotics, and grown overnight at 37°C. 2.2.2.8 Preparation of pGEX-KG-TDP-43     The full-length (FL) TDP-43 was provided as a pGEX-6P-1 vector from Dr. Jean-Pierre Julien (Université Laval). The FL TDP-43 was transferred to the pGEX-KG vector that would have a poly-glycine linker between the GST-tag/thrombin cut site and the protein of interest. A PCR fragment of TDP-43 was generated using pGEX-6P-1-TDP-43 as the template. XbaI-TDP43-5F-mod and TDP43-HindIII-3R were used as 5’ and 3’ primers respectively (See Appendix B for primers details). The 5’ primer contained a XbaI site and the 3’ primer contained a HindIII site. The PCR product was digested with XbaI and HindIII to form a fragment ~1200 bp long, and was ligated into the ~5000 bp fragment of pGEX-KG-Fyn-SH3, which had also been digested with XbaI and HindIII (Figure 2-1).   26  TDP-43 pGEX-6P-1-TDP-43  TDP-43-HindIII primer XbaI-TDP-43-5F-mod primer Fyn-SH3 pGEX-KG-Fyn-SH3 TDP-43 XbaIII HindIII 1200bp PCR reaction XbaI GST-tag Thrombin cut site HindIII XbaI/HindIII RE digest Extract 5000bp fragment Ligation of  1200 bp pGEX-6P-1-TDP43 fragment  and 5000bp pGEX-KG-Fyn-SH3 fragment TDP-43 pGEX-KG-TDP-43 GST-tag Thrombin cut site XbaI/HindIII RE digest  Figure 2-1  Construction of pGEX-KG-TDP-43 Full length TDP-43 were excised from the original pGEX-6P-1 vector and ligated into the pGEX-KG vector that houses a poly-glycine linker   27  2.2.2.9 Expression of Protein via IPTG Induction After successfully obtaining colonies from pGEX-transformed competent BL21 (DE3) cells, 5 mL liquid culture was prepared by inoculation with a single bacterial colony. The culture medium was LB with 100 µg/ml ampicillin. After incubating at 37˚C for 8 h with shaking at 225 rpm, 0.5 mL of this starter culture was the inoculum for a 10 mL LB / Amp media. After an overnight growth at the same conditions, this 10 mL O/N culture was the inoculum for a 1 L LB / Amp media. When this culture had reached 0.5-0.7 at an optical density of 595nm, the expression of the transgene was induced with IPTG with 225 rpm shaking incubation at various temperature and times as discussed in section 2.3 2.2.2.10 GST-tagged Protein Purification After induction with IPTG, a 1 L liquid cell culture was centrifuged at 3000 x g for 10 min. All of the supernatant was aspirated and the cell pellet was either stored at -80˚C or immediately resuspended in 50 ml of lysis buffer  (PBS pH 7.4, 1 mM DTT, 1% Triton X-100, 4 ug/mL DNase I, and Sigma protease inhibitor cocktail as per the manufacturer instructions). The sample was then sonicated on ice with cycle of 5 sec sonication, 10 sec pause for 2 min in total with a Branson Digital Sonifier and 13mm stepped horn with threaded body at 37% amplitude output. This whole cell homogenate was then centrifuged at 16,000 x g for 30 min to separate the supernatant and the pellet fraction. The supernatant was filtered by passing it through a syringe with a Millipore 0.45µm filter attached. A 1 mL bed volume of Glutathione Sepharose 4B beads were equilibrated in a 15 mL Falcon tube using binding buffer (PBS pH 7.4, 1 mM DTT, 1% Triton X-100) as per the  28  manufacturer instructions. The supernatant was combined with the glutathione sepharose beads and transferred into a 50 mL Falcon tube. The resulting mixture was rotated end-over-end for 2 h at RT to ensure complete binding of GST-tagged protein to the glutathione sepharose beads. After binding, the bead/supernatant mixture was centrifuged at 500 x g for 5 min and the supernatant was collected as the flow through sample. The beads were washed for a total of six times: first three washes with 5 mL binding buffer each time (W1a-c), then three washes (W2a-c) with 5 mL wash buffer each time (PBS pH 7.4, 1 mM DTT) in order to remove as much detergent as possible. A sample of each fraction was run on a 10% SDS-PAGE to estimate the proteins obtained in each step. 2.2.2.11 HRV 3C Digestion HRV 3C protease (100U) was mixed with 1.8 mL wash buffer and 0.2 mL 10X Reaction Buffer. This mixture was then added directly to the glutathione sepharose beads for a batch mode cleavage and purification. HRV 3C digestion was carried out with end-over-end rotation at 4 ˚C for 18 h. The HRV 3C recognition sequence is LEVLFQGP where the protease will cleave between Gln and Gly residues. The sample was centrifuged at 500 x g for 5 minutes and the supernatant consisting of the untagged protein of interest was recovered, along with the HRV 3C used. TDP-43 and RRM-1 expressed from the pGEX-6P-1 plasmid would carry an extra five amino acids (GPLGS) at the N-terminus after HRV 3C digestion.  29  2.2.2.12 Thrombin Digestion Thrombin (100U) was diluted with 0.9 mL PBS and added directly to the glutathione sepharose beads for batch mode cleavage and purification. Thrombin digestion was carried out with end-over-end rotation at RT for 18 h. The thrombin cut site amino acid sequence is LVPRGS where the thrombin will digest after the arginine residue. The sample was centrifuged at 500 x g for 5 min and the supernatant consisting of the untagged protein of interest was recovered, along with the thrombin. TDP-43 expressed from the pGEX-KG plasmid would carry an N-termini linker composed of fourteen amino acids (GSPGISGGGGGILD) after thrombin digestion. 2.2.2.13 Gel Electrophoresis Proteins were separated on 10% SDS-polyacrylamide gels using Bio-Rad’s Mini-PROTEAN II gel apparatus at an initial 75V in the stacking gel and  150 V in the separating gel, for a total of ~75 min in Running Buffer (25 mM Tris pH 8.3, 192 mM glycine, 0.1% (w/v) SDS) [56]. After electrophoresis, gels were fixed in Coomassie stain (0.3% (w/v) Coomassie Brilliant Blue R, 45% (v/v) MeOH, 10% (v/v) HAc) for at least 1 h, destained for 10-15 min in Fast Destain (40% (v/v) MeOH, 10% (v/v) HAc), and followed by several hours in Slow Destain (7% (v/v) MeOH, 5% (v/v) HAc). The proportions of aliquots loaded onto the SDS-polyacrylamide gel are as follows: homogenate, supernatant, pellet, flow-through, and wash fractions were 0.05% of total volume; thrombin digestion fractions were 0.1% of total volume.   30  2.3 Results and Discussion 2.3.1 Expression of N-terminal GST-tagged TDP-43 RRM-1 In my thesis work, the construct encompassing the RNA recognition motif-1 domain of TDP-43 (residues 42-176) are referred to as RRM-1. This original pGEX construct was provided by the group of Dr. Jean-Pierre Julien, who is the Chief Scientific Officer at ImStar Therapeutics. The initial trial expression utilized a GST-tagged TDP-43 RRM-1 construct, pGEX-6P-1-TDP43-RRM1 (provided by Dr. Jean-Pierre Julien). The construct was transformed into BL21 (DE3) strain of E. coli. Expression of this vector was driven by the tac promoter located upstream of the N-terminal GST-tag sequence. The GST-tag is followed by a HRV 3C protease cleavage sequence upstream of the TDP-43 RRM-1 protein sequence. The tag and any potential interference can be removed following purification via affinity chromatography The temperature conditions tested for BL21 (DE3) were 18°C and 37°C. The recommended temperature for E. coli growth is at 37°C. Expression of this recombinant protein, GST-RRM-1 (42.2kDa), at 37°C (induced with 0.5 mM IPTG for 4 h) resulted in appreciable yield. However, upon lysis and centrifugation to separate the soluble and insoluble fractions, GST-RRM-1 was found predominately insoluble, making protein purification difficult (Figure 2-2a). In an attempt to raise yields of properly folded soluble proteins, inducing protein expression at a lower temperature, namely 18°C, was trialed (induced with   31  0.5 mM IPTG for 20 h). Even though the expression level was comparable to the previous method, lowering the temperature did not aid in the solubility of this protein (Figure 2-2b).   Figure 2-2  Expression of GST-RRM-1 (A) Expression of GST-RRM-1 inducing with 0.5 mM IPTG for 4 hr at 37 °C in E.coli BL21 (DE3). GST-RRM-1 (#) is over-expressed in the homogenate fraction (H) as compared to the pre-IPTG induced homogenate fraction (PH). After cell lysis, GST-RRM-1 is predominately in the pellet fraction (P) and negligible in the soluble fraction (S). (B) Expression of GST-RRM-1 inducing with 0.5 mM IPTG for 20 h at  18 °C in E.coli BL21 (DE3). Again here GST-RRM-1 (#) is predominately found in the pellet fraction (P). While looking for further ways to optimize the protocol, we have identified that the construct for RRM-1 provided was incorrect. In literature, the RRM-1 should only encompass residues 101-191 of TDP-43 [57]. However, the construct provided comprises of residues 42 to 176 of TDP-43, which extends the N-terminal portion (which is less structures) and deletes residues in the C-terminus. Due to the sequence error and the difficulty in expressing this  32  construct, the protocol optimization was halted and focus was placed on the expression and purification of the full length TDP-43.   