Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigations of the origin, regulation, and substrate specificities of protein kinases in the human… Lai, Shenshen 2015

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2015_may_lai_shenshen.pdf [ 7.57MB ]
Metadata
JSON: 24-1.0167195.json
JSON-LD: 24-1.0167195-ld.json
RDF/XML (Pretty): 24-1.0167195-rdf.xml
RDF/JSON: 24-1.0167195-rdf.json
Turtle: 24-1.0167195-turtle.txt
N-Triples: 24-1.0167195-rdf-ntriples.txt
Original Record: 24-1.0167195-source.json
Full Text
24-1.0167195-fulltext.txt
Citation
24-1.0167195.ris

Full Text

    INVESTIGATIONS OF THE ORIGIN, REGULATION AND SUBSTRATE SPECIFICITIES OF PROTEIN KINASES IN THE HUMAN KINOME by Shenshen Lai B.Sc. Peking University, China, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate and Postdoctoral Studies (Experimental Medicine) THE UNIVERSITY OF BRITHSH COLUMBIA (Vancouver)  April 2015 © Shenshen Lai, 2015 ii  Abstract The eukaryotic protein kinases (ePKs) constitute one of the largest families of enzymes encoded by eukaryotic genomes. They regulate all cellular processes by transducing inter- and intracellular signals via the phosphorylation of specific protein substrates. The primary sequences of ePK catalytic domains are highly conserved, indicating a common ancestry. They all share a conserved catalytic core for their phosphotransferase function and often a common activation mechanism by phosphorylation of a variable segment known as the activation T-loop. Starting from a manually aligned and annotated map of 492 typical human protein kinase domains, I first explored the origin of ePKs using a modified BLAST method in various species. Comparisons of primary, secondary and tertiary structures supported the hypothesis that protein kinases and choline kinases evolved from an ancient aminoacyl-tRNA ligase. Secondly, I studied the functional roles of two extremely conserved phosphosites that exist ubiquitously in the activation T-loops of most protein-serine/threonine kinases. The extensively examined extracellular signal-regulated kinases (ERKs) 1/2 from the mitogen-activated protein kinase family were used as a model in this study. I discovered that both Thr-207 and Tyr-210 from human ERK1 were essential for regulation of its phosphotransferase activity. Autophosphorylation of Thr-207 played an inhibitory role to control ERK1 activity after the initial phosphorylation and activation by MEK1. This may serve as a general mechanism for kinase autoinhibition. I also examined the substrate specificities of more than 200 human protein kinases with peptide microarrays populated with semi-optimal substrate sequences. The resultant  iii  data were used to expand the training datasets for development of next generation protein kinase substrate predictive algorithms. Additionally, I describe a novel method to produce a more unbiased polyclonal generic phosphotyrosine antibody than the monoclonal antibodies that are commonly used to enrich and track tyrosine-phosphorylated proteins. The tools and knowledge resulting from this research should enable improvements in the characterization of protein kinases and establishing their linkages to a variety of human diseases for the development of better diagnostic tests and therapeutic drugs. This work also provides basic insights into the evolution of life and the specific architecture of protein kinase-based signalling systems.  iv  Preface The research presented in this dissertation was conducted in Dr. Steven Pelech's laboratory at Kinexus Bioinformatics Corporation. Under the direct supervision of Dr. Pelech, I performed the majority of the experiments and data analysis presented in this dissertation. I was the primary investigator for the projects in Chapters 3, 4 and 5, where I was responsible for major areas of technique optimization, data collection, analysis, and manuscript composition. Dr. Dirk Winkler synthesized all the peptides that were used in Chapters 4and 5. Mr. Dominik Sommerfeld assisted me with part of the microarray experiments and data analysis in Chapter 4. The manuscript is currently in revision. The work described in Appendix 7 was performed by my colleagues and I at Kinexus. Dr. Pelech designed the methods of generic phosphotyrosine antibody production. Dr. Jane Shi purified the PY-K antibody. Dr. Hong Zhang performed the JPT peptide microarray experiment (Figure A7.3) and part of the analysis. I was responsible for the rest of the experiments, data analysis, and composition of the manuscript. A portion of Chapter 4 was prepared as a methodology paper on which I was the primary author. The manuscript [Lai S, Winkler D, Zhang H, and Pelech S. Determination of the Substrate Specificity of Protein Kinases with Peptide Micro- and Macroarrays.] is currently under review.  v  Table of Contents  Abstract ......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents .......................................................................................................................... v List of Tables .................................................................................................................................. x List of Figures ............................................................................................................................... xi List of Abbreviations .................................................................................................................. xiii Acknowledgements .................................................................................................................. xviii  Chapter 1: Introduction ............................................................................................................... 1 1.1  Protein phosphorylation and phosphorylation sites .......................................... 1 1.2  Protein kinases ................................................................................................... 3 1.2.1 Kinome evolution......................................................................................... 4 1.2.2 Eukaryotic protein kinase catalytic domain ................................................. 7 1.2.3 Activation and regulation of protein kinases .............................................. 11 1.2.4 Substrate specificity of protein kinases...................................................... 15 1.2.5 Gatekeepers and protein kinase inhibitors ................................................. 17 1.3  Methodology in phosphorylation studies ........................................................ 18 1.3.1 Radioactive kinase assays .......................................................................... 21  vi  1.3.2 Phosphosite-specific antibodies ................................................................. 22 1.3.3 High-throughput microarrays..................................................................... 24 1.3.4 Mass spectrometry ..................................................................................... 28 1.3.5 Bioinformatics and computational methods .............................................. 29 1.4  Protein kinases and signalling pathways........................................................... 30 1.5  MAP kinases ERK1/2 ....................................................................................... 32 1.5.1 MAP kinases .............................................................................................. 32 1.5.2 Fundamentals of ERK1/2 ........................................................................... 34 1.5.3 The Ras-Raf-MEK-ERK signalling cascade ............................................. 36 1.5.4 ERK substrates ........................................................................................... 45 1.5.5 Regulation of ERK1/2................................................................................ 50 1.5.6 Targeting the ERK signalling pathway ...................................................... 56 1.6  Rationale and objectives .................................................................................. 61 Chapter 2: Materials and Methods ........................................................................................... 63 2.1  Molecular biology techniques ......................................................................... 63 2.1.1 Plasmids ..................................................................................................... 63 2.1.2 Mutagenesis ............................................................................................... 63 2.1.3 Expression and purification of recombinant proteins ................................ 65 2.2  Antibodies and immunological detection reagents ......................................... 66 2.2.1 Commercial antibodies .............................................................................. 66  vii  2.2.2 Phospho-specific antibody production and validation ............................... 66 2.2.3 Immunological detection reagents ............................................................. 68 2.3 Kinase assays .................................................................................................... 68 2.4  Kinase substrate profiling technique ............................................................... 69 2.4.1 Kinase substrate peptide microarray .......................................................... 69 2.4.2 SPOT membrane ........................................................................................ 70 2.5 Cell lines and cell culture .................................................................................. 71 2.6  Cell stimulation or treatment ........................................................................... 71 2.6.1 EGF or serum stimulation .......................................................................... 71 2.6.2 Phosphatase inhibitor and protease inhibitor treatment ............................. 72 2.7 Cell lysis, immunoprecipitation and immunoblot analysis ............................... 72 2.8 Databases and online resources ........................................................................ 73 2.9 Data analysis ..................................................................................................... 74 Chapter 3: Ancestry of Eukaryotic Protein Kinases ................................................................ 75 3.1 Rationale ........................................................................................................... 75 3.2 Alignment of catalytic domains of typical human protein kinases ................... 77 3.3 Proteins most closely related to protein kinases ............................................... 79 3.4 Ancestry of ePK, ChK and GlnRS .................................................................... 86 3.5 The most ancient STKs and PTKs .................................................................... 93 3.6 Discussion ......................................................................................................... 95  viii  Chapter 4: Investigation of the Substrate Specificities of 200 Human Protein Kinases with Peptide Microarrays ................................................................................................................... 99 4.1 Rationale ........................................................................................................... 99 4.2 Kinase substrate peptide microarray technique .............................................. 101 4.3 Determination of substrate specificity of human protein kinases ................... 103 4.4  Comparisons of substrate specificity of different kinases .............................. 108 4.5 Kinase specificity determination with peptide macroarrays ............................ 110 4.6 Discussion ........................................................................................................ 115 Chapter 5: Regulatory Roles of Conserved Phosphorylation Sites in the Activation T-Loop      of the MAP Kinase ERK1......................................................................................................... 120 5.1 Rationale ......................................................................................................... 120 5.2 T207 and Y210 are highly conserved phosphorylation sites of ERK1 ........... 121 5.3 ERK1 slowly autophosphorylates T207 in vitro ............................................. 122 5.4    T207 and Y210 play critical roles in the regulation of ERK1 phosphotransferase  activity ............................................................................................................ 126 5.5 Mutation at T207 does not affect the specificity of ERK1 towards peptide substrates ........................................................................................................ 127 5.6 Phosphorylation at T207 may reduce the stability of activated ERK1 ........... 131 5.7 Phosphorylation of T207 is regulated by protein phosphatases ...................... 135 5.8 Discussion ....................................................................................................... 136  ix  Chapter 6: Discussion and Future Directions ........................................................................ 141 6.1 The alignment of human protein kinases ........................................................ 141 6.2 Common regulation mechanisms of protein kinases ...................................... 142 6.3 Specificity determining residues and the Kinase Substrate Predictor algorithms ........................................................................................................................ 145 6.4   Conclusion ....................................................................................................... 147  Bibliography .............................................................................................................................. 149 Appendices ................................................................................................................................. 177 Appendix 1. Amino acid frequency in the human STK alignment ............................ 177 Appendix 2. Kinase substrate consensus sequences from literature .......................... 180 Appendix 3. Protocol of the kinase substrate microarray technique .......................... 181 Appendix 4. Specificities of 214 human protein kinases tested on kinase substrate         peptide microarrays ............................................................................... 183 Appendix 5. Specificity of ERK1-pT207 antibody .................................................... 189 Appendix 6. S74 near Subdomain II is an autophosphorylation site ......................... 190 Appendix 7. Generation and evaluation of a new polyclonal generic phosphotyrosine antibody ................................................................................................. 192  x  List of Tables Table 1.1 FDA-approved small molecule protein kinase inhibitors and their therapeutic           t a rgets……………………………………………………… .………… . .  19 Table 3.1 Comparison of average BLAST scores of top subject proteins………………..… 85 Table 3.2 Alignment of consensus sequences of protein-serine/threonine kinases, choline           kinases and glutaminyl-tRNA synthetases………………………………… 89 Table 3.3 List of most probable ancient protein kinases……………………………...…….. 94 Table 4.1 List of the purified recombinant human protein kinases tested on the peptide           microarrays………………………………………… . . .………… .… . . .  106 Table 4.2 Determination of the optimal substrate amino acid specificity of PKAα using           substrate peptide microarrays……………………………………………. 107 Table 4.3 Substrate specificity of multiple MAPK members in vitro………………..…… 111 Table 4.4 Comparison of human protein-tyrosine kinase specificities…………………… 112 Table 5.1 List of ERK1 mutants…………………………………………………………... 128 Table 5.2 List of publications about mutants of the two conserved phosphorylation sites in           activation T-loop……………………………………………… . . . .……  138  xi  List of Figures Figure 1.1 The phylogenetic tree of human kinome……………….………………………….. 6 Figure 1.2 Primary structure of kinase domain…………...…………………………………. 10 Figure 1.3 Structure of conserved protein kinase core………………………………………. 12 Figure 1.4 Mitogen-activated protein kinase cascades……………………………………….. 33 Figure 1.5 Domain structures of ERK1 and ERK2……………………………..……………. 37 Figure 1.6 Domain structures of the Raf isoforms…………………………………………… 42 Figure 1.7 Domain structures of the MEK1 and MEK2……………………...……………… 43 Figure 1.8 The ERK signalling networks…………………..………………………………… 52 Figure 1.9 Primary resistance to Raf-B inhibition……………….…………………………... 60 Figure 3.1 Alignment of human protein kinase catalytic domains…………..……………….. 81 Figure 3.2 Structural comparisons of ePK, ChK and GlnRS……………………..….………. 90 Figure 3.3 Comparison of 3D structures of ePK, ChK and GlnRS………………………..…. 91 Figure 3.4 Predicted secondary structures of STK and GlnRS consensus sequences….……. 92 Figure 4.1 The protein kinase substrate peptide microarray technique……………..….…. 105 Figure 4.2 Screening of positive determinants at -1 and -2 amino acid positions of ERK1           substrates with peptide macroarrays…………………………………… 113 Figure 5.1 Phosphorylation sites in human protein kinase catalytic domains……..…….… 123 Figure 5.2 Phosphorylation of ERK1 T207 and Y210 in vitro…………..…………………. 125 Figure 5.3 Phosphorylation and activity of ERK1 and its mutants……………………..….. 129  xii  Figure 5.4 Kinase specificity of ERK1-WT, T207A and T207E on the Kinex™ Kinase           Substrate Microarray………………………………………………………..…. 130 Figure 5.5 Phosphorylation and activity of ERK1-WT, T207A and T207E in HEK293           cells……………………………………………………………….………….. 133 Figure 5.6 Phosphorylation of ERK T207/T188 in multiple cell lines…………………..… 134    xiii  List of Abbreviations A, Ala   alanine AMPK   AMP-dependent protein kinase Bad    Bcl-2-associated death promoter BSA   bovine serum albumin C, Cys   cysteine CD motif  common docking motif CDK   cyclin-dependent kinase ChK   choline kinase CREB   cyclic-AMP responsive element-binding protein CRS   cytoplasmic-retention sequence D, Asp   aspartic acid DAPK   death-associated protein kinase DMEM   Dulbecco’s modified Eagle medium DTT   dithiothreitol DuSP   dual-specificity phosphatase E, Glu   glutamic acid EDTA   ethylenediaminetetraacetic acid EGF   epidermal growth factor EGFR   epidermal growth factor receptor ELISA   enzyme-linked immunosorbent assay eIF    eukaryotic initiation factor ePK   eukaryotic protein kinase F, Phe   phenylalanine ERK   extracellular signal-regulated kinase FAK   focal adhesion kinase  xiv  FBS   fetal bovine serum FDA   (U.S.) food and drug administration FTase   farnesyltransferase G, Gly   glycine GAP   GTPase-activating protein GEF   guanine nucleotide exchange factor GGTase   geranylgeranyltransferase GlnRS   glutaminyl-tRNA synthetase GPCR   G protein-coupled receptor GRK   G protein-coupled receptor kinase GST   glutathione S-transferase H, His   histidine HGF   hepatocyte growth factor HRP   horseradish peroxidase I, Ile   isoleucine ICC   immunocytochemistry Ig    immunoglobulin IHC   immunohistochemistry IMAC   immobilized metal affinity chromatography JNK   c-Jun N-terminal kinase K, Lys   lysine KAMK2D  calmodulin-dependent protein kinase 2-delta KD    kinase dead KID   kinase insert domain KLH   keyhole limpet hemocyanin L, Leu   leucine LC-MS   liquid chromatography-mass spectrometry  xv  LPS   lipopolysaccharide LSP   local spatial pattern M, Met   methionine MAP2   microtubule-associated protein 2 MAPK   mitogen-activated protein kinase MAST2   microtubule-associated serine/threonine kinase 2 MBP   myelin basic protein MEF   mouse embryonic fibroblast MEK   MAPK kinase MEK1-ΔN3EE active mutant of MEK1 with deletion of the amino terminus (amino acids 32-51) and replacement of serine by glutamic acid at position 218 and 222 MEKK MAPK kinase kinase MEM   minimum essential medium MKP   MAPK phosphatase MORG   mitogen-activated protein kinase organizer MP1   MEK partner 1 MS    mass spectrometry N, Asn   asparagine NES   nuclear export signal P, Pro   proline P70S6K   p70 ribosomal protein S6 kinase PAO   phenylarsine oxide PBS   phosphate buffered saline PCR   polymerase chain reaction PH domain  pleckstrin homology domain PhK   phosphorylase kinase PI3K   phosphoinositide 3 kinase  xvi  PKA   cAMP-dependent protein kinase PMA   phorbol 12-myristate 13-acetate pMBP   phospho-myelin basic protein PP2A   protein phosphatase 2A PTB   phosphotyrosine binding PTK   protein-tyrosine kinase PTM   post-translational modification PTP   protein-tyrosine phosphatase Q, Gln   glutamine R, Arg   arginine RSK   ribosomal protein S6 kinase RTK   receptor-tyrosine kinase S, Ser   serine SAS   saturated ammonium sulphate SDR   specificity-determining residue SDS-PAGE  sodium dodecyl sulphate polyacrylamide gel electrophoresis SH2   Src homology 2 SOS   son of sevenless SRF   serum response factor STK   protein-serine/threonine kinase STP   protein-serine/threonine phosphatase T, Thr   threonine TBS   Tris-buffered saline TCF   ternary complex factor TIF    transcription initiation factor TLR   Toll-like receptor TNF-α   tumour necrosis factor-alpha  xvii  V, Val   valine VEGF   vascular endothelial growth factor W, Trp   tryptophan WT    wild type Y, Tyr   tyrosine    xviii  Acknowledgements First of all, I would like to express my deepest gratitude to my research supervisor, Dr. Steven Pelech, for his continual encouragement, guidance and support throughout my PhD study. He inspired me with his passion, dedication and integrity in the pursuit of scientific discoveries. He provided opportunities for me to attend and participate in scientific conferences, to be involved in training of students in the lab, and to be exposed to research and development in the biotech industry. I am extremely grateful for this great opportunity to finish my graduate studies under Dr. Pelech’s supervision.  I want to sincerely thank my supervisory committee members, Drs. Vincent Duronio, Michael Cox and Leonard Foster, and my Ph.D. thesis examiners, Drs. Rony Seger, Roger Brownsey and Catherine Pallen for their invaluable advice and critiques. I offer my genuine gratitude to all my colleagues from our academic group and Kinexus Bioinformatics, especially to Drs. Dirk Winkler, Hong Zhang, Jane Shi, and Javad Safaei, and Mr. Dominik Sommerfeld and Ms. Winnie So, for their companionship, help and inspiring comments. I would like to thank our industrial collaborators, Drs. Jun Yan and Gary Yalloway, for their technical help. I am also grateful for training for the virus facility and microscopy facility given by Ms. Melanie Bertrand from Brain Research Centre. I am deeply grateful for all my family and friends, especially my mother, who was always there to listen and support. I would not have been able to complete the study without their care and understanding. Last but not least, I would like to express my appreciation to The University of British  xix  Columbia and the Experimental Medicine Program for their financial support.  1  Chapter 1: Introduction  1.1  Protein phosphorylation and phosphorylation sites One fundamental objective of biological research is to understand how extracellular and intracellular signals are transduced and integrated to regulate the behavior and fate of a cell. Reversible phosphorylation of proteins on serine, threonine, and tyrosine residues provides an efficient means to regulate most physiological activities including metabolism, transcription, DNA repair, cell growth, division, and apoptosis (Hunter, 2000). Dysregulation of protein phosphorylation events is implicated in over 400 types of human diseases, including cancer, diabetes, cardiovascular, neurological and immunological disorders. In 1955, Edmond Fischer and Edwin Krebs first described protein phosphorylation on a specific serine residue in glycogen phosphorylase using γ-32P-labeled ATP (Fischer and Krebs, 1955; Krebs, 1993). This event is essential to convert inactive phosphorylase b into active phosphorylase a. Protein-tyrosine phosphorylation was not reported until much later in 1979 (Eckhart et al., 1979). This phosphorylation on tyrosine residues was linked to malignant transformation. Hundreds of different types of covalent modifications of proteins have been identified, including acetylation, methylation, glycosylation, myristoylation, palmitoylation, farnesylation and ubiquitination (Deribe et al., 2010). Some of these forms of post-translational modifications (PTMs) are also very widespread. For example, more than 600 ubiquitin ligases  2  have been identified in the human genome, which regulate a host of biological processes (Pickart and Eddins, 2004; Teixeira and Reed, 2013).  Nevertheless, phosphorylation of serine, threonine and tyrosine residues remains one of the most extensively studied forms of reversible regulation of proteins.  Protein phosphorylation dynamically controls the signalling network in cells through different mechanisms. First, phosphorylation on a regulatory site often induces conformational change of the target protein. For an enzyme, this can be stimulatory (Barford and Johnson, 1989) or inhibitory (Parker et al., 1992) to its catalytic activity. Moreover, phosphorylation sometimes creates specific binding sites for a variety of phospho-specific recognition domains, including Src homology 2 (SH2) domains, phosphotyrosine-binding (PTB) domains for phosphotyrosine residues, and 14-3-3 domains for phosphoserine/phosphothreonine residues. These interactions enable the target protein to recruit specific substrates or interaction partners upon phosphorylation. Adaptor proteins with multiple interaction domains can assemble several phosphoproteins into a functional complex in response to specific stimulations (Pawson, 2007; Seet et al., 2006). Recent studies also found cross-talk between phosphorylation and other kinds of post-translational modifications (PTMs), such as acetylation, ubiquitination and methylation, to integrate the upstream signals (Deribe et al., 2010).  Protein phosphorylation is very common. More than 80% of all the different proteins encoded by the human genome are known to be phosphorylated. Based on one of the most comprehensive online resources, PhosphoSitePlus (http://www.phosphosite.org/), so far nearly 240,000  3  non-redundant phosphorylation sites in diverse species have been identified in total (Hornbeck et al., 2012). Only 5% of these sites were detected using site-specific methods, with most identified by mass spectrometry (MS). Consequently, functional data about these phosphorylation sites is limited to only a few thousand sites. Under normal physiological conditions, most of these confirmed sites are phosphorylated at very low stoichiometry. They can be detected by MS, which is highly sensitive, but not quantitative. One speculation is that these phosphosites are weakly phosphorylated due to the off-target effects of some protein kinases, and thus are not functionally important. Provided that such phosphorylation events are not deleterious to cells, they might have accumulated during eukaryotic evolution, but are not well conserved from species to species. Another possibility is that the hyperphosphorylation of proteins may induce their unfolding and proteolysis as a general mechanism for their down-regulation after initial activation (Steven Pelech, personal communication).  1.2  Protein kinases The protein kinase superfamily is comprised of a large number of enzymes that catalyze the phosphorylation of proteins in eukaryotic species. The complement of diverse protein kinase genes or their encoded proteins within a species is known as its kinome. With the complete sequencing of the human genome, 518 protein kinases were originally identified (Figure 1.1) (Manning et al., 2002b). More recently, with more atypical protein kinases being discovered, this figure has been determined to be closer to 568 human protein kinases (Anthony Hunter, personal communication),  4  which corresponds to 2.7% of the genes in the human genome. About 479 of these protein kinases are typical protein kinases with usually one (and in 13 cases two) conserved eukaryotic protein kinase (ePK) domain that encompassed around 247 amino acid residues. The others are atypical protein kinases (aPKs) that have been shown to possess phosphotransferase activity, but lack sequence similarity to the ePK catalytic domain. Since all of the aPK families are very small in number with no homology to the ePKs, these protein kinases are believed to have distinct ancestry from the typical protein kinases (Manning et al., 2002b). Dynamic regulation of protein kinases by phosphorylation and other mechanisms allows inter- and intracellular signal transduction and integration. Sequentially interacting protein kinases also form cascades as an efficient way to integrate and amplify upstream signalling and direct downstream responses.  1.2.1 Kinome evolution Over a dozen full kinomes have been analyzed (Manning et al., 2002a). The availability of full genome sequences of humans and many other model organisms have allowed researchers to identify most of their protein kinases and to study kinome evolution from yeast to human. In each kinome, kinases have been classified into groups, families and subfamilies. By comparisons of kinomes from multiple eukaryotes, the birth, expansion and contraction of kinase subfamilies have been tracked. Among these highly divergent species, around 50 distinct subfamilies were identified linked to their early common eukaryotic ancestors (http://www.kinase.com/evolution). Reversible phosphorylation on serine, threonine or tyrosine residues is widely adopted in  5  eukaryotic organisms and has provided a powerful means to control cells that feature complicated intracellular compartments.  Prokaryotic organisms use two-component signal transduction systems to respond to environmental stimuli (Hoch, 2000). Phosphorylation on histidine residues induces conformational change of the output domain, which target specific element of downstream genes (Stock et al., 2000). Unlike the eukaryotic signal transduction systems that form complex network, the two-component phosphorylation systems prevent cross-talk to enable tight regulation (Laub and Goulian, 2007). Considering the smaller size of bacterial genomes, these systems provided more economical strategies for prokaryotic signalling transduction. There are two major classes of ePKs: serine/threonine kinases (STKs) and protein-tyrosine kinases (PTKs), although several protein kinases are capable of both types of phosphorylation and some are described as dual-specificity kinases. It is believed PTKs arose from STKs, which was a critical event in early metazoan evolution (Darnell, 1997; Rokas et al., 2005). For example, plants appear to lack PTKs (Rudrabhatla et al., 2006). PTKs were only known in metazoans until the discovery of phosphotyrosine signalling in the choanoflagellate Monosiga brevicollis (King et al., 2008). As among the closest known living relative to animals, the unicellular Monosiga possesses a genome with more protein-tyrosine kinase-related genes than any other identified metazoan (Manning et al., 2008). Orthologues of Src, FGFR and Eph family PTKs were observed in this organism. Some choanoflagellates form colonies under certain conditions, which is consistent with a previous discovery of phosphotyrosine-SH2 signalling system and cell determination in    6   Figure 1.1. The phylogenetic tree of human kinome. Individual kinase groups are coloured differently. Adapted from Manning, Whyte et al., 2002, and Cell Signaling Technology (www.cellsignal.com/common/content/content.jsp?id=kinases-human-protein). Some kinase groups are named after their members (AGC: protein kinase A, G and C families; CAMK: calcium and calmodulin-regulated kinases).  7  slime molds (Kawata et al., 1997). There is insufficient evidence to ascertain whether choanoflagellates evolved from a metazoan to a unique species that behaves like a unicellular organism under most circumstances. However, all observations from Monosiga point to a relationship between phosphotyrosine signalling and intracellular communications, which was already established in the common ancestor of the modern choanoflagellates and metazoans.  