2.3.2 Expression and Purification of N-terminal GST-tagged TDP-43 The full length version of the TDP-43 constructs composed of residues 1-414 are referred to as TDP-43. The initial trial utilized a GST-tagged TDP-43 wild-type construct, pGEX-6P-1-TDP-43 (provided by Dr. Jean-Pierre Julien). It was first transformed into the BL21 (DE3) strain of E. coli. The expression system and the construct arrangement are identical to that of pGEX-6P-1-TDP43-RRM1. Over-expression of this recombinant protein, GST-TDP-43, was observed when the cells were induced with 0.5 mM IPTG for 20 hr at 18 °C. However, the GST-TDP-43 protein was mainly in insoluble form with only approximately 20% being in the supernatant fraction (Figure 2-3a) after being lysed in PBS containing 1% Triton X-100 via sonication. The experiment was carried out with GST affinity chromatography but resulted with most of the GST-TDP-43 being eluted in the flow-through fraction. Only a small percentage of the recombinant protein was bound onto the glutathione sepharose which was then washed with PBS and cleaved with on-column Human rhinovirus 3C (HRV3 C) protease cleavage to release pure TDP-43 (45.1kDa) (Figure 2-3b). Furthermore, optimization towards the over-expression of this protein was done extensively. By reducing IPTG concentration to 0.3 mM and changing the induction time to 16 hr at 16 °C, the protein was over-expressed and the solubility was found to improve (Figure 2-3c). However, the issue with GST-TDP-43 not binding onto glutathione sepharose limited the overall yield of TDP-43 (Figure 2-3d). Additionally, the molecular weight  33  of HRV 3C enzyme used to cleave TDP-43 from the GST-tag is 47.8 kDa, which is relatively similar to the molecular weight of TDP-43. Hence it would be difficult to further separate the two proteins to obtain pure TDP-43.   Figure 2-3  Expression and Purification of GST-TDP-43 (A) Expression and Purification of GST-TDP-43 inducing with 0.5 mM IPTG for 20 hr at 18 °C in E.coli BL21 (DE3). GST-TDP-43 (#) is over-expressed in the homogenate fraction (H) as compared to the pre-IPTG induced homogenate fraction (fraction not shown). After cell lysis, GST-TDP-43 is predominately in the pellet fraction (P) and only approximately 20% remained in the soluble fraction (S). The soluble fraction was filtered through a 0.45 µm filter to remove any cell debris (FS). The FS fraction was loaded onto glutathione sepharose column but most of the GST-TDP-43 eluted in the flow-through (FT) and wash (W) fractions. An on-column HRV 3C cleavage was performed and TDP-43 (*; 45.1kDa) was obtained in the HRV 3C cleaved FT. (B) A mixture of TDP-43 (majority) and HRV 3C (minority) was obtained after on-column cleavage via HRV 3C protease. (C) Inducing with 0.3 mM IPTG for 16 hr at 16 °C helped GST-TDP-43 become more soluble as more was found in the supernatant fraction. (D) GST-TDP-43 extracted from condition mentioned in (C) did not enhance binding between the recombinant protein and glutathione sepharose.    34  2.3.3 Expression and Purification of N-terminal GST-tagged TDP-43 with Glycine Linker Between the GST tag and TDP-43 Since there was minimal binding of GST-TDP-43 onto the glutathione sepharose as indicated in section 2.3.2, it was hypothesized to be due to TDP-43 being closely linked to GST and affecting the GST-tag binding affinity. Hence, the pGEX-KG plasmid, which consists of a glycine linker (five additional glycine residues) following the GST-tag and thrombin cleavage sequence, was used. The use of this plasmid also introduced the use of a thrombin cleavage site to replace the HRV 3C protease cleavage sequence to avoid the use of HRV 3C (see section 2.3.2). The approach was to perform standard PCR using the plasmid pGEX-6p-1-TDP-43 as mentioned above (sections 2.2.2.1 and 2.2.2.8) to obtain a 1.2 kbp fragment which encompasses the TDP-43 coding sequence to be subsequently inserted into the pGEX-KG plasmid. The recombinant protein, denoted as GST-gly-TDP-43, was overexpressed and had good solubility in E.coli BL21 (DE3) strain when the cells were induced with 0.3 mM IPTG for 16 hr at   16 °C (Figure 2-4). Upon addition of the lysate supernatant onto the glutathione sepharose medium, this version of the GST-tagged protein (GST-gly-TDP-43) was more susceptible to bind onto the beads, as compared to GST-TDP-43 discussed in section 2.3.2. This result indicated that the poly-glycine linker which lengthens the distance between the GST-tag and TDP-43 had a positive effect on binding. The protein and sepharose mixture were then washed three times with binding buffer which includes detergent to remove any non-specific binding. Then, the mixture was washed three times with wash buffer that lacks detergent to remove Triton X-100 since its presence may have an impact on thrombin  35  digestion. As shown in Figure 2-4, minimal quantities of GST-gly-TDP-43 were eluted during the wash fractions. Upon thrombin cleavage, free TDP-43 (46 kDa) was observed in the flow-through along with lower molecular weight bands which may represent the thrombin (37 kDa) and the free GST tag (25 kDa).    Figure 2-4  Expression and Purification of GST-gly-TDP-43 Protein expression induced with 0.3 mM IPTG for 16 hr at 16 °C in E.coli BL21 (DE3). GST-gly-TDP-43 (#) is over-expressed in the homogenate fraction (H) as compared to the pre-IPTG induced homogenate fraction (fraction not shown). After cell lysis, GST-gly-TDP-43 is equally distributed in the pellet fraction (P) and the supernatant fraction (S). The soluble fraction was filtered through a 0.45 µm filter to remove any cell debris (FS). The FS fraction was loaded onto glutathione sepharose column with minimal GST-gly-TDP-43 eluted in the flow-through (FT) and wash (W) fractions. An on-column thrombin cleavage was performed and TDP-43 (*; 45.9kDa) was obtained in the thrombin cleaved FT.  36  2.4 Conclusion The work presented in this chapter has demonstrated that TDP-43 can be expressed and purified from an E. coli system, albeit at present still in low yields. Nevertheless, strained metabolic conditions such as low temperature and low concentration of IPTG induction were found to enhance yields. The methodology here may serve as a foundation for further optimized protocols when working with highly insoluble proteins. However, there are potential pitfalls in the use of the nonionic detergent Triton X-100 which should be addressed, such as its inability to dialyze out of samples due to its low critical micelle concentration and the interference during light scattering and NMR spectroscopy. However, with appropriate methods of detergent removal such as Thermo Scientific Pierce Detergent Removal Resin, or isolating the protein via acetone precipitation, the amount of Triton X-100 present can be minimized in the final solution. The latest working protocol for TDP-43 could serve as a basis for the expression and purification of the correct version of the RRM-1 protein and also for CysSer mutants for TDP-43.   37  CHAPTER 3: DEVELOPMENT OF A SOLID PHASE PEPTIDE SYNTHESIS PROTOCOL FOR THE EIGHT P65 PEPTIDES  3.1 Introduction  The other objective of this thesis was to produce the eight different peptides from the p65 subunit of NF-κB that were identified to interact with TDP-43 (Table 3-1). Since subsequent interaction studies may require large quantities of these peptides, an in-house approach using solid phase peptide synthesis (SPPS) using Fmoc chemistry was carried out to help reduce cost. The potential challenges when preparing peptides are the various shortened variants produced in the crude products and the purification of the peptides using reverse phase HPLC due to their low hydrophobicities. The unique sequences of the eight peptides affect the potential challenges differently and must be addressed individually. However, once the workup is optimized for each peptide, there will be small margin of error for subsequent peptide synthesis and purification. The account that follows describes the trial and error to produce some of these peptides. One important note is that these sequences were provided by Dr. Jean-Pierre Julien/ Dr. William Jie’s laboratory, and the initial sequence of NF1 was incorrect so will be designated as NF1-error (PGRRSTDT) (section 1.2.5).    38  Table 3-1  Eight peptides from p65 subunit of NF-κB that were found to interact with TDP-43. The full length sequence of human p65 can be found in Uniprot entry Q04206. Name Sequence Residues on p65 NF1 PGERSTDT 47-54 NF2 DLEQAISQRIQT 125-136 NF3 EEKRKRTY 299-306 NF4 PFSGPTDPRP 317-326 NF5 APQPYPFT 345-352 NF6 ISQASALA 373-380 NF7 LLQLQFDDED 439-448 NF8 RLVTGAQRPP 501-510  3.2 Material and Methods 3.2.1 Materials O-Benzotriazole-N,N,N’,N’-tetramethyl-uronium-hexafluoro-phosphate (HBTU, >98%), N,N-Dimethylformamide (DMF), dichloromethane (DCM, ACS reagent 99.5%), piperidine (biotech grade 99.5%), N,N-Diisopropylethylamine (DIEA, ReagentPlus 99%), trifluoroacetic acid (TFA, ReagentPlus 99%), 1,2-Ethanedithiol (EDT, 98% GC), Triethylsilane (TES, 97%), and acetonitrile (LC-MS Ultra CHROMASOLV® grade, >99.9%) were purchased from Sigma-Aldrich. Methanol and ethyl ether were from Fisher Scientific. Various Fmoc protected amino acids used in this thesis were purchased from different companies including AnaSpec Inc., Peptides International, PerSeptive Biosystem Inc., and Advanced ChemTech. Wang resin and preloaded Fmoc amino acid Wang resins were purchased from Advanced ChemTech.   