The ePKs were found to be evolutionarily related with some metabolite kinases like the choline kinases (ChKs) (Scheeff and Bourne, 2005), another broadly expanded family in eukaryotes. In metazoans, choline kinases catalyze the phosphorylation of choline and ethanolamine, which is the first step of CDP-choline pathway for phosphatidylcholine biosynthesis. Phosphatidylcholine, is the predominant phospholipid of most eukaryotic cell membranes. This phospholipid can also be cleaved into some second messengers that serve as a substrate or activator for phospholipases in signal transduction pathways (Exton, 1994; Kent, 2005). The three-dimensional x-ray structure of choline kinase has been reported to share strikingly high similarity with ePKs (Peisach et al., 2003). The appearance of choline kinases could also be a major advancement in eukaryote evolution, considering the importance of phosphatidylcholine molecules for the creation of internal compartments within cells.  1.2.2 Eukaryotic protein kinase catalytic domain The first protein kinases characterized, glycogen phosphorylase kinase (PhK) and its regulatory kinase cAMP-dependent kinase (PKA), were purified by Krebs and his colleagues  8  (Krebs, 1993; Walsh et al., 1968). In the “pre-genome” era, scientists applied molecular approaches to isolate cDNAs of many novel protein kinases genes (Hanks, 1987). In 1988, Hanks and Hunter aligned the amino acid sequences of 65 protein kinases and described the conserved features of the typical protein kinase domains (Hanks et al., 1988). Protein kinase catalytic domains were identified as sequences that spanned from 250 to 300 amino acids with eleven conserved subdomains that featured 15 particularly conserved amino acid residues at these locations (Figure 1.2). With alignments of the catalytic domains of most of the human protein kinases, the number of defined conserved subdomains was increased to twelve with an additional subdomain near the C-terminal end (Hanks and Hunter, 1995).  The functional roles of these subdomains were initially assigned using the results from site-directed mutagenesis (Gibbs and Zoller, 1991) and analyses of the crystal structure of the catalytic subunit of PKA (Knighton et al., 1991a; Knighton et al., 1991b). In 1994, Taylor and Radzio-Andzelm compared the structure of PKA with cyclin-dependent kinase 2 (CDK2) and extracellular signal-regulated kinase 2 (ERK2), and characterized a common motif of typical protein kinases (Taylor and Radzio-Andzelm, 1994). This catalytic core features a relatively small N-lobe and a bigger C-lobe. Several conserved β-strands and α-helices were shown to be essential for kinase activity (Figure 1.3). The N-lobe contains five β-strands and a universally conserved αC-helix, which serves as an anchor to the C-lobe (Tsigelny et al., 1999). One functionally important motif from the N-lobe is the glycine-rich loop (GxGxxG; Subdomain I). This flexible loop is responsible for positioning the γ-phosphate of ATP. An invariant lysine residue from AxK  9  (Subdomain II) motif helps to couple the α- and β-phosphates of ATP to the C-helix (Zheng et al., 1993). The C-lobe is dominated by α-helices plus one β-sheet. The helical structure forms a stable hydrophobic core of the entire catalytic domain. Anchored to the core, the β-sheet includes strands that function as catalytic and regulatory machinery for the kinase. The catalytic loop (HRD; Subdomain VI) contains a catalytic aspartate residue, which acts as a base to remove a proton from the hydroxyl group in the substrate (Johnson et al., 1996). The DFG motif (Subdomain VII) is important for Mg2+ recognition (Zheng et al., 1993). There are also two conserved residues far from the active site, Glu-208 (in the APE motif (Subdomain VIII)) and R280 in PKA. They form an ion pair that is conserved in all ePK kinase structures. To identify spatially conserved residues from various active and inactive kinases, a local spatial pattern (LSP) alignment was developed and applied (Kornev et al., 2006). This analysis revealed the concepts of the regulatory (R) spine and the catalytic (C) spine that span both lobes, as well as the importance of the F-helix. The R spine contains four non-consecutive hydrophobic residues that link the two lobes, while the C spine is comprised of residues that dock onto the adenine ring of ATP (Kornev et al., 2008). The F-helix is an extremely hydrophobic fragment that not only serves as an anchor for the R- and C-spines, but also promotes the proper conformation of the whole kinase domain (Taylor and Kornev, 2011). Taken together, the F-helix, R-spine and C-spine define the overall architecture of the hydrophobic core of the protein kinase catalytic domain (McClendon et al., 2014).   10    Figure 1.2. Primary structure of kinase domain. The sequences of protein kinase domains of human PKAα catalytic subunit (Uniprot ID: P17612), ERK1 (Uniprot ID: P27361), and Src (Uniprot ID: P12931) were aligned based on the evolutionary conserved subdomains described by Hanks et al. (1988). Colour-coding is based on the ability of certain selected amino acids to show similar structural properties. Pink/red: aspartic acid and glutamic acid (acidic amino acids); blue: histidine, arginine and lysine (basic amino acids); green: leucine, isoleucine, phenylalanine and valine (hydrophobic amino acids with bulky side chains); brown: proline.   11  1.2.3 Activation and regulation of protein kinases Protein kinases are highly regulated to ensure the efficient transduction of inter- and intracellular signals. In some cases, protein kinases are stimulated through binding to allosteric modulators such as second messengers like calcium, cyclic nucleotides and lipid as well as the other proteins. However, activation of many protein kinases is commonly achieved through a phosphorylation event, which leads to dynamic reorganization of its structure. In most protein kinases, especially STKs, activating phosphorylation sites are clustered in a highly variable loop, the activation segment, which extends from the DFG motif to the APE motif. Reversible phosphorylation controls the conformation of this flexible segment, thereby promoting the correct domain orientation for catalysis (Johnson et al., 1996; Kornev et al., 2006). The HRD motif is extremely conserved in all typical protein kinases with an essential aspartate residue for catalytic activity. As is mentioned above, this residue is believed to activate the substrate hydroxyl group that becomes phosphorylated. In most kinases, the residue that immediately precedes the aspartate residue is an arginine residue. The phosphorylation in the activation segment provides a phosphoamino acid to establish an ion interaction and neutralize the charge of the arginine residue from HRD (Johnson et al., 1996). This interaction helps to reorganize the disordered domain structures, coordinate the two lobes and stabilize the active conformation of kinases (Taylor and Kornev, 2011).      12         Figure 1.3. Structure of conserved protein kinase core. The structure of PKAα catalytic subunit is shown complexed with a pseudosubstrate peptide (PDB ID: 2CPK). The conserved structure features a smaller N-terminal lobe dominated by β-sheets and a larger C-terminal lobe with mostly α-helices. Locations of kinase subdomains are indicated with Roman numerals. The protein structure was obtained from the RCSB protein data bank (PDB) (http://www.rcsb.org/pdb/) and visualized using PDB Protein Workshop.     13  There are also other strategies that protein kinases apply to trigger downstream signals. For most receptor-tyrosine kinases, stimulation with ligands is often followed by homo- or heterodimerization and trans-autophosphorylation on specific cytoplasmic residues. These phosphotyrosine sites may recruit target proteins or adaptors to facilitate downstream signals (Pawson, 2004). Adaptors, or scaffold proteins, can also constrain the availabilities of kinases to certain group of substrates by regulating their subcellular localization. Different outputs can be delivered through one adaptor that employs distinct combinations of interaction partners (Pawson, 2007). For example, the A kinase anchor proteins (AKAPs) represent a group of scaffold proteins that bind the regulatory subunit of PKA to control the localization of the holoenzyme. The AKAPs facilitate the assembling of signalling complex by recruiting binding partners including phosphodiesterases (PDEs), specific phosphatases and protein kinases (Skalhegg and Tasken, 2000; Schwartz, 2001). Moreover, protein kinases can be activated by switching off inhibitory mechanisms. For example, the proto-oncogene-encoded kinase Src is inhibited by another protein-tyrosine kinase CSK through phosphorylation at Y530 at its C-terminus. The CSK-inhibited Src can be partially reactivated by CDK1 phosphorylation at a different site before it can be completely reactivated by the protein-tyrosine phosphatase PTPRC (Stover et al., 1994). For a cell to commit to a significant change in its behaviour, it is likely that a sufficient portion of the population of protein kinases must be regulated in a likewise manner to achieve a critical threshold for commitment. Protein phosphatases often have much higher specific  14  enzyme activities than their protein kinase counterparts (S. Pelech, personal communication), and may act to buffer weak stimulations of cells with extracellular mediators. Consequently, the activation of protein kinases is usually transient. These enzymes work like dynamic switches that are tightly regulated. Various mechanisms are employed to turn protein kinases off shortly after their initial activation and prevent prolonged downstream signalling unless there is a sustained exposure to extracellular stimuli. Firstly, there are inhibitory phosphorylation sites in some protein kinases. As mentioned above, Src can be repressed by CSK-dependent phosphorylation on the inhibitory Y530 residue. The internal SH2 domain of Src specifically binds to the phospho-Y530 residue, which leads to a conformational change in this kinase (Kaplan et al., 1994). A more common way to deactivate a kinase is via dephosphorylation of activating phosphosites by protein phosphatases. For example, the tumour suppressor protein phosphatase 2A (PP2A) deactivates Akt protein kinase isoforms by dephosphorylation of multiple activating phosphosites (Rodgers et al., 2011). The PH domain-containing phosphatases (PHLPPs) also act as negative regulators of the Akt kinases (Brognard et al., 2007; Gao et al., 2005), as well as the related protein kinase C (PKC) α and β isoforms (Gao et al., 2008). Akt’s are important components of the PI3K cell survival pathway (Duronio, 2008), and are also negatively regulated by other PTMs, including ubiquitination by ZNRF1 (Medina et al., 2005) and O-linked N-acetylglucosamine glycosylations at activating phosphosites (Wang et al., 2012). In addition, interaction partners sometimes affect kinase phosphotransferase activity by blocking the active sites, competing with the substrates or  15  disturbing the active conformation of the kinases. For instance, p27Kip1 acts as an inhibitor of CDK2, which makes it a key regulator in cell cycle progression and cancer development (Belletti et al., 2005). Additionally, kinases are also regulated at the transcriptional level by transcription factors and other related proteins. In turn, these kinase regulators are often modulated by protein kinases via phosphorylation. All the connections between cell signalling proteins establish robust networks that integrate diverse signals and govern all cellular behaviours.  1.2.4 Substrate specificity of protein kinases The human kinome is comprised of more than 500 protein kinases. Every kinase may contribute in a unique way to support the exquisite control that exists in the signalling networks within which it participates. Besides their differential expression levels in different cell types and their strict regulation in multiple ways, each protein kinase selectively phosphorylates a subset of targets to preserve the high fidelity that exists in cell signalling.  Several molecular mechanisms contribute to protein kinase-substrate pairing specificity, such as the co-expression, co-localization and co-activation of the kinase and its substrates, and protein-protein interactions mediated through direct docking sites or indirectly mediated via adaptor/scaffolding proteins. However, the selectivity of a particular protein kinase for its substrates is influenced to a high degree by molecular recognition of the amino acid sequences surrounding the phosphorylation sites (Kreegipuu et al., 1998; Songyang, 1999). Based on the  16  structure of protein kinases, one of the regions that determine the substrate sequence specificity is located in the same variable segment as the activation loop (Knighton et al., 1991b). The flexible P+1 loop accommodates the residue at the +1 position of the phosphosite, while the peptide binding groove on the surface is responsible for docking the substrate peptide. Differences in primary sequences of this region and shapes of the surface groove contribute to the diverse substrate profiles of protein kinases. Simple alignment of the amino acid residues in phosphorylation sites of known substrate proteins of many protein kinases provides insights into the amino acids preferences of individual protein kinases. This defines what is termed as a consensus substrate recognition sequence for a protein kinase. Favored amino acid residues in substrates show up at the highest frequencies at particular positions surrounding the phosphosite. Exploring the specificity of protein kinases using synthetic peptides has been a fruitful endeavour in cell signalling research for about 40 years (Kemp et al., 1976; Kemp et al., 1977; Casnellie and Krebs, 1984; Tegge et al., 1995; Manning and Cantley, 2002; Schutkowski et al., 2005). Still, only a relatively small fraction of the human protein kinases have been systematically and comprehensively studied for their substrate selectivity (Appendix 2), leaving lots of “orphans” with unknown connections to the rest of the cell signalling apparatus. Recently, algorithms have been developed to predict protein kinase substrate specificity based on the primary amino acid sequence of a kinase’s catalytic domain. In our laboratory, our protein kinase substrate specificity predictor was recently trained with empirical data for over  17  14,000 known kinase-protein substrate pairs and 8,000 kinase-peptide substrate pairs (Safaei et al., 2011). The application of this kinase substrate specificity predictor is an excellent starting place for identifying candidate peptide substrates for assaying target protein kinases of interest that have not been previously well-studied.  1.2.5 Gatekeepers and protein kinase inhibitors In every typical eukaryotic protein kinase, there is a gatekeeper residue positioned between the C- and R-spines. The side chain of this gatekeeper residue has substantial influence on the accessibility of ATP binding pocket (Liu et al., 1998). In the human kinome, about two thirds of the kinases use relatively large residues (Leu, Met, Tyr, Phe) at this position, whereas the rest of them have smaller (Ala, Ser, Thr, Val) gatekeepers (Zuccotto et al., 2010). A smaller gatekeeper residue allows binding of bulky ATP analogues, which provides a useful strategy for designing selective kinase inhibitors. As of 2014, there were 27 small molecule protein kinase inhibitors that have been approved by the U.S. Food and Drug Administration (FDA) (Table 1.1), most of which are ATP-competitive. Despite the fact that all typical protein kinases share similar conformations after being activated, the inactive kinases have distinct features that make them better targets for more selective drugs. However, targeting active protein kinases has potential advantages in minimizing resistance, since activating mutants usually preserve their conformations for catalysis. In addition, there are allosteric kinase inhibitors that target additional regions, or outside of  18  catalytic domains to achieve specificity (Dar and Shokat, 2011; Noble et al., 2004). Monoclonal antibodies are also used to target protein kinases, especially for the extracellular domains of receptor-tyrosine kinases that are amplified or overexpressed in particular types of cancers (Messersmith and Ahnen, 2008). Mutations and dysregulation of protein kinases have been implicated in various human diseases. These enzymes are the second largest group of drug targets, after the G protein-coupled receptors (Cohen, 2002). Personalized, targeted therapies have been the trend for clinical oncology in the past decade (Hayes et al., 2014). Currently, the two major obstacles are the limited availability of drugs and the development of resistance. Resistance to targeted treatments is usually achieved through reactivation of the target kinase by mutations, or activation of alternative signalling pathways. In some cases, drug resistance can be overcome by rational combinatorial targeted therapy, which has recently emerged as a fast growing field in clinical research (Al-Lazikani et al., 2012; Li et al., 2014).   1.3  Methodology in phosphorylation studies Since protein phosphorylation was first reported in early 1900s (Pawson and Scott, 2005), this crucial type of PTM has become one of the most investigated regulatory mechanisms in cell signalling research, with over 200,000 human phosphorylation sites detected by numerous techniques. Phosphosite prediction algorithms developed in our laboratory project the existence of nearly a million different phosphosites [Javad Safaei, personal communication]  19     Table 1.1. FDA-approved small molecule protein kinase inhibitors and their therapeutic targets.      20     Adapted from Roskoski (2012a) with updates from www.accessdata.fda.gov/scripts/cder/drugsatfda/index.cfm     21  (http://www.phosphonet.ca/). Development of protein phosphorylation detection methods has been a productive driving force for advancement in biomedical research. However, in many studies involving the analysis of protein phosphorylation, it is still particularly challenging to develop tools for the sensitive and highly specific detection of phosphorylation sites, and even to assay any protein kinase from cell and tissue specimens with specificity. In this section, several methodologies that are currently used will be reviewed.  1.3.1 Radioactive kinase assays The radioactive kinase assay is the earliest technique scientists applied to detect phosphoproteins or peptides (Fischer and Krebs, 1955). This method involves incubation of a purified or immunoprecipitated protein kinase and its substrate in the presence of radioisotope-labeled ATP (usually [γ-32P] ATP) (Hastie et al., 2006). The measurement of radioactivity allows quantification of phosphotransferase activity of the enzyme. Non-radioactive detection methods were also developed that monitored the production of ADP in the reaction system, although trace amounts of contaminating ADP in ATP preparations can generate high backgrounds (Lowery et al., 2010). Although the in vitro kinase assay is a sensitive and reliable technique to track kinase phosphotransferase activity, it does not provide information on the specific phosphorylation sites that are targeted unless subsequent protein sequencing by the Edman degradation method is also performed (Yu et al., 1998). Additionally, this method is limited to protein kinases with known substrates.   22  The [γ-32P]ATP is also used with purified preparations of protein kinases to label lysates from cells after the cellular proteins are extracted and before they are separated by SDS-PAGE. Lysate proteins may also be radio-labeled with radioactive inorganic phosphate in vivo, and then resolved by SDS-PAGE. In both cases, a film is exposed to visualize radioactive targets on the SDS-PAGE gel or it is subjected to scanning on a phosphorimager. Another classical method used to separate phosphoproteins is 2-dimensional gel electrophoresis, which also reveals alterations in protein mobility after protein phosphorylation (de Graauw et al., 2006). A major limitation of 1- and even 2-dimensional gel electrophoresis is that cell and tissue lysates feature so many thousands of diverse proteins with so many different states of covalent modifications at widely differing concentrations that the resolution of phosphoproteins is insufficient to permit their specific detection and quantitation.  1.3.2 Phosphosite-specific antibodies In biomedical research, purified antibodies are widely used to detect specific protein targets or particular modifications. Traditionally, antibodies are produced by injecting an antigen into a host animal. Sera from these animals, which contain the elicited antibodies, are collected after a period of weeks to months. Polyclonal antibodies, mixed antibodies that bind to the same antigenic peptide but recognize slightly different epitopes, are isolated from the serum. To make monoclonal antibodies, B lymphocytes from the host animal are isolated and fused with a cancer cell line to create a hybridoma. Cell clones that produce exactly the same antibody are then  23  individually selected and propagated (Cole et al., 1984).  The first phosphoamino acid-specific antibody was produced in rabbits immunized with benzonyl phosphate conjugated to keyhole limpet hemocyanin (KLH) in 1980s (Ross et al., 1981). This antibody recognized phosphotyrosine residues in spite of the surrounding amino acid sequence context. These generic phosphotyrosine antibodies have been powerful tools to study cell signalling in response of cytokines and growth factors. Monoclonal phosphotyrosine antibodies were developed, such as 4G10, pY20 and p-TYR-100 (PY100) (Glenney et al., 1988). Recently, a comprehensive comparison of these monoclonal antibodies was done using high throughput approach (Tinti et al., 2012). The results indicated low coverage rates and sequence preferences for all three of the aforementioned generic antibodies that were tested. Considering the wide application of phosphotyrosine antibodies in research, a better generic antibody is much needed. Generic phosphoserine- and phosphothreonine-specific antibodies are also available from multiple suppliers. However, because of the smaller side chains of serine and threonine residues comparing to tyrosine, it is much more difficult to produce a generic phospho-antibody that is selective for phosphoserine or phosphothreonine residues. Commercial antibodies that react with phosphothreonine residues in proteins also commonly bind to phosphotyrosine residues as well [Steven Pelech, personal communication].  A sequence specific phosphosite-antibody against a phosphosite of interest usually can be raised in rabbits and other animals such as mice, goats and chickens following immunization with a synthetic phosphopeptide patterned after the sequence of the phosphorylation site (Ginty  24  et al., 1993; Yung et al., 1997). The antigen peptide is later coupled to an affinity column for purification of polyclonal antibodies. Sometimes peptide competition assay with the unphosphorylated version of the antigen is needed for optimal phospho-specificity. Phospho-specific antibodies have useful applications in many antibody-based techniques including Western blotting, enzyme-linked immunosorbent assays (ELISA) and immunocytochemistry/immunohistochemistry (ICC/IHC), and thus are fundamental tools in cell signalling research (Brumbaugh et al., 2011).  1.3.3 High-throughput microarrays Based on the printing technologies originally developed for oligonucleotide microarrays, peptide/protein microarrays provide cost-efficient solutions for high-throughput screening. Microarrays commonly used in cell signalling studies, include in antibody microarrays, lysate/tissue microarrays, recombinant protein microarrays and synthetic peptide microarrays.  Antibody microarrays There are two basic formats of protein microarrays, the forward phase arrays and reverse phase arrays. The antibody microarray is a most broadly used forward phase array in which antibodies are printed at defined positions on a glass or plastic slide and tested for binding various mixtures of proteins, usually in cell lysates or tissue samples. This method allows a rapid screening for an extensive number of targets for their protein levels and PTMs under diverse  25  experimental perturbations or pathological conditions (Borrebaeck, 2006).  Major drawbacks of this technique are the false positives caused by cross-reactivity and detection methods, as well as the false negatives caused by inaccessibility of some target epitopes in protein complexes (Angenendt, 2005). It is also possible that apparent changes in protein expression or phosphorylation of target proteins may also arise from alterations in the association of the target proteins with other proteins. Scientists have taken great efforts to improve sensitivity and overcome these limitations. Direct labeling of protein samples with one single-colored fluorescence dye has proven to be preferred in most cases (Wingren and Borrebaeck, 2008). However, even with the most robust protocols, the resultant data should still be handled cautiously. Follow-up validation of the top hits using Western blots or reverse phase arrays is usually required. This can be challenging, as antibody microarrays operate at a magnitude or greater sensitivity than Western blots (Zhang and Pelech, 2012). For example, with immunoblots, there are more limitations with respect to the total volume of protein that can be loaded on an SDS-PAGE gel, and the immunoreactive protein is dispersed over a larger area in a gel band (typically 4 mm or wider) than a microarray spot (around 0.1 mm in diameter).   Lysate/tissue microarrays Lysate/tissue microarrays are reverse phase arrays that have multiple cell lysates or issue samples spotted individually on microarray slides. This method works like a high throughput ELISA, and requires only microgram amount of samples. Basically, one antibody will be used to  26  probe for changes in expression or post-translational modification of one target protein of interest in all printed lysates on the array (Speer et al., 2005). Control of the sample concentration for each spot is a key factor when producing lysate microarrays. Often a lysate will be printed on different spots at different concentrations to allow for detection and quantification in a linear range. This technique is a powerful strategy for biomarker discovery and translational research as it can permit the simultaneous screening of hundreds of biological specimens (Mueller et al., 2010; VanMeter et al., 2007).  Peptide macro- and microarrays When compared to proteins and cell lysates, peptides are easier to synthesize and are usually more stable, which makes them especially suitable for large-scale arrays. Although peptide arrays can be used to test for protein-protein interactions and compound binding, one of the most fruitful applications of this method in cell signalling studies is for determination of the substrate specificity of protein kinases (Reimer et al., 2002; Reineke et al., 2001). It can also be a useful tool for epitope mapping of antibodies for signalling proteins. The SPOT technique allows flexible and inexpensive synthesis of large number of peptides on cellulose membranes (Frank, 2002; Winkler and Campbell, 2008). These peptides can be released and immobilized on microarray slides, or used directly for kinase assays as peptide macroarrays. Different strategies can be taken to select a library of peptides for protein kinase profiling (Manning and Cantley, 2002; Schutkowski et al., 2004; Schutkowski et al., 2005). A  27  knowledge-based peptide library of confirmed physiological phosphorylation sites can be chosen in a microarray screening, while sequences modified from known protein substrates are also applicable. Quantitative analysis of the microarray image permits derivation of a consensus substrate sequence, which can later be used as a wild type template of a substitutive library to be tested on the macroarray. In other cases, combinatorial libraries are employed for de novo screenings on the SPOT membrane (Uttamchandani et al., 2003). Solid-phase phosphorylation approach using rcombinant proteins expressed in phage plaques from a cDNA library can also be applied to detect protein substrates of a kinase in vitro. However, this method cannot identify specific sites that are phosphorylated in the substrate protein sequences (Fukunaga and Hunter, 2004). Kinase substrate determination with peptide macro- and microarrays has been demonstrated to very successfully identify an overall preference of specific amino acids at each given position surrounding a phosphosite (Appendix 2). One limitation of this technique is the potential loss of contextual information unique to proteins that reflects higher ordered structures. This is, however, overweighed by the advantages of their flexibility, easiness, convenience and ever-decreasing costs. A specific substrate peptide can serve as a powerful investigative tool to assay protein kinase activity in vitro with high sensitivity and selectivity, and to provide insight in the regulation of a target protein kinase in cell signalling networks.   28  1.3.4 Mass spectrometry The development of mass spectrometry (MS) has been a major driving force in proteomics and phosphoproteomics studies. This technique allows comprehensive assessment of phosphoproteins in complex biological systems with high resolution and sensitivity (Mann et al., 2002). The application of liquid chromatography-mass spectrometry (LC-MS) or tandem mass spectrometry (LC-MS/MS) enables separation, identification and sequencing of trypsin digested peptide products from cell lysates (Xie et al., 2011). For the purpose of phosphopeptide detection, one major problem is the low abundance of phosphoproteins comparing to the non-phosphorylated protein background. Phosphopeptide enrichment by immobilized metal affinity chromatography (IMAC) or phosphospecific antibodies helps to overcome this issue (Brill et al., 2004). Other strategies involving chemical modifications were also developed to improve efficiency in phosphorylation detection (Oda et al., 2001; Zhou et al., 2001). So far, over 95% of all the confirmed human phosphorylation sites were detected by MS (http://www.phosphosite.org/), and this has made a tremendous contribution to our knowledge about the composition of phosphoproteomes. However, due to its high costs, the large amount of starting cell lysate protein required, the need for many separate MS runs and its non-quantitative nature, this technique is just not practical for analysis of large numbers of specimens. In view of this, MS is a powerful means to identify novel phosphosites, but is not really desirable for tracking large numbers of specific phosphosites of interest in many cell or tissue samples.   29  1.3.5 Bioinformatics and computational methods The development of phosphorylation detection techniques has dramatically increased our understanding of the complexity of cell siganlling network. Bioinformatics methods play critical roles in analysis, storage and presentation of massive empirical data (Chen and Eschrich, 2014; Liu and Chance, 2014). For example, several computational tools have been developed to analyze MS proteomics data (Matzke et al., 2013; Mueller et al., 2008), including MaxQuant (Cox and Mann, 2008), LFQuant (Zhang et al., 2012) and SuperHirn (Mueller et al., 2007). Statistical tools are applied for normalization in diverse high-throughput platforms to remove systematic biases (Callister et al., 2006; Leek et al., 2010). Databases have provided comprehensive online resources for these annotated data (Xue et al., 2010). Furthermore, with all the phosphoproteomics data available, bioinformaticians have developed predictive algorithms such as Scansite (Obenauer et al., 2003), NetworkKIN (Linding et al., 2008), NetPhorest (Miller et al., 2008), and the Kinexus Kinase Substrate Predictor used in PhosphoNET (Safaei et al., 2011), which predict protein-protein interactions, especially kinase-substrate interactions based on short sequence motifs. Bioinformatics and computational approaches facilitate the identification of protein connections in cell signalling networks, which leads to improved knowledge of physiological processes and human diseases, as well as discoveries of potential biomarkers and drug targets.   30  1.4  Protein kinases and signalling pathways Intercellular and intracellular signals are usually transduced though signalling pathways. Protein phosphorylation mediated by protein kinases provides a most common mechanism of communication to govern all aspects of cellular activities (Krebs, 1993; Pawson and Scott, 2005). At the surface of cells, a variety of transmembrane-spanning receptors mediate responses to different signalling molecules, including cytokines, growth factors, hormones and neurotransmitters. The binding of ligands to the specific receptors results in the activation of intracellular domains and/or the release of second messengers, leading to a myriad of physiological responses. Different signalling pathways cross-talk to form complex signalling networks that control the behaviour of cells. The G protein-coupled receptors (GPCRs), also known as seven-transmembrane domain (7TM) receptors, constitute the largest family of cell surface receptors in all eukaryotic species examined to date (Lefkowitz, 2004). Ligand binding induces conformational changes of the GPCRs, which allows the exchange of GDP for GTP on associate G proteins. The Gα subunit of the trimeric G protein and the bound GTP will then disassociate from the β and γ subunits to trigger downstream signals. One of the best studied classical pathways involving GPCRs involves cyclic-AMP (cAMP) signalling (Gilman, 1995). The activated Gαs subunit stimulates adenylyl cyclase, which produces cAMP from ATP. Four molecules of cAMP bind to the two regulatory subunits of PKA, which causes the two catalytic subunits to dissociate from the inhibitory regulatory subunit dimer and manifestation of their phosphotransferase (Taylor et al.,  31  1990). PKA phosphorylates a multitude of diverse protein substrates, including for example, the cAMP-responsive element-binding protein (CREB), which is a transcription factor that is activated upon PKA phosphorylation and recognizes genes with a specific promoter for transcription activation (Sands and Palmer, 2008).  The receptor-tyrosine kinases (RTKs) are another important group of receptors for many growth factors, cytokines and hormones. Activation of RTKs results in autophosphorylation of their intracellular domains and phosphorylation of intracellular substrates. Tyrosine phosphorylation provides docking site for adaptor proteins or downstream signalling proteins (Lemmon and Schlessinger, 2010). For example, the epidermal growth factor receptor (EGFR) chains form dimers when activated, and trans-autophosphorylate on multiple tyrosine residues. Adaptor proteins like Grb2 are then recruited to activate, via the guanine nucleotide exchange protein SOS, the subsequent Ras-Raf-MEK-ERK kinase cascade (Pullikuth and Catling, 2007). This important signalling arm amplifies upstream signals from diverse receptors and regulates various cellular processes through phosphorylation of hundreds of cytosolic and nuclear substrates including protein kinases, phosphatases, transcription factors, cytoskeleton proteins, and many others (Pearson et al., 2001; Plotnikov et al., 2011; Roskoski, 2012a). The following section will be focused on this extensively investigated cascade and the highly related protein kinases ERK1 and ERK2.   32  1.5  MAP kinases ERK1/2 1.5.1 MAP kinases Mitogen-activated protein kinases (MAPKs) play fundamental regulatory roles in diverse cellular processes, including differentiation, proliferation, cell movement and apoptosis. This family of protein-serine/threonine kinases is ubiquitous and conserved in all eukaryotes.  In mammals, the best studied groups are the extracellular signal-regulated kinases 1, 2 and 5 (ERK1, 2 and 5), p38 MAPKs (α, β, γ and δ) and c-Jun N-terminal kinase (JNK1, 2 and 3) (Pearson et al., 2001; Raman et al., 2007; Schaeffer and Weber, 1999). MAPKs lie within multi-tiered signalling modules (Figure 1.4). MAPKs are usually phosphorylated and activated by upstream MAPK-kinases (MAPKKs or MAP2Ks), MAPKK-kinases (MAPKKKs or MAP3Ks) and MAPKKKKs (MAP4Ks). MAP3K and MAP4Ks respond directly to the external stimuli, or through small GTPases and other kinases. The ERKs are often indirectly stimulated by mitogens and growth factors, whereas p38 and JNKs are recruited in response to stress or inflammatory cytokines (Plotnikov et al., 2011). After being activated, MAPKs phosphorylate a variety of substrate proteins in both the cytoplasm and nucleus. All MAPKs are tightly regulated under different mechanisms. Crosstalk between the MAPK signalling pathways are also observed (Junttila et al., 2008). Dysregulation of these kinases has been reported in many human diseases including various cancers (Dhillon et al., 2007). 33     Figure 1.4. Mitogen-activated protein kinase cascades. MAPK cascades are comprised of multi-tiered modules. In response to various stimuli, activated small GTPases or MAP4Ks mediate the phosphorylation and activation of MAP3Ks. MAP3Ks phosphorylate MAP2Ks, which in turn phosphorylate specific MAPKs. MAPKs then phosphorylate diverse cytosolic and nuclear substrates to regulate cell growth, differentiation, apoptosis, etc. Adapted from Plotnikov et al. (2011).  34  1.5.2 Fundamentals of ERK1/2 ERK1 and ERK2 are the two best studied MAPKs and were discovered more than 25 years ago as insulin-stimulated protein-serine/threonine kinases that phosphorylated microtubule-associated protein 2 (MAP2) (Hoshi et al., 1989; Ray and Sturgill, 1987) and myelin basic protein (Cicirelli et al., 1988; Pelech et al., 1988). Human ERK1 and ERK2 are 43 kDa and 41 kDa proteins, respectively, which share 84% identity (Figure 1.5). The larger ERK1 has an N-terminal insertion before the protein kinase domain. When compared to the other eukaryotic protein kinases, ERK1/2 contain a 31-amino-acid kinase insert domain (KID) within the catalytic domain, which is believed to provide functional specificity.  Both ERK1 and ERK2 are commonly, but not exclusively, co-expressed in most cell types. Numerous studies have so far revealed extremely similar behaviours and functions for both isoforms. Firstly, all the stimulations of ERK1/2 pathway that were observed appear to activate both ERK1 and ERK2 (Lefloch et al., 2009). Their upstream MAP2Ks, i.e. MEK1 and MEK2, appear to phosphorylate both ERKs equally. Secondly, researchers found the contribution of ERK1 and ERK2 to growth signalling depended on their total expression levels. The ratio of activated ERK1/ERK2 corresponds with the ratio of their protein levels. The output downstream signalling reflects the total amount of ERK1 and ERK2 combined, which means the two ERKs are activated in parallel (Lefloch et al., 2008). Moreover, in vitro experiments have indicated identical substrate specificities for recombinant ERK1 and ERK2 (Robbins et al., 1993).  35  While many efforts have failed to differentiate the functions of ERK1 and ERK2, the two isoforms cannot completely compensate for each other in vivo. Ablation of erk2 gene in mice was found to be embryonic lethal due to the defects in mesoderm differentiation (Yao et al., 2003). It was also reported that the placental development in erk2-knockout mice was severely impaired (Hatano et al., 2003). However, in erk2-null embryonic stem (ES) cells, growth and proliferation was not affected, because of the augmented activity of ERK1. Although the overall ERK activity is reduced based on the decreased phosphorylation level of the downstream RSK substrate in these cells, ERK1 can partially compensate for ERK2 loss in vitro (Yao et al., 2003). By contrast, the erk1 gene seems dispensable for development, since the erk1-deficient mice are viable and fertile. Nonetheless, researchers found that thymocyte maturation beyond the CD4+CD8+ stage was reduced significantly in these mice, indicating an important role of ERK1 in thymocyte development (Pages et al., 1999). Recent studies using mathematical modeling have shed some light on the functional differences between the two isoforms, including the different nuclear shuttling rates of ERK1 and ERK2 (Harrington et al., 2012). The N-terminal domain of ERK1 is believed to account for the slower trafficking and lower signalling capacity of ERK1 comparing to the smaller ERK2 (Marchi et al., 2008). Schilling et al. (2009) also reported the isoform-specific ERK signalling to cell fate decisions based on a cytokine-induced ERK activation model. Mathematical analysis revealed that dually phosphorylated ERK1 attenuated proliferation beyond a certain activation level, whereas activated ERK2 was able to stimulate proliferation even with saturation kinetics.  36  Despite of the minor differences between ERK1 and ERK2, it is still believed that they are activated and regulated in parallel under most circumstances. The most important factor when studying cellular regulation and drug effects is the total levels of both of these kinases.  1.5.3 The Ras-Raf-MEK-ERK signalling cascade The dual-specificity kinases MEK1 and MEK2 are the upstream MAP2Ks for ERK1/2. MEK1/2 activate ERK1/2 by phosphorylating one tyrosine and one threonine residue in a TEY motif in the activation loops of ERK1/2 between their Subdomain VII and VIII motifs. The MAP3Ks that activate MEK1/2 are the Raf isoforms Raf-A, Raf-B and c-Raf/Raf-1 as well as Cot/Tpl2 and Mos. The activation of Raf kinases is induced by small G proteins from the Ras family such as H-Ras, K-Ras and N-Ras through homo- or heterodimerization and phosphorylation. All these components form the Ras-Raf-MEK-ERK signalling cascade, which serves as a paradigm for cell signal transmission through sequentially activating protein kinases.  Ras small G proteins Ras proteins are molecular switches that cycle between an inactive Ras-GDP conformation and an active Ras-GTP conformation. The human genome encodes three highly homologous Ras genes, H-Ras, K-Ras and N-Ras, which are members of a much larger superfamily of Ras-related genes. Like other guanine nucleotide binding proteins, the status of these Ras proteins is controlled by guanine nucleotide exchange factors (GEFs), which activate Ras by promoting the  37           Figure 1.5. Domain structures of ERK1 and ERK2. ERK1 and ERK2 are 43 and 41 kDa proteins with 84% identity within their primary sequences. ERK1 contains a 17-amino acid N-terminal insertion comparing to ERK2. Both kinases feature a kinase insert domain (KID) that contributes to substrate specificity. The activation T-loop contains multiple confirmed phosphorylation sites, including the activating TEY motif.     38  exchange of GTP for GDP, and by GTPase-activating proteins (GAPs), which stimulate the intrinsic GTPase activity of the Ras proteins to hydrolyze GTP, returning it to its deactivated GDP-bound state (Pylayeva-Gupta et al., 2011; Vigil et al., 2010). The Ras activation by GEFs is usually mediated by receptor protein kinases, such as epidermal growth factor (EGF) receptor family members, vascular endothelial growth factor (VEGF) receptors, and insulin receptors (Lemmon and Schlessinger, 2010). Binding of the ligands induces dimerization and trans-autophosphorylation on tyrosine sites, which creates binding sites for adapter/scaffold proteins such as Shc and Grb2 (Pullikuth and Catling, 2007). Grb2 in turn recruits the GEFs Sos1 and Sos2 to the plasma membrane to activate Ras and the MAPK signalling cascade. Other ways to activate Ras involve G protein-coupled receptors (Luttrell, 2002) and integrins (Yee et al., 2008). Shortly after the activation of Ras, Ras-GAPs will accelerate the GTPase activity of Ras to prevent prolonged downstream pro-growth signalling. Mutations that interfere in this deactivation process are oncogenic. In fact, the Ras isoforms are among the most commonly mutated genes in human cancers. About 30% of all cancers harbor oncogenic Ras mutations. The most frequent mutations are located at G12, G13 and Q61, which are critical residues involved in GTP binding and the hydrolysis reaction (Pylayeva-Gupta et al., 2011). The Q61 mutant of Ras fails to coordinate the water molecule for GTP hydrolysis (Buhrman et al., 2010; Scheidig et al., 1999), whereas G12 and G13 mutations disturb the interaction of Ras with GAP and the proper orientation of the Q61 catalytic residue (Scheffzek et al., 1997). As a result, all these oncogenic  39  mutations lead to a persistent GTP binding state of Ras, and therefore constitutively active Ras downstream signals.  The Ras-like small GTPase Rap1 was first reported due to its ability to restore the morphological phenotype of K-Ras-transformed fibroblasts (Kitayama et al., 1989; Kitayama et al., 1990). Rap1 is involved in the inhibition of Ras-dependent downstream signalling, and the differential regulation of Raf isoforms by cAMP (Erhardt et al., 1995). Raf-B can be activated by Rap1 (Ohtsuka et al., 1996; Vossler et al., 1997). However, binding of Rap1 to Raf-1 interferes with the activation of the kinase due to much stronger interaction of Rap1-GTP to the cysteine-rich domain (Hu et al., 1997). Rap1 is regulated by a variety of stimuli including growth factors, cytokines, as well as stress and physical forces (Hattori and Minato, 2003; Gloerich and Bos, 2011). For example, Rap1 mediates the crosstalk between PKA pathway and Raf-MEK-ERK pathway through the GEF EPAC (exchange protein directly activated by cAMP) (de Rooij et al., 1998). Rap1 is also phosphorylated by PKA, which abolishes its binding activity to the cysteine-rich domain of Raf-1 and activates Raf-1 downstream signals (Hu et al., 1999).  Raf isoforms In humans, there are three Raf isoforms, Raf-A, Raf-B and Raf-1. These kinases are 75-90 kDa proteins with an N-terminal regulatory domain and a C-terminal catalytic domain (Figure 1.6). Raf-1, which is also known as c-Raf, was the first MAP3K identified upstream of ERK1/2 (Rapp et al., 2006). The activation of Raf isoforms by Ras includes multiple steps. Firstly,  40  binding with Ras-GTP releases the inhibitory N-terminal domain from the kinase domain. The initial conformation changes then lead to homo- or hetero-dimerization, followed by the phosphorylation in the activation loops to fully activate the kinases. Activating mutations in Raf isoforms, especially Raf-B, are very frequent in various types of human cancers, including melanoma, thyroid, pituitary and large intestinal cancers. In particular, the V600E mutation in Raf-B occurs in about 11.5% of all human cancers according to the COSMIC database for somatic mutations in human cancer  (Forbes et al., 2014) (http://cancer.sanger.ac.uk/cosmic/gene/analysis?ln=BRAF#histo). A conserved acidic motif (NtA) that precedes the kinase domain is responsible for the main difference between Raf-B and the other two isoforms (Hu et al., 2013; Shaw et al., 2014). There are one or two serine residues in the NtA of all three Rafs. In Raf-A and Raf-1, the serine sites are only phosphorylated after the activation of the kinases. In Raf-B, the NtA (SSDDW) is constitutively phosphorylated on the both serine residues (Mason et al., 1999), which serves as a capping motif to coordinate the kinase domain in Raf dimers for allosteric activation. Therefore, Raf-B, but not Raf-A or Raf-1, can function as an allosteric activator through homo- or heterodimerization even in the kinase-inactive form (Wellbrock et al., 2004). Besides the Raf isoforms, other MAP3Ks also activate ERK1/2 under some specific stimulation or in more restricted cell types. For example, Tpl2, originally identified as the Cot proto-oncogene, is required for ERK1/2 activation in tumour necrosis factor-alpha (TNF-α) treated mouse embryonic fibroblasts (MEFs) (Das et al., 2005), macrophages and B cells  41  stimulated by lipopolysaccharide (LPS) (Dumitru et al., 2000) or upstream Toll-like receptor (TLR) signals (Babu et al., 2006). Mos is another MAP3K that mediates ERK activation during oocyte maturation. The activation of this pathway depends on Mos translation during the initial burst of protein synthesis that is triggered at the time of germinal vesicle breakdown. Paxillin, a scaffold protein that is often found in focal adhesions, is also required in the maturation process (Rasar et al., 2006).   MEK1/2 MEK1 and MEK2 are the MAPKKs downstream of Raf in the MAPK signalling cascade (Figure 1.7). MEK1 mutations have been implicated in the etiology of some melanomas and colon cancers (Murugan et al., 2009) as well as cardio-facio-cutaneous syndrome (Rodriguez-Viciana et al., 2006). These dual-specificity kinases are activated though phosphorylation of two serine sites (S218 and S222) in their activation segments (Zheng and Guan, 1994). Activated MEK1/2 then catalyze the phosphorylation of the TEY activation motif in ERK1/2 (Roskoski, 2012b). It was generally believed that ERK1/2 are the only physiological substrates of MEK1/2. However, MEK1 can provide positive feedback phosphorylation of Raf1 through induction of S338 phosphorylation (Hu et al., 2013). Recently, MEK1 was also reported to phosphorylate HSF1, the master regulator of the proteotoxic stress response, to guard proteome stability and suppress amyloidogenesis (Tang et al., 2015). Data from our lab   42         Figure 1.6. Domain structures of the Raf isoforms. The Raf isoforms share three conserved regions, the regulatory domains CR1 and CR2, and the protein kinase domain (CR3). CR1 contains the Ras-binding domain and the cysteine-rich domain, which are both essential for membrane recruitment and activation of Raf. The inhibitory acidic motif (NtA) features multiple phosphosites. S445 of Raf-B is constitutively phosphorylated. The indicated serine sites in CR2 and at the C-terminus are responsible for 14-3-3 binding. Adapted from Wellbrock et al. (2004).      43              Figure 1.7. Domain structures of the MEK1 and MEK2. MEK1 and MEK2 consist of a N-terminal regulatory region, a dual-specific protein kinase domain, and a short C-terminal sequence. The N-terminal region contains the ERK docking site, and a nuclear export sequence (NES) that overlaps the autoinhibitory segment. Multiple phosphorylation sites have been identified in the activation loop, including activating sites (S218 and S222) and inhibitory sites (S212). MEK1/2 are also phosphorylated in the proline-rich domain, which facilitates the binding of Raf isoforms. Phosphorylation of T292 of MEK1 serves as a negative feedback mechanism. The phosphosites at the C-terminus are responsible for regulation of nuclear localization.     44  has demonstrated that MEK1/2 may indeed have other unidentified physiological substrates (Steven Pelech, personal communication). MEK1/2 serve as a convergence point for the upstream signals. They can be activated by a variety of MAP3Ks in response to different stimulations or in diverse cell types. The substrate specificity of MEK1/2 is very restricted. MEK1/2 do not appear to phosphorylate denatured ERK1/2 nor substrate peptides with the corresponding sequences (Seger et al., 1992). An N-terminal ERK binding domain is required for the interaction and phosphorylation of ERK1/2.  MEK1/2 are also regulated by additional phosphorylation sites. Gopalbhai et al. 2003) reported that S212 was a negative regulatory phosphosite of MEK1 Substitution of the serine to alanine increased basal activity of MEK1, while phospho-mimicking mutation to aspartic acid completely suppressed the activity of both wild type and constitutively active MEK1. In addition, MEK1, but not MEK2, is phosphorylated and inhibited by ERK1/2 at T292 in a negative feedback mechanism (Rossomando et al., 1994). Phosphorylation of T292 facilitates the dephosphorylation of the S218 and S222 activating sites in MEK1 and MEK2 homo- and heterodimers. Knocking down MEK1 would disable this negative feedback inhibition and cause prolonged ERK1/2 activation by phosphorylated MEK2 (Catalanotti et al., 2009). Finally, phosphorylation at the C-terminus of MEK1 induces nuclear accumulation of the kinase (Chuderland et al., 2008). Scaffold proteins, for example, the kinase suppressor of Ras isoforms KRS1 and KRS2, are also supportive regulators of MEK1/2 (Brennan et al., 2011). KSR1/2 induces a conformational  45  switch of MEK1/2 in the Raf-KSR-MEK complex, thus facilitating MEK phosphorylation and activation by Raf isoforms. KSR also interacts with many other proteins, including G protein γ subunits, 14-3-3, Hsp90 and c-TAKs, all of which participate in the regulation of Raf-MEK-ERK signalling cascades (Morrison and Davis, 2003).  1.5.4 ERK substrates Unlike the Rafs and MEKs, which have very restricted specificities, close to 900 known phosphosites distributed in membrane-bound, cytoplasmic and nuclear proteins have been documented as substrates of ERK1/2 so far (Yoon and Seger, 2006; Xue et al., 2014;). ERK1/2 execute their functions though these substrates under different upstream stimulations, and in various subcellular locations. The optimal consensus sequence for ERK1/2 substrate recognition was deduced as P-X-S/T-P (Clark-Lewis et al., 1991; Gonzalez et al., 1991). Binding of ERK1/2 to their substrate is induced by phosphorylation of TEY in the activation loop and substantial conformation change upon activation (Canagarajah et al., 1997). The P+1 loops of ERK1/2 specifically recognize Pro at the +1 position. The Pro at the -2 position is less conserved, but preferable as well in vitro. However, this interaction between the phospho-acceptor region and ERK catalytic site is not able to provide sufficient affinity and specificity in vivo. Secondary docking interaction is usually required, which typically involves a common docking (CD) motif (or cytoplasmic-retention sequence/CRS) in ERK1/2 (Rubinfeld et al., 1999; Tanoue et al., 2000)  46  and a D domain in the substrate (Garai et al., 2012; Holland and Cooper, 1999; Sharrocks et al., 2000). The CD motif is located at the C-terminus of ERK1/2 immediately after the protein kinase domain. Featured by three acidic residues and two tyrosine residues, the CD motif binds to the positively charged and hydrophobic amino acids from the D domain. D domains exist not only in ERK1/2 substrates, but also in MEKs and MAPK phosphatases (MKPs), indicating the functional role of this motif is not limited to substrate specificity of the ERK kinases. For instance, MEK1/2 can function as cytosolic anchors of inactive ERK1/2 through docking interactions (Rubinfeld et al., 1999). Once activated, MEKs and ERKs disassociate from each other and localize into nucleus separately. MEK1/2 then selectively bind to ERK1/2 that is dephosphorylated by MKPs and facilitate nuclear exportation of the inactive ERKs (Adachi et al., 2000). Besides D domains, some other ERK substrates use an additional docking site, termed the DEF motif to obtain higher affinity (Fantz et al., 2001; Garai et al., 2012; Jacobs et al., 1999). The DEF motif has a common sequence F-X-F-P, which binds to a hydrophobic pocket in the large lobe of ERK1/2 (Lee et al., 2004). Unlike the D domain, DEF motif only interacts with active ERK1/2, since the binding site is only exposed after phosphorylation of the activation loops. Collectively, these interactions motifs collaborate to achieve a higher order of specificity of ERK1/2 signalling. Protein substrates that are downstream of ERK1/2 were until just recently mostly discovered by individual study of growth factor modulated proteins. These substrates include various kinds of proteins, such as transcription factors, protein kinases, protein phosphatases,  47  cytoskeleton proteins, receptors, apoptosis related proteins, and many others (Yoon and Seger, 2006). Some of the best studied substrates will be discussed below.  RSK The ribosomal S6 kinases (RSKs) are protein kinases that involve in both cytoplasmic and nuclear ERK signalling. Four human isoforms, RSK1-4 have been identified with high similarity. They play key roles downstream of ERK1/2 in control of cell cycle progression, apoptosis inhibition, and cell migration (Frodin and Gammeltoft, 1999; Lara et al., 2013). RSKs also participate in cancer cell invasion and metastasis (Sulzmaier and Ramos, 2013). All four RSK isoforms feature in tandem two active protein kinase catalytic domains with distinct functions. The C-terminal kinase domain is involved in autophosphorylation and activation by ERK1/2, whereas the N-terminal kinase domain catalyzes the phosphorylation of exogenous RSK substrates. The activation of RSKs requires sequential phosphorylation events (Dalby et al., 1998). Taking human RSK1 as an example, T573 in the activation loop is first phosphorylated by ERK1/2, followed by the phosphorylation of the nearby S363 and S359. Phosphorylation of the three sites leads to activation of the C-terminal kinase domain and trans-autophosphorylation of S380, which then recruits PDK1 to phosphorylate S221 in the activation loop of the N-terminal kinase domain (Frodin et al., 2000). Additionally, a phosphosite near the ERK docking site, S733 is necessary for the dissociation of RSKs from the ERKs, and allows the dephosphorylation and deactivation by phosphatases (Roux et al., 2003).  48  The RSK isoforms phosphorylate a number of transcription factors including CREB, transcription initiation factor 1A (TIF1A), nuclear factor-κB (NF-κB), and serum response factor (SRF) (Anjum and Blenis, 2008). RSKs enhance protein synthesis through ribosomal protein S6 and eukaryotic initiation factor 4B (eIF4B). RSK1 and RSK2 are reported to promote cell cycle progression by suppressing CDK inhibitor p27Kip via phosphorylation. Also, RSK isoforms phosphorylate and inhibit pro-apoptosis proteins such as Bcl-2-associated death promoter (Bad) and death-associated protein kinase (DAPK) (Frodin and Gammeltoft, 1999).   Elk1 Elk1 is one of the most extensively studied ERK1/2 targets in the nucleus. Elk1 belongs to the ternary complex factors (TCFs) subfamily of ETS-domain transcription factors. Multiple MAPK consensus phosphorylation sites were identified in the C-terminal transcriptional activation domain (Buchwalter et al., 2004; Hollenhorst et al., 2011), including S324, T336, T353, T363, S383, S389 and S422 (Cruzalegui et al., 1999; Gille et al., 1995). Out of these sites, in particular phosphorylation of S383 and S389 by ERK1/2 induces the activation of Elk1. The interaction of Elk1 with ERK1/2 is mediated by the D domain as well as the DEF region of Elk1 (Jacobs et al., 1999). Binding of each docking domain is required for a distinct subset of phosphosites. Specifically, DEF domain is essential for phosphorylation of S383 and its nearby sites, whereas D domain interaction is responsible for the phosphorylation of the other sites (Fantz et al., 2001). This precise regulatory mechanism makes Elk1 an attractive model for  49  investigation of phosphorylation-dependent transcriptional activation.  Cytoskeleton proteins Cytoskeleton and cytoskeleton-binding proteins can be regulated by phosphorylation. This process controls the remodeling of cytoskeleton, and thus involves in mitogenesis, cell migration and morphology determination. Palladin, an actin-binding protein, was found to be phosphorylated by ERK1/2 at S77 and S197. This ERK-dependent phosphorylation of palladin is believed to have anti-migratory function (Asano et al., 2011; Roskoski, 2012a). ERK1/2 are also reported to phosphorylate vinexin (Mitsushima et al., 2004) and calnexin (Chevet et al., 1999). In some other cases, cytoskeleton elements, for example tubulin (Reszka et al., 1995) and vimentin (Perlson et al., 2005), recruit ERK1/2 and induce their activity without being phosphorylated. Paxillin is a focal adhesion protein involved in integrin signalling to control cell adhesion and migration (Brown and Turner, 2004). Paxillin works as a multifunctional docking protein, which is well regulated by a set of phosphorylation events. Phosphorylation of paxillin by ERK1/2 was first characterized in phorbol 12-myristate 13-acetate (PMA) treated EL4 thymoma cells (Ku and Meier, 2000). PMA induces the activation of PKC isoforms, and these can phosphorylate Raf1 at S499 in its T-activation loop to stimulate the ERK1/2 pathway (Kolch et al., 1993). ERK1/2 also induce paxillin phosphorylation under hepatocyte growth factor (HGF) stimulation (Liu et al., 2002). In this case, paxillin also promote the local activity of ERK1/2 by  50  assisting the recruitment of Raf-1 and ERK1/2 to MEK1/2 (Ishibe et al., 2004; Ishibe et al., 2003). Upon phosphorylation, paxillin recruits protein-tyrosine kinases including focal adhesion kinase (FAK) and Src, which also target paxillin for tyrosine phosphorylation. These phosphotyrosines provide additional docking sites for adaptors and other signalling proteins, and facilitate downstream signal propagation (Yoon and Seger, 2006).  1.5.5 Regulation of ERK1/2 The Raf-MEK-ERK cascade provides a central theme of the cell signalling network (Figure 1.8). Activation of receptor-tyrosine kinases, G protein-coupled receptors, integrins, or elevation of intracellular calcium level can all stimulate the activity of ERK1/2 pathway (McKay and Morrison, 2007; Zhang and Dong, 2007). With the extremely narrow specificity, Rafs and MEK1/2 converge the upstream stimulus and activate ERK1/2. These ERKs serves as a diverging point by phosphorylating a large variety of protein substrates. The signalling cascade is regulated in each tier so that different physiological conditions would lead to different durations of ERK1/2 activation, as well as various outcomes downstream of ERK1/2. One important factor that influences ERK1/2 signalling is the differential expression of each component in specific cell types. Nonetheless, regulation of ERK1/2 mainly depends on local availability of these ERKs controlled by scaffold and anchor proteins, as well as deactivation by MAP kinase phosphatases (MKPs). Researchers have also discovered that protein levels of ERK1/2 may be regulated by ubiquitination (Lu et al., 2002).  51  ERK1/2 scaffolds  Scaffolds are proteins that specifically bind to multiple signalling molecules through docking interactions. They play pivotal roles in regulate and integrate signal transduction of ERK1/2 (Dhanasekaran et al., 2007; Kolch, 2005; Roskoski, 2012a). Scaffolds assist assembling signalling complex by recruiting components together, promote propagation of signal, mediate crosstalk between pathways, and provide insulation against undesirable crosstalk with other pathways. Protein-protein interactions by scaffolds also provide specificity to signalling. KSR1/2 are among the best studied scaffold proteins of the ERK1/2 cascade. KSR is localized primarily in the cytoplasm, and associates with MEK1/2 constitutively in resting cells (Matheny et al., 2004; Morrison and Davis, 2003). Upon stimulation, KSR recruits activated Raf-1 and ERKs, and thus inducing activation of the ERK1/2 cascade. KSR was thought to be catalytically inactive, because of the absence of a critical arginine from the conserved HRD motif (Subdomain VI). However, recent investigations indicated both KSR1 and KSR2 possess kinase activity, which is essential for the activation of MEK1/2 by Raf (Brennan et al., 2011). MEK partner 1 (MP1) is a 13.5 kDa protein that specifically facilitates the interaction between ERK1 and MEK1, but not ERK2 or MEK2, and this may contribute to differences in ERK1 and ERK2 signalling (Schaeffer et al., 1998). MP1 was observed to operate together with mitogen-activated protein kinase organizer 1 (MORG1), which stabilizes the assembly of each component in the complex. The function of MORG1 is mediated by a WD domain (Vomastek et al., 2004).  52       Figure 1.8. The ERK signalling networks. The ERK1/2 signalling pathway can be activated by different types of receptors under various stimuli, some of which are represented here. Recruitment of adaptor proteins leads to activation of guanine exchange factor SOS, which activates the Ras-Raf-MEK1/2-ERK1/2 cascade. Activated ERK1/2 phosphorylate a large variety of cytosolic and nucleus substrates. The pathway is also negatively regulated by feedback loops and MAPK phosphatases. Summarized from Luttrell (2002); McKay and Morrison (2007); and Zhang and Dong (2007).    