39  3.2.1 Methods 3.2.1.1 Solid Phase Peptide Synthesis The different NF-κB p65 peptides were synthesized (using Fmoc chemistry) using a CS135XT solid phase peptide synthesizer from CS Bio Co. For many of the syntheses, Wang resin was purchased with the first residue at the C-terminus pre-loaded to improve the yield of the C-terminally carboxylated peptides. The appropriate amount of resin corresponding to 0.25 mmol was weighed and added into the reaction vessel to soak. Other amino acids were prepared by making 0.2 M amino acid stocks in 40 mL of Solution A (0.2 M HBTU in 500 mL DMF). Approximately 5 mL of the amino acid solutions are needed per residue for each coupling step. The amino acids were added step by step to the resin via single coupling of 2 hr each (except for the proline residues which double coupled for 2 hr each). For peptides that are difficult to synthesize, double coupling is preferred such that in case the first coupling is not complete, a quick wash and drain of the free amino acid and old buffer followed by a second coupling step of the same amino acid with fresh buffer will be performed. The synthesis was completed with a final de-blocking step. The resin was then transferred into a reaction funnel and put into a vacuum desiccator to dry overnight.  3.2.1.2 Peptide Cleavage from Resin and De-protection of Side Chains A cleavage mixture (23.60 mL TFA, 0.125 mL TES, 1.25 mL ddH2O, and 1.25 mL EDT) was prepared and added to the reaction vessel containing the resin. The mixture was stirred for 5 hours at moderate speed.  40  3.2.1.3 Peptide Precipitation The solution from section 3.2.2.2 above was filtered and transferred into a 100 mL round bottom flask. The filtrate contains the peptide and the retentate contains the resin. The reaction funnel was washed three times with 5 mL of TFA and the filtrates added to the same 100 mL round bottom flask. The volume of the solution was reduced by rotary evaporation (Büchi Rotavapor R-200 and Heating Bath B-490) until only oily droplets remained. The round bottom flask was placed in an ice bath and cold diethyl ether was added drop-wise for the first 20 mL, and then pipette-wise until a total of 100 mL was added. The precipitate was filtered using a vacuum suction sinister glass crucible unit and washed with cold diethyl ether. The precipitate (peptide) was transferred into a 50 mL Falcon tube and dissolved in 30 mL of ddH2O via water bath sonication. The solution was store in a freezer overnight and then lyophilized to obtain the crude product form of peptide. The identity of the crude peptide was confirmed by MALDI-time of flight mass spectrometry (MALDI-ToF MS). 3.2.1.4 High Performance Liquid Chromatography The crude products were purified by preparative reverse phase high performance liquid chromatography on a Waters 600 system with a 229 nm UV detector (Waters 2996 Photodiode Array Detector) and using either a Phenomenex C4 preparative column (10 µm, 2.1 cm x 25.0 cm) or a C18 preparative column (10 µm, 2.1 cm x 25.0 cm). The mobile phase was composed of two buffers (A and B) with gradient flow. Buffer A was composed of 90% ddH20,  41  10% acetonitrile, 0.1% TFA and buffer B was made up of 10% ddH2O, 90% acetonitrile, 0.1% TFA.   3.3 Results and Discussion 3.3.1 Development of Peptide Synthesis Protocol Initial attempts in preparing the peptides utilized the unloaded Wang resin. Hence, the C-terminal amino acid of each peptide was loaded onto the Wang resin as the first step. To help ensure the first amino acid is loaded onto the resin, a double coupling step (24 hr for each coupling step) was used; without a second coupling reaction, a larger mixture of products was found. This helped maximize the probability that the resin was loaded with the desired C-terminal amino acid. Next, the Fmoc protecting group for the amino groups that form the peptide bond was removed. Then the second amino acid carboxy termini was coupled onto the amino group of the first amino acid linked to the resin via a 2 hr single coupling reaction (except for proline residues which required doubling coupling of 2 hr each). The removal of the Fmoc protecting group and the addition of new amino acids were repeated until the full peptide sequence was completed. However, when the crude product of the prepared peptides (NF1-error, NF2, NF3, and NF8) was examined using MALDI-ToF MS, it contained the full length peptide (*) and a number of shortened variants (#). A number of these shorter variants were clearly the result of poor coupling between the desired C-terminal amino acid and the Wang resin (Figure 3-1, Figure 3-2, Table 3-1).   42   Figure 3-1  MALDI-ToF MS analysis of NF1-error and NF2 crude peptides prepared via SPPS The expected peptides are denoted as (*) whereas the shortened variants are denoted as (#). All other MW from MS analysis which are not specified could not be identified. The expected sequences for (A) NF1-error, (B) NF2 and their predicted shortened sequences are listed in Table 3-2.    43   Figure 3-2  MALDI-ToF MS analysis of NF3 and NF8 crude peptides prepared via SPPS The expected peptides are denoted as (*) whereas the shortened variants are denoted as (#). All other MW from MS analysis which are not specified could not be identified. The expected sequences for (A) NF3, and (B) NF8 and their predicted shortened sequences are listed in Table 3-2.   44  Table 3-2  List of p65 peptides made using solid phase peptide synthesis Peptide Name Sequence Theoretical MW MALDI-ToF MS (positive ion mode) MW Predicted Sequences NF1-error PGRRSTDT 888.94 889.7 788.6 572.5 485.4 329.3 PGRRSTDT PGRRSTD PGRRS PGRR PGR NF2 DLEQAISQRIQT 1401.54 1401.7 1300.7 1172.6 1059.5 DLEQAISQRIQT DLEQAISQRIQ DLEQAISQRI DLEQAISQR NF3 EEKRKRTY 1109.25 1109.7 EEKRKRTY NF8 RLVTGAQRPP 1094.28 1094.7 997.6 900.5 744.4 616.4 RLVTGAQRPP RLVTGAQRP RLVTGAQR RLVTGAQ RLVTGA Preloaded NF1-error PGRRSTDT 888.94 889.4 PGRRSTDT  For example, looking at peptide NF1-error (PGRRSTDT), there are 4 shortened variants shown in the mass spectrum, where threonine was to couple onto the Wang resin via in-house SPPS (Figure 3-1a). However, when using purchased threonine preloaded Wang resin, the synthesis produced only the anticipated NF1-error peptide (Figure 3-3). This difference suggests that for the sequences studied, coupling of amino acid onto the resin can be difficult. Hence, amino acid preloaded resins should be used in subsequent peptide synthesis to minimize contamination from truncated products.    45   Figure 3-3  MALDI-ToF MS analysis of NF1-error peptide crude product made by SPPS using threonine preloaded Wang resin The use of threonine preloaded Wang resin generated only the expected PGRRSTDT peptide sequence (*) with no lower MW shortened variants. The higher MW impurities could not be identified.  3.3.2 Reverse Phase HPLC Purification After obtaining the crude products from SPPS, the desired peptides needed to be purified to remove any impurities. The purification strategy exploited was by reverse phase high performance liquid chromatography (RP-HPLC), which is the most widely used HPLC method. Here, solute retention is mainly due to hydrophobic interactions between the solute and the hydrocarbon stationary phase surface. A polar mobile phase, usually water mixed with acetonitrile is used for elution, such that solutes elute in the order of increasing hydrophobicity.  However, when the four crude products made from SPPS (NF1-error, NF2, NF3, and NF8) were subjected to RP-HPLC, the peptides were eluted in the void volume when trialed with both the C4 and the C18 column (NF3 RP-HPLC data  46  shown in Figure 3-4; other peptide RP-HPLC data not shown). This result suggested that the peptides produced were hydrophilic. This was supported by the grand average of hydropathicity (GRAVY) index indicating the calculated hydropathy values of the peptides to be negative, indicating that they are polar (Table 3-3). Hence, modification on the RP-HPLC method should be carried out in the future, possibly experimenting on various mixtures of polar mobile phase used for gradient elution or utilizing a normal phase HPLC approach.  Figure 3-4  RP-HPLC purification of SPPS prepared crude NF3 peptide  Analysis of peptide purification using RP-HPLC utilizing a gradient elution showed that NF3 eluted from the (A) C4 column and (B) C18 column in the void volume. Insets show minor peaks in the HPLC traces. Table 3-3  GRAVY indices of the eight original p65 peptides and the NF1-error peptide p65 peptide name Sequence GRAVY index NF1-error PGRRSTDT -2.087 NF1 PGERSTDT -1.962 NF2 DLEQAISQRIQT -0.742 NF3 EEKRKRTY -3.225 NF4 PFSGPTDPRP -1.350 NF5 APQPYPFT -0.713 NF6 ISQASALA -1.075 NF7 LLQLQFDDED -0.680 NF8 RLVTGAQRPP -0.700   47  3.4 Conclusion The work presented in this chapter regarding the synthesis of NF-κB p65 peptides demonstrated that an in-house approach is feasible to generate large quantity of peptides, yet many obstacles still need to be tackled. The use of amino acid preloaded Wang resin was found to be essential for making these peptides by Fmoc SPPS. The advantage of this pre-loaded resin was evidenced by the nearly pure crude samples obtained, which contained mainly the anticipated peptide of interest. However, due to the unique sequences of each peptide, the methodology will need to be adapted for each individual peptide, especially the ones with multiple proline residues in the sequence (due to its cyclic and therefore relatively rigid ‘side chain’ leading to inefficient coupling reaction). Moreover, the purification of these p65 peptides was not feasible with the use of RP-HPLC utilizing a gradient elution method. The several peptides where purification was attempted all eluted in the void volume indicating they do not have enough hydrophobic moieties to bind onto the C4 and C18 columns. These trials suggest that in future, a C8 column should be tried. This column may be better at capturing small hydrophilic peptides. Alternatively, a completely different system such as size exclusion chromatography may be useful.     