53  In the cases of ERK1/2 cascade activation by G protein-coupled receptors (GPCRs), β-arrestin1/2 are involved. GPCR kinases (GRKs) catalyze the phosphorylation of agonist-occupied receptors and create binding sites for β-arrestin1/2, which are recruited to deactivate the receptors and induce endocytosis of the complex (Luttrell and Gesty-Palmer, 2010). In addition, β-arrestin1/2 can also function as scaffolds for Raf-1, MEK1 and ERK1/2 (DeFea et al., 2000). Interestingly, active ERK1/2 is restricted in the cytosol by β-arrestin1/2, causing inhibition of ERK1/2-mediated transcriptional activation in nucleus (Tohgo et al., 2002). Some other scaffold proteins recruit ERK1/2 to certain subcellular locations to activate specific downstream effectors. For example, Sef-1 attracts MEK-ERK complexes to Golgi membranes (Torii et al., 2004), and paxillin induces ERK1/2 activity at the cytoskeleton machinery (Ishibe et al., 2003).  Subcellular localization of ERK1/2 In resting cells, unphosphorylated ERK1/2 can enter the nucleus by an energy-independent mechanism that is facilitated by direct interaction with nucleoporins (Matsubayashi et al., 2001). Although lacking nuclear export signals (NESs), ERKs were found to exit from nucleus by either energy-independent or energy-dependent processes (Adachi et al., 2000). Moreover, some interacting proteins play important role in ERK1/2 cytoplasmic retention by anchoring them in the cytoplasm or blocking its nuclear import process. Upon stimulation, ERK1/2s are phosphorylated by MEKs at the TEY motif in their  54  activation loops. Rapid nuclear accumulation of ERK1/2 was observed following activation. Activated ERK1/2 are imported into nucleus mainly via an active transport process (Ranganathan et al., 2006). This translocation enables ERK1/2 to phosphorylate transcription factors and other nuclear substrates. The duration of ERK1/2 phosphotransferase activity in the nucleus is regulated in a stimulus-dependent manner. In addition to the cytoplasmic scaffolds that suppress nuclear localization of ERK1/2, MEK1 also plays an important role in regulation of ERK1/2 localization. MEK1 and MEK2 each harbour a very strong NES. Thus, they escalate the nuclear export of ERK1/2 that are dephosphorylated by phosphatases (Adachi et al., 2000). Overexpression of ERK1/2 in cells often leads to nuclear accumulation of the kinases, which is presumably from overwhelming cytosolic anchors (Rubinfeld et al., 1999). Stimulus-dependent distribution of active ERK1/2 cascade components to certain subcellular compartments is one of the most important mechanisms to generate specific phenotypic output downstream of ERK1/2 (Raman et al., 2007).  MAP kinase phosphatases Since all three tiers of protein kinases in the MAPK cascades are activated by phosphorylation, protein phosphatases can impact all aspects of these signalling cascades. Indeed, protein-serine/threonine phosphatases (STPs), protein-tyrosine phosphatases (PTPs) and dual specificity phosphatases (DuSPs, or MAP kinase phosphatases/MKPs) were all found to participate in down-regulation of ERK1/2 (Roskoski, 2012a; Yoon and Seger, 2006).  55  The DuSPs/MKPs share similarities in primary sequences and domain structures with the PTPs, both featuring a conserved HCX5R motif and an aspartate as the base for catalysis (Mustelin, 2007). Comparing the PTPs, MKPs have shallower active clefts, which allow the phosphatases to accommodate either phosphotyrosine or phosphothreonine (Farooq and Zhou, 2004). In the MKPs, the MAPK binding domain, or the kinase interaction domain (KID), plays a key role in determining substrate specificity through docking interactions (Patterson et al., 2009). The best studied MAPK phosphatase, MKP3, is highly specific to phosphorylated ERK1/2. The expression of MKP3 is induced after ERK1/2 activation, and the phosphatase only becomes catalytically active after binding to its ERK1/2 substrates (Camps et al., 1998). Marchetti et al., (2005), discovered that MKP3 was phosphorylated by ERK1/2 at two serine sites, which promoted the proteasomal degradation of this phosphatase. Similarly to the DuSPs, tyrosine phosphatase PTP-SL also selectively binds to ERK1/2 through a KIM domain, and negatively regulates ERK1/2 activation by dephosphorylating the tyrosine site at the TEY motif (Pulido et al., 1998). Protein phosphatase 2A (PP2A), an extensively studied STP, is another important regulator of the ERK1/2 cascade. PP2A is shown to dephosphorylate ERK2 at T183 and deactivate the kinase in vitro (Anderson et al., 1990) and in EGF-stimulated rat PC12 cells (Alessi et al., 1995). However, PP2A also positively regulates ERK1/2 signalling by dephosphorylating KSR and Raf-1 at their inhibitory sites (Dougherty et al., 2005). The activity of PP2A is required for Raf-1 activation by Ras, while PP2A restrict the downstream ERK1/2 phosphotransferase activity by  56  dephosphorylating the activation motif. Considering the inhibitory effects of these phosphatases to ERK signalling cascade, they may function as tumour suppressor proteins. Actually, down-regulation of MKP3 is believed to associate with ovarian cancer (Keyse, 2008). However, recent large scale analysis indicated PTPs can act as either tumour suppressor proteins or oncoproteins. Amplification or overexpression of some PTPs was found to be implicated in various types of human cancers (Julien et al., 2011), although this may actually be a protective counter response to regain regulatory control of cell proliferation in cancer cells.  1.5.6 Targeting the ERK signalling pathway Functioning as a pro-growth and pro-survival pathway, the Ras-Raf-MEK1/2-ERK1/2 signalling cascade is activated in a large variety of human cancers. The Ras and Raf isoforms are among the most frequently mutated genes in cancer. Somatic mutations of human K-Ras occur in about 58% of pancreatic cancers, 33% of colorectal cancers and 31% of biliary cancers, whereas N-Ras mutations are detected in around 18% of melanomas (Vakiani and Solit, 2011). B-Raf is the most commonly mutated of the Raf isoforms, which transpires in about 11.5% of all human tumours. Raf-B mutations occur in 40-60% of melanomas, 40% of thyroid cancers, and 30% of ovarian cancers (Davies et al., 2002). Dysregulation of the ERK1/2 cascade signalling activity has also been observed in absence of any oncogenic mutations in some cases (Roskoski, 2012a).   57  Ras inhibitors Significant efforts have been devoted to targeting the Ras-Raf-MEK-ERK cascade for anti-cancer and other therapies. However, development of Ras inhibitors has fallen short of expectations due to the high affinity of the Ras-GTP interaction. Instead, researchers have attempted to suppress PTMs that are required for Ras activation (Ahearn et al., 2012). For example, the protein farnesyltransferase (FTase) and protein geranylgeranyltransferase (GGTase), which are enzymes that carry out the isoprenylation of Ras proteins, have been targeted by small molecule inhibitors in Ras mutant tumour cells (Roskoski, 2003). Great efforts have been put in identification of other prenylated proteins that are affected by FTase and GGTase inhibitors, which will promote the design of better clinical trials (Berndt et al., 2011). However, due to the importance of isoprenylation of a wide range of diverse proteins, FTase and GGTase inhibitors are likely to have severe side-effects, including blindness, which may mitigate their ultimate use.  Raf-B inhibitors Mutationally activated Raf-B has been identified as driver in 40-60% of malignant melanomas, as well as various other human cancers including thyroid, lung, ovary and colon cancers (Vakiani and Solit, 2011). A single point mutation, V600E, accounts for about 80% of all the Raf-B mutations that have been identified. This substitution mimics the phosphorylation of the activation loop, thus resulting in a constitutively active Raf-B kinase (Garnett and Marais, 2004). Additionally, gene fusion events are also reported to activate Raf-B and Raf-1 (Jones et al.,  58  2009; Palanisamy et al., 2010).  The first Raf inhibitor to enter clinical trials is sorafenib, which was screened from a chemical library for inhibitors against recombinant activated Raf-1 (Lyons et al., 2001). Sorafenib is an ATP-competitive kinase inhibitor, which lacks selectivity and meaningful clinical activity in most cancers harboring Raf mutants (Eisen et al., 2006). Second generation ATP-competitive Raf inhibitors were developed specifically to target Raf-B V600E mutant. More than half of all cases of melanomas carry V600E mutation of Raf-B, and are addicted to ERK1/2 activity, making them ideal test models for these inhibitors (Flaherty et al., 2010). The oncoprotein binding selectivity has translated into an unusually high therapeutic index for the Raf-B V600E inhibitors, which allows for high exposures of the drug while avoiding the acute toxicities associated with Raf inhibition (Bollag et al., 2012). However, recent studies indicate that all ATP-competitive inhibitors were not only poor inhibitors for wild-type Rafs, but also induced ERK activation in Raf-B wild type cells, which revealed the complexity of Raf actions (Hatzivassiliou et al., 2010; Poulikakos et al., 2010).  Paradoxical activation of Raf-1 has been observed, in which case inhibitors promote Raf-1 homo- and heterodimerization and association of Raf with Ras-GTP (Gibney et al., 2013; Heidorn et al., 2010). The Raf-B inhibitors also induce the formation of Raf-KSR1-MEK complex as a scaffold (McKay et al., 2011). Moreover, Raf-B inhibition activates ERK signalling though wild-type Rafs by relieving the negative feedback and increasing the level of Ras-GTP (Lito et al., 2012). This discovery partially explains the mechanisms of primary drug resistance  59  to Raf inhibition (Figure 1.9). Resistance can also be acquired through activation of PI3K survival pathway (Lo and Shi, 2014; Shi et al., 2014), Raf-B overexpression, or alternative splicing (Poulikakos et al., 2011).   MEK and ERK inhibitors Since therapeutic benefit of using Raf inhibitor alone only lasts shortly before the development of resistance, combinational therapies with MEK or ERK inhibitors are strongly desired to achieve greatest efficacy (Holderfield et al., 2014). FDA approval was recently granted for the combination of dabrafenib (Raf inhibitor) and trametinib (MEK inhibitor) in advanced melanoma. It was observed that the tolerability of either agent improved in the combination group (Flaherty et al., 2012). However, ERK1/2-independent resistance is still a concern for this strategy. Inhibitors against RTKs and PI3Ks have also been considered to combat the increased signalling through IGF1R/PI3K pathway (Villanueva et al., 2010).            60        Figure 1.9. Primary resistance to Raf-B inhibition. Primary resistance to the Raf-B inhibitors is due to the loss of feedback from ERK (red) and the increased activity of receptor-tyrosine kinases (RTKs), Ras and PI3K pathways (blue). GEFs, guanine nucleotide exchange factors. Adapted from Holderfield et al. (2014).   61  1.6  Rationale and objectives It is conceivable that all of the human typical protein kinases arose from a common ancestor as they are highly homologous and share several invariant amino acids residues (Manning et al., 2002a), but the identity of such a protein has been elusive. The evolution and expansion of this super family has played critical roles in evolution of eukaryotic species, so this common ancestor has made a very significant contribution to the establishment of multicellular organisms on our planet. The successful identification of this ancestral protein could shed new light on the evolution of life. Besides the similarity in the amino acid sequences and crystal structures of all the ePKs, there are also common themes in distribution of phosphorylation sites within the protein kinase catalytic domains. Phosphosites that exist in the most heavily phosphorylated regions of the kinase domain, such as the activation T-loop segment between kinase Subdomains VII and VIII and the ATP-binding motif in Subdomain I, have well recognized general regulatory functions for kinase activation and deactivation. However, within the activation T-loop, there resides many other highly conserved phosphosites of undefined functions. These additional phosphosites may offer additional regulatory controls that have thus far been largely unappreciated.  The more variable parts in the kinase catalytic domain may further contribute towards increased specificity in upstream regulation and substrate selection. Expansion of our knowledge concerning the specificity of individual protein kinases could improve our ability to integrate these regulatory proteins into signalling pathways that have remained obscure.    62  Our ability to detect the phosphoprotein substrates of kinases and phosphatases has greatly benefitted from the availability of generic antibodies that recognize different types of phosphorylation sites, especially the mouse monoclonal antibodies such as 4G10, PY20 and PY100, which target tyrosine-phosphorylated proteins. However, the repertoire of phosphotyrosine sites that are strongly recognized by these monoclonal antibodies is actually significantly limited (Tinti et al., 2012). Better immunological tools are needed to expand the range of phosphotyrosine sites that can be tracked in the phosphoproteomes of cells. Therefore, the primary objectives of my thesis research included: (i) exploration of the possible origin of ePKs; (ii) definition of the optimal substrate specificities of about 200 diverse ePKs using a high throughput peptide microarray technique; (iii) investigation of the regulatory roles of evolutionary conserved phosphorylation sites in the activation loop of ERK1/2 as a paradigm to better understand the regulation of ePKs in general; and (iv) development of a better generic phosphotyrosine antibody as a reagent for tracking the phosphorylation of ePKs such as ERK1/2 and their upstream regulators. 63  Chapter 2: Materials and Methods  2.1  Molecular biology techniques 2.1.1 Plasmids The plasmids pGEX-2T-GST-ERK1, pGEX-2T-GST-ERK1-K71A and pGEX-2T-GST-MEK1-ΔN3EE were constructed by Dr. David Charest from our lab (Charest et al., 1993; Mansour et al., 1994). The plasmid containing full-length wild-type ERK1 (pGEX-2T-GST-ERK1) was used as the template in a coupled PCR reaction (primers F, 5’-CGGGATCCATGGCGGCGGCGGCGGCT-3’, R1, 5’- ATCGTCGTCCTTGTAGTCGGGGGCCTCCAGCAC-3’, and R2, 5’- CCCGAATTCCTACTTGTCATCGTCGTCCTTGTAGTC-3’) to add a C-terminal flag tag (DYKDDDDK), 5’ BamHI site and 3’ EcoRI site to the ERK1 sequence. The underlined sequences represent bases that were added at 3’ end of the original sequence. The PCR product was cloned into pcDNA3.0 vector using the BamHI and EcoRI sites.   2.1.2 Mutagenesis The Phusion Site-Directed Mutagenesis Kit (New England BioLabs) was used to create mutations at phosphorylation sites of interest in ERK1 based on the  64  pGEX-2T-GST-ERK1 template. Back to back primers with 5’ phosphorylation were synthesized by NAPS-IDT (University of British Columbia) as listed below (mutated bases underlined). K72R_F: 5’-AGGATCAGCCCCTTCGAACATCAGACCT-3’ K72R_R: 5’-CTTGATGGCCACGCGAGTCTTGCGCAC-3’ T198A_F: 5’-GTATGTGGCTACGCGCTGGATCCGGGCCCCAGAGAT-3’ T198A_R: 5’-TCCGTCAGGAAGCCGGCGTGGTCATGCTCAGGATC-3’ T207A/E_F: 5’-CTGGTACCGGGCCCCAGAGATCATGCT-3’ T207A_R: 5’-CGCGCAGCCACATACTCCGTCAGGAAGC-3’ T207E_R: 5’-CGCTCAGCCACATACTCCGTCAGGAAGC-3’ Y210F_F: 5’-CTGGTTCCGGGCCCCAGAGATCATGCT-3’ Y210E_F: 5’-CTGGGAACGGGCCCCAGAGATCATGCT-3’ Y210E/F_R: 5’-CGCGTAGCCACATACTCCGTCAGGAAGC-3’ S74E_F: 5’-GAGCCCTTCGAACATCAGACCTACTGCCA-3’ S74E_R: 5’-GATCTTCTTGATGGCCACGCGAGTCTTGC-3’ After exponential amplification of the entire plasmid by PCR, products were circularized by Quick T4 DNA Ligase (New England BioLabs) at room temperature for 5 min and tranformed into E. coli DH5α cells. Plasmids were then purified from selected colonies and sequenced to confirm the mutations. The T207A and T207E mutants were  65  cloned into pcDNA3.0 vector with C-terminal flag tags as described above.  2.1.3 Expression and purification of recombinant proteins Expression of recombinant ERK1 proteins was induced with 0.1 mM IPTG at 30 °C for 4 hours in E. coli BL21 cells. After incubation, cells were collected, washed with cold phosphate buffered saline (PBS) and resuspended in bacteria lysis buffer (PBS [pH 7.4], 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM EDTA, and 1% (v/v) Triton X-100). Sonication was performed for 10-20 rounds, 15 seconds each on ice. Lysates were then incubated with glutathione-agarose beads slurry (lysate/beads (v/v) = 10:1) at 4 °C for 30 min. The beads were washed in PBST (PBS with 1% (v/v) Triton X-100) three times. GST-ERK1 proteins were eluted with 5 mM reduced glutathione in 50 mM Tris-HCl [pH 8.0]. The concentration and purity of the recombinant protein would be examined following SDS-PAGE and Coomassie-blue G-250 staining. Eluted beads were regenerated by washing in cleansing Buffer I (0.1 M boric acid and 0.5 M NaCl [pH 8.5]) and cleansing Buffer II (0.1 M sodium acetate and 0.5 M NaCl [pH 4.5]) before storage in 20% ethanol at 4 °C.   66  2.2  Antibodies and immunological detection reagents 2.2.1 Commercial antibodies The following antibodies were used for Western blotting (WB) or immunoprecipitation (IP): flag tag antibody (2368, Cell Signaling Technology), ERK1/2-CT (06-182, Upstate Biotechnology), ERK1/2 (SPC-120, Stressmarq Biosciences), ERK2 (KAP-MA007, Stressgen), dual phospho-ERK1/2 (KAP-MA021, Stressgen, discontinued), phospho-ERK1/2 (pT202/pT204) (9101, Cell Signaling Technology), phospho-MBP (05-429, Millipore), MEK1 (610122, BD Transduction Laboratories), Actin (sc-1616, Santa Cruz Biotechnology), phospho-threonine-proline antibody (9391, Cell Signaling Technology), 4G10 (05-321, Millipore), PY20 (ab10321, Abcam), pY-100 (9411, Cell Signaling Technology).  Secondary antibodies that were used included donkey anti-rabbit IgG-HRP (sc-2077, Santa Cruz Biotechnology), sheep anti-mouse IgG-HRP (NA931, GE Healthcare Life Sciences), and bovine anti-goat IgG-HRP (sc-2352, Santa Cruz Biotechnology).  2.2.2 Phospho-specific antibody production and validation Antigenic phosphopeptides were synthesized with an additional C-terminal beta-alanine (as spacer) and a cysteine residue (for conjugation). The syntheses were  67  carried out according to Fmoc solid-phase synthesis strategy (Fields and Noble, 1990). The peptide products were confirmed by mass spectrometry and purified to a final purity between 70% and 98%. Conjugation of the peptides with keyhole limpet hemocyanin (KLH, Sigma) was performed according to the manufacturer instructions. Conjugated peptides were injected into New Zealand White rabbits for immunization based on an extended 4-month protocol (Pacific Immunology Corp., Ramona, CA, USA). Immunoglobulins were precipitated from the rabbit sera with 50% saturated ammonium sulphate (SAS) before subjected to affinity purification on agarose columns were coupled through a thio-ester linkage with phosphotyrosine- beta-alanine-cysteine. The captured antibodies were eluted with a reduction in the pH using the column elution buffer (0.1 M Tris-glycine [pH 2.5]) and collected in 0.5 mL fractions into ice-cooled tubes containing 50 µL of neutralization buffer (1 M Tris-HCl [pH 8.8]) so that the final pH was ~7.4 as determined with pH indicator paper (MColorpHast™, Merck). The peak fractions as determined with the Bradford Protein Assay (Bachmann et al., 2004) were pooled and concentrated with Centricon plus-70 centrifugal filters (Millipore). To validate the specificity of phospho-antibodies, purified antigenic peptides and unphosphorylated peptides with the same sequences were spotted on nitrocellulose membrane (1-5 ng/spot) and used to test purified antibodies using the Western blotting protocol. Antibodies that showed strong immunoreactivity and phosphosite-specificity on  68  peptide dot blots were further evaluated on Western blots prepared with a lysates from panel of cell lines subjected to various treatment conditions.  2.2.3 Immunological detection reagents Amersham ECL Plus Western blotting detection reagent (GE Healthcare Life Sciences, discontinued) was used for WB detection. Pro-Q® Diamond phosphoprotein gel stain (Life Technologies) was also used to detect phosphoproteins or phosphopeptides. The destain buffer consisted of 20% acetonitrile in 50 mM sodium acetate [pH 4.0].  2.3 Kinase assays Purified myelin basic protein (MBP, Sigma) was used to test the phosphotransferase activity of wild-type or mutant ERKs in vitro. For recombinant ERK1, 1 μg purified GST-ERK1 was pre-incubated with 0.1 μg of constitutively active MEK1-ΔN3EE (Mansour et al., 1994) in 20 μL kinase assay buffer (20 mM MOPS [pH 7.2], 2 mM EDTA, 5 mM EGTA, 25 mM β-glycerophosphate, 20 mM MgCl2, 1 mM Na3VO4, and 0.25 mM dithiothreitol (DTT) supplemented with 50 μM ATP. After the pre-incubation at 30 °C for 5-15 min, 5 μg of MBP were added and incubated with the reaction mix for another 2 min. At the end of the incubation, 7 μL of 4x SDS sample buffer were mixed  69  with the solution to terminate the reactions. The samples were then boiled and subjected to SDS-PAGE. To test the kinase activity of immunoprecipitated ERK1, anti-ERK immunoprecipitates bound to protein A-beads were further washed twice in kinase assay buffer before incubation with MBP substrate. In this case, 10 μg of MBP were used for a longer incubation time of 15 min. Reactions were stopped by mixing with 4x SDS sample buffer for Western blotting analysis. In preliminary experiments, radioactive kinase assays were also performed. Basically, GST-ERK1 was pre-activated by MEK1 using non-radioactive ATP, followed by incubation with MBP and [γ-33P] ATP. The reactions were terminated by spotting the solution onto P81 phosphocellulose paper (Whatman). After three washes in 1% H3PO4, the captured radioactivity on the P81 phosphocellulose papers, which corresponded to the phosphorylated protein product, was quantitated by liquid scintillation counting.  2.4  Kinase substrate profiling technique 2.4.1 Kinase substrate peptide microarray The Kinex™ Kinase Substrate Peptide Microarray (Kinexus) was used to test the specificity of purified active human protein kinases in vitro. The microarray slides were  70  printed with 445 kinase substrate peptides in triplicates. Purified kinases were diluted in kinase assay buffer (as described above) supplemented with 100 μM ATP to 200 μL. The reaction mix was then loaded to the pre-blocked microarray slide for 2 hours incubation at 30 °C with gentle agitation. The slide was washed in 0.5% SDS solution and TBST buffer before it was stained with Pro-Q® Diamond for detection of phosphorylated peptides. The destained slides were dried and scanned at 546 nm using a ScanArray Gx microarray scanner (PerkinElmer, Inc.). A detailed protocol for this technique is available in Appendix 3.  2.4.2 SPOT membrane Peptides were also synthesized on cellulose membrane as SPOT membranes. These membranes were hydrated in methanol and equilibrated in kinase assay buffer supplemented with 1 mg/mL BSA. After the phosphorylation reactions, membranes were washed and blocked in TBST with 5% sucrose and 4% skim milk. Phosphorylated peptides were detected using proper generic phospho-antibodies and secondary antibodies conjugated to horseradish peroxidase (HRP). Visualization of the signal was carried out by colorimetric detection technique with 4-Chloro-1-naphthol (4CN, Acros Organics) (Hawkes et al., 1982).    71  2.5 Cell lines and cell culture A431 (human epidermoid carcinoma) cells and human embryonic kidney/HEK 293 cells were cultured in Dulbecco’s modified Eagle medium (DMEM, Gibco) and minimum essential medium (MEM, Gibco), respectively, supplemented with 10% fetal bovine serum (FBS). All cell lines were cultured based on conditions recommended by ATCC (http://www.atcc.org/). HEK-293 (human embryonic kidney) cells were used as transfection hosts of flag-ERK1 and its mutants. Culture dishes were coated with 2% poly-L-lysine solution (Sigma) for 20 min prior to each use. At 50-60% confluency, cells were transfected with pcDNA3.0-ERK1 plasmids using Lipofectamine 2000 transfection reagent (Invitrogen, Life Technologies). Transfection medium was replaced 6 hours after transfection. Cells were cultured for another 48 hours before selected with 500 μg/mL of G418 for positively transfected clones. Clones stably expressing flag-ERK1 were tested and expanded for further experiments.  2.6  Cell stimulation or treatment 2.6.1 EGF or serum stimulation Cells at about 80% confluency were starved in serum free medium for 16-18 hours  72  before stimulation. EGF was added to a final concentration of 10 ng/mL for 5 min incubation at 37 °C. For serum stimulation, cells were incubated with medium in presence of 10% FBS for 10 min at 37 °C.  2.6.2 Phosphatase inhibitor and protease inhibitor treatment Cells were cultured to around 80% confluency. Culture medium that was warmed up to 37 °C was used to dilute phosphatase inhibitors to proper final concentrations (25 μM for phenylarsine oxide (PAO), 50 μM for sodium orthovanadate (Na3VO4), and 30 mM for sodium fluoride (NaF)). The solution was then added to the cells with 30 min incubation at 37 °C. Control groups were incubated with DMSO at comparable concentration. Treatment with the protease inhibitor MG132 (Calbiochem) was performed similarly with the final concentration of 10 μM, and incubation of 4 hours at 37 °C.  2.7 Cell lysis, immunoprecipitation and immunoblot analysis Cells were washed with ice cold PBS before harvested in lysis buffer (20 mM MOPS [pH 7.2], 5 mM EDTA, 2 mM EGTA, 0.5% (v/v) Triton X-100, 30 mM NaF, 20 mM Na4P2O7, 1 mM Na3VO4, 40 mM β-glycerophosphate, 1 mM DTT, 1mM PMSF, 3  73  mM benzamidine, 5 μM pepstatin A, and 10 μM leupeptin). The NaF, Na4P2O7, Na3VO4 and β-glycerophosphate are phosphatase inhibitors that prevent protein dephosphorylation during lysate processing. The protease inhibitors (PMSF, benzamidine, pepstatin and leupeptin) can be substituted with Roche COMPLETE protease inhibitor cocktail tablets. Concentration of cell lysates was determined using the Bradford protein assay (Harlow and Lane, 2006).  Procedures for immunoprecipitation were described previously (Zhang et al., 2001). The immunoprecipitated flag-ERK1 was used for kinase assays as described above, or directly mixed with SDS sample buffer for Western blot analysis.   2.8 Databases and online resources Protein kinase sequences of human and other species that were used for evolutionary conservation studies are based on UniProt database (http://www.uniprot.org/) and NCBI RefSeq (http://www.ncbi.nlm.nih.gov/refseq/). Protein structures were from RCSB protein data bank (PDB) (http://www.rcsb.org/pdb/) and visualized using PDB Protein Workshop.  Information on protein phosphorylation sites was based on data in the UniProt (http://www.uniprot.org/), PhosphoSitePlus (http://www.phosphosite.org/), and PhosphoNET (http://www.phosphonet.ca/) databases.   74  2.9 Data analysis Quantification of immunoblots was performed using Quantity One software (Bio-Rad Laboratories). The Student’s t-test was used for statistical analysis based on at least three independent experiments.  Microarray analysis software ImaGene Version 9.0 was used to analyze microarray images. The net signal median was calculated to represent signal strength of each spot. Standard deviation of triplicates was used to omit inconsistent signals. Short lists were generated using a certain threshold percentage comparing to the strongest signal. A specificity matrix and a consensus substrate sequence were calculated from the sequences of the top hits. The online tool Two Sample Logo (http://www.twosamplelogo.org/) was used to calculate and visualize kinase specificity or antibody selectivity.     75  Chapter 3: Ancestry of Eukaryotic Protein Kinases  3.1 Rationale Protein kinases play pivotal roles in communicating intracellular signals in eukaryotes. The human genome encodes at least 568 protein kinases, which account for up to 2.7 % of the protein-encoding genes in the entire human genome (Manning, Whyte et al., 2002, Anthony Hunter, personal communication). Of these, 479 are eukaryotic typical protein kinases (ePKs) that are highly conserved in both their primary amino acid sequences (Hanks et al., 1988) and the 3D structures (Taylor and Radzio-Andzelm, 1994) of their catalytic domains. Because of their central regulatory roles and their high conservation in widely diverse species, the ancestry of ePKs has become an important question in studying the evolution of eukaryotic organisms. The majority of the kinases among the ePKs are responsible for the phosphorylation of proteins on serine or threonine residues, while a smaller group of protein kinases catalyze tyrosine phosphorylation. This branch of protein-tyrosine kinases (PTKs) arose from protein-serine/threonine kinases (STKs), which is believed to be an important development in early metazoan evolution (Darnell, 1997; Rokas et al., 2005). Of all the STKs, there is another lumped group of diverse kinases that are called atypical protein  76  kinases. With little sequence identity and structural similarity to typical protein kinases, theses atypical protein kinases have been suggested to have diverged early in evolution and have distinct evolutionary histories (Leonard et al., 1998; Middelbeek et al., 2010). Apart from the atypical protein kinases and recently derived PTKs, the rest of the typical protein kinases constitute a major lineage in protein kinase evolution.  Eukaryotic life on Earth is believed to have evolved over 1.7 to 2.7 billion years ago and no living representatives of the earliest eukaryotes survive today. Consequently, the actual origin of protein kinases is difficult to establish with a high degree of confidence. Firstly, protein sequences are highly degenerate, which makes the detection of sequence similarities difficult even at the superfamily level (Murzin et al., 1995). Secondly, the ePKs comprise a group of very broadly expanded proteins. Loss and expansion of kinase-relatedness tree branches occurs in various species, as well as insertions and deletions inside their catalytic domains. To investigate these problems, I developed novel strategies using consensus sequences from precise amino acid sequence alignments as the initial query in BLAST searches and compared top hits from multiple species. My conclusions are ultimately supported by protein primary and tertiary structure comparisons. These findings offer considerable insight into the evolution of ePKs and choline kinases (ChKs) in ancient eukaryotes. The molecular paleontology approach undertaken in this study also provides  77  a novel and broadly applicable method to generally investigate the origins of large protein domain families.  3.2 Alignment of catalytic domains of typical human protein kinases The amino acid sequences of all the 492 typical catalytic domains in 479 human protein kinases were collected and precisely aligned (Figure 3.1A). The alignment contained 12 catalytic subdomains made up of about 30 highly conserved amino acids, and 10 gaps, which represented more variable regions responsible for the specificity of individual kinases. The initial alignment was facilitated by the early work of Hanks et al., (1988), which was further refined with more secondary structure information that has arisen from x-ray crystallographic structures of more than 50 protein kinases. The STKs and the PTKs were separated into two groups. Despite the preponderance of conserved residues in both STKs and PTKs, major differences between the conserved features of two groups often occur near Subdomains VI (HRD) and VIII (APE), which determine the phosphotransferase activity of a kinase towards serine/threonine or tyrosine (Figure 3.1B). To explore the origin of ePKs, the alignment of the 393 human STK catalytic domains were used to generate a consensus sequence (Appendix 1, Figure 3.1B, STK). Although all of these STK sequences originated from one species, the human kinome  78  featured one of the largest groups of diverse STKs at the time this work was undertaken. With such a diversity of STK catalytic domain primary structures, the rationale was that those amino acid residues that were the most conserved might also be shared with the ancestor of STKs. The frequency of each of the 20 common amino acids at each position was calculated. The average frequency of the most common amino acid at each position was 36%, and two thirds of them were higher than 20%, indicating very high conservation among the catalytic domains of these protein kinases. An STK consensus sequence of 247 amino acids in length was defined using the amino acid with highest frequency at each position.  To evaluate whether the derived human STK consensus sequence was not particularly bias for humans and other mammals, a protein kinase domain alignment with 56,691 sequences from Pfam database (http://pfam.sanger.ac.uk/) was also downloaded (Finn et al., 2010; Sonnhammer et al., 1997). This alignment included the catalytic domains of both protein-serine/threonine and protein-tyrosine kinases, which were not easily resolvable in view of the diversity of species represented and the similarity shared between these subgroups of ePKs. Nevertheless, the consensus sequence of protein kinase domains from all species was shown to be highly similar to the human STK consensus sequence with 85% overall homology. Since the development of the protein-tyrosine kinase group from protein-serine/threonine kinases is proposed to be a  79  relatively late event during evolution with the emergence of metazoans (Darnell, 1997; Rokas et al., 2005), the human STK consensus sequence was chosen as the representation of the earliest protein kinases. It should be appreciated that while the human STK consensus may be representative of lowest common denominator of all of the modern STKs, it cannot be ruled out that through evolution the catalytic efficiency of protein kinases may have continued to improve from cumulative modifications to the primary structures of the most ancient typical protein kinases.    3.3 Proteins most closely related to protein kinases To identify the proteins that were most closely related to protein kinases, our STK consensus sequence was employed as the query in BLAST searches performed in six diverse species including Escherichia coli, Saccharomyces cerevisiae, Arabidopsis thaliana, Caenorhabditis elegans, Drosophila melanogaster and Homo sapiens. For each organism, the top 100 non-protein kinase subjects with a BLAST score higher than 18.0 and an alignment length longer than 35 amino acids were considered as top hits and compared among species. The cutoff score of 18.0 represented the average BLAST score of the 100th subject in each of the six organisms. In search of the most likely ancestor of protein kinases, the average BLAST scores were calculated for proteins present among the top hits in more than one species to define candidates with consistent similarity to our  80  consensus sequence. As listed in Table 3.1, 11 proteins were identified with the range of average score from 19.9 to 22.6. In humans, the BLAST scores of the STK consensus sequence in all typical STKs range from 28.1 to 206. Given that the alignment of these distant genes might be missed by automatic BLAST search due to insertions and deletions, all these candidates listed in Table 3.1 were manually checked for their alignment with STK consensus sequence. Among the 11 proteins, only choline/ethanolamine kinase (ChK) and glutaminyl-tRNA synthetase (GlnRS) showed particular similarities at the kinase catalytic subdomains, which are highly conserved and critical for phosphotransferase activity. ChKs have already been reported to share great similarity in 3D structure with protein kinases (Peisach et al., 2003). As a group of proteins, tRNA synthetase appeared among the top hits in all the six species employed for BLAST search. Glutaminyl-tRNA synthetase, as well as glutamyl- and alanyl-tRNA synthetase, belongs to class-I aminoacyl-tRNA synthetase family, which was the group most similar with the STK consensus sequences. Like ePKs and ChKs, aminoacyl-tRNA synthetases utilize ATP to accomplish their functions. These results pointed to a possible evolutionary relationship between tRNA synthetases, choline/ethanolamine kinases and protein kinases.   81    A. Alignment of human protein kinase catalytic domains (selected)        82           83       84    B. Alignment of human STKs and PTK consensus sequences    Figure 3.1. Alignment of human protein kinase catalytic domains. (A) Alignment of selected human typical protein kinase catalytic domains. Colour-coating is based on the properties of amino acids as described in Figure 1.2. (B) Alignment of human STKs and PTK consensus sequences. To generate the consensus, the frequency of the 20 amino acids at each alignment position was calculated. The STK and PTK consensus sequences were defined using the amino acid with highest frequency at each position. “X”s stand for positions where the highest frequency is below 15%. Major differences in the conserved features of STKs and PYKs near Subdomains VI and VIII are highlighted in yellow.    85    Table 3.1. Comparison of average BLAST scores of top subject proteins.   Arabidopsis thaliana Escherichia coli Saccharomyces cerevisiae Caenorhabditis elegans Drosophila melanogaster Homo sapiens Average Score Choline/ethanolamine kinase    18.7  26.4 22.6  Phenylalanine-4-hydroxylase    21.9 21.5  21.7  Dynein heavy chain    21.2  20.5 20.9  Glutaminyl-tRNA synthetase  21.5 19.8    20.7  Actin-related protein 20.5  20.8    20.7  Peptidyl-tRNA hydrolase  21.5  19.4   20.5  Alanyl-tRNA synthetase 22.2 18.4     20.3  Cullin protein    20.8 19.4 20.8 20.3  3',5'-cyclic phosphodiesterase    19.8 19.1 21.5 20.1  DNA topoisomerase  18.7   20.5  19.6  Nucleolar GTP-binding protein       19.1 19.1   19.1    A consensus sequence of the human protein-serine/threonine kinase catalytic domain was used to search for most closely related non-kinase proteins in six diverse species. The STK consensus sequence used was shown in Figure 3.1B and Appendix 1. Top subjects from BLAST results of each organism (score≥18.0) were compared, and those present among the top hits in more than one species were listed above. Cells were left empty when no significant similarities were detected and no BLAST score could be calculated by the BLAST algorithm.    86  3.4 Ancestry of ePK, ChK and GlnRS Among the three tRNA synthetases from class-I aminoacyl-tRNA synthetase family, glutaminyl-tRNA synthetase (GlnRS) is believed to be evolved from glutamyl-tRNA synthetase (GluRS) (Lamour et al., 1994). Most bacteria employ an alternative two-step pathway to synthesize glutaminyl-tRNA without GlnRS. Phylogenetic analyses indicate that GlnRS arose from the duplication of ancient GluRS after the split of bacteria and archaea/eukarya branches and acquired an N-terminal non-specific RNA binding domain later during evolution (Saha et al., 2009). GlnRS in a few bacterial species is the result of horizontal transfer from eukaryotes before the domain acquisition event (Siatecka et al., 1998). Glutaminyl-tRNA synthetase (GlnRS) was the only candidate that was found with particular conservation at the protein kinase catalytic subdomains from the BLAST results. A consensus sequence for this protein was created using sequences from various species and compared with the STK consensus sequence. From the alignment (Table 3.2), the catalytic domain of GlnRS showed strong similarities with the kinase subdomains near the activation loop, including the LxxLH and DFG motifs. The GlnRS N-terminal domain aligned with the ATP-binding subunits of ePKs. The same strategy was also applied to generate glutamyl- and alanyl-tRNA synthetase (GluRS and AlaRS) consensus sequences. Both of the two genes lack the N-terminal fragment that aligns with kinase Subdomain I to  87  V. At the same time, their catalytic domains also share much lower similarities with kinase Subdomains VI to IX when compared to GlnRS. All these results reveal a closer relationship of ePK to GlnRS than to the other aminoacyl-tRNA synthetases.  To investigate the evolutionary relationships between GlnRS, ePK and ChK, the consensus sequence of ChK was also aligned (Table 3.2). It was noticed that the GlnRS and STK consensus sequences shared the highest identity of 24%, and 18 of the 30 conserved amino acids were identical. The GlnRS consensus sequence showed 20% identity with ChK. The similarities of the two pairs were both around 34%. The identity between STK and ChK was 16%, comparable to the identity of two randomly chosen human protein kinases. These numbers strongly support the possibility that ePK, ChK and GlnRS evolved from a common ancestor, which probably functioned as an aminoacyl-tRNA synthetase. To further characterize the evolutionary bonds between GlnRS, ePK and ChK, 3D structure comparison tools from RCSB Protein Data Bank (http://www.pdb.org/) were employed to align the structures of these three groups of proteins (Ye and Godzik, 2003). Available human structures of GlnRS and ChK were used for comparison. For STKs, the top 30 human candidates that shared the most similar amino acid sequences with the STK consensus sequence were chosen for 3D structure comparisons. The STK structures of highest similarity scores with ChK and GlnRS are shown in Figures 3.2 and 3.3. The human choline kinase displayed high structural similarity with human PKA (p = 0.024,  88  calculated by the algorithm). The human GlnRS had slightly lower scores with both ePK and ChK (p = 0.1~0.2, calculated by the algorithm). For the human GlnRS, only part of the 3D structure was available. The lower scores may have partly reflected the absence of part of the N-terminal RNA binding domain in the human GlnRS 3D structure, which is the portion that has high primary amino acid sequence identity with ePK and ChK. The Phyre server (Kelley and Sternberg, 2009) was also used to predict the secondary structure of the part of GlnRS that aligns with ePK catalytic domain (Figure 3.4). Although some of the important beta strands are missing in the predicted GlnRS secondary structure, the overall pattern is similar, especially near Subdomains VI to VIII, which is also the most conserved region in GlnRS. For the regions with more dissimilar secondary structures, most of the key amino acid residues critical for maintaining the kinase catalytic core are conserved in GlnRS sequences, including E91 and H158 in human cAMP-dependent protein kinase (Zheng et al., 1993; Johnson et al., 2001), which is consistent with the possibility of obtaining the ePK catalytic structure through a series of point mutation events starting with a duplicated GlnRS gene. In summary, these results indicated that ePKs and ChKs may have both emerged from an ancient aminoacyl-tRNA synthetase, which was also the ancestor of GlnRS.   89   Table 3.2. Alignment of consensus sequences of protein-serine/threonine kinases, choline kinases and glutaminyl-tRNA synthetases. Sequences of glutaminyl-tRNA sythetases and choline/ethanolamine kinases from diverse species were aligned to generate consensus sequences. Twelve gaps (G1 – G12) were created at the same positions where the sequences were more variable in protein kinases alignment and labeled as dashes. “X”s stand for positions that were not conserved in the original consensus sequences. Conserved protein kinase Subdomains (I - XII) are marked, and the “LxxLH” motif is highlighted between the G5 and G6 gaps. 90            Figure 3.2. Structural comparisons of ePK, ChK and GlnRS. The identity and similarity of each pair (orange text) were calculated based on primary amino acid consensus sequences. The PDB ID of each 3D structure used for comparison is shown in blue circles and linked with scores generated from RCSB PDB Protein Comparison Tool.     91     Figure 3.3. Comparison of 3D structures of ePK, ChK and GlnRS. (A) glutaminyl-tRNA synthetase (PDB ID: 4R3Z Chain C, orange) with choline kinase (PDB ID: 2IG7, blue); (B) glutaminyl-tRNA synthetase (PDB ID: 4R3Z Chain C, orange) with microtubule affinity-regulating kinase (PDB ID: 2R0I, blue); (C) cAMP-dependent protein kinase PKA (PDB ID: 3DND, orange) with choline/ ethanolamine kinase (PDB ID: 2IG7, blue). All structures were generated from human proteins. The comparisons were done by RCSB PDB Protein Comparison Tool using jFatCat – Java Web Start (rigid) method. Two different angles of each alignment are shown. BA 92      Figure 3.4 Predicted secondary structures of STK and GlnRS consensus sequences. The consensus sequences of STK and GlnRS were used for secondary structure prediction, respectively, by the Phyre server. The gaps from primary amino acid sequence alignment were substituted with “X”s. The predicted structures were aligned with the sequences for comparison. Locations of the conserved kinase subdomains (I-XII) are indicated.  93  3.5 The most ancient STKs and PTKs The most ancient protein kinases are more likely to be closely related to other PKs in primary sequence, and also conserved across species. However, scores from BLAST search using STK consensus sequence were very close mainly due to the high similarity among all the human STKs, which rendered the result lacking resolution. Therefore, catalytic domains of all the human protein kinases were also aligned with the GlnRS consensus sequence by BLAST. The top 20 hits aligned with GlnRS consensus sequence have the scores ranging from 19.4 to 23.3. Seven of them, including AMPKs, ribosomal protein S6 kinases (p70 S6 kinase and RSK isoforms), calmodulin-dependent protein 2-delta (KAMK2D) and microtubule-associated serine/threonine kinase 2 (MAST2), also appear among the top 100 hits in the BLAST search against STK consensus sequence (Table 3.3). Among the seven candidates, AMPKs (AMPK1 and AMPK2), the metabolic stress-sensing protein kinases switching off biosynthetic pathways when AMP level rises due to fuel limitation or hypoxia (Lamour et al., 1994), had the highest conservation scores. Additionally, the AMPKs are also consistently found in kinomes from yeast to human (Manning et al., 2002a), indicating that these kinases are the most closely related to the ancient protein kinases. Similarly, a human PTK catalytic domain consensus sequence was generated from the alignments to possibly identify the nature of the earliest protein-tyrosine kinases.  94     Table 3.3 List of most probable ancient protein kinases.   Consensus sequences of human protein-serine/threonine kinases and glutaminyl-tRNA synthetases were used as queries in BLAST searches separately to identify the most closely related human protein kinases. Human protein kinases presented in both of the BLAST results were listed and sorted by their conservation scores, which are generated by summing up the identity of each human protein kinase with its homologs in ten diverse species (data from HomoloGene).      95  From BLAST search with this consensus sequence, the modern EPH and Src families were identified as potentially most resembling the most ancient PTKs. In fact, these two families appear most close to the merging point of receptors and non-receptors in the evolutionary tree of human kinome (Manning et al., 2002b). Moreover, they are also the most broadly expanded PTK families, with 221 EPH receptors and 172 Src family members identified from 37 metazoans. Thus, it may be concluded that the EPH and Src kinases were the most ancient receptor and non-receptor PTKs, respectively.  3.6 Discussion A few eukaryotic protein kinase-like genes have been identified in archaebacteria (Smith and King, 1995) and prokaryotes (Perez et al., 2008; Tyagi et al., 2010). The widespread distribution of protein kinase genes has led to suggestions that the ancestry of these catalytic domains predated the divergence of archaebacteria, prokaryotes and eukaryotes. (Leonard et al., 1998). However, these eukaryotic-like protein kinases miss some of the essential motifs of ePKs, and are often inactive. Signal transduction in prokaryotes is mainly conducted through the two-component system by histidine kinases instead of by protein-serine/threonine or protein-tyrosine kinases (Hoch, 2000). Other studies have indicated some of the eukaryotic-like protein kinases had distinct evolutionary histories, which might be even more ancient than ePKs (Kannan et al., 2007;  96  Scheeff and Bourne, 2005). With the recent capability to sequence whole genomes, it is believed genes actively undergo horizontal transfers across species, which contribute significantly to the flows of genes in evolution (Keeling and Palmer, 2008; Syvanen, 2012). Horizontal gene transfers most likely account for many of the eukaryotic-like protein kinases that were identified in bacteria. These proteins, such as the PknB kinases (Kennelly, 2002; Ortiz-Lombardia et al., 2003) and the aminoglycoside phosphotransferase APH(3’)-IIIa (Daigle et al., 1999; Walsh, 2000), are usually limited to just a few branches of the entire bacterial kingdom. Thus, eukaryotic-like protein kinases are still likely to have eukaryotic origin. The human protein kinase complement is a well studied group of regulatory enzymes that is expanded broadly in relatedness trees in all investigated eukaryotes. As a result, all of the human STK catalytic domains were selected and precisely aligned them to generate a consensus sequence. This STK consensus sequence represented a “lowest common denominator” of all modern ePKs, and was used to search for the protein groups that were most closely related to ePKs. The conserved kinase subdomains represented the essential core of ePKs that was probably required in a functional ancient protein kinase. The strategy of comparing BLAST results from various well studied organisms and aligning the conserved key residues made it possible to detect long distant relationships. The supportive results from my primary sequence analyses and structural comparison  97  have permitted the first recognition of the evolutionary linkages between glutaminyl-tRNA synthetase and the protein and choline/ethanolamine kinases.The results indicated that ePKs and ChKs share a common ancestor, which is consistent with previous 3D structure studies on these proteins (Peisach et al., 2003). GlnRS exhibited higher sequence identities with ePKs and ChKs than these did with each other, as well as moderate structural similarities. GlnRS appears to be the contemporary gene most closely related to the ancestor of both ePK and ChK. Although GlnRS appears exclusively in eukarya and archaea, the aminoacyl-tRNA synthetases comprise a most ancient group of genes that are believed to have undergone horizontal transfers early in evolution and given birth to many of the contemporary genes (Brown et al., 2001; Theobald, 2010). The ePKs and ChKs also have an early eukaryotic origin and both play an important part in early evolution of highly complex eukaryotic cells. It is proposed that ePKs and ChKs arose from a common ancestor that is an ancient gene involved in the mRNA translation process as an aminoacyl-tRNA ligase. The emergence of ChKs offered additional phospholipid constituents for construction of more complex membrane structures that provide for intracellular compartmentalization as well as sources of intracellular mediators (Exton, 1994; Kent, 2005). The larger size and increased compartmentalization in eukaryotic as compared to prokaryotic cells required the development of new regulatory mechanisms to propagate and amplify intracellular  98  signals and better coordinate diverse cellular processes. The ePKs may have made communication between different compartments within cells more specific and efficient, which could have contributed to the production and specialization of various organelles (Kannan and Neuwald, 2005). The emergence of protein kinases and choline/ethanolamine kinases may have been critical events for the development and success of eukaryotic organisms.  99  Chapter 4: Investigation of the Substrate Specificities of 200 Human Protein Kinases with Peptide Microarrays  4.1 Rationale Protein kinases achieve their unique physiological roles in signal transduction by selectively recognizing and phosphorylating their substrates. Elucidation of the key determinants for the phosphorylation site specificities of protein kinases facilitates identification of their physiological substrates, and serves to better define their critical roles in the signalling systems that regulate a multitude of cellular activities. Many substrates for protein kinases are other protein kinases, so definition of their consensus phosphorylation site recognition sequences could help facilitate the linkage of individual protein kinases within networks. In the catalytic domain of protein kinases, specificity-determining residues (SDRs) are involved in direct interactions with side chains of amino acids surrounding the target phosphosites (Miller et al., 2008). Recently, algorithms have been developed to predict a protein kinase’s specificity based on the primary amino acid sequence of its catalytic domain (Linding et al., 2007; Saunders et al., 2008), including our Protein Kinase Substrate Specificity Predictor Version 1.0 (KSPv1.0), which was trained with over  100  14,000 known kinase-protein substrate pairs (i.e., a protein kinase linked to the phosphorylation of a specific amino acid residue in a substrate protein) that were identified from the scientific literature (Safaei et al., 2011). However, more empirical data was still needed to match phosphosites to specific kinases with higher accuracy.  Consensus substrate sequences for individual kinases have been determined through alignment of known phosphosites, systematic mutation of peptide sequences, and screening of peptide libraries in solution and arrays. For most protein kinases, the alignment of the sequences of phosphosites in known substrates has been the primary route for definition of consensus sequences. However, many of the kinase substrates in these alignments have very diverse affinities for their upstream kinases, and often a significant portion are suboptimal in their primary structures surrounding the phosphosites. Albeit with some apparent limitations, such as for example the lack of contextual information for secondary substrate binding sites, the synthetic peptide-based approach has been adopted widely for kinase specificity profiling studies. When used in an array format, this permits the screening of large numbers of potential peptide substrates in parallel and the identification of the most efficiently phosphorylated substrates  In this chapter, methodology for determining protein kinase substrate specificity with peptide microarrays is presented. Over 200 purified human protein kinases were  101  tested against 445 semi-optimized substrate peptide sequences on substrate peptide microarrays. These substrate peptides were predicted using the KSPv1.0 algorithm.  4.2 Kinase substrate peptide microarray technique Dr. Steven Pelech and Kinexus Bioinformatics have developed the KSPv1.0 algorithm based on the alignment of all typical human ePK domains with training from over 14,000 known kinase-protein substrate pairs (Safaei et al., 2011). With given sequences of any typical protein kinase domain, the application of this algorithm allowed the generation of predicted kinase specificity matrices that featured the expected probability frequencies of each of the 20 amino acids, 7 residues before and after the substrate phosphosites. Based on the crystal structures of several protein kinases with substrate peptides (Knighton et al., 1991; Zheng et al., 1993) and in vitro specificity studies using peptide substrates (Appendix 2), amino acids beyond the -6 and +6 positions (phosphoacceptor amino acid residue = 0 position) normally do not appear to effect kinase-substrate recognition at the primary binding site. While the accuracy of this algorithm has been tested in silico by comparing the computer-predicted consensus sequences with deduced consensus sequences by alignment of known protein substrates, it had yet to be validated empirically by in vitro assays. Using the KSPv1.0 algorithm, 445 semi-optimum human protein kinase substrate  102  sequences were generated. Glycine was used in situations where KSPv1.0 predicted that the amino acid residue was less critical. This was to avoid the introduction of amino acids that might increase the recognition of the peptides by off-target protein kinases. These sequences were individually produced as 13 amino acid-long peptides by SPOT synthesis (Reineke et al., 2001) with both an extra glycine at the C-terminus and a beta-alanine at the N-terminus, which was also biotinylated. Each peptide was printed in triplicate on streptavidin-coated microarray slides and orientated based on the biotin interaction with streptavidin. Each array had four identical fields and permitted the testing of three different protein kinases and one control incubation that monitored the background binding of the fluorescent Pro-Q Diamond phosphorylation sensor dye to unphosphorylated peptides (Steinberg et al., 2003). Some kinases with well-documented specificity data including PKAα, ERK2 and CDK1 were used to optimize the experimental conditions. All of the active preparations of protein kinases used in this study were primarily obtained from commercial sources. The presence of the GST tag was determined by the suppliers not to affect the phosphotransferase activities of these kinase preparations. After blocking in 1% BSA/HEPES buffer, the slides were incubated with reaction mix containing 0.1~0.2 μg purified active kinase and 100 μM ATP at 30 °C for 2 hours. The arrays in 0.5% SDS solution were washed to remove the kinases, and then in TBST buffer to remove the SDS. The slides were blocked again before staining  103  for phosphorylated peptides with Pro-Q Diamond stain for 1 hour. Destained slides were dried with N2 flow and scanned at 546 nm in a microarray scanner to visualize the phosphorylation of peptides (Figure 4.1). The images were analyzed using the advanced microarray image analysis software ImaGene 8.0.   4.3 Determination of substrate specificity of human protein kinases Following the protocols described above, 214 active recombinant human protein kinases for their phosphotransferase activity against the 445 peptides on the microarray slides. These kinases were from 75 different families, which provided wide coverage of all of the typical protein kinase groups in the human kinome (Table 4.1). Overall, about 80% of the kinases showed appreciable activity to phosphorylate peptides on the microarray slides when compared to control fields that were incubated with only kinase buffer and ATP.  Based on the raw data generated by the analysis software for each kinase tested, signal median values of each spot were quantified, and the average and standard deviation of each peptide triplicate were calculated. Peptides exhibiting average signal intensities above 20% of the highest signal that was detected with the same kinase were aligned and used to derive a consensus sequence (peptide consensus). This peptide consensus sequence was then compared with the protein consensus that was created by  104  aligning all the known substrates, and the predicted sequences based on our kinase substrate predictor algorithm. For example, the substrate specificity of cyclic AMP-dependent protein kinase (PKA) in vitro was studied in the 1970s (Kemp et al., 1976; Kemp et al., 1977), and the resultant peptide, Kemptide (LRRASLG), has been widely used to assay the activity of PKA. The specificity of the catalytic domain of PKAα was tested with the substrate peptide microarray. After applying the cut-off criteria, 20 peptides were selected as top hits and used to generate the optimal peptide consensus sequence. As shown in Table 4.2, the PKA peptide amino acid consensus was very similar to both the experimentally determined protein substrate and the in silico predicted substrate amino acid consensus sequences, especially at positions close to the phosphosite. Except for the phenylalanine at the +1 position, other positions at the C-terminus are less important to the substrate recognition for this kinase. The preference for arginine residues at the -2 and -3 positions is consistent with the sequence of Kemptide, indicating that these basic residues at the N-terminus are critical determinants of the substrate specificity of PKA. There was also high preponderance of histidine residues at the +6 position of the top peptides phosphorylated by PKA, even though only 10% of the peptides printed on the microarray featured a histidine residue at this position. A weak preference for histidine at the -6 position was also predicted with the KSPv1.0 algorithm, and it was also somewhat favoured amongst 1115 known protein substrates of PKA that have been   105       Figure 4.1. The protein kinase substrate peptide microarray technique. (A) Peptide microarray setup. Every peptide microarray has 4 fields that separately contain 445 peptides (plus control spots and orientation markers) printed in triplicate. This setup allows for the testing of three different kinases with one control per slide. (B) Phosphorylation and detection of phospho-peptides on microarray slides. Pro-Q Diamond® (Invitrogen) is a proprietary fluorescent stain that binds directly to the highly negatively charged phosphate group and allows for the detection of phosphopeptides without the requirement for antibodies or radioisotopes. Image adapted from Gao et al. (2010).  106  Table 4.1. List of the purified recombinant human protein kinases tested on the peptide microarrays.   A total of 214 human protein kinases from 75 kinase families were tested for their peptide substrate specificities on the microarray slides. Detailed information is available in Appendix 4.  107   Table 4.2. Determination of the optimal substrate amino acid specificity of PKAα using substrate peptide microarrays.   Based on the average value from triplicate measurements, 20 peptides were phosphorylated with signal strengths higher than 20% of the maximum observed value with 445 peptides. A peptide consensus was generated based on these 20 sequences (upper case: strong positive determinants; lower case: weak positive determinants; x: not a determinant). The peptide consensus is compared with the consensus sequences generated from experimentally determined protein substrates reported in the literature and from application of the Kinase Substrate Predictor Version 1.0 algorithm.    108  reported, similar to arginine and hydrophobic amino acid residues (Steven Pelech, personal communication)  4.4 Comparisons of substrate specificity of different kinases Using the data that was derived from the in vitro testing of more than 200 purified human protein kinases, the substrate selectivities between different protein kinase families or subfamilies were compared (a complete table is provided in Appendix 4). The MAPK family is an extensively-studied protein kinase family in the CMGC group as described by Hanks et al. (1988), and plays fundamental roles in the regulation of cell growth, differentiation, proliferation and apoptosis (Raman et al., 2007). There are three major subfamilies in the MAPK family, i.e. ERK, p38 and JNK, each comprising multiple members of the proline-directed protein-serine/threonine kinases. MAP kinases display very similar substrate specificities. With the kinase substrate peptide microarrays, members from all three subfamilies of MAP kinases were tested. As shown in Table 4.3, the KSPv1.0 algorithm was able to reveal slight differences in selectivity of the subfamilies. Peptide consensus sequences derived from the microarray data supported the essential roles of proline at +1 position, as well as leucine or isoleucine at -1 position for all five MAP kinases listed. Despite the occurrence of multiple proline residues in the protein consensus and predicted sequences as positive determinants, the results from  109  peptide microarrays did not support adjacent prolines in substrate peptides, which was possibly due to the impact of multiple proline residues on peptide structures. Some differences across subfamilies were observed. ERK1 and ERK2 showed preference of proline at the -2 position, whereas p38 isoforms and apparently ERK2 positively selected for basic residues at N-terminus of the phosphosites. Although substrate specificity of MAP kinases is mainly determined by secondary docking interactions and subcellular localization under physiological conditions (Biondi and Nebreda, 2003; Tanoue and Nishida, 2003), our data may be particularly useful to further improve the sensitivity of in vitro kinase assays. More specific kinase substrate peptides can be designed and synthesized for use in in vitro kinase assays to more accurately assess the phosphotransferase activity of these enzymes instead of indirectly using the phosphospecific antibody against phosphosites in the activation loops to monitor kinase activities. As discussed later in Chapter 5, phosphorylation state in the activation loop may not always correlate with kinase activity. Sixty one of the 90 known human protein-tyrosine kinases were also tested for their ability to phosphorylate peptides on microarray slides (a complete table is provided in Appendix 4). In general, PTKs were found to be less specific for their peptide substrates when compared to STKs. All PTKs were able to phosphorylate tyrosine, and to a lesser extent serine and threonine residues on the microarrays, while STKs usually  110  phosphorylated only serine and threonine residues. Table 4.4 lists some of the PTKs that were tested in this study. Based on the resulting data, PTKs displayed very similar specificities for their peptide substrates, which featured a strong preference for acidic residues at positions on the N-terminal side of the peptides. A proline residue at the +3 position might also be a positive determinant for PTKs. In cells, many PTKs autophosphorylate on tyrosine residues and recruit substrates with specific phosphotyrosine binding domains such as SH2 and PTB domains (Pawson, 2004). Thus, substrate specificities of protein-tyrosine kinases in vivo would largely depend on protein-protein interactions and dimerization of the kinases. Formation of such stable complexes effectively renders the substrate concentrations of their component proteins as infinite, so precise amino acid substrate sequence specificity for receptor-tyrosine kinases may be less crucial than for protein-serine/threonine kinases in general.  