48  CHAPTER 4: MOLECULAR DYNAMIC SIMULATIONS 4.1 Introduction In this thesis, MD simulation was used to look at the two different models by which TDP-43 may lead to ALS: 1) the aggregation model and 2) the deregulation hypothesis. These have been described in detail in section 1.2.4. and 1.2.5, respectively. In the following, MD will be used to shed light on these hypotheses.    First, MD will be used to determine if there are any differences between WT TDP-43 RRM-1 and RRM-1 where the two cysteine residues (Cys173 and Cys175) are mutated into serine residues. With the cysteine residues substituted with serine residues, it is expected that any bonds associated with the side chain atoms of residues 173/175 would change and hence the secondary structure of its nearby region may alter. This may consequently affect the conformation of RRM-1 causing it to be more rigid or vice versa. Hence, MD simulation would be able to determine not only the conformation of the different RRM-1 variants (WT and mutants) but also the effect towards the flexibility of the domain due to residue substitution. In a second instance, MD simulations were used to examine the similarities and differences between the NF-κB p65 peptides; in particular the NF1 peptide (AGSIPGERSTDTTK; residues 43-56) and NF2 peptide (KRDLEQAISQRIQT; residues 123-136) since their secondary structures are known (Figure 1-7). As the  49  native secondary structures of NF1 and NF2 are identified to be a loop and a helix respectively, it will be interesting to determine if structure plays a role in the interaction between TDP-43 RRM-1 and p65 peptides. 4.2 General Methods All simulations were performed using the GROMOS96 [58], [59] biomolecular simulation package and the 53A6 force field [60], [61]. The various proteins were solvated in explicit SPC water [62] or methanol [63], as indicated in the detailed sections below. Rectangular periodic boundary conditions were imposed. Simulations were performed in the NPT ensemble (T = 300 K, P = 1 atm) using the Berendsen weak coupling methods [64]. Covalent bonds were constrained using the SHAKE method [65], with a relative geometric tolerance of 10-4 . A reaction field [66] long-range correction to the truncated Coulomb potential was applied. A twin-range cut-off of 0.8/1.4nm was used for all non-bonded interactions.  There were several major steps in preparing for a molecular dynamics simulations: 1) Finding the coordinates (experimental) of the protein/peptides atoms and determine if there were atoms missing or if multiple occupancies occurred. The protein database (PDB) [67] was the main source of protein structures. 2) Placing the protein into a periodic box. Solvate the protein/peptide by fitting solvent molecules around the protein/peptide. 3) Minimizing the overall energy of the system by adjusting the position of the solvent molecules. The solute atom coordinates were positionally restrained.  50  4) If the total charge of the system was non-zero, ions were added to neutralize it. This was achieved by replacing a solvent molecule with a charged ion. Re-minimization was necessary after this substitution. 5) Assign initial velocities (υ) to the individual atom via the Maxwell-Boltzmann distribution:  where m is the particle mass and kT is the product of the Boltzmann’s constant and thermodynamic temperature. 6) Equilibrating to achieve the required, steady temperature and pressure of the molecular system. The solute atom coordinates were positionally restrained 7) Simulating the system, collecting coordinates and energies for later analysis. 4.2.1 Scalar Atomic B-Factors Analysis The atomic scalar B-factors are a measure of an atom’s mobility. This can be calculated from a computer simulation from the mean square fluctuation of the atom’s position:   after the structures have been superimposed. 4.2.2 Clustering Analysis We applied the concept of data clustering to the analysis of the coordinate trajectories. The general idea behind data clustering is to reduce a large complex  51  data set, here consisting of atomic coordinates as a function of time, into a sequence of subsets each containing structurally similar structures. Here, all clustering analyses were performed using an algorithm previously developed for this purpose [68]–[71] and a similarity measure based on Cα atoms. The algorithm orders the clusters from zero onwards in decreasing size. Therefore, cluster zero represents the dominant set of structures observed during the course of the simulation. From a given clustering, the entropic contribution to the free energy difference between any two clusters i and j of size Ni and Nj can be calculated from:   where kB is the Boltzmann constant and T is the equilibrium temperature of the system in Kelvin. Here, the method is also applied to trajectories from different simulations of the same protein in order to see whether the same structures occur in the different simulations 4.2.3 Hydrogen Bonding Analysis .The strength of a hydrogen bond is assessed by the percentage of simulated time that the following hydrogen bond criterion is met: the distance between the heavy atoms must be 3Å or less and the angle between the donor atom-proton-acceptor atom must be 135° or more (Figure 4-1).  52   Figure 4-1  Hydrogen bonding analysis A hydrogen bond is considered to exist if the following conditions are met. The heavy atoms must be (A) 3 angstroms or less apart and (B) the angle (α) between the donor atom – proton – acceptor atom must be greater or equal to 135°. 4.3 TDP-43 RRM-1 Simulations As mentioned above, in this thesis, MD simulation was used to look at differences between WT TDP-43 RRM-1 and when the two cysteine residues (Cys173 and Cys175) in RRM-1 were mutated into serine residues. 4.3.1 Methods The coordinates for each TDP-43 RRM-1 variant were taken from the PDB file model of 2CQG shown in Figure 4-2 [72]. The single and double CysSer mutants were introduced using UCSF Chimera [32], resulting in four different simulations (Table 4-1). These simulations were done in water only at 1 atm. Equilibration time was 1 ns and simulation time was 14 ns.   53   Figure 4-2  The initial structures of Wild-Type and single/double mutants of TDP-43-RRM-1 (A) WT-TDP-43-RRM-1 with Cys-173 and Cys-175 denoted as RRM1-CC. (B) Single mutant TDP-43-RRM-1 with Cys-173 and Ser-175 denoted as RRM1-CS. (C) Single mutant TDP-43-RRM-1 with Ser-173 and Cys-175 denoted as RRM1-SC. (D) Double mutant TDP-43-RRM-1 with Ser-173 and Ser-175 denoted as RRM1-SS.    54  Table 4-1  Overview of the four TDP-43-RRM1 variants  Please refer to Figure 4-2 for notation Simulation Name Residue 173 Residue 175 Comment RRM1-CC Cysteine Cysteine Wild type RRM1-CS Cysteine  Serine Single mutant RRM1-SC Serine Cysteine Single mutant RRM1-SS Serine Serine Double mutant  4.3.2 Results and Discussion As shown in Table 4-1, four simulations of the TDP-43-RRM1 variants, each of 14 ns in length after equilibration, were performed. The four starting structures are shown in Figure 4-2.  The Cα isotropic B-factors of the four systems indicate that the wild-type RRM1-CC has a higher mobility between residues 178-183 compared to its serine substituted variants (Figure 4-3). This may suggest that the two CysSer mutations at residues 173 and 175 are likely to accentuate aggregation in the nearby region due to its reduction in freedom of movement. This correlates with the in vivo experimental results that examined C/S substitution leading to increase cytoplasmic and nuclear inclusions of TDP-43 [44]. This also correlates well with the in vitro NMR study looking at high pressure treatment as the stress factor on RRM-1 [44]. Transient high pressure treatment showed irreversible chemical shifts in region C (Figure 1-6) which contained Cys173 and Cys175.     55   Figure 4-3  The calculated Cα Isotropic B-factors of the respective TDP-43 RRM-1 variants  Please refer to Figure 4-2 and Table 4-1 for name designation. The four different TDP-43 RRM-1s are listed as follows: (Red) RRM1-CC, (Green) RRM1-CS, (Blue) RRM1-SC, and (Purple) RRM1-SS. The inset shows residues 170-190, where the most important data lies between residues 178-183.  