4.5 Kinase specificity determination with peptide macroarrays Although peptide microarrays have the advantage of requiring very little amount of materials, peptide macroarrays prepared using the SPOT synthesis technique are powerful tools for the further optimization of lead sequences identified from peptide microarrays. 111      Table 4.3. Substrate specificity of multiple MAPK members in vitro.   MAP kinases from all three major subfamilies were tested on peptide microarray for their selectivity to phosphorylate peptide substrates. The “Peptide Consensus” was determined from our peptide microarray studies, whereas the “Protein Consensus” was generated from alignment of the phosphosites of known protein substrates reported in the scientific literature (upper case: strong positive determinants; lower case: weak positive determinants; x: not a determinant). Slight differences (at positions -3, -2, +3, +4) were observed among subfamilies, which may help to facilitate further optimization of substrate peptides.     112      Table 4.4. Comparison of human protein-tyrosine kinase specificities.     Over two thirds of all the human protein-tyrosine kinases were tested with our peptide substrate microarrays. Despite the differences between their physiological substrate sequences, most protein-tyrosine kinases displayed similar specificities for peptide substrates in vitro. Acidic residues located N-terminal to the phosphoacceptor site often appear to be critical for tyrosine phosphorylation.    113         Figure 4.2. Screening of positive determinants at -1 and -2 amino acid positions of ERK1 substrates with peptide macroarrays. The SPOT membrane was prepared in a combinatorial manner such that each spot represented a mixture of peptides with a fixed combination of amino acids at -2 (B1) and -1 (B2) positions of the phosphosite. Purified ERK1 was used to phosphorylate the membrane. Based on the resultant image, I identified L/V as positive determinants for -1 position, and L/P for -2 position.    114  Several strategies are described for the optimization of peptides on peptide macroarrays, e.g. substitution analysis (systematic substitution of all residues of the sequence by other amino acids), truncation analysis (systematic reduction of the length of a peptide) and loop scan (systematic variation/insertion of cyclization of a peptide) (Schutkowski et al., 2005; Winkler and Campbell, 2008). Moreover, the combinatorial library strategy can be employed for de novo screenings of kinase peptide substrate sequences (Uttamchandani et al., 2003).  To verify the results from the peptide microarray with an independent method, SPOT membranes were prepared using a combinatorial strategy to screen for positive determinants at the -1 and -2 amino acid positions of ERK1 substrates. A substrate peptide combinatorial library was synthesized on cellulose membrane such that each spot represented a mixture of peptides with a fixed combination of amino acids at -2 and -1 positions of the phosphosite (Figure 4.2). Peptides on the membrane were phosphorylated by 1 μg of purified ERK1 for 1 hour. From the colorimetric staining result, L and V were identified as positive determinants for the -1 position, and L and P for -2 position (Figure 4.2). This sequence was consistent with the consensus sequences of ERK1/2 from peptide microarray analyses and the application of the KSPv1.0 algorithm. Given the bias of antibody binding affinity of peptides with different sequence context, the signal intensity cannot be used alone to determine the optimal substrate sequence in a quantitative manner. In this respect, the use of radioactive [-32P]ATP or [-33P]ATP for  115  phosphorylation of the immobilized peptides would provide a more reliable readout, but this approach is much less convenient and more hazardous to perform. More rounds of screening can be performed in the future using the same basic strategy to further define the most optimal amino acid residues as positive determinants at the surrounding positions.  4.6 Discussion The peptide substrate phosphorylation data obtained from testing over 200 purified human protein kinases against our peptide microarrays was used to expand the input datasets for training of our protein kinase substrate specificity predictive algorithm. The fruits of these effects contributed towards the creation of the Kinase Predictor Module of our open-access PhosphoNET website.  This appears to be the largest panel of diverse protein kinases that have ever been tested with peptide arrays. The largest previous study in this regard was the examination of the specificity of 61 yeast protein kinases individually with a positional peptide scanning library that featured 200 different degenerate mixtures of biotinylated peptides in which each mixture had one of the amino acid residues fixed at one of the amino acid positions surrounding the phosphoacceptor site and equimolar mixtures of 17 different amino acids at all of the other positions. Following incubation of the individual  116  recombinant yeast kinases with radiolabeled ATP in solution in a 1536-well microtitre plate, the phosphorylated peptides were then spotted with a capillary pin-based liquid transfer device onto a streptavidin-coated membrane and later exposed to a phosphor screen (Mok et al., 2010). One of the disadvantages of this approach is that each mixture could contain different amino acid substitutions that simultaneously both facilitate and inhibit peptide substrate recognition. With our methodology, the semi-optimized substrate peptides printed on our microarrays were based on in silico predictions and they were not degenerate. As different but related protein kinases have overlapping substrate specificities and often recognize basic amino acid residues at the -2 and -3 amino acid positions and/or a proline or hydrophobic amino acid residue at the +1 position, this strategy can still provide sufficient variations in amino acid sequences of the peptides to identify preferred amino acids at key positions surrounding the phosphoacceptor site. Our findings experimentally demonstrated that 132 of 153 tested STKs prefer an arginine residue at the -3 position, and this is in keeping with the observation that an arginine residue at the -3 position is the second most frequent amino acid residue in 13.2% of 125,456 human serine and threonine phosphosites. Interestingly, a proline residue at the +1 position is the most frequent amino acid residue in 17.6% of 125,456 human serine and threonine phosphosites (http://www.phosphonet.ca/). However, only 22 of 153 STK’s tested actually favoured a proline residue at the +1 phosphosite.  117  In the derivation of consensus substrate peptide sequences for the protein kinases tested in this study, it is important to also consider the frequency of each of the 20 amino acids at each position surrounding the phosphoacceptor site. Certain amino acids were over represented and others under represented in the 445 peptides printed on the microarrays. Therefore, some apparent preferences for certain amino acids at different positions could simply reflect the high preponderance of particular amino acids in the population of peptides that were used. Nevertheless, the consensus sequences derived from the peptide array studies were very similar to what observed from the alignment of known phosphosites in substrate proteins phosphorylated by protein kinases in vitro as shown in Appendix 4.  Another long term goal of this work was to eventually develop a sensitive microarray system that could be used to track multiple protein kinases simultaneously in more complex biological samples. To fully achieve this objective, the efficiencies of the microarray still need to be improved in different aspects. First of all, the production of peptide substrates was carried out using the SPOT synthesis technique. To evaluate the quality of these peptides, 5% of the 450 peptides were randomly selected and tested their purity with MS and HPLC. However, due to the differential chemical properties of peptides with different sequences, the actual amount of peptides in each spot still varied from 70 to 95%. Given the variation of these peptides and their immobilized state on the microarray slides, it was difficult to assess the real concentration of each peptide.  118  Designed as a platform to test the substrate specificities of different protein kinases, the microarrays were incubated with active kinases for 2 hours to ensure that all the phosphorylation reactions were saturated. The phosphorylation of each peptide was considered “all or none”, and a list of top peptide sequences were used to generate the peptide consensus sequence. In future studies, the protein kinases and their purified optimized substrate peptides should be tested in solution to determine the actual Km and Kcat values. It should also be appreciated that the Km values for peptides are usually orders of magnitude higher than phosphosites in proteins, in part because these sequences are less constrained and peptides have more entropy with respect to their 3D structures. Furthermore, it is possible that the immobilization of the short peptides on the glass slides may have reduced the accessibility of these substrates through steric inhibition.   Our detection of phosphorylated peptides on the microarray was based on their reactivity with the Pro-Q Diamond reagent. Ongoing studies in our laboratory with phosphopeptide arrays spotted with highly purified synthetic peptides have revealed that this reagent exhibits differential binding to phosphopeptides. Therefore, some of the differences between peptides as kinase substrates could have also partly reflected differences their reactivities with the Pro-Q Diamond.  In the future, more accurate kinase peptide substrate microarrays could be created by using the Kinase Substrate Predictor Version 2.0 algorithm, the use of more highly purified peptides, and other phosphopeptide detection reagents such as the pIMAGO  119  reagent from Tymora. Moreover, given the importance of secondary docking sites to substrates binding of many protein kinases, including additional binding sequences in the substrates may also be a future direction to pursue. The lengths and relative spacing of the docking sites and phosphorylation sites in these substrate peptides will need to be carefully evaluated. The definition of the D domain in ERK1/2 substrates (Garai et al., 2012; Holland and Cooper, 1999; Sharrocks et al., 2000), which bind to the CD motif in ERK1/2 (Rubinfeld et al., 1999; Tanoue et al., 2000), offers the opportunity of using D domain sequences in combination with optimal ERK1/2 consensus recognition sequences to develop even higher affinity and specificity substrate peptides to assay these MAP kinases. Understanding the molecular determinants that guide substrate recognition by a protein kinase is essential to understanding the role of individual protein kinases in particular cellular processes. The knowledge of substrate specificities that can be gained through our work has significant implications to the mapping of phosphorylation sites in the human proteome, the identification of previously uncharacterized substrates, and the generation of model substrates for small-molecule inhibitor design for drug development. The identification of novel phosphosites and substrates for kinases may ultimately lead to high-resolution maps of the kinase-dependent phosphorylation signalling networks that operate inside the cell, and a greater understanding of their role in the pathology of many human diseases, including cancer, diabetes, neurodegenerative and immune disorders.  120  Chapter 5: Regulatory Roles of Conserved Phosphorylation Sites in the Activation T-Loop of the MAP Kinase ERK1  5.1 Rationale In most typical eukaryotic protein kinases, reversible phosphorylation and dephosphorylation of the activation T-loop plays a key role in regulating their phosphotransferase activity (Taylor and Kornev, 2011; Taylor and Radzio-Andzelm, 1994). This activation loop is a variable segment extending from Subdomain VII (DFG) to Subdomain VIII (APE). Comparisons of active and inactive protein kinase structures indicate that conformational changes of this flexible region via phosphorylation promote adoption of the correct orientation for catalysis (Kornev et al., 2006).  Mitogen-activated protein kinases (MAPKs) play fundamental roles in the regulation of a diversity of functions including meiosis, cell cycle progression and stress responses from yeast to humans (Raman et al., 2007; Schaeffer and Weber, 1999). The Ras-Raf-MEK1/2-ERK1/2 signalling cascade has served as a paradigm for the study of cell signal transduction and protein kinase regulation (Wortzel and Seger, 2011). The activity of this cascade is commonly up-regulated in various types of cancers (Dhillon et al., 2007). In extracellular signal-regulated protein kinases 1 and 2 (ERK1/2), the dual phosphorylation on a threonine-glutamic acid-tyrosine (TEY) motif in the activation T  121  loop by the upstream kinases MAPK/ERK-1 and 2 (MEK1/2) was described as a critical event in the stimulation of ERK1/2 more than 20 years ago (Ahn et al., 1991; Anderson et al., 1990; Payne et al., 1991). Phosphosite-specific antibodies targeting the TEY site have been widely used to monitor the activation state of ERK1/2. However, besides the TEY motif, three flanking phosphorylation sites were identified by mass spectrometry in multiple cell lines and tumour samples as documented in the PhosphoSitePlus (www.phosphosite.org) and PhosphoNET (www.phosphonet.ca) knowledgebases.  In this study, the functional roles of the three phosphosites flanking the TEY motif in human ERK1 were investigated. The possible regulatory mechanisms of these phosphorylation events were also explored by comparing the primary sequences of all the human typical protein-serine/threonine kinases. The influences of these highly evolutionarily conserved phosphosites on ERK1 conformation and phosphotransferase activity may reflect a general mechanism for tight regulation of typical protein-serine/threonine kinases, which represent the largest class of highly ubiquitous protein kinases.   5.2 T207 and Y210 are highly conserved phosphorylation sites of ERK1 To investigate the regulation of protein kinases, the primary amino acid sequences of 491 human protein kinase domains were aligned along with information for all their  122  experimentally confirmed phosphosites. Out of the over 300 phosphosites that were reported to induce protein kinase phosphotransferase activity, 75% were located in the activation T-loop (Figure 5.1 A). From our further evolutionary analysis, phosphosites at -5 and -2 positions of APE  motif in Subdomain VIII were identified as the most conserved sites in many human protein-serine/threonine kinases, including MAPKs, CDKs, PKCs and PKB/AKTs (Figure 5.1 B). Their corresponding sites, T207 and Y210 in ERK1, are commonly conserved across 20 diverse species. However, the functional relevance of these sites flanking the TEY activation sites is still unclear. Interestingly, in human protein-tyrosine kinases, 100% of them featured a tryptophan residue instead of tyrosine residue at the -5 position of APE motif, and 98% had a proline residue (alanine and leucine for the exceptions) instead of a threonine residue at the -2 position, indicating possible general roles of these phosphosites specifically in protein-serine/threonine kinases.  5.3 ERK1 slowly autophosphorylates T207 in vitro Polyclonal antibodies were raised in rabbits against synthetic peptides that featured the phospho-T207 and phospho-Y210 sites of ERK1 (pT188 and pY191 of ERK2), and removed the non-phosphospecific antibodies with peptide columns with peptide ligands that corresponded to the unphosphorylated version of these sites following by affinity    123      Figure 5.1. Phosphorylation sites in human protein kinase catalytic domains. (A) Distribution of experimentally confirmed phosphosites in human protein kinase domains. Phosphosites in human protein kinase domain were mapped on the alignment. The total number of phosphosites was 1950 with 304 activation sites (green) mostly clustering at activation loop. (B) Examples of phosphosites in activation T-loop of serine/threonine kinases. Green: confirmed activating phosphosites; Yellow: phosphosites identified by mass spectrometry; Grey: predicted phosphosites.    124  purification with columns that featured the original immunizing phosphopeptides. Furthermore, unphosphorylated peptide with the corresponding sequence was added in the antibody incubation solution to block non-phosphospecific bindings. The specificity of each antibody was confirmed on peptide dot blots (Appendix 5). To evaluate whether the phosphorylation of these sites was due to autocatalysis, a kinase-dead (KD) mutant of ERK1 was created by substituting the lysine from the AxK motif (Subdomain II) with alanine (K71A). Purified GST fusion proteins of ERK1 wildtype (WT) and KD mutant were incubated with or without a constitutively active version of the upstream kinase MEK1 (MEK1-ΔN3EE) (Mansour et al., 1994) in presence of ATP (Figure 5.2A). On the one hand, phosphorylation of T207 was detected in WT but not KD, indicating autophosphorylation of the site in vitro. On the other hand, phosphorylation of Y210 was induced by MEK1 in both WT and KD ERK1. To determine the correlation between these phosphorylation events and ERK1 activation by MEK1, a time course experiment was performed (Figure 5.2B). Over 60 minutes of incubation time, T207 phosphorylation increased slowly while the phospho-signal from the TEY site saturated after 20 minutes. Activation of ERK1 by MEK1-ΔN3EE may induce the ability of the kinase to autophosphorylate on the T207 site. Observations from mass spectrometry analyses also supported the contention that phosphorylation of the T207 site was achieved by autophosphorylation. Samples of trypsin-digested peptides were prepared from ERK1-WT and KD that were  125       Figure 5.2 Phosphorylation of ERK1 T207 and Y210 in vitro. (A) ERK1-WT and KD phosphorylation by MEK1-ΔN3EE. The reactions were carried out at 30 °C for 15 min. (B) Time course experiment of ERK1-WT phosphorylation. At each time point, an aliquot of the incubation mix was taken and mixed with SDS-PAGE sample buffer to terminate the reaction. The samples were subsequently probed with phosphosite-specific antibodies (pT07, pY210, pTEpY) or the pan-expression ERK-CT antibody, and the Western blots from the region of the migration of the ERK1 are shown. Specificity data of pT207 antibody is shown in Appendix 5.    126  phosphorylated by MEK1-ΔN3EE in vitro and then subjected them to MS analysis. In the ERK1-KD sample, a tryptic peptide from the activation segment (190-208) with the TEY motif (T202 and Y204) dually phosphorylated was the only phosphopeptide detected. The unphosphorylated form of this peptide was also detected in the KD sample, but not the WT sample. Surprisingly, the tryptic peptide corresponding to the activation loop was not detected at all in the ERK1-WT from two independent experiments. The probable explanation of these results is that the additional phosphate groups of this target peptide made it difficult for MS detection either by interfering with ionization or the resultant fragment had too high of a charge to mass ratio and traveled too quickly through the mass spectrometer (Xie et al., 2011). 5.4 T207 and Y210 play critical roles in the regulation of ERK1 phosphotransferase activity To further characterize the roles of these phosphosites, T207 and Y210 were mutated, as well as another confirmed threonine phosphosite at the N-terminus of TEY (T198) in ERK1 activation loop (Table 5.1). Coupled kinase assays with recombinant protein of MEK1-ΔN3EE, ERK1 and myelin basic protein (MBP) revealed functional effects of T207 and Y210, but not T198 on ERK1 activation and its downstream phosphotransferase activity in vitro (Figure 5.3).  127  Substitution of T207 to Ala (T207A) markedly increased the autophosphorylation of the TEY phosphosites. Surprisingly, the T207A mutant only preserved about 20% of the phosphotransferase activity towards MBP when compared to WT. Moreover, the T207E mutant was phosphorylated at the TEY sites by MEK1-ΔN3EE to a similar extent with WT and T207A, but it completely failed to phosphorylate MBP. These findings indicated that the autophosphorylation of T207 is independent from TEY phosphorylation by MEK. Furthermore, phosphorylation at the TEY site does not necessarily correlate with ERK1 activation. All mutants with the Y210 substituted by just Phe or Glu (Y210F,or Y210E) or Phe in combination with Ala residue replacements of T198 and T207 sites (2AF) were not recognized by MEK1-ΔN3EE for phosphorylation, indicating an important role of this tyrosine residue in providing the proper conformation of the activation T-loop of the kinase for recognition by MEK1.  5.5 Mutation at T207 does not affect the specificity of ERK1 towards peptide substrates To confirm the effects of T207 phosphorylation on ERK1 phosphotransferase activity, ERK1-WT, T207A and T207E were tested on a KinexTM kinase substrate peptide microarray, which can help assess the phosphorylating activity of a kinase towards 445   128         Table 5.1. List of ERK1 mutants.     Six ERK1 mutants were created to characterize the functional roles of the three phosphosites near TEY motif. Reagents and methods used for mutagenesis are described in Chapter 2.     129  Figure 5.3. Phosphorylation and activity of ERK1 and its mutants. Purified recombinant ERK1 and its mutants were incubated with MEK1-ΔN3EE (red) or kinase dilution buffer (blue) in presence of 50 μM ATP at 30 °C for 15 min. An aliquot of each reaction mix was mixed with 2.5 μg MBP and incubated for another 2 min. Samples were mixed with SDS-PAGE sample buffer and analyzed by Western blotting using ERK-pTEpY antibody (A) and phospho-MBP antibody (B). Name of the mutants are listed in Table 5.1. The results are averaged from 3 to 5 separate experiments with the standard deviations indicated by bars. (**p<0.005)  130       Figure 5.4. Kinase specificity of ERK1-WT, T207A and T207E on the Kinex™ Kinase Substrate Microarray. Purified ERK1 and its mutants were activated by MEK1-ΔN3EE before incubated with the kinase substrate peptide microarray at 30 °C for 2 hours. MEK1 activity was suppressed by UO126. The control field was incubated with MEK1/UO126 only.  131  different peptides patterned after optimal protein kinase consensus substrate sequences. Recombinant ERK1 and its mutants were pre-activated by incubation with MEK1, and MEK1 activity was inhibited by adding the compound UO126 at the end of the pre-incubation step. After analyzing the microarray image, no phosphotransferase activity of the ERK1-T207E mutant was observed when compared to the MEK1/UO126 control field (Figure 5.4). The T207A and WT preparations showed the same selectivity in phosphorylating the substrate peptides on the chip. The strongest phosphorylation detected in both fields was from the same substrate peptide with the sequence (GGSFPLSPGKKGG). The ratio of net signal strength between WT and T207A from this peptide was 10:3. Among the top hits from T207A mutant, 14 out of 16 peptides were also strongly phosphorylated by ERK1 WT. These results are consistent with the in vitro kinase assays described above (Figure 5.3), and supported the conclusion that an Ala mutation at T207 of ERK1 did not affect the specificity towards peptide substrates, but decreased the overall phosphotransferase activity of the kinase by about 70%.  5.6 Phosphorylation at T207 may reduce the stability of activated ERK1 To study the importance of the T207 phosphorylation site in a physiologically relevant cell-based system, ERK1-WT, T207A and T207E constructs with flag tags were  132  transfected into the human embryonic kidney HEK293 cell line. After stimulation of overnight serum-deprived cells with 10% fetal bovine serum for 10 minutes to obtain maximal ERK1/2 activation, Flag-ERK1 was immunoprecipitated from the lysates of the harvested cells. It was observed that with similar amount of the kinase, the T207A mutant had a significantly higher level of TEY phosphorylation than WT (Figure 5.5B). However, only a weak phospho-TEY signal was detected from the T207E phosphorylation mimicking mutant. This indicated that the phosphorylation of T207 might have interfered with MEK1 phosphorylation and activation of ERK1 in vivo and/or enhance the dephosphorylation of the TEY phosphosites, since the absolute levels of the Flag-ERK1 constructs were comparable. Considering that all three forms of ERK1 were phosphorylated to an equal level by MEK1-ΔN3EE at the TEY site in vitro, the weaker TEY phosphorylation of ERK1-T207E in vivo was unlikely to be due to decreased phosphorylation by MEK1.  The immunoprecipitated ERK1 proteins were also tested for their kinase activity towards MBP. Despite the stronger phosphorylation of ERK1-T207A at the TEY site after incubated with ATP, neither the Ala nor Glu mutants were able to phosphorylate MBP as a substrate. Substitution of a nonphosphorylatable alanine residue at T207 only induced the autophosphorylation activity of ERK1, indicating that the unphosphorylated threonine residue is important for the kinase to maintain its active conformation. Once phosphorylated at T207, the phosphotransferase activity of the kinase would be   133       Figure 5.5. Phosphorylation and activity of ERK1-WT, T207A and T207E in HEK293 cells. (A) Phosphorylation of TEY motif of ERK1 under serum stimulation. HEK293 cells stably expressing Flag-ERK1 were starved overnight before stimulation with 10% FBS for 10 min. (B) Activity of immunoprecipitated Flag-ERK1. After serum stimulation, Flag-ERK1 was immunoprecipitated by Flag-tag antibody and incubated with 5 μg myelin basic protein (MBP) and 50 μM ATP at 30 °C for 15 min. The samples were subsequently subjected to SDS-PAGE and Western blotting with phosphosite-specific antibodies for the ERK1/2 TEY phosphosite (pTEpY) and phospho-MBP (pMBP), and the flag tag (Flag). Each image is representative of three independent experiments, and the averages of the phosphorylation of the pTEpY site and MBP from the three separate experiments are shown with the standard deviations indicated by bars. (**p<0.005)   134     Figure 5.6. Phosphorylation of ERK T207/T188 in multiple cell lines. (A) Phosphorylation of ERK T207/T188 was tested in NIH-3T3 cells, A431 cells, A549 cells and HeLa cells under various treatment conditions (AsO2-, arsenite; PMA, phorbol myristyl acetate; Noco., nocodazole). (B) A431 cells were treated with 10 μM proteasome inhibitor MG132 for 4 hours followed by stimulation with 100 ng/mL EGF for 5 min. (C) A431 cells were treated with 0.025% DMSO (control), protein-tyrosine phosphatase (PTP) inhibitors (25 μM PAO and 50 μM Na3VO4), or protein-serine/threonine phosphatase (STP) inhibitors (30 mM NaF) for 30 min. The samples were subsequently subjected to SDS-PAGE and Western blotting with phosphosite-specific antibodies or the pan-expression ERK-CT antibody, and the immunoblots from the region of the migration of the ERK1 are shown.    135  suppressed, and this may also possibly serve to increase recognition of the TEY site for dephosphorylation by phosphatases.  5.7 Phosphorylation of T207 is regulated by protein phosphatases The phosphorylation of T207 in multiple cell lines with various treatments that activated ERK1/2 was also examined (Figure 5.6A). None of these conditions were able to induce phosphorylation of T207 of ERK1 (or T188 of ERK2), indicating ERK1/2 that was phosphorylated on this threonine site was quickly removed from the system by degradation or dephosphorylation. Protein levels of ERK1/2 have been reported to be negatively regulated by the ubiquitin/proteasome pathway (Lu et al., 2002). However, treatment with a specific proteasome inhibitor, MG132, only increased the phosphorylation of the TEY site, but not the T207/T188 sites of ERK1/2 in A431 cells (Figure 5.6B). Given the earlier results that expression of ERK1-T207E did not affect the protein level of ERK1/2 in HEK293 cells, the T207 phosphorylation status of ERK1 is not likely to be related to degradation of the kinase in vivo. To investigate the effect of protein phosphatases on ERK1/2 T207/T188 phosphorylation, human A431 cervical carcinoma cells were treated with protein-tyrosine phosphatase (PTP) or protein-serine/threonine phosphatase (STP) inhibitors. Both sets of inhibitors were able to elevate the TEY phosphorylation level, but the  136  phospho-T207/T188 signal only increased in PTP inhibitor treated cells (Figure 5.6 C). These PTP inhibitors (PAO and Na3VO4) can suppress the activities of both protein-tyrosine phosphatases and dual-specificity phosphatases/MAPK phosphatases. Since the expression of the phospho-mimicking T207E mutant of ERK1 resulted in decreased levels of TEY phosphorylation, I conclude that phosphorylation of T207 may induce the dephosphorylation of ERK1 by MAPK phosphatases.  5.8 Discussion Phosphorylation of the activation T-loop is the most common mechanism of kinase activation. Protein kinases are dynamically regulated in this flexible segment to control downstream signalling. Under physiological conditions, activation of protein kinases is usually transient and robust treatments are required to ensure elicitation of the appropriate physiological responses. Various processes including phosphorylation at inhibitory sites, dephosphorylation at activatory sites and enhanced proteolysis have been reported to account for negative regulation of the protein phosphotransferase activity of different individual kinases. However, no common mechanism has been proposed for negative regulation of protein-serine/threonine kinases in general.  