In order to detect any structural similarity in the structures observed during the course of the four simulations, a distance matrix clustering analysis was performed (Table 4-2). The data indicates that while all variants adopt cluster-0 conformation, the RRM1-SS mutant occurs in cluster-0 almost exclusively, indicating that cluster-0 exhibits the most rigid conformation. This result suggests that the RRM1-SS mutant conformation induces/ accelerates aggregation; other mutants aggregate slower because they adopt the cluster-0 conformation less often. This is supported by the structural analysis of cluster-0 as its representative  56  structure is stabilized by a hydrogen bond between Asn-179 and Gln-182 (Figure 4-4). Table 4-2  Combined clustering results incorporating conformations from all four TDP-43-RRM1 variants simulations.  N denotes cluster, cum denotes cumulative, C denotes cysteine, and S denotes serine. Refer to Figure 4-7 for notation. N # Size N % Cum.% RRM1- CC % RRM1- CS % RRM1- SC % RRM1- SS % 0 26 46.4 46.4 34.6 11.5 3.8 50.0 1 10 17.9 64.3 0.0 0.0 100.0 0.0 2 5 8.9 73.2 20.0 80.0 0.0 0.0 3 4 7.1 80.4 25.0 25.0 25.0 25.0 4 3 5.4 85.7 100.0 0.0 0.0 0.0 5 2 3.6 89.3 0.0 100.0 0.0 0.0 6 2 3.6 92.9 0.0 0.0 100.0 0.0 7 1 1.8 94.6 0.0 100.0 0.0 0.0 8 1 1.8 96.4 0.0 100.0 0.0 0.0 9 1 1.8 98.2 0.0 100.0 0.0 0.0 10 1 1.8 100.0 0.0 100.0 0.0 0.0    57   Figure 4-4  Representative structure of cluster-0  Proposed reason for stability in RRM1-SS mutant is the hydrogen bond (dashed line) between residues Asn179 and Gln182.  4.4   NF-Kappa B p65 Peptide Simulations As mentioned above, MD simulations were used to examine the similarities and differences between the NF1 peptide (AGSIPGERSTDTTK; residues 43-56) and NF2 peptide (KRDLEQAISQRIQT; residues 123-136). These two peptides were analyzed in detail because their secondary structures are known (Figure 1-7). As the native secondary structures of NF1 and NF2 are identified to be a loop and a helix respectively in the full length p65 protein, it will be interesting to determine whether this structure is preserved when the peptides are studied individually. The simulation results will provide insight into whether the peptides  58  can maintain a particular secondary structure and whether this could be important for the interaction with TDP-43. Recall that Jie et al. (unpublished) studied short peptides attached to a cellulose surface and found specific interactions between NF1, NF2, and the other 6 peptides mentioned in Chapter 1.   4.4.1 Methods The X-ray structure of NF-Kappa B, PDB entry 1IKN [73] formed the basis of all simulations (Figure 4-5) . Two peptides sequences of equal length were chosen from the protein, Table 4-3. These two peptide fragments are structurally very different in the PDB file.  An additional two peptides were generated by changing the sequence of one peptide to that of the other in the coordinate file using UCSF Chimera [32] (Table 4-4, Figure 4-5).The simulation protocol was identical for all for simulations (see section 4.1). The peptides were solvated in methanol and energy minimized with position restraints in place on the peptide atoms. In the case of simulations with NF2 which is charged, a neutralizing Cl- ion was added to the system and the system minimized again. The systems were equilibrated for 1ns, after which data collection simulations were performed.   59   Figure 4-5  Structure of 1IKN  The Structure of NF-κB p65 subunit, PDB entry 1IKN [73], showing the two parts of the structure used in the simulations (red: sequence NF1, purple: sequence NF2). Picture made using UCSF Chimera [32]  Table 4-3   The sequences of the two peptidic fragments simulated, showing their residue numbers in the whole protein.  NF1 is a loop region, while NF2 is a helix in the X-ray structure (see Figure 4-5). Sequence number Sequence  Residue number in p65 NF1 AGSIPGERSTDTTK 43-56 NF2 KRDLEQAISQRIQT 123-136  Table 4-4  Overview of four simulations performed in methanol.  Native refers to the sequence-conformation combination found in the X-ray structure. Simulation Name Sequence Starting Structure Comment P65-S1-L-METH-01 NF1 Loop Native P65-S2-H-METH-01 NF2 Helix Native P65-S1-H-METH-01 NF1 Helix Non-native P65-S2-L-METH-01 NF2 Loop Non-native  60  4.4.2 Results and Discussion As can be seen from Table 4-4, four simulations, each of 100 ns in length after equilibration, were performed. The two starting structures are shown in Figure 4-6.   Figure 4-6  The helix and loop starting structures of the four simulations for NFkB p65 peptide Prior to the start of MD simulation, different parameters were provided to the program and one of these was to set the starting conformation. Shown here is the NF-kB p65 peptide NF1 sequence in (A) the native loop conformation and (B) in the non-native helix conformation. The NF2 sequence starting structures are identical apart from the side chains and its native state (not shown).  61  The Cα root mean square distance (RMSD) to the respective starting structures as a function of simulation time, Figure 4-7, indicates that all of the peptides undergo conformational changes, and do not simply remain in the starting X-ray conformation. This may also suggest that even though the NF2 sequence exists as a helix in p65 subunit, it will lose its structure as a peptide alone.  Figure 4-7  The Cα RMSD to the respective starting structures as a function of time Please refer to Table 4-3 for name designation. (A) P65-S1-L-METH-01: NF-kB p65 peptide NF1 starting in the random loop conformation. (B) P65-S1-H-METH-01: NF-kB p65 peptide NF1 starting in the helix conformation. (C) P65-S2-H-METH-01: NF-kB p65 peptide NF2 starting in the helix conformation. (D) P65-S2-L-METH-01: NF-kB p65 peptide NF2 starting in the random loop conformation In order to detect any structural similarity in the structures observed during the course of the four simulations, a combined clustering analysis was performed  62  (Table 4-5, Figure 4-8). The clustering results indicate that both NF1 simulations, even though they start from completely different structures (i.e. random loop and helix conformation), both sample a dominant, common conformational area, the representative structure of which is shown in Figure 4-9. This structure is stabilized by hydrogen bonds and has the lowest relative entropic contribution to its free energy (Figure 4-10).   Table 4-5  Combined clustering results incorporating conformations from all four peptide simulations.  N denotes cluster, Cum denotes cumulative, L denotes random loop, and H denotes helix. Only cluster-0 has data points from more than one simulation. Refer to Figure 4-7 for name notation. N # Size N % Cum.% NF1-L % NF2-H % NF1-H % NF2-L % 0 563 14.1 14.1 19.9 0.0 80.1 0.0 1 248 6.2 20.3 0.0 0.0 100.0 0.0 2 232 5.8 26.1 0.0 0.0 0.0 100.0 3 198 5.0 31.0 100.0 0.0 0.0 0.0 4 150 3.8 34.8 100.0 0.0 0.0 0.0 5 132 3.3 38.1 0.0 100.0 0.0 0.0 6 115 2.9 41.0 0.0 100.0 0.0 0.0 7 111 2.8 43.7 0.0 100.0 0.0 0.0 8 107 2.7 46.4 0.0 0.0 0.0 100.0 9 104 2.6 49.0 0.0 0.0 0.0 100.0 10 83 2.1 51.1 100.0 0.0 0.0 0.0   63   Figure 4-8  Combined clustering of the four p65 peptide simulations Y-axis is the cumulative percentage of the cluster size (for clarity, only clusters larger than 2% of the overall data set are shown). X-axis is the percentage contribution of each peptide in a given cluster. These results show that only cluster zero (the largest one) has data points from more than one simulation. Both of these are from simulations with the NF1 sequence.  64   Figure 4-9  The representative structure of cluster-0 of the combined cluster analysis shows a loop-like structure stabilized by hydrogen bonds (cyan) In this representation, the length of the lines are not representative of bond lengths.  Figure 4-10 The simulated data points versus the entropic part of the free energy difference between cluster i and cluster-0. For the purpose of visualization, the simulated data points are subjected to a principle component analysis (PCA). The x- and y-coordinates are derived from the projection of a data point onto the two largest eigenvectors, V1 and V2. The data points from the simulation are plotted as dots. The representative points determined from clustering are represented as circles. The entropically lowest cluster-0 is visible in black in the foreground.   65  On the other hand, the two NF2 simulations are never observed to adopt the conformation shown in Figure 4-9 during the simulations, nor do the NF2 simulations share any common conformation. Figure 4-11 shows representative structures from these two simulations. Even though there is no common conformation for the two NF2, the native helix NF2 show a loss of helical content by the end of the simulation, becoming more random loop-like in structure.    66   Figure 4-11 The representative structures of cluster-0 of the two NF2 simulations. These structures are reminiscent of their respective starting structures. In this representation, the length of the lines are not representative of bond lengths.     67  4.5 Conclusion The work presented in this chapter focused on the structural characteristics of TDP-43 RRM-1 domain and NF-κB p65 peptides as analysed by molecular dynamic simulations. The major restraint in MD simulation is the limited sampling time for each individual protein or peptide sample. For example, it would take approximately 3 days to generate 1 ns of data for the TDP-43 RRM-1 domain simulation. Fortunately, valuable information was extracted from these simulation data such that predictions and conclusions can be made.  Examination of the wild-type TDP-43 RRM-1 domain and its single and double C/S mutants (residue 173 and 175) indicated that the double mutant (RRM1-SS) is more likely to adopt a cluster-0 conformation which exhibits a rigid C-terminal region of the domain. This regional rigidity is most likely the outcome of a single hydrogen bond between residues Asn-179 and Gln-182 that is found in the representative structure of cluster-0. And more importantly, this result correlates well with previous experimental findings that double serine substitutions are more prone to aggregation and inclusion bodies formation in vivo. If such hypothesis is valid, then it would suggest that the rigidity from one hydrogen bond could play a role in forming a nucleus for aggregation. Hence, disruption of cluster-0 structure may be considered to be a future therapeutic approach.  MD simulations of p65 peptides, NF1 and NF2, provided results that are more difficult to interpret. When looking at NF1 simulations, with starting conformations as loop (native) and helix (non-native), the two converged into a  68  lowest relative free energy intermediate at cluster-0 that is stabilized by hydrogen bonds. However, NF2 does not adopt this cluster-0 conformation, nor do the two simulations (loop and helix starting structure) share a common conformation within the 100 ns of sampling time. These results would suggest that the secondary structure of the peptides does not necessarily play a role in the interaction between RRM-1 and p65. But again, the sampling time of these peptides may be too short to detect any significant outcome in the similarities between NF1 and NF2 intermediate conformations. Moreover, there are no direct experimental results linked to this set of simulations. Hence it will be important in future work to compare structural data obtained from the actual peptides prepared from Chapter 3 to MD simulations data.    69  CHAPTER 5: CONCLUSION AND FUTURE PLANS The focus of this thesis rests upon TDP-43, a protein which is suggested to play a role in the pathogenesis of amyotrophic lateral sclerosis. The motivation for initiating research was to examine the two proposed mechanism of TDP-43, aggregation and dysregulation, leading to proteinopathy at a molecular level. Hence, to look at the aggregation model, the work presented in this thesis included the formation of a recombinant form of TDP-43 protein, and its RRM-1 domain variant. When examining the dysregulation model, binding partners such as peptides from p65 subunit of NF-κB were synthesized in order to look at interaction with the recombinant protein. In addition, MD simulations were performed and used as models to predict outcomes of the two mechanisms.  Chapter 1 of the thesis provided background information to introduce the hypothesis that TDP-43 RRM-1 domain and p65 subunit of NF-κB have a relationship to the disease amyotrophic lateral sclerosis (ALS) on the basis that RRM-1 is the domain susceptible to protein aggregation and that p65 subunit is a binding partner to TDP-43 which is likely linked to the dysregulation model. Also reviewed in this chapter were the general methods employed in the thesis work. Chapter 2 focused on the development of methods to obtain TDP-43 and its RRM-1 domain variant. Chapter 3 explored the methodology of solid phase peptide synthesis (SPPS) to prepare the 8 peptides from p65 subunit known to interact with TDP-43. Finally, Chapter 4 discussed the results of molecular  70  dynamics simulations performed on the RRM-1 domain of TDP-43 (WT and C/S mutants) and two of the eight peptides from p65.  Large quantities of TDP-43 protein had not previously been isolated for study using a heterologous E. coli expression system. The main goal was to acquire a large amount of pure TDP-43 for biophysical studies and interaction studies. As of now, the expression of TDP-43 was successful with high yield. However, the purification of the protein deemed difficult partly due to the possibility that TDP-43 may affect the GST affinity tag, lowering its binding ability to the resin. But the major issue lies on TDP-43 itself, as over-expression of the protein would lead to aggregation and the formation of inclusion bodies in the cell. This limits the amount of soluble protein available for isolation. Further optimization of the protocol would be necessary such as the use of a poly-His tag to avoid the need for well-structured GST in order to make GST-glutathione affinity chromatography successful. In addition, finding the balance of minimizing the expression time and maximizing soluble protein yield would reduce nucleation cascade of aggregation in the cell. Upon establishing the methodology for the full-length TDP-43, it will serve as a foundation in obtaining the RRM-1 domain and various RRM-1 C/S mutants so that further study can be achieved by performing NMR structural studies to look at molecular differences that may lead to different aggregation propensities. This prediction was demonstrated in the MD simulations comparing the four RRM-1 constructs (RRM1-CC, RRM1-CS, RRM1-SC, and RRM1-SS). Data analyses showed that the wild-type RRM-1 (RRM1-CC) possess the most mobile and disordered region following residues  71  173 and 175, whereas the double substitution mutant (RRM1-SS) was the most rigid. This is predicted to be due to residue induced conformational change leading to the formation of a hydrogen bond between the side chains of Asn179 and Gln182. This specific conformation implements a turn or a β-hairpin in region C as shown in Figure 1-6 which suggests that it could lead the protein into aggregation.  The eight peptide sequences that were predicted to interact with TDP-43 RRM-1 were subjected to solid phase peptide synthesis. Despite the fact that most of these sequences were able to be synthesized, they cannot be purified via reverse phase high performance liquid chromatography because these peptides are non-polar. However, once pure peptides can be obtained, structural studies using circular dichroism can compare the secondary structure of the peptide alone and when the sequence is within the p65 subunit. This may be an important factor since we know the conformation of two sequences within the subunit; hence it would inform us to whether the structure will be retained in the short peptide sequence and if the structure plays a crucial role in binding with RRM-1 during interaction studies. MD simulations were performed on two peptide sequences, NF1 and NF2, with known native secondary structure within the subunit, loop and helix, respectively. Analysis of the data revealed that NF1, starting either as a loop or helix, converged into a similar low energy intermediate stabilized by hydrogen bonds. On the contrary, NF2 did not show any low energy intermediate. But by the end of the simulations, NF2 lost most of its helical content and adopted a random coil configuration. Hence, secondary structure of free short peptides in solution  72  may not reflect its native configuration in the subunit. Hence, even if the peptide interacts or does not interact with TDP-43, the results may not be conclusive as the structure may or may not play a role in the interaction.  Many of the experiments described in this thesis remain to be optimized. The first step in future work would be to complete these optimizations. Once these are achieved, future studies would involve looking at the structure of TDP-43 in the aggregated state and binding interactions between TDP-43 and the p65 peptides. The former could be achieved by using methods such as CD spectroscopy or solid state NMR. In the latter case, methods such as solution state NMR and ITC could be used. Overall, the fundamental experiments described herein and future work will serve to expand our understanding of ALS and potentially lead to a cure.   73  REFERENCES [1] D. W. Mulder, L. T. Kurland, K. P. Offord, and C. M. Beard, “Familial adult motor neuron disease: amyotrophic lateral sclerosis.,” Neurology, vol. 36, no. 4, pp. 511–7, Apr. 1986. [2] D. R. Rosen, T. Siddique, D. Patterson, D. A. Figlewicz, P. Sapp, A. Hentati, D. Donaldson, J. Goto, J. P. O’Regan, and H. X. Deng, “Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis.,” Nature, vol. 362, no. 6415, pp. 59–62, Mar. 1993. [3] M. Neumann, D. M. Sampathu, L. K. Kwong, A. C. Truax, M. C. Micsenyi, T. T. Chou, J. Bruce, T. Schuck, M. Grossman, C. M. Clark, L. F. McCluskey, B. L. Miller, E. Masliah, I. R. Mackenzie, H. Feldman, W. Feiden, H. A. Kretzschmar, J. Q. Trojanowski, and V. M.-Y. Lee, “Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis.,” Science, vol. 314, no. 5796, pp. 130–3, Oct. 2006. [4] T. Arai, M. Hasegawa, H. Akiyama, K. Ikeda, T. Nonaka, H. Mori, D. Mann, K. Tsuchiya, M. Yoshida, Y. Hashizume, and T. Oda, “TDP-43 is a component of ubiquitin-positive tau-negative inclusions in frontotemporal lobar degeneration and amyotrophic lateral sclerosis.,” Biochem. Biophys. Res. Commun., vol. 351, no. 3, pp. 602–11, Dec. 2006. [5] J. Sreedharan, I. P. Blair, V. B. Tripathi, X. Hu, C. Vance, B. Rogelj, S. Ackerley, J. C. Durnall, K. L. Williams, E. Buratti, F. Baralle, J. de Belleroche, J. D. Mitchell, P. N. Leigh, A. Al-Chalabi, C. C. Miller, G. Nicholson, and C. E. Shaw, “TDP-43 mutations in familial and sporadic amyotrophic lateral sclerosis.,” Science, vol. 319, no. 5870, pp. 1668–72, Mar. 2008. [6] E. Kabashi, P. N. Valdmanis, P. Dion, D. Spiegelman, B. J. McConkey, C. Vande Velde, J.