In this study, alignment of activation segment of human protein kinases permitted the identification of the two most ubiquitous phosphorylation sites in all  137  protein-serine/threonine kinases. Over 40 articles have been published regarding the functional studies of these two phosphosites using the site-directed mutagenesis technique in a variety of protein kinases (Table 5.2). All of the mutants created, including substitutions with nonphosphorylatable residues or phospho-mimicking residues, were reported to show decreased activity, and it has been repeatedly interpreted that these phosphosites may play critical roles in maintaining the active conformation of kinases. One of the most fundamental and well-studied kinases, ERK1, was used as a model to investigate the functional roles of these highly conserved phosphorylation sites in activation T-loop. Based on the in vitro data, T207 in ERK1 was autophosphorylated, whereas the phosphorylation of Y210 was catalyzed by MEK1. All of the mutants of these two sites (T207A, T207E, Y210F, Y210E) showed significantly decreased kinase activity toward MBP regardless of the level of TEY phosphorylation, which was consistent with the results from studies of many other kinases. Neither the nonphosphorylatable (Y210F) nor the phospho-mimicking (Y210E) mutant of Y210 site was recognized by MEK1, indicating the importance of this residue in its unphosphorylated form for the maintaining the structure of ERK1 so that it can be targeted by MEK1 and MEK2. Both the T207 and Y210 sites were phosphorylated slowly in vitro, and their phosphorylation was usually stimulated with TEY phosphorylation and kinase activation.  In contrast, in cell lysates, very low level of T207/T188 phosphorylation was 138     Table 5.2. List of publications about mutants of the two conserved phosphorylation sites in activation T-loop.   The positions of the two phosphosites are marked as T and Y in the header. The sites that were mutated are indicated in yellow.    139  observed even when the ERK1/2 pathway was activated. While this might be concluded to indicate that these sites might not be phosphorylated in cells at all, but in HEK293 cells stably expressing the phospho-mimicking ERK1-T207E mutant, the phosphorylation of TEY under serum stimulation was lower when compared to WT and T207A, suggesting that the hyper-phosphorylated forms of ERK1/2 were quickly removed from the system. This hypothesis was further supported by increased level of both TEY and T207/T188 phosphorylation in PTP inhibitor-treated A431 cells. Based on these results, it is reasonable to assume that the phosphorylation of TEY in ERK1-T207E would be restored when inhibiting or knocking down specific phosphatases. However, with the general PTP inhibitors that were used in this study, TEY phosphorylation would increase in all wild type and mutant ERK1 proteins. More specific inhibitors or knocking down experiments should be performed in the future to investigate this matter. Moreover, phosphospecific antibody can be used to isolate the TEY dually phosphorylated form of ERK1/2 to test if T207 site is phosphorylate in this population. Time course experiments can also be performed track the phosphorylation of TEY and T207 site in cells. Autophosphorylation of the T188 site in ERK2 was previously reported to promote cardiac hypertrophy in both mice and human without affecting the phosphotransferase activity of the kinase (Lorenz et al., 2009). In contrast, the present work with ERK1 mutants of the corresponding T207 site indicated that phosphorylation of this site inhibited the  140  phosphotransferase activity of ERK1 towards exogenous substrates. Any mutation of the T207 residue resulted in a dramatic decrease of ERK1 phosphotransferase activity even when the TEY activating motif was phosphorylated. This conclusion was also supported by the independent studies of Dr. Engelberg’s group with constitutively active mutants of ERK1/2 (David Engelberg, personal communication). In summary, ERK1 slowly undergoes autophosphorylation at the T207 site. Phosphorylation at this site inhibits the phosphotransferase activity of ERK1 towards other substrates, and it is involved in the dephosphorylation and deactivation of ERK1 by MAPK phosphatases. This may be an important general mechanism by which protein-serine/threonine kinases negatively regulate their activity after their initial activation. This may ensure that only when a sufficient portion of the population of a particular protein-serine/threonine kinase is activated by upstream stimuli over a sustained period will the signalling be advanced downstream to commit cells to evoking major changes in their cellular processes.     141  Chapter 6: Discussion and Future Directions  6.1 The alignment of human protein kinases One of the important advantages of studying the eukaryotic protein kinases is the huge size of this family of enzymes and the intense scrutiny that they have been subjected to in hundreds of laboratories and many diverse species over the last half century. The availability of a wealth of data about the primary structures of thousands of protein kinases and over a hundred tertiary structures, knowledge of natural variations in amino acid sequence between cognate protein kinases within and between species, site-directed mutagenesis studies, and careful examination of the biochemical properties of purified preparations of these enzymes has facilitated detailed structure-function analyses. Consolidation of this information in kinome-wide studies has permitted predictions of protein kinase regulation and actions amongst closely related members of these important enzymes. This has all been spurred on by the recognition that protein kinases play major roles in hundreds of different diseases, including cancer, diabetes, neurological and immunological disorders.   In the present study, I started with a carefully aligned map of all the human protein kinase catalytic domains. The alignment was created based on the well-characterized conserved subdomains of eukaryotic protein kinases (Hanks et al., 1988). Our bioinformatics group  142  analyzed data of confirmed phosphorylation sites including their functional effects and conservation scores, as well as point mutations and disease relevance of mutations that have been discovered within human protein kinase domains. All the information was mapped to the original alignment to investigate the general theme of human protein kinase regulation, and also to identify specific elements that differentiate functions of individual kinase. With more and more data appearing in the literature, updated versions of this annotated map will serve a powerful tool to predict functional effects of new phosphosites and point mutations in catalytic domain of any eukaryotic protein kinases.  In this chapter, I will discuss regulation mechanisms that are potentially applicable to most protein kinases, as well as the improved protein kinase specificity predictive algorithm with the peptide microarray screening data.  6.2 Common regulation mechanisms of protein kinases Most protein kinases, especially those that are serine/threonine-specific, are usually activated by phosphorylation of the activation T-loop segment. Phosphorylation of this flexible loop of a kinase promotes the active conformation for catalysis (Johnson et al., 1996; Taylor and Kornev, 2011). From our human protein kinase catalytic domain alignments and evolutionary analysis of confirmed phosphorylation sites in this region, two additional phosphosites were identified nearby within activation segment of most human protein-serine/threonine kinases. In  143  Chapter 5, human ERK1 was used as a model to investigate the functional effects of these conserved phosphosites. It was demonstrated that the autophosphorylation of T207 after the initial activation of ERK1 inhibited its phosphotransferase activity towards other substrates, and it may be involved in dephosphorylation and deactivation of the kinase.  Given the high conservation and the ubiquity of the phosphosite in various protein-serine/threonine kinases, these results may support an important general mechanism of negative self-regulation of protein-serine/threonine kinases. Autoinhibition after the initial activation ensures that only when a sufficient portion of the population of a particular protein-serine/threonine kinase is activated over a sustained period will the cell permit the downstream signalling events that commit it to advance to the next steps. For many cellular processes such a cell cycle progression in response to growth factors, the subsequent inactivation of protein kinases is just as critical as their initial recruitment. Other such processes include, amongst others, relief from checkpoint controls with availability of nutrients and repair of DNA damage (Elledge, 1996; Johnson and Walker, 1999; Kastan and Bartek, 2004), resumption of anabolic reactions and inhibition of catabolic reactions with restoration of ATP levels following build up of AMP and cAMP, and depression of the immune system after an infection has subsided.  The ability of signalling proteins to auto-inactivate is not uncommon. For example, G proteins feature an intrinsic GTPase activity that curtails their actions on effector proteins. Many  144  GTPase activating proteins act to stimulate the rates of GTP hydrolysis by G proteins. It may be speculated that some regulatory proteins may directly act on protein kinases to accelerate or depress their rates of autoinactivation by stimulation of phosphorylation of the inhibitory phosphosites upstream of the Kinase Subdomain VIII APE sequence. Such kinase inactivating proteins might specifically recognize the phosphorylated region just before the APE region, which in the phosphorylated state would be very conserved amongst most protein-serine/threonine kinases. These regions may also act as docking sites for members of the large family of dual-specificity protein phosphatases to facilitate their dephosphorylation of the activating phosphorylation sites that are located further downstream in the activation T-loop.  The finding that a tyrosine residue equivalent to the Y210 position of ERK1 does not occur in any of the 90 human protein-tyrosine kinases may reflect the need to prolong the sustained activation of this class of kinases. Protein-tyrosine kinases would be more likely than protein-serine/threonine kinases to undergo rapid autophosphorylation and inactivation with a Tyr residue just before the APE region. Interestingly, a third of the human protein-tyrosine kinases feature a Tyr residue that is located just 4 amino acid residues C-terminal to the APE sequence, and phosphorylation has been confirmed for two thirds of these Tyr residues by mass spectrometry. It is feasible that this highly conserved Tyr phosphosite services as an inactivating autophosphorylation site for protein-tyrosine kinases. While little is known from experimental studies about the physiological role of this Tyr phosphorylation site, this residue at Y701 in TrkA  145  is known to be inhibitory for this kinase (de Pablo et al., 2008).  6.3 Specificity determining residues and the Kinase Substrate Predictor algorithms Besides the conserved catalytic core and general activation mechanism shared by most protein kinases, there are variable regions in the sequences of their catalytic domains that determine the unique physiological roles of individual kinases by influencing their substrate selectivity. From the alignment of human protein kinase domains, protein-serine/threonine kinases can be differentiated from protein-tyrosine kinases based on the sequence of several residues near Subdomain VI and VIII (Figure 3.1 B). Some of these residues serve as specificity-determining residues (SDRs), which determine whether a protein kinase selectively phosphorylates serine/threonine or tyrosine sites. There appear to be many more SDRs that directly interact with side chains of amino acids surrounding the target phosphosites and thus determine the optimal substrate sequence context of each protein kinase (Miller et al., 2008; Safaei et al., 2011). In Chapter 4, I described a study that used peptide microarrays to test the specificities of over 200 human protein kinases. This method allows fast screening against more peptide sequences with less material when compared to the kinase assays on SPOT membranes or in solution. The peptide microarrays were designed with 445 semi-optimum kinase substrate sequences, which provide higher efficiency than arrays with random sequences and less bias than  146  arrays based on physiological phosphosites. It should be appreciated that physiological phosphorylation sites are likely to have evolved so they can be individually recognized by a variety of different protein kinases, protein phosphatases and interacting proteins. Thus, the amino acid sequences surrounding a physiological site are unlikely to be optimally recognized by any one kinase. This is why alignment of the sequences of even physiological phosphosites may be insufficient to define the consensus substrate sequence for most protein kinases. The employment of arrays of substrate peptide analogues is necessary to elucidate the most optimal substrate sequences. In this thesis, I have shown that this optimizing process can be facilitated by in silico prediction of kinase substrate peptides and their incorporation into peptide microarrays. After the initial specificity profiling using the peptide microarray, subsequent optimization with peptide macroarray technique and validation with in solution kinase assay can be applied to develop a sensitive and specific substrate peptide sequence. The peptide substrate phosphorylation data obtained from the peptide microarrays was used to expand the training datasets of our protein kinase specificity predictive algorithm, which in turn could lead to the design of even better substrate peptides for incorporation into improved kinase substrate peptide arrays. The acquisition of 3D structural information of kinase catalytic domains, the incorporation of a larger number of SDRs, and the adoption of an empirically-derived amino acid interaction table were also included in the development of the second generation KSPv2.0 algorithm with improved accuracy and predictive power for mapping  147  phosphoproteome interactions. The fruits of these efforts to improve the KSPv2.0 algorithm are now accessible on the PhosphoNET website (www.phosphonet.ca) with kinase-substrate predictions available for over 960,000 known and predicted human phosphosites. Understanding the in vitro kinase substrate specificities provides an important first step to study protein kinases with unknown function or connections in the cell signalling networks. However, as is discussed in this thesis, protein kinases are tightly regulated in cells by a variety of mechanisms including expression, post-translational modifications, subcellular localization, protein-protein interaction, protein degradation and many others. Under specific physiological stimuli, the output from protein kinases is decided by the location of the protein, as well as the availability of its interaction partners, regulators and substrates. Different techniques need to be applied together to study the functional roles of protein kinases and their pathways.  6.4 Conclusion As one of the largest group of cell signalling proteins, protein kinases play pivotal role in regulating all aspects of physiological activities. In the present study, I explored the commonalities in protein kinases, including their evolutionary origin and general regulatory mechanisms. Based on the results, eukaryotic protein kinases evolved from an ancient aminoacyl-tRNA ligase. The conserved phosphorylation sites in activation T-loops of many protein-serine/threonine kinases may serve as autoinhibitory sites to regulate kinase activity in  148  addition to the activating phosphosites. Together with my colleagues, I tested the substrate specificities of over 200 human protein kinases with our peptide microarray technique, which helped to improve our predictive abilities for specificity-determining residues and kinase substrate selectivities. The knowledge derived from this work complement our current understanding of the protein kinase superfamily, and should contribute to the development of improved techniques for cell signalling research that will hopefully paved the way to development of better diagnostics and therapeutics to combat the wide range of diseases that still confront humanity and other creatures that co-inhabit planet Earth.    149  Bibliography Adachi M, Fukuda M, Nishida E (2000) Nuclear export of MAP kinase (ERK) involves a MAP kinase kinase (MEK)-dependent active transport mechanism. The Journal of cell biology 148: 849-856  Ahearn IM, Haigis K, Bar-Sagi D, Philips MR (2012) Regulating the regulator: post-translational modification of RAS. Nature Reviews Molecular Cell Biology 13: 39-51  Ahn NG, Seger R, Bratlien RL, Diltz CD, Tonks NK, Krebs EG (1991) Multiple components in an epidermal growth factor-stimulated protein kinase cascade. In vitro activation of a myelin basic protein/microtubule-associated protein 2 kinase. J Biol Chem 266: 4220-4227  Al-Lazikani B, Banerji U, Workman P (2012) Combinatorial drug therapy for cancer in the post-genomic era. Nat Biotechnol 30: 679-692  Alessi DR, Gomez N, Moorhead G, Lewis T, Keyse SM, Cohen P (1995) Inactivation of p42 MAP kinase by protein phosphatase 2A and a protein tyrosine phosphatase, but not CL100, in various cell lines. Current Biology : CB 5: 283-295  Anderson NG, Maller JL, Tonks NK, Sturgill TW (1990) Requirement for integration of signals from two distinct phosphorylation pathways for activation of MAP kinase. Nature 343: 651-653  Angenendt P (2005) Progress in protein and antibody microarray technology. Drug Discovery Today 10: 503-511  Anjum R, Blenis J (2008) The RSK family of kinases: emerging roles in cellular signalling. Nature Reviews Molecular Cell Biology 9: 747-758  Asano E, Maeda M, Hasegawa H, Ito S, Hyodo T, Yuan H, Takahashi M, Hamaguchi M, Senga T (2011) Role of palladin phosphorylation by extracellular signal-regulated kinase in cell migration. PLoS One 6: e29338  Babu GR, Jin W, Norman L, Waterfield M, Chang M, Wu X, Zhang M, Sun SC (2006) Phosphorylation of NF-kappaB1/p105 by oncoprotein kinase Tpl2: implications for a novel mechanism of Tpl2 regulation. Biochimica et Biophysica Acta 1763: 174-181  Bachmann M, Hennemann H, Xing PX, Hoffmann I, Moroy T (2004) The oncogenic serine/threonine kinase Pim-1 phosphorylates and inhibits the activity of Cdc25C-associated  150  kinase 1 (C-TAK1): a novel role for Pim-1 at the G2/M cell cycle checkpoint. J Biol Chem 279: 48319-48328  Barford D, Johnson LN (1989) The allosteric transition of glycogen phosphorylase. Nature 340: 609-616  Belletti B, Nicoloso MS, Schiappacassi M, Chimienti E, Berton S, Lovat F, Colombatti A, Baldassarre G (2005) p27(kip1) functional regulation in human cancer: a potential target for therapeutic designs. Current Medicinal Chemistry 12: 1589-1605  Berndt N, Hamilton AD, Sebti SM (2011) Targeting protein prenylation for cancer therapy. Nature Reviews Cancer 11: 775-791  Biondi RM, Nebreda AR (2003) Signalling specificity of Ser/Thr protein kinases through docking-site-mediated interactions. Biochem J 372: 1-13  Blume-Jensen P, Hunter T (2001) Oncogenic kinase signalling. Nature 411: 355-365  Bollag G, Tsai J, Zhang J, Zhang C, Ibrahim P, Nolop K, Hirth P (2012) Vemurafenib: the first drug approved for BRAF-mutant cancer. Nature Reviews Drug Discovery 11: 873-886  Borrebaeck CA (2006) Antibody microarray-based oncoproteomics. Expert Opinion on Biological Therapy 6: 833-838  Brennan DF, Dar AC, Hertz NT, Chao WC, Burlingame AL, Shokat KM, Barford D (2011) A Raf-induced allosteric transition of KSR stimulates phosphorylation of MEK. Nature 472: 366-369  Brill LM, Salomon AR, Ficarro SB, Mukherji M, Stettler-Gill M, Peters EC (2004) Robust phosphoproteomic profiling of tyrosine phosphorylation sites from human T cells using immobilized metal affinity chromatography and tandem mass spectrometry. Analytical Chemistry 76: 2763-2772  Brognard J, Sierecki E, Gao T, Newton AC (2007) PHLPP and a second isoform, PHLPP2, differentially attenuate the amplitude of Akt signaling by regulating distinct Akt isoforms. Molecular Cell 25: 917-931  Brown JR, Douady CJ, Italia MJ, Marshall WE, Stanhope MJ (2001) Universal trees based on large combined protein sequence data sets. Nat Genet 28: 281-285   151  Brown MC, Turner CE (2004) Paxillin: adapting to change. Physiological Reviews 84: 1315-1339  Brumbaugh K, Johnson W, Liao WC, Lin MS, Houchins JP, Cooper J, Stoesz S, Campos-Gonzalez R (2011) Overview of the generation, validation, and application of phosphosite-specific antibodies. Methods Mol Biol 717: 3-43  Buchwalter G, Gross C, Wasylyk B (2004) Ets ternary complex transcription factors. Gene 324: 1-14  Buhrman G, Holzapfel G, Fetics S, Mattos C (2010) Allosteric modulation of Ras positions Q61 for a direct role in catalysis. Proc Natl Acad Sci U S A 107: 4931-4936  Callister SJ, Barry RC, Adkins JN, Johnson ET, Qian WJ, Webb-Robertson BJ, Smith RD, Lipton MS (2006) Normalization approaches for removing systematic biases associated with mass spectrometry and label-free proteomics. J Proteome Res 5: 277-286  Camps M, Nichols A, Gillieron C, Antonsson B, Muda M, Chabert C, Boschert U, Arkinstall S (1998) Catalytic activation of the phosphatase MKP-3 by ERK2 mitogen-activated protein kinase. Science 280: 1262-1265  Canagarajah BJ, Khokhlatchev A, Cobb MH, Goldsmith EJ (1997) Activation mechanism of the MAP kinase ERK2 by dual phosphorylation. Cell 90: 859-869  Casnellie JE, Krebs EG (1984) The use of synthetic peptides for defining the specificity of tyrosine protein kinases. Adv Enzyme Regul 22: 501-515  Catalanotti F, Reyes G, Jesenberger V, Galabova-Kovacs G, de Matos Simoes R, Carugo O, Baccarini M (2009) A Mek1-Mek2 heterodimer determines the strength and duration of the Erk signal. Nature Structural and Molecular Biology 16: 294-303  Charest DL, Mordret G, Harder KW, Jirik F, Pelech SL (1993) Molecular cloning, expression, and characterization of the human mitogen-activated protein kinase p44erk1. Mol Cell Biol 13: 4679-4690  Chen YA, Eschrich SA (2014) Computational methods and opportunities for phosphorylation network medicine. Transl Cancer Res 3: 266-278  Chevet E, Wong HN, Gerber D, Cochet C, Fazel A, Cameron PH, Gushue JN, Thomas DY, Bergeron JJ (1999) Phosphorylation by CK2 and MAPK enhances calnexin association with ribosomes. The EMBO Journal 18: 3655-3666  152   Chuderland D, Konson A, Seger R (2008) Identification and characterization of a general nuclear translocation signal in signaling proteins. Molecular Cell 31: 850-861  Cicirelli MF, Pelech SL, Krebs EG (1988) Activation of multiple protein kinases during the burst in protein phosphorylation that precedes the first meiotic cell division in Xenopus oocytes. J Biol Chem 263: 2009-2019  Clark-Lewis I, Sanghera JS, Pelech SL (1991) Definition of a consensus sequence for peptide substrate recognition by p44mpk, the meiosis-activated myelin basic protein kinase. J Biol Chem 266: 15180-15184  Cohen P (2002) Protein kinases--the major drug targets of the twenty-first century? Nature Reviews Drug Discovery 1: 309-315  Cole SP, Campling BG, Atlaw T, Kozbor D, Roder JC (1984) Human monoclonal antibodies. Molecular and Cellular Biochemistry 62: 109-120  Cox J, Mann M (2008) MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat Biotechnol 26: 1367-1372  Cruzalegui FH, Cano E, Treisman R (1999) ERK activation induces phosphorylation of Elk-1 at multiple S/T-P motifs to high stoichiometry. Oncogene 18: 7948-7957  Daigle DM, McKay GA, Thompson PR, Wright GD (1999) Aminoglycoside antibiotic phosphotransferases are also serine protein kinases. Chem Biol 6: 11-18  Dalby KN, Morrice N, Caudwell FB, Avruch J, Cohen P (1998) Identification of regulatory phosphorylation sites in mitogen-activated protein kinase (MAPK)-activated protein kinase-1a/p90rsk that are inducible by MAPK. J Biol Chem 273: 1496-1505  Dar AC, Shokat KM (2011) The evolution of protein kinase inhibitors from antagonists to agonists of cellular signaling. Annual Review of Biochemistry 80: 769-795  Darnell JE, Jr. (1997) Phosphotyrosine signaling and the single cell:metazoan boundary. Proc Natl Acad Sci U S A 94: 11767-11769  Das S, Cho J, Lambertz I, Kelliher MA, Eliopoulos AG, Du K, Tsichlis PN (2005) Tpl2/cot signals activate ERK, JNK, and NF-kappaB in a cell-type and stimulus-specific manner. J Biol Chem 280:  153  23748-23757  Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, Teague J, Woffendin H, Garnett MJ, Bottomley W, Davis N, Dicks E, Ewing R, Floyd Y, Gray K, Hall S, Hawes R, Hughes J, Kosmidou V, Menzies A, Mould C, Parker A, Stevens C, Watt S, Hooper S, Wilson R, Jayatilake H, Gusterson BA, Cooper C, Shipley J, Hargrave D, Pritchard-Jones K, Maitland N, Chenevix-Trench G, Riggins GJ, Bigner DD, Palmieri G, Cossu A, Flanagan A, Nicholson A, Ho JW, Leung SY, Yuen ST, Weber BL, Seigler HF, Darrow TL, Paterson H, Marais R, Marshall CJ, Wooster R, Stratton MR, Futreal PA (2002) Mutations of the BRAF gene in human cancer. Nature 417: 949-954  de Graauw M, Hensbergen P, van de Water B (2006) Phospho-proteomic analysis of cellular signaling. Electrophoresis 27: 2676-2686  de Pablo Y, Perez-Garcia MJ, Georgieva MV, Sanchis D, Lindqvist N, Soler RM, Comella JX, Llovera M (2008) Tyr-701 is a new regulatory site for neurotrophin receptor TrkA trafficking and function. J Neurochem 104: 124-139  de Rooij J, Zwartkruis FJ, Verheijen MH, Cool RH, Nijman SM, Wittinghofer A, Bos JL (1998) Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature 396: 474-477  DeFea KA, Zalevsky J, Thoma MS, Dery O, Mullins RD, Bunnett NW (2000) beta-arrestin-dependent endocytosis of proteinase-activated receptor 2 is required for intracellular targeting of activated ERK1/2. The Journal of Cell Biology 148: 1267-1281  Deribe YL, Pawson T, Dikic I (2010) Post-translational modifications in signal integration. Nature Structural and Molecular Biology 17: 666-672  Dhanasekaran DN, Kashef K, Lee CM, Xu H, Reddy EP (2007) Scaffold proteins of MAP-kinase modules. Oncogene 26: 3185-3202  Dhillon AS, Hagan S, Rath O, Kolch W (2007) MAP kinase signalling pathways in cancer. Oncogene 26: 3279-3290  Dougherty MK, Muller J, Ritt DA, Zhou M, Zhou XZ, Copeland TD, Conrads TP, Veenstra TD, Lu KP, Morrison DK (2005) Regulation of Raf-1 by direct feedback phosphorylation. Molecular Cell 17: 215-224  Dumitru CD, Ceci JD, Tsatsanis C, Kontoyiannis D, Stamatakis K, Lin JH, Patriotis C, Jenkins NA,  154  Copeland NG, Kollias G, Tsichlis PN (2000) TNF-alpha induction by LPS is regulated posttranscriptionally via a Tpl2/ERK-dependent pathway. Cell 103: 1071-1083  Duronio V (2008) The life of a cell: apoptosis regulation by the PI3K/PKB pathway. Biochem J 415: 333-344  Eckhart W, Hutchinson MA, Hunter T (1979) An activity phosphorylating tyrosine in polyoma T antigen immunoprecipitates. Cell 18: 925-933  Eisen T, Ahmad T, Flaherty KT, Gore M, Kaye S, Marais R, Gibbens I, Hackett S, James M, Schuchter LM, Nathanson KL, Xia C, Simantov R, Schwartz B, Poulin-Costello M, O'Dwyer PJ, Ratain MJ (2006) Sorafenib in advanced melanoma: a Phase II randomised discontinuation trial analysis. British Journal of Cancer 95: 581-586  Elledge SJ (1996) Cell cycle checkpoints: preventing an identity crisis. Science 274: 1664-1672  Erhardt P, Troppmair J, Rapp UR, Cooper GM (1995) Differential regulation of Raf-1 andB-Raf and Ras-dependent activation of mitogen-activated protein kinase by cyclic AMP in PC12 cells. Mol Cell Biol 15: 5524–5530  Exton JH (1994) Phosphatidylcholine breakdown and signal transduction. Biochimica et Biophysica Acta 1212: 26-42  Fantz DA, Jacobs D, Glossip D, Kornfeld K (2001) Docking sites on substrate proteins direct extracellular signal-regulated kinase to phosphorylate specific residues. J Biol Chem 276: 27256-27265  Farooq A, Zhou MM (2004) Structure and regulation of MAPK phosphatases. Cellular Signalling 16: 769-779  Fields GB, Noble RL (1990) Solid phase peptide synthesis utilizing 9-fluorenylmethoxycarbonyl amino acids. International Journal of Peptide and Protein research 35: 161-214  Finn RD, Mistry J, Tate J, Coggill P, Heger A, Pollington JE, Gavin OL, Gunasekaran P, Ceric G, Forslund K, Holm L, Sonnhammer EL, Eddy SR, Bateman A (2010) The Pfam protein families database. Nucleic Acids Research 38: D211-222  Fischer EH, Krebs EG (1955) Conversion of phosphorylase b to phosphorylase a in muscle extracts. J Biol Chem 216: 121-132   155  Flaherty KT, Infante JR, Daud A, Gonzalez R, Kefford RF, Sosman J, Hamid O, Schuchter L, Cebon J, Ibrahim N, Kudchadkar R, Burris HA, 3rd, Falchook G, Algazi A, Lewis K, Long GV, Puzanov I, Lebowitz P, Singh A, Little S, Sun P, Allred A, Ouellet D, Kim KB, Patel K, Weber J (2012) Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. The New England Journal of Medicine 367: 1694-1703  Flaherty KT, Puzanov I, Kim KB, Ribas A, McArthur GA, Sosman JA, O'Dwyer PJ, Lee RJ, Grippo JF, Nolop K, Chapman PB (2010) Inhibition of mutated, activated BRAF in metastatic melanoma. The New England Journal of Medicine 363: 809-819  Forbes SA, Beare D, Gunasekaran P, Leung K, Bindal N, Boutselakis H, Ding M, Bamford S, Cole C, Ward S, Kok CY, Jia M, De T, Teague JW, Stratton MR, McDermott U, Campbell PJ (2014) COSMIC: exploring the world's knowledge of somatic mutations in human cancer. Nucleic Acids Research  Frank R (2002) The SPOT-synthesis technique. Synthetic peptide arrays on membrane supports--principles and applications. J Immunol Methods 267: 13-26  Frodin M, Gammeltoft S (1999) Role and regulation of 90 kDa ribosomal S6 kinase (RSK) in signal transduction. Molecular and Cellular Endocrinology 151: 65-77  Frodin M, Jensen CJ, Merienne K, Gammeltoft S (2000) A phosphoserine-regulated docking site in the protein kinase RSK2 that recruits and activates PDK1. The EMBO Journal 19: 2924-2934  Fukunaga R, Hunter T (2004) Identification of MAPK substrates by expression screening with solid-phase phosphorylation. Methods Mol Biol 250: 211-236  Gao L, Sun H, Yao SQ (2010) Activity-based high-throughput determination of PTPs substrate specificity using a phosphopeptide microarray. Biopolymers 94: 810-819  Gao T, Brognard J, Newton AC (2008) The phosphatase PHLPP controls the cellular levels of protein kinase C. J Biol Chem 283: 6300-6311  Gao T, Furnari F, Newton AC (2005) PHLPP: a phosphatase that directly dephosphorylates Akt, promotes apoptosis, and suppresses tumor growth. Molecular Cell 18: 13-24  Garai A, Zeke A, Gogl G, Toro I, Fordos F, Blankenburg H, Barkai T, Varga J, Alexa A, Emig D, Albrecht M, Remenyi A (2012) Specificity of linear motifs that bind to a common mitogen-activated protein kinase docking groove. Sci Signal 5: ra74   156  Garnett MJ, Marais R (2004) Guilty as charged: B-RAF is a human oncogene. Cancer Cell 6: 313-319  Giard DJ, Aaronson SA, Todaro GJ, Arnstein P, Kersey JH, Dosik H, Parks WP (1973) In vitro cultivation of human tumors: establishment of cell lines derived from a series of solid tumors. J Natl Cancer Inst 51: 1417-1423  Gibbs CS, Zoller MJ (1991) Rational scanning mutagenesis of a protein kinase identifies functional regions involved in catalysis and substrate interactions. J Biol Chem 266: 8923-8931  Gibney GT, Messina JL, Fedorenko IV, Sondak VK, Smalley KS (2013) Paradoxical oncogenesis--the long-term effects of BRAF inhibition in melanoma. Nat Rev Clin Oncol 10: 390-399  Gille H, Kortenjann M, Thomae O, Moomaw C, Slaughter C, Cobb MH, Shaw PE (1995) ERK phosphorylation potentiates Elk-1-mediated ternary complex formation and transactivation. The EMBO Journal 14: 951-962  Gilman AG (1995) Nobel Lecture. G proteins and regulation of adenylyl cyclase. Bioscience Reports 15: 65-97  Ginty DD, Kornhauser JM, Thompson MA, Bading H, Mayo KE, Takahashi JS, Greenberg ME (1993) Regulation of CREB phosphorylation in the suprachiasmatic nucleus by light and a circadian clock. Science 260: 238-241  Glenney JR, Jr., Zokas L, Kamps MP (1988) Monoclonal antibodies to phosphotyrosine. J Immunol Methods 109: 277-285  Gloerich M, Bos JL (2011) Regulating Rap small G-proteins in time and space. Trends Cell Biol 21: 615-623  Gonzalez FA, Raden DL, Davis RJ (1991) Identification of substrate recognition determinants for human ERK1 and ERK2 protein kinases. J Biol Chem 266: 22159-22163  Gopalbhai K, Jansen G, Beauregard G, Whiteway M, Dumas F, Wu C, Meloche S (2003) Negative regulation of MAPKK by phosphorylation of a conserved serine residue equivalent to Ser212 of MEK1. J Biol Chem 278: 8118-8125  Hanks SK (1987) Homology probing: identification of cDNA clones encoding members of the protein-serine kinase family. Proc Natl Acad Sci U S A 84: 388-392  157   Hanks SK, Quinn AM, Hunter T (1988) The protein kinase family: conserved features and deduced phylogeny of the catalytic domains. Science 241: 42-52  Hanks SK, Hunter T (1995) Protein kinases 6. The eukaryotic protein kinase superfamily: kinase (catalytic) domain structure and classification. FASEB J 9:576-596  Harlow E, Lane D (2006) Bradford assay. CSH Protocols 2006  Harrington HA, Komorowski M, Beguerisse-Diaz M, Ratto GM, Stumpf MP (2012) Mathematical modeling reveals the functional implications of the different nuclear shuttling rates of Erk1 and Erk2. Physical Biology 9: 036001  Hastie CJ, McLauchlan HJ, Cohen P (2006) Assay of protein kinases using radiolabeled ATP: a protocol. Nat Protoc 1: 968-971  Hatano N, Mori Y, Oh-hora M, Kosugi A, Fujikawa T, Nakai N, Niwa H, Miyazaki J, Hamaoka T, Ogata M (2003) Essential role for ERK2 mitogen-activated protein kinase in placental development. Genes to Cells : Devoted to Molecular and Cellular Mechanisms 8: 847-856  Hattori M, Minato N (2003) Rap1 GTPase: Functions, Regulation, and Malignancy. J Biochem 134: 479-484  Hatzivassiliou G, Song K, Yen I, Brandhuber BJ, Anderson DJ, Alvarado R, Ludlam MJ, Stokoe D, Gloor SL, Vigers G, Morales T, Aliagas I, Liu B, Sideris S, Hoeflich KP, Jaiswal BS, Seshagiri S, Koeppen H, Belvin M, Friedman LS, Malek S (2010) RAF inhibitors prime wild-type RAF to activate the MAPK pathway and enhance growth. Nature 464: 431-435  Hawkes R, Niday E, Gordon J (1982) A dot-immunobinding assay for monoclonal and other antibodies. Anal Biochem 119: 142-147  Hayes DF, Markus HS, Leslie RD, Topol EJ (2014) Personalized medicine: risk prediction, targeted therapies and mobile health technology. BMC Medicine 12: 37  Heidorn SJ, Milagre C, Whittaker S, Nourry A, Niculescu-Duvas I, Dhomen N, Hussain J, Reis-Filho JS, Springer CJ, Pritchard C, Marais R (2010) Kinase-dead BRAF and oncogenic RAS cooperate to drive tumor progression through CRAF. Cell 140: 209-221  Hoch JA (2000) Two-component and phosphorelay signal transduction. Current Opinion in Microbiology 3: 165-170  158   Holderfield M, Deuker MM, McCormick F, McMahon M (2014) Targeting RAF kinases for cancer therapy: BRAF-mutated melanoma and beyond. Nature reviews Cancer 14: 455-467  Holland PM, Cooper JA (1999) Protein modification: docking sites for kinases. Current Biology : CB 9: R329-331  Hollenhorst PC, McIntosh LP, Graves BJ (2011) Genomic and biochemical insights into the specificity of ETS transcription factors. Annual Review of Biochemistry 80: 437-471  Hornbeck PV, Kornhauser JM, Tkachev S, Zhang B, Skrzypek E, Murray B, Latham V, Sullivan M (2012) PhosphoSitePlus: a comprehensive resource for investigating the structure and function of experimentally determined post-translational modifications in man and mouse. Nucleic Acids Research 40: D261-270  Hoshi M, Nishida E, Sakai H (1989) Characterization of a mitogen-activated, Ca2+-sensitive microtubule-associated protein-2 kinase. Eur J Biochem 184: 477-486  Hu CD, Kariya K, Kotani G, Shirouzu M, Yokoyama S, Kataoka T (1997) Coassociation of Rap1A and Ha-Ras with Raf-1 N-terminal region interferes with ras-dependent activation of Raf-l. J Biol Chem 272: 11702-11705  Hu CD, Kariya K, Okada T, Qi X, Song C, Kataoka T (1999) Effect of phosphorylation on activities of Rap1A to interact with Raf-1 and to suppress Ras-dependent Raf-1 activation. J Biol Chem 274: 48-51  Hu J, Stites EC, Yu H, Germino EA, Meharena HS, Stork PJ, Kornev AP, Taylor SS, Shaw AS (2013) Allosteric activation of functionally asymmetric RAF kinase dimers. Cell 154: 1036-1046  Hunter T (1989) Protein modification: phosphorylation on tyrosine residues. Current Opinion in Cell Biology 1: 1168-1181  Hunter T (2000) Signaling--2000 and beyond. Cell 100: 113-127  Ishibe S, Joly D, Liu ZX, Cantley LG (2004) Paxillin serves as an ERK-regulated scaffold for coordinating FAK and Rac activation in epithelial morphogenesis. Molecular Cell 16: 257-267  Ishibe S, Joly D, Zhu X, Cantley LG (2003) Phosphorylation-dependent paxillin-ERK association mediates hepatocyte growth factor-stimulated epithelial morphogenesis. Molecular Cell 12: 1275-1285  159   Jacobs D, Glossip D, Xing H, Muslin AJ, Kornfeld K (1999) Multiple docking sites on substrate proteins form a modular system that mediates recognition by ERK MAP kinase. Genes and Development 13: 163-175  Johnson DG, Walker CL (1999) Cyclins and cell cycle checkpoints. Annu Rev Pharmacol Toxicol 39: 295-312  Johnson LN, Noble ME, Owen DJ (1996) Active and inactive protein kinases: structural basis for regulation. Cell 85: 149-158  Jones DT, Kocialkowski S, Liu L, Pearson DM, Ichimura K, Collins VP (2009) Oncogenic RAF1 rearrangement and a novel BRAF mutation as alternatives to KIAA1549:BRAF fusion in activating the MAPK pathway in pilocytic astrocytoma. Oncogene 28: 2119-2123  Julien SG, Dube N, Hardy S, Tremblay ML (2011) Inside the human cancer tyrosine phosphatome. Nature reviews Cancer 11: 35-49  Junttila MR, Li SP, Westermarck J (2008) Phosphatase-mediated crosstalk between MAPK signaling pathways in the regulation of cell survival. FASEB J 22: 954-965  Kannan N, Neuwald AF (2005) Did protein kinase regulatory mechanisms evolve through elaboration of a simple structural component? J Mol Biol 351: 956-972  Kannan N, Taylor SS, Zhai Y, Venter JC, Manning G (2007) Structural and functional diversity of the microbial kinome. PLoS Biol 5: e17  Kaplan KB, Bibbins KB, Swedlow JR, Arnaud M, Morgan DO, Varmus HE (1994) Association of the amino-terminal half of c-Src with focal adhesions alters their properties and is regulated by phosphorylation of tyrosine 527. The EMBO Journal 13: 4745-4756  Kastan MB, Bartek J (2004) Cell-cycle checkpoints and cancer. Nature 432: 316-323  Kawata T, Shevchenko A, Fukuzawa M, Jermyn KA, Totty NF, Zhukovskaya NV, Sterling AE, Mann M, Williams JG (1997) SH2 signaling in a lower eukaryote: a STAT protein that regulates stalk cell differentiation in dictyostelium. Cell 89: 909-916  Keeling PJ, Palmer JD (2008) Horizontal gene transfer in eukaryotic evolution. Nat Rev Genet 9: 605-618   160  Kelley LA, Sternberg MJ (2009) Protein structure prediction on the Web: a case study using the Phyre server. Nat Protoc 4: 363-371  Kemp BE, Benjamini E, Krebs EG (1976) Synthetic hexapeptide substrates and inhibitors of 3':5'-cyclic AMP-dependent protein kinase. Proc Natl Acad Sci U S A 73: 1038-1042  Kemp BE, Graves DJ, Benjamini E, Krebs EG (1977) Role of multiple basic residues in determining the substrate specificity of cyclic AMP-dependent protein kinase. J Biol Chem 252: 4888-4894  Kennelly PJ (2002) Protein kinases and protein phosphatases in prokaryotes: a genomic perspective. FEMS Microbiol Lett 206: 1-8  Kent C (2005) Regulatory enzymes of phosphatidylcholine biosynthesis: a personal perspective. Biochimica et Biophysica Acta 1733: 53-66  Keyse SM (2008) Dual-specificity MAP kinase phosphatases (MKPs) and cancer. Cancer Metastasis Reviews 27: 253-261  King N, Westbrook MJ, Young SL, Kuo A, Abedin M, Chapman J, Fairclough S, Hellsten U, Isogai Y, Letunic I, Marr M, Pincus D, Putnam N, Rokas A, Wright KJ, Zuzow R, Dirks W, Good M, Goodstein D, Lemons D, Li W, Lyons JB, Morris A, Nichols S, Richter DJ, Salamov A, Sequencing JG, Bork P, Lim WA, Manning G, Miller WT, McGinnis W, Shapiro H, Tjian R, Grigoriev IV, Rokhsar D (2008) The genome of the choanoflagellate Monosiga brevicollis and the origin of metazoans. Nature 451: 783-788  Kitayama H, Sugimoto Y, Matsuzaki T, Ikawa Y, Noda M (1989) A ras-related gene with transformation sup pressor activity. Cell 56: 77-84  Kitayama H, Matsuzaki T, Ikawa Y, Noda M (1990) Genetic analysis of the Krsten-ras-revertant 1 gene: potentiatior of its tumor suppressor activity by specific point mutaions. Proc Natl Acad Sci U S A 87: 4284–4288  Knighton DR, Zheng JH, Ten Eyck LF, Ashford VA, Xuong NH, Taylor SS, Sowadski JM (1991a) Crystal structure of the catalytic subunit of cyclic adenosine monophosphate-dependent protein kinase. Science 253: 407-414  Knighton DR, Zheng JH, Ten Eyck LF, Xuong NH, Taylor SS, Sowadski JM (1991b) Structure of a peptide inhibitor bound to the catalytic subunit of cyclic adenosine monophosphate-dependent protein kinase. Science 253: 414-420  161   Kolch W, Heidecker G, Kochs G, Hummel R, Vahidi H, Mischak H, Finkenzeller G, Marmé D, Rapp UR (1993) Protein kinase C alpha activates RAF-1 by direct phosphorylation. Nature 364: 249-52  Kolch W (2005) Coordinating ERK/MAPK signalling through scaffolds and inhibitors. Nature reviews Molecular Cell Biology 6: 827-837  Kornev AP, Haste NM, Taylor SS, Eyck LF (2006) Surface comparison of active and inactive protein kinases identifies a conserved activation mechanism. Proc Natl Acad Sci U S A 103: 17783-17788  Kornev AP, Taylor SS, Ten Eyck LF (2008) A helix scaffold for the assembly of active protein kinases. Proc Natl Acad Sci U S A 105: 14377-14382  Krebs EG (1993) Nobel Lecture. Protein phosphorylation and cellular regulation I. Bioscience Reports 13: 127-142  Kreegipuu A, Blom N, Brunak S, Jarv J (1998) Statistical analysis of protein kinase specificity determinants. FEBS Lett 430: 45-50  Ku H, Meier KE (2000) Phosphorylation of paxillin via the ERK mitogen-activated protein kinase cascade in EL4 thymoma cells. J Biol Chem 275: 11333-11340  Lamour V, Quevillon S, Diriong S, N'Guyen VC, Lipinski M, Mirande M (1994) Evolution of the Glx-tRNA synthetase family: the glutaminyl enzyme as a case of horizontal gene transfer. Proc Natl Acad Sci U S A 91: 8670-8674  Lara R, Seckl MJ, Pardo OE (2013) The p90 RSK family members: common functions and isoform specificity. Cancer Research 73: 5301-5308  Laub MT, Goulian M (2007) Specificity in Two-Component Signal Transduction Pathways. Annu Rev Genet 41: 121-145  Lee T, Hoofnagle AN, Kabuyama Y, Stroud J, Min X, Goldsmith EJ, Chen L, Resing KA, Ahn NG (2004) Docking motif interactions in MAP kinases revealed by hydrogen exchange mass spectrometry. Molecular Cell 14: 43-55  Leek JT, Scharpf RB, Bravo HC, Simcha D, Langmead B, Johnson WE, Geman D, Baggerly K, Irizarry RA (2010) Tackling the widespread and critical impact of batch effects in high-throughput  162  data. Nat Rev Genet 11: 733-739  Lefkowitz RJ (2004) Historical review: a brief history and personal retrospective of seven-transmembrane receptors. Trends Pharmacol Sci 25: 413-422  Lefloch R, Pouyssegur J, Lenormand P (2008) Single and combined silencing of ERK1 and ERK2 reveals their positive contribution to growth signaling depending on their expression levels. Mol Cell Biol 28: 511-527  Lefloch R, Pouyssegur J, Lenormand P (2009) Total ERK1/2 activity regulates cell proliferation. Cell Cycle 8: 705-711  Lemmon MA, Schlessinger J (2010) Cell signaling by receptor tyrosine kinases. Cell 141: 1117-1134  Leonard CJ, Aravind L, Koonin EV (1998) Novel families of putative protein kinases in bacteria and archaea: evolution of the "eukaryotic" protein kinase superfamily. Genome Res 8: 1038-1047  Li F, Zhao C, Wang L (2014) Molecular-targeted agents combination therapy for cancer: developments and potentials. International Journal of Cancer Journal International du Cancer 134: 1257-1269  Linding R, Jensen LJ, Ostheimer GJ, van Vugt MA, Jorgensen C, Miron IM, Diella F, Colwill K, Taylor L, Elder K, Metalnikov P, Nguyen V, Pasculescu A, Jin J, Park JG, Samson LD, Woodgett JR, Russell RB, Bork P, Yaffe MB, Pawson T (2007) Systematic discovery of in vivo phosphorylation networks. Cell 129: 1415-1426  Linding R, Jensen LJ, Pasculescu A, Olhovsky M, Colwill K, Bork P, Yaffe MB, Pawson T (2008) NetworKIN: a resource for exploring cellular phosphorylation networks. Nucleic Acids Research 36: D695-699  Lito P, Pratilas CA, Joseph EW, Tadi M, Halilovic E, Zubrowski M, Huang A, Wong WL, Callahan MK, Merghoub T, Wolchok JD, de Stanchina E, Chandarlapaty S, Poulikakos PI, Fagin JA, Rosen N (2012) Relief of profound feedback inhibition of mitogenic signaling by RAF inhibitors attenuates their activity in BRAFV600E melanomas. Cancer Cell 22: 668-682  Liu Y, Chance MR (2014) Integrating phosphoproteomics in systems biology. Comput Struct Biotechnol J 10: 90-97  Liu Y, Shah K, Yang F, Witucki L, Shokat KM (1998) A molecular gate which controls unnatural  163  ATP analogue recognition by the tyrosine kinase v-Src. Bioorganic and Medicinal Chemistry 6: 1219-1226  Liu ZX, Yu CF, Nickel C, Thomas S, Cantley LG (2002) Hepatocyte growth factor induces ERK-dependent paxillin phosphorylation and regulates paxillin-focal adhesion kinase association. J Biol Chem 277: 10452-10458  Lo RS, Shi H (2014) Detecting mechanisms of acquired BRAF inhibitor resistance in melanoma. Methods Mol Biol 1102: 163-174  Lorenz K, Schmitt JP, Schmitteckert EM, Lohse MJ (2009) A new type of ERK1/2 autophosphorylation causes cardiac hypertrophy. Nat Med 15: 75-83  Lowery RG, Vogel K, Till J (2010) ADP detection technologies. Assay and Drug Development Technologies 8: 1-2; author reply 1  Lu Z, Xu S, Joazeiro C, Cobb MH, Hunter T (2002) The PHD domain of MEKK1 acts as an E3 ubiquitin ligase and mediates ubiquitination and degradation of ERK1/2. Molecular Cell 9: 945-956  Luttrell LM (2002) Activation and targeting of mitogen-activated protein kinases by G-protein-coupled receptors. Canadian Journal of Physiology and Pharmacology 80: 375-382  Luttrell LM, Gesty-Palmer D (2010) Beyond desensitization: physiological relevance of arrestin-dependent signaling. Pharmacological Reviews 62: 305-330  Lyons JF, Wilhelm S, Hibner B, Bollag G (2001) Discovery of a novel Raf kinase inhibitor. Endocrine-related Cancer 8: 219-225  Machida K, Mayer BJ, Nollau P (2003) Profiling the global tyrosine phosphorylation state. Mol Cell Proteomics 2: 215-233  Mann M, Ong SE, Gronborg M, Steen H, Jensen ON, Pandey A (2002) Analysis of protein phosphorylation using mass spectrometry: deciphering the phosphoproteome. Trends Biotechnol 20: 261-268  Manning BD, Cantley LC (2002) Hitting the target: emerging technologies in the search for kinase substrates. Sci STKE 162: pe49  Manning G, Plowman GD, Hunter T, Sudarsanam S (2002a) Evolution of protein kinase signaling  164  from yeast to man. Trends Biochem Sci 27: 514-520  Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S (2002b) The protein kinase complement of the human genome. Science 298: 1912-1934  Manning G, Young SL, Miller WT, Zhai Y (2008) The protist, Monosiga brevicollis, has a tyrosine kinase signaling network more elaborate and diverse than found in any known metazoan. Proc Natl Acad Sci U S A 105: 9674-9679  Mansour SJ, Matten WT, Hermann AS, Candia JM, Rong S, Fukasawa K, Vande Woude GF, Ahn NG (1994) Transformation of mammalian cells by constitutively active MAP kinase kinase. Science 265: 966-970  Marchi M, D'Antoni A, Formentini I, Parra R, Brambilla R, Ratto GM, Costa M (2008) The N-terminal domain of ERK1 accounts for the functional differences with ERK2. PLoS One 3: e3873  Mason CS, Springer CJ, Cooper RG, Superti-Furga G, Marshall CJ, Marais R (1999) Serine and tyrosine phosphorylations cooperate in Raf-1, but not B-Raf activation. The EMBO Journal 18: 2137-2148  Matheny SA, Chen C, Kortum RL, Razidlo GL, Lewis RE, White MA (2004) Ras regulates assembly of mitogenic signalling complexes through the effector protein IMP. Nature 427: 256-260  Matsubayashi Y, Fukuda M, Nishida E (2001) Evidence for existence of a nuclear pore complex-mediated, cytosol-independent pathway of nuclear translocation of ERK MAP kinase in permeabilized cells. J Biol Chem 276: 41755-41760  Matzke MM, Brown JN, Gritsenko MA, Metz TO, Pounds JG, Rodland KD, Shukla AK, Smith RD, Waters KM, McDermott JE, Webb-Robertson BJ (2013) A comparative analysis of computational approaches to relative protein quantification using peptide peak intensities in label-free LC-MS proteomics experiments. Proteomics 13: 493-503  McClendon CL, Kornev AP, Gilson MK, Taylor SS (2014) Dynamic architecture of a protein kinase. Proc Natl Acad Sci U S A 111: 4623-4631  McKay MM, Morrison DK (2007) Integrating signals from RTKs to ERK/MAPK. Oncogene 26: 3113-3121   165  McKay MM, Ritt DA, Morrison DK (2011) RAF inhibitor-induced KSR1/B-RAF binding and its effects on ERK cascade signaling. Current Biology : CB 21: 563-568  Medina EA, Afsari RR, Ravid T, Castillo SS, Erickson KL, Goldkorn T (2005) Tumor necrosis factor-{alpha} decreases Akt protein levels in 3T3-L1 adipocytes via the caspase-dependent ubiquitination of Akt. Endocrinology 146: 2726-2735  Messersmith WA, Ahnen DJ (2008) Targeting EGFR in colorectal cancer. The New England Journal of Medicine 359: 1834-1836  Middelbeek J, Clark K, Venselaar H, Huynen MA, van Leeuwen FN (2010) The alpha-kinase family: an exceptional branch on the protein kinase tree. Cellular and Molecular Life Sciences : CMLS 67: 875-890  Miller ML, Jensen LJ, Diella F, Jorgensen C, Tinti M, Li L, Hsiung M, Parker SA, Bordeaux J, Sicheritz-Ponten T, Olhovsky M, Pasculescu A, Alexander J, Knapp S, Blom N, Bork P, Li S, Cesareni G, Pawson T, Turk BE, Yaffe MB, Brunak S, Linding R (2008) Linear motif atlas for phosphorylation-dependent signaling. Sci Signal 1: ra2  Mitsushima M, Suwa A, Amachi T, Ueda K, Kioka N (2004) Extracellular signal-regulated kinase activated by epidermal growth factor and cell adhesion interacts with and phosphorylates vinexin. J Biol Chem 279: 34570-34577  Mok J, Kim PM, Lam HY, Piccirillo S, Zhou X, Jeschke GR, Sheridan DL, Parker SA, Desai V, Jwa M, Cameroni E, Niu H, Good M, Remenyi A, Ma JL, Sheu YJ, Sassi HE, Sopko R, Chan CS, De Virgilio C, Hollingsworth NM, Lim WA, Stern DF, Stillman B, Andrews BJ, Gerstein MB, Snyder M, Turk BE (2010) Deciphering protein kinase specificity through large-scale analysis of yeast phosphorylation site motifs. Sci Signal 3: ra12  Morrison DK, Davis RJ (2003) Regulation of MAP kinase signaling modules by scaffold proteins in mammals. Annual Review of Cell and Developmental biology 19: 91-118  Mueller C, Liotta LA, Espina V (2010) Reverse phase protein microarrays advance to use in clinical trials. Molecular Oncology 4: 461-481  Mueller LN, Brusniak MY, Mani DR, Aebersold R (2008) An assessment of software solutions for the analysis of mass spectrometry based quantitative proteomics data. J Proteome Res 7: 51-61  Mueller LN, Rinner O, Schmidt A, Letarte S, Bodenmiller B, Brusniak MY, Vitek O, Aebersold R, Muller M (2007) SuperHirn - a novel tool for high resolution LC-MS-based peptide/protein  166  profiling. Proteomics 7: 3470-3480  Murugan AK, Dong J, Xie J, Xing M (2009) MEK1 mutations, but not ERK2 mutations, occur in melanomas and colon carcinomas, but none in thyroid carcinomas. Cell Cycle 8:2122-2124  Murzin AG, Brenner SE, Hubbard T, Chothia C (1995) SCOP: a structural classification of proteins database for the investigation of sequences and structures. J Mol Biol 247: 536-540  Mustelin T (2007) A brief introduction to the protein phosphatase families. Methods Mol Biol 365: 9-22  Noble ME, Endicott JA, Johnson LN (2004) Protein kinase inhibitors: insights into drug design from structure. Science 303: 1800-1805  Obenauer JC, Cantley LC, Yaffe MB (2003) Scansite 2.0: Proteome-wide prediction of cell signaling interactions using short sequence motifs. Nucleic Acids Research 31: 3635-3641  Oda Y, Nagasu T, Chait BT (2001) Enrichment analysis of phosphorylated proteins as a tool for probing the phosphoproteome. Nat Biotechnol 19: 379-382  Ohtsuka T, Shimizu K, Yamamori B, Kuroda S, Takai Y (1996) Activation of brain B-Raf protein kinase by Rap1B small GTPbinding protein. J Biol Chem 271: 1258-1261  Ortiz-Lombardia M, Pompeo F, Boitel B, Alzari PM (2003) Crystal structure of the catalytic domain of the PknB serine/threonine kinase from Mycobacterium tuberculosis. J Biol Chem 278: 13094-13100  Pages G, Guerin S, Grall D, Bonino F, Smith A, Anjuere F, Auberger P, Pouyssegur J (1999) Defective thymocyte maturation in p44 MAP kinase (Erk 1) knockout mice. Science 286: 1374-1377  Palanisamy N, Ateeq B, Kalyana-Sundaram S, Pflueger D, Ramnarayanan K, Shankar S, Han B, Cao Q, Cao X, Suleman K, Kumar-Sinha C, Dhanasekaran SM, Chen YB, Esgueva R, Banerjee S, LaFargue CJ, Siddiqui J, Demichelis F, Moeller P, Bismar TA, Kuefer R, Fullen DR, Johnson TM, Greenson JK, Giordano TJ, Tan P, Tomlins SA, Varambally S, Rubin MA, Maher CA, Chinnaiyan AM (2010) Rearrangements of the RAF kinase pathway in prostate cancer, gastric cancer and melanoma. Nat Med 16: 793-798  Parker LL, Atherton-Fessler S, Piwnica-Worms H (1992) p107wee1 is a dual-specificity kinase that phosphorylates p34cdc2 on tyrosine 15. Proc Natl Acad Sci U S A 89: 2917-2921  167   Patterson KI, Brummer T, O'Brien PM, Daly RJ (2009) Dual-specificity phosphatases: critical regulators with diverse cellular targets. Biochem J 418: 475-489  Pawson T (2004) Specificity in signal transduction: from phosphotyrosine-SH2 domain interactions to complex cellular systems. Cell 116: 191-203  Pawson T (2007) Dynamic control of signaling by modular adaptor proteins. Current Opinion in Cell Biology 19: 112-116  Pawson T, Scott JD (2005) Protein phosphorylation in signaling--50 years and counting. Trends Biochem Sci 30: 286-290  Payne DM, Rossomando AJ, Martino P, Erickson AK, Her JH, Shabanowitz J, Hunt DF, Weber MJ, Sturgill TW (1991) Identification of the regulatory phosphorylation sites in pp42/mitogen-activated protein kinase (MAP kinase). The EMBO Journal 10: 885-892  Pearson G, Robinson F, Beers Gibson T, Xu BE, Karandikar M, Berman K, Cobb MH (2001) Mitogen-activated protein (MAP) kinase pathways: regulation and physiological functions. Endocrine Reviews 22: 153-183  Peisach D, Gee P, Kent C, Xu Z (2003) The crystal structure of choline kinase reveals a eukaryotic protein kinase fold. Structure 11: 703-713  Pelech SL, Tombes RM, Meijer L, Krebs EG (1988) Activation of myelin basic protein kinases during echinoderm oocyte maturation and egg fertilization. Developmental Biology 130: 28-36  Perez J, Castaneda-Garcia A, Jenke-Kodama H, Muller R, Munoz-Dorado J (2008) Eukaryotic-like protein kinases in the prokaryotes and the myxobacterial kinome. Proc Natl Acad Sci U S A 105: 15950-15955  Perlson E, Hanz S, Ben-Yaakov K, Segal-Ruder Y, Seger R, Fainzilber M (2005) Vimentin-dependent spatial translocation of an activated MAP kinase in injured nerve. Neuron 45: 715-726  Pickart CM, Eddins MJ (2004) Ubiquitin: structures, functions, mechanisms. Biochimica et Biophysica Acta 1695: 55-72  Plotnikov A, Zehorai E, Procaccia S, Seger R (2011) The MAPK cascades: signaling components, nuclear roles and mechanisms of nuclear translocation. Biochimica et Biophysica Acta 1813:  168  1619-1633  Poulikakos PI, Persaud Y, Janakiraman M, Kong X, Ng C, Moriceau G, Shi H, Atefi M, Titz B, Gabay MT, Salton M, Dahlman KB, Tadi M, Wargo JA, Flaherty KT, Kelley MC, Misteli T, Chapman PB, Sosman JA, Graeber TG, Ribas A, Lo RS, Rosen N, Solit DB (2011) RAF inhibitor resistance is mediated by dimerization of aberrantly spliced BRAF(V600E). Nature 480: 387-390  Poulikakos PI, Zhang C, Bollag G, Shokat KM, Rosen N (2010) RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF. Nature 464: 427-430  Pulido R, Zuniga A, Ullrich A (1998) PTP-SL and STEP protein tyrosine phosphatases regulate the activation of the extracellular signal-regulated kinases ERK1 and ERK2 by association through a kinase interaction motif. The EMBO Journal 17: 7337-7350  Pullikuth AK, Catling AD (2007) Scaffold mediated regulation of MAPK signaling and cytoskeletal dynamics: a perspective. Cellular Signalling 19: 1621-1632  Pylayeva-Gupta Y, Grabocka E, Bar-Sagi D (2011) RAS oncogenes: weaving a tumorigenic web. Nature Reviews Cancer 11: 761-774  Raman M, Chen W, Cobb MH (2007) Differential regulation and properties of MAPKs. Oncogene 26: 3100-3112  Ranganathan A, Yazicioglu MN, Cobb MH (2006) The nuclear localization of ERK2 occurs by mechanisms both independent of and dependent on energy. J Biol Chem 281: 15645-15652  Rapp UR, Gotz R, Albert S (2006) BuCy RAFs drive cells into MEK addiction. Cancer Cell 9: 9-12  Rasar M, DeFranco DB, Hammes SR (2006) Paxillin regulates steroid-triggered meiotic resumption in oocytes by enhancing an all-or-none positive feedback kinase loop. J Biol Chem 281: 39455-39464  Ray LB, Sturgill TW (1987) Rapid stimulation by insulin of a serine/threonine kinase in 3T3-L1 adipocytes that phosphorylates microtubule-associated protein 2 in vitro. Proc Natl Acad Sci U S A 84: 1502-1506  Reimer U, Reineke U, Schneider-Mergener J (2002) Peptide arrays: from macro to micro. Curr Opin Biotechnol 13: 315-320   169  Reineke U, Volkmer-Engert R, Schneider-Mergener J (2001) Applications of peptide arrays prepared by the SPOT-technology. Curr Opin Biotechnol 12: 59-64  Reszka AA, Seger R, Diltz CD, Krebs EG, Fischer EH (1995) Association of mitogen-activated protein kinase with the microtubule cytoskeleton. Proc Natl Acad Sci U S A 92: 8881-8885  Robbins DJ, Zhen E, Owaki H, Vanderbilt CA, Ebert D, Geppert TD, Cobb MH (1993) Regulation and properties of extracellular signal-regulated protein kinases 1 and 2 in vitro. J Biol Chem 268: 5097-5106  Rodgers JT, Vogel RO, Puigserver P (2011) Clk2 and B56beta mediate insulin-regulated assembly of the PP2A phosphatase holoenzyme complex on Akt. Molecular Cell 41: 471-479  Rodriguez-Viciana P, Tetsu O, Tidyman WE, Estep AL, Conger BA, Cruz MS, McCormick F, Rauen KA (2006) Germline mutations in genes within the MAPK pathway cause cardio-facio-cutaneous syndrome. Science 311: 1287-1290  Rokas A, Kruger D, Carroll SB (2005) Animal evolution and the molecular signature of radiations compressed in time. Science 310: 1933-1938  Roskoski R, Jr. (2003) Protein prenylation: a pivotal posttranslational process. Biochemical and Biophysical Research Communications 303: 1-7  Roskoski R, Jr. (2012a) ERK1/2 MAP kinases: structure, function, and regulation. Pharmacological Research : The Official Journal of the Italian Pharmacological Society 66: 105-143  Roskoski R, Jr. (2012b) MEK1/2 dual-specificity protein kinases: structure and regulation. Biochemical and Biophysical Research Communications 417: 5-10  Ross AH, Baltimore D, Eisen HN (1981) Phosphotyrosine-containing proteins isolated by affinity chromatography with antibodies to a synthetic hapten. Nature 294: 654-656  Rossomando AJ, Dent P, Sturgill TW, Marshak DR (1994) Mitogen-activated protein kinase kinase 1 (MKK1) is negatively regulated by threonine phosphorylation. Mol Cell Biol 14: 1594-1602  Roux PP, Richards SA, Blenis J (2003) Phosphorylation of p90 ribosomal S6 kinase (RSK) regulates extracellular signal-regulated kinase docking and RSK activity. Mol Cell Biol 23: 4796-4804  170   Rubinfeld H, Hanoch T, Seger R (1999) Identification of a cytoplasmic-retention sequence in ERK2. J Biol Chem 274: 30349-30352  Rudrabhatla P, Reddy MM, Rajasekharan R (2006) Genome-wide analysis and experimentation of plant serine/ threonine/tyrosine-specific protein kinases. Plant Mol Biol 60: 293-319  Safaei J, Manuch J, Gupta A, Stacho L, Pelech S (2011) Prediction of 492 human protein kinase substrate specificities. Proteome Science 9 Suppl 1: S6  Saha R, Dasgupta S, Basu G, Roy S (2009) A chimaeric glutamyl:glutaminyl-tRNA synthetase: implications for evolution. Biochem J 417: 449-455  Sands WA, Palmer TM (2008) Regulating gene transcription in response to cyclic AMP elevation. Cellular Signalling 20: 460-466  Saunders NF, Brinkworth RI, Huber T, Kemp BE, Kobe B (2008) Predikin and PredikinDB: a computational framework for the prediction of protein kinase peptide specificity and an associated database of phosphorylation sites. BMC Bioinformatics 9: 245  Schaeffer HJ, Catling AD, Eblen ST, Collier LS, Krauss A, Weber MJ (1998) MP1: a MEK binding partner that enhances enzymatic activation of the MAP kinase cascade. Science 281: 1668-1671  Schaeffer HJ, Weber MJ (1999) Mitogen-activated protein kinases: specific messages from ubiquitous messengers. Mol Cell Biol 19: 2435-2444  Scheeff ED, Bourne PE (2005) Structural evolution of the protein kinase-like superfamily. PLoS Computational Biology 1: e49  Scheffzek K, Ahmadian MR, Kabsch W, Wiesmuller L, Lautwein A, Schmitz F, Wittinghofer A (1997) The Ras-RasGAP complex: structural basis for GTPase activation and its loss in oncogenic Ras mutants. Science 277: 333-338  Scheidig AJ, Burmester C, Goody RS (1999) The pre-hydrolysis state of p21(ras) in complex with GTP: new insights into the role of water molecules in the GTP hydrolysis reaction of ras-like proteins. Structure 7: 1311-1324  Schutkowski M, Reimer U, Panse S, Dong L, Lizcano JM, Alessi DR, Schneider-Mergener J (2004) High-content peptide microarrays for deciphering kinase specificity and biology. Angew  171  Chem Int Ed Engl 43: 2671-2674  Schutkowski M, Reineke U, Reimer U (2005) Peptide arrays for kinase profiling. Chembiochem 6: 513-521  Schwartz JH (2001) The many dimensions of cAMP signaling. Proc Natl Acad Sci U S A 98: 13482–13484  Seet BT, Dikic I, Zhou MM, Pawson T (2006) Reading protein modifications with interaction domains. Nature Reviews Molecular Cell Biology 7: 473-483  Seger R, Ahn NG, Posada J, Munar ES, Jensen AM, Cooper JA, Cobb MH, Krebs EG (1992) Purification and characterization of mitogen-activated protein kinase activator(s) from epidermal growth factor-stimulated A431 cells. J Biol Chem 267: 14373-14381  Sharrocks AD, Yang SH, Galanis A (2000) Docking domains and substrate-specificity determination for MAP kinases. Trends Biochem Sci 25: 448-453  Shaw AS, Kornev AP, Hu J, Ahuja LG, Taylor SS (2014) Kinases and pseudokinases: lessons from RAF. Mol Cell Biol 34: 1538-1546  Shi H, Hugo W, Kong X, Hong A, Koya RC, Moriceau G, Chodon T, Guo R, Johnson DB, Dahlman KB, Kelley MC, Kefford RF, Chmielowski B, Glaspy JA, Sosman JA, van Baren N, Long GV, Ribas A, Lo RS (2014) Acquired resistance and clonal evolution in melanoma during BRAF inhibitor therapy. Cancer Discovery 4: 80-93  Siatecka M, Rozek M, Barciszewski J, Mirande M (1998) Modular evolution of the Glx-tRNA synthetase family--rooting of the evolutionary tree between the bacteria and archaea/eukarya branches. Eur J Biochem 256: 80-87  Skalhegg BS, Tasken K (2000) Specificity in the cAMP/PKA signaling pathway. Differential expression,regulation, and subcellular localization of subunits of PKA. Front Biosci 5: 678-693  Smith RF, King KY (1995) Identification of a eukaryotic-like protein kinase gene in Archaebacteria. Protein Sci 4: 126-129  Songyang Z (1999) Recognition and regulation of primary-sequence motifs by signaling modular domains. Progress in Biophysics and Molecular Biology 71: 359-372  Sonnhammer EL, Eddy SR, Durbin R (1997) Pfam: a comprehensive database of protein domain  172  families based on seed alignments. Proteins 28: 405-420  Speer R, Wulfkuhle JD, Liotta LA, Petricoin EF, 3rd (2005) Reverse-phase protein microarrays for tissue-based analysis. Current Opinion in Molecular Therapeutics 7: 240-245  Steinberg TH, Agnew BJ, Gee KR, Leung WY, Goodman T, Schulenberg B, Hendrickson J, Beechem JM, Haugland RP, Patton WF (2003) Global quantitative phosphoprotein analysis using Multiplexed Proteomics technology. Proteomics 3:1128-1144  Stock AM, Robinson VL, Goudreau PN (2000) Two-component signal transduction. Annu Rev Biochem 69: 183-215  Stover DR, Liebetanz J, Lydon NB (1994) Cdc2-mediated modulation of pp60c-src activity. J Biol Chem 269: 26885-26889  Sulzmaier FJ, Ramos JW (2013) RSK isoforms in cancer cell invasion and metastasis. Cancer Research 73: 6099-6105  Syvanen M (2012) Evolutionary implications of horizontal gene transfer. Annu Rev Genet 46: 341-358  Tang Z, Dai S, He Y, Doty RA, Shultz LD, Sampson SB, Dai C (2015) MEK guards proteome stability and inhibits tumor-suppressive amyloidogenesis via HSF1. Cell 160: 729-744  Tanoue T, Adachi M, Moriguchi T, Nishida E (2000) A conserved docking motif in MAP kinases common to substrates, activators and regulators. Nature Cell Biology 2: 110-116  Tanoue T, Nishida E (2003) Molecular recognitions in the MAP kinase cascades. Cellular Signalling 15: 455-462  Taylor SS, Buechler JA, Yonemoto W (1990) cAMP-dependent protein kinase: framework for a diverse family of regulatory enzymes. Annual Review of Biochemistry 59: 971-1005  Taylor SS, Kornev AP (2011) Protein kinases: evolution of dynamic regulatory proteins. Trends Biochem Sci 36: 65-77  Taylor SS, Radzio-Andzelm E (1994) Three protein kinase structures define a common motif. Structure 2: 345-355  Teixeira LK, Reed SI (2013) Ubiquitin Ligases and Cell Cycle Control. Annu Rev Biochem 82:  173  387-414  Tegge W, Frank R, Hofmann F, Dostmann WR (1995) Determination of cyclic nucleotide-dependent protein kinase substrate specificity by the use of peptide libraries on cellulose paper. Biochemistry 34: 10569-10577  Theobald DL (2010) A formal test of the theory of universal common ancestry. Nature 465: 219-222  Tinti M, Nardozza AP, Ferrari E, Sacco F, Corallino S, Castagnoli