-P. Bouchard, L. Lacomblez, K. Pochigaeva, F. Salachas, P.-F. Pradat, W. Camu, V. Meininger, N. Dupre, and G. A. Rouleau, “TARDBP mutations in individuals with sporadic and familial amyotrophic lateral sclerosis.,” Nat. Genet., vol. 40, no. 5, pp. 572–4, May 2008. [7] W. Robberecht and T. Philips, “The changing scene of amyotrophic lateral sclerosis.,” Nat. Rev. Neurosci., vol. 14, no. 4, pp. 248–64, Apr. 2013. [8] J. Sreedharan and R. H. Brown, “Amyotrophic lateral sclerosis: Problems and prospects.,” Ann. Neurol., vol. 74, no. 3, pp. 309–16, Sep. 2013. [9] K. Talbot and O. Ansorge, “Recent advances in the genetics of amyotrophic lateral sclerosis and frontotemporal dementia: common pathways in  74  neurodegenerative disease.,” Hum. Mol. Genet., vol. 15 Spec No, pp. R182–7, Oct. 2006. [10] L. K. Kwong, K. Uryu, J. Q. Trojanowski, and V. M.-Y. Lee, “TDP-43 Proteinopathies: Neurodegenerative Protein Misfolding Diseases without Amyloidosis,” Neurosignals, vol. 16, no. 1, pp. 41–51, Jan. 2008. [11] E. Buratti and F. E. Baralle, “Multiple roles of TDP-43 in gene expression, splicing regulation, and human disease.,” Front. Biosci., vol. 13, pp. 867–78, Jan. 2008. [12] C. Fallini, G. J. Bassell, and W. Rossoll, “The ALS disease protein TDP-43 is actively transported in motor neuron axons and regulates axon outgrowth.,” Hum. Mol. Genet., vol. 21, no. 16, pp. 3703–18, Aug. 2012. [13] S. H. Ou, F. Wu, D. Harrich, L. F. García-Martínez, and R. B. Gaynor, “Cloning and characterization of a novel cellular protein, TDP-43, that binds to human immunodeficiency virus type 1 TAR DNA sequence motifs.,” J. Virol., vol. 69, no. 6, pp. 3584–96, Jun. 1995. [14] Y. M. Ayala, T. Misteli, and F. E. Baralle, “TDP-43 regulates retinoblastoma protein phosphorylation through the repression of cyclin-dependent kinase 6 expression,” Proc. Natl. Acad. Sci., vol. 105, no. 10, pp. 3785–3789, Feb. 2008. [15] Y. M. Ayala, F. Pagani, and F. E. Baralle, “TDP43 depletion rescues aberrant CFTR exon 9 skipping.,” FEBS Lett., vol. 580, no. 5, pp. 1339–44, Feb. 2006. [16] E. Buratti, T. Dörk, E. Zuccato, F. Pagani, M. Romano, and F. E. Baralle, “Nuclear factor TDP-43 and SR proteins promote in vitro and in vivo CFTR exon 9 skipping.,” EMBO J., vol. 20, no. 7, pp. 1774–84, Apr. 2001. [17] E. Buratti and F. E. Baralle, “Characterization and functional implications of the RNA binding properties of nuclear factor TDP-43, a novel splicing regulator of CFTR exon 9.,” J. Biol. Chem., vol. 276, no. 39, pp. 36337–43, Sep. 2001. [18] M. Polymenidou, C. Lagier-Tourenne, K. R. Hutt, S. C. Huelga, J. Moran, T. Y. Liang, S.-C. Ling, E. Sun, E. Wancewicz, C. Mazur, H. Kordasiewicz, Y. Sedaghat, J. P. Donohue, L. Shiue, C. F. Bennett, G. W. Yeo, and D. W. Cleveland, “Long pre-mRNA depletion and RNA missplicing contribute to neuronal vulnerability from loss of TDP-43.,” Nat. Neurosci., vol. 14, no. 4, pp. 459–68, Apr. 2011. [19] E. Buratti, A. Brindisi, F. Pagani, and F. E. Baralle, “Nuclear factor TDP-43 binds to the polymorphic TG repeats in CFTR intron 8 and causes skipping of exon 9: a functional link with disease penetrance.,” Am. J. Hum. Genet., vol. 74, no. 6, pp. 1322–5, Jun. 2004. [20] J. R. Tollervey, T. Curk, B. Rogelj, M. Briese, M. Cereda, M. Kayikci, J. König, T. Hortobágyi, A. L. Nishimura, V. Zupunski, R. Patani, S. Chandran,  75  G. Rot, B. Zupan, C. E. Shaw, and J. Ule, “Characterizing the RNA targets and position-dependent splicing regulation by TDP-43.,” Nat. Neurosci., vol. 14, no. 4, pp. 452–8, Apr. 2011. [21] M. J. Strong, K. Volkening, R. Hammond, W. Yang, W. Strong, C. Leystra-Lantz, and C. Shoesmith, “TDP43 is a human low molecular weight neurofilament (hNFL) mRNA-binding protein.,” Mol. Cell. Neurosci., vol. 35, no. 2, pp. 320–7, Jun. 2007. [22] N. H. Alami, R. B. Smith, M. A. Carrasco, L. A. Williams, C. S. Winborn, S. S. W. Han, E. Kiskinis, B. Winborn, B. D. Freibaum, A. Kanagaraj, A. J. Clare, N. M. Badders, B. Bilican, E. Chaum, S. Chandran, C. E. Shaw, K. C. Eggan, T. Maniatis, and J. P. Taylor, “Axonal transport of TDP-43 mRNA granules is impaired by ALS-causing mutations.,” Neuron, vol. 81, no. 3, pp. 536–43, Feb. 2014. [23] D. C. Diaper, Y. Adachi, B. Sutcliffe, D. M. Humphrey, C. J. H. Elliott, A. Stepto, Z. N. Ludlow, L. Vanden Broeck, P. Callaerts, B. Dermaut, A. Al-Chalabi, C. E. Shaw, I. M. Robinson, and F. Hirth, “Loss and gain of Drosophila TDP-43 impair synaptic efficacy and motor control leading to age-related neurodegeneration by loss-of-function phenotypes.,” Hum. Mol. Genet., vol. 22, no. 8, pp. 1539–57, Apr. 2013. [24] C. Colombrita, E. Zennaro, C. Fallini, M. Weber, A. Sommacal, E. Buratti, V. Silani, and A. Ratti, “TDP-43 is recruited to stress granules in conditions of oxidative insult.,” J. Neurochem., vol. 111, no. 4, pp. 1051–61, Nov. 2009. [25] S. Lindquist, “Regulation of protein synthesis during heat shock.,” Nature, vol. 293, no. 5830, pp. 311–4, Sep. 1981. [26] H.-Y. Wang, I.-F. Wang, J. Bose, and C.-K. J. Shen, “Structural diversity and functional implications of the eukaryotic TDP gene family.,” Genomics, vol. 83, no. 1, pp. 130–9, Jan. 2004. [27] G. Dreyfuss, M. J. Matunis, S. Piñol-Roma, and C. G. Burd, “hnRNP proteins and the biogenesis of mRNA.,” Annu. Rev. Biochem., vol. 62, pp. 289–321, Jan. 1993. [28] C. Maris, C. Dominguez, and F. H.-T. Allain, “The RNA recognition motif, a plastic RNA-binding platform to regulate post-transcriptional gene expression.,” FEBS J., vol. 272, no. 9, pp. 2118–31, May 2005. [29] Y. M. Ayala, S. Pantano, A. D’Ambrogio, E. Buratti, A. Brindisi, C. Marchetti, M. Romano, and F. E. Baralle, “Human, Drosophila, and C.elegans TDP43: nucleic acid binding properties and splicing regulatory function.,” J. Mol. Biol., vol. 348, no. 3, pp. 575–88, May 2005. [30] E. Buratti, A. Brindisi, M. Giombi, S. Tisminetzky, Y. M. Ayala, and F. E. Baralle, “TDP-43 binds heterogeneous nuclear ribonucleoprotein A/B through its C-terminal tail: an important region for the inhibition of cystic  76  fibrosis transmembrane conductance regulator exon 9 splicing.,” J. Biol. Chem., vol. 280, no. 45, pp. 37572–84, Nov. 2005. [31] Y. M. Ayala, P. Zago, A. D’Ambrogio, Y.-F. Xu, L. Petrucelli, E. Buratti, and F. E. Baralle, “Structural determinants of the cellular localization and shuttling of TDP-43.,” J. Cell Sci., vol. 121, no. Pt 22, pp. 3778–85, Nov. 2008. [32] E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch, D. M. Greenblatt, E. C. Meng, and T. E. Ferrin, “UCSF Chimera--a visualization system for exploratory research and analysis.,” J. Comput. Chem., vol. 25, no. 13, pp. 1605–12, Oct. 2004. [33] L. K. Kwong, M. Neumann, D. M. Sampathu, V. M.-Y. Lee, and J. Q. Trojanowski, “TDP-43 proteinopathy: the neuropathology underlying major forms of sporadic and familial frontotemporal lobar degeneration and motor neuron disease.,” Acta Neuropathol., vol. 114, no. 1, pp. 63–70, Jul. 2007. [34] A. S. Chen-Plotkin, V. M.-Y. Lee, and J. Q. Trojanowski, “TAR DNA-binding protein 43 in neurodegenerative disease.,” Nat. Rev. Neurol., vol. 6, no. 4, pp. 211–20, Apr. 2010. [35] C.-F. Tan, H. Eguchi, A. Tagawa, O. Onodera, T. Iwasaki, A. Tsujino, M. Nishizawa, A. Kakita, and H. Takahashi, “TDP-43 immunoreactivity in neuronal inclusions in familial amyotrophic lateral sclerosis with or without SOD1 gene mutation.,” Acta Neuropathol., vol. 113, no. 5, pp. 535–42, May 2007. [36] L. M. Igaz, L. K. Kwong, Y. Xu, A. C. Truax, K. Uryu, M. Neumann, C. M. Clark, L. B. Elman, B. L. Miller, M. Grossman, L. F. McCluskey, J. Q. Trojanowski, and V. M.-Y. Lee, “Enrichment of C-terminal fragments in TAR DNA-binding protein-43 cytoplasmic inclusions in brain but not in spinal cord of frontotemporal lobar degeneration and amyotrophic lateral sclerosis.,” Am. J. Pathol., vol. 173, no. 1, pp. 182–94, Jul. 2008. [37] N. J. Cairns, M. Neumann, E. H. Bigio, I. E. Holm, D. Troost, K. J. Hatanpaa, C. Foong, C. L. White, J. A. Schneider, H. A. Kretzschmar, D. Carter, L. Taylor-Reinwald, K. Paulsmeyer, J. Strider, M. Gitcho, A. M. Goate, J. C. Morris, M. Mishra, L. K. Kwong, A. Stieber, Y. Xu, M. S. Forman, J. Q. Trojanowski, V. M.-Y. Lee, and I. R. A. Mackenzie, “TDP-43 in familial and sporadic frontotemporal lobar degeneration with ubiquitin inclusions.,” Am. J. Pathol., vol. 171, no. 1, pp. 227–40, Jul. 2007. [38] J. Brettschneider, K. Del Tredici, J. B. Toledo, J. L. Robinson, D. J. Irwin, M. Grossman, E. Suh, V. M. Van Deerlin, E. M. Wood, Y. Baek, L. Kwong, E. B. Lee, L. Elman, L. McCluskey, L. Fang, S. Feldengut, A. C. Ludolph, V. M.-Y. Lee, H. Braak, and J. Q. Trojanowski, “Stages of pTDP-43 pathology in amyotrophic lateral sclerosis.,” Ann. Neurol., vol. 74, no. 1, pp. 20–38, Jul. 2013.  77  [39] M. J. Winton, L. M. Igaz, M. M. Wong, L. K. Kwong, J. Q. Trojanowski, and V. M.-Y. Lee, “Disturbance of nuclear and cytoplasmic TAR DNA-binding protein (TDP-43) induces disease-like redistribution, sequestration, and aggregate formation.,” J. Biol. Chem., vol. 283, no. 19, pp. 13302–9, May 2008. [40] N. R. Stallings, K. Puttaparthi, C. M. Luther, D. K. Burns, and J. L. Elliott, “Progressive motor weakness in transgenic mice expressing human TDP-43.,” Neurobiol. Dis., vol. 40, no. 2, pp. 404–14, Nov. 2010. [41] L. M. Igaz, L. K. Kwong, E. B. Lee, A. Chen-Plotkin, E. Swanson, T. Unger, J. Malunda, Y. Xu, M. J. Winton, J. Q. Trojanowski, and V. M.-Y. Lee, “Dysregulation of the ALS-associated gene TDP-43 leads to neuronal death and degeneration in mice.,” J. Clin. Invest., vol. 121, no. 2, pp. 726–38, Mar. 2011. [42] Y.-F. Xu, T. F. Gendron, Y.-J. Zhang, W.-L. Lin, S. D’Alton, H. Sheng, M. C. Casey, J. Tong, J. Knight, X. Yu, R. Rademakers, K. Boylan, M. Hutton, E. McGowan, D. W. Dickson, J. Lewis, and L. Petrucelli, “Wild-type human TDP-43 expression causes TDP-43 phosphorylation, mitochondrial aggregation, motor deficits, and early mortality in transgenic mice.,” J. Neurosci., vol. 30, no. 32, pp. 10851–9, Aug. 2010. [43] V. Swarup, D. Phaneuf, N. Dupré, S. Petri, M. Strong, J. Kriz, and J.-P. Julien, “Deregulation of TDP-43 in amyotrophic lateral sclerosis triggers nuclear factor κB-mediated pathogenic pathways.,” J. Exp. Med., vol. 208, no. 12, pp. 2429–47, Nov. 2011. [44] A. Shodai, T. Morimura, A. Ido, T. Uchida, T. Ayaki, R. Takahashi, S. Kitazawa, S. Suzuki, M. Shirouzu, T. Kigawa, Y. Muto, S. Yokoyama, R. Takahashi, R. Kitahara, H. Ito, N. Fujiwara, and M. Urushitani, “Aberrant Assembly of RNA Recognition Motif 1 Links to Pathogenic Conversion of TAR DNA-binding Protein of 43 kDa (TDP-43),” J. Biol. Chem., vol. 288, no. 21, pp. 14886–14905, 2013. [45] P.-H. Kuo, L. G. Doudeva, Y.-T. Wang, C.-K. J. Shen, and H. S. Yuan, “Structural insights into TDP-43 in nucleic-acid binding and domain interactions.,” Nucleic Acids Res., vol. 37, no. 6, pp. 1799–808, Apr. 2009. [46] C. Chang, M. Chiang, E. K.-W. Toh, C.-F. Chang, and T. Huang, “Molecular mechanism of oxidation-induced TDP-43 RRM1 aggregation and loss of function,” FEBS Lett., vol. 587, no. 6, pp. 575–582, 2013. [47] S. Boillée, C. Vande Velde, and D. W. Cleveland, “ALS: a disease of motor neurons and their nonneuronal neighbors.,” Neuron, vol. 52, no. 1, pp. 39–59, Oct. 2006. [48] R. Sen and D. Baltimore, “Multiple nuclear factors interact with the immunoglobulin enhancer sequences.,” Cell, vol. 46, no. 5, pp. 705–16, Aug. 1986.  78  [49] A. Kumar, Y. Takada, A. M. Boriek, and B. B. Aggarwal, “Nuclear factor-kappaB: its role in health and disease.,” J. Mol. Med. (Berl)., vol. 82, no. 7, pp. 434–48, Jul. 2004. [50] U. Siebenlist, G. Franzoso, and K. Brown, “Structure, regulation and function of NF-kappa B.,” Annu. Rev. Cell Biol., vol. 10, pp. 405–55, Jan. 1994. [51] Y. Q. Chen, L. L. Sengchanthalangsy,  a Hackett, and G. Ghosh, “NF-kappaB p65 (RelA) homodimer uses distinct mechanisms to recognize DNA targets.,” Structure, vol. 8, no. 4, pp. 419–28, 2000. [52] G. L. Rosano and E. A. Ceccarelli, “Recombinant protein expression in Escherichia coli: advances and challenges.,” Front. Microbiol., vol. 5, p. 172, Jan. 2014. [53] D. M. Francis and R. Page, “Strategies to optimize protein expression in E. coli.,” Curr. Protoc. Protein Sci., vol. Chapter 5, p. Unit 5.24.1–29, Aug. 2010. [54] B. Merrifield, “Solid phase synthesis.,” Science, vol. 232, no. 4748, pp. 341–7, Apr. 1986. [55] G. B. Fields, Current Protocols in Protein Science, vol. Chapter 18. Hoboken, NJ, USA: John Wiley & Sons, Inc., 2001. [56] U. K. LAEMMLI, “Cleavage of Structural Proteins during the Assembly of the Head of Bacteriophage T4,” Nature, vol. 227, no. 5259, pp. 680–685, Aug. 1970. [57] C. Chang, M. Chiang, E. K.-W. Toh, C.-F. Chang, and T. Huang, “Molecular mechanism of oxidation-induced TDP-43 RRM1 aggregation and loss of function.,” FEBS Lett., vol. 587, no. 6, pp. 575–82, Mar. 2013. [58] W. R. P. Scott, P. H. Hünenberger, I. G. Tironi, A. E. Mark, S. R. Billeter, J. Fennen, A. E. Torda, T. Huber, P. Krüger, and W. F. van Gunsteren, “The GROMOS Biomolecular Simulation Program Package,” J. Phys. Chem. A, vol. 103, no. 19, pp. 3596–3607, May 1999. [59] W. van Gunsteren, S. Billeter, A. Eising, P. Hünenberger, P. Krüger, A. Mark, W. Scott, and I. Tironi, “Biomolecular Simulation: The {GROMOS96} manual and userguide,” 1996. [60] C. Oostenbrink, A. Villa, A. E. Mark, and W. F. van Gunsteren, “A biomolecular force field based on the free enthalpy of hydration and solvation: the GROMOS force-field parameter sets 53A5 and 53A6.,” J. Comput. Chem., vol. 25, no. 13, pp. 1656–76, Oct. 2004. [61] C. Oostenbrink, T. A. Soares, N. F. A. van der Vegt, and W. F. van Gunsteren, “Validation of the 53A6 GROMOS force field,” Eur. Biophys. J., vol. 34, no. 4, pp. 273–284, Apr. 2005.  79  [62] H. J. C. Berendsen, J. P. M. Postma, W. F. van Gunsteren, and J. Hermans, “Interaction models for water in relation to protein hydration,” Intermol. Forces, pp. 331–342, 1981. [63] R. Walser, A. E. Mark, W. F. van Gunsteren, M. Lauterbach, and G. Wipff, “The effect of force-field parameters on properties of liquids: Parametrization of a simple three-site model for methanol,” J. Chem. Phys., vol. 112, no. 23, pp. 10450–10459, 2000. [64] H. J. C. Berendsen, J. P. M. Postma, W. F. van Gunsteren, A. DiNola, and J. R. Haak, “Molecular dynamics with coupling to an external bath,” J. Chem. Phys., vol. 81, no. 8, pp. 3684–3690, 1984. [65] J.-P. Ryckaert, G. Ciccotti, and H. J. C. Berendsen, “Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes,” J. Comput. Phys., vol. 23, no. 3, pp. 327–341, 1977. [66] I. G. Tironi, R. Sperb, P. E. Smith, and W. F. van Gunsteren, “A generalized reaction field method for molecular dynamics simulations,” J. Chem. Phys., vol. 102, no. 13, p. 5451, 1995. [67] H. Berman, K. Henrick, and H. Nakamura, “Announcing the worldwide Protein Data Bank.,” Nat. Struct. Biol., vol. 10, no. 12, p. 980, Dec. 2003. [68] X. Daura, B. Jaun, D. Seebach, W. F. van Gunsteren, and A. E. Mark, “Reversible peptide folding in solution by molecular dynamics simulation,” J. Mol. Biol., vol. 280, no. 5, pp. 925–932, 1998. [69] S. W. Ho, D. Jung, J. R. Calhoun, J. D. Lear, M. Okon, W. R. P. Scott, R. E. W. Hancock, and S. K. Straus, “Effect of divalent cations on the structure of the antibiotic daptomycin,” Eur. Biophys. J., vol. 37, no. 4, pp. 421–433, Oct. 2007. [70] W. R. P. Scott, S.-B. Baek, D. Jung, R. E. W. Hancock, and S. K. Straus, “NMR structural studies of the antibiotic lipopeptide daptomycin in DHPC micelles.,” Biochim. Biophys. Acta, vol. 1768, no. 12, pp. 3116–26, Dec. 2007. [71] W. R. P. Scott and S. K. Straus, “Determining and visualizing flexibility in protein structures.,” Proteins, vol. 83, no. 5, pp. 820–6, May 2015. [72] S. Suzuki, Y. Muto, M. Inoue, T. Kigawa, T. Terada, M. Shirouzu, and S. Yokoyama, “Solution structure of the RNA binding domain of TAR DNA-binding protein-43,” To be Publ. [73] T. Huxford, D. B. Huang, S. Malek, and G. Ghosh, “The crystal structure of the IkappaBalpha/NF-kappaB complex reveals mechanisms of NF-kappaB inactivation,” Cell, vol. 95, no. 6, pp. 759–770, 1998.  80  [74] M. J. Guerrero-Muñoz, J. Gerson, and D. L. Castillo-Carranza, “Tau Oligomers: The Toxic Player at Synapses in Alzheimer’s Disease.,” Front. Cell. Neurosci., vol. 9, p. 464, Jan. 2015. [75] D. Olivares, X. Huang, L. Branden, N. H. Greig, and J. T. Rogers, “Physiological and pathological role of alpha-synuclein in Parkinson’s disease through iron mediated oxidative stress; the role of a putative iron-responsive element.,” Int. J. Mol. Sci., vol. 10, no. 3, pp. 1226–60, Mar. 2009.   81  APPENDICES Appendix A: PCR And Thermocycling Reaction Conditions   Table A 1 PCR Reaction Components Template DNA Various 10 ηg Forward Primer Various 0.5 µM Reverse Primer Various 0.5 µM dNTPs dATP, dTTP dCTP,dGTP 250 µM each (final) 5X Phusion HF Buffer  10 µl ddH2O  to 50 µl polymerase Phusion 2.5 U    Table A 2 PCR conditions     x 25 cycles                   98°C 30 sec 98°C 10 sec 60°C 30 sec 72°C 30 sec/kb template length 72°C 10 min 4°C ∞  82  Appendix B: Oligonucleotide Primers    Primer Design for PCR  - analyzed using Premiere Biosoft NetPrimer software NAME: XbaI-TDP-43-F DESCRIPTION: Forward primer for the construction of pGEX-KG-TDP43                                                                 5’ T AGT CTA GAC ATG TCT GAA TAT ATT CGG GTA ACC G 3’                           M   S   E   Y   I    R   V   T                 ↑ inserted XbaI site  Annealing sequence to TDP43  # of Bases in Annealing Sequence: 25 Total # of Bases: 35   GC Content in Annealing Sequence:  42.3% Total GC Content:  40% Annealing Sequence TM:  56.6°C Total TM:  62.9°C  NAME:  DESCRIPTION: Reverse primer complementary to                                      5’ CAT AAG CTT CTA CAT TCC CCA GCC AGA AG 3’  ↑ inserted HindIII site Annealing sequence to TDP43  # of Bases in Annealing Sequence: 20  Total # of Bases: 29 GC Content in Annealing Sequence:  55% Total GC Content:  48.3% Annealing Sequence TM:  53.3°C Total TM:  62.4°C     

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0224505/manifest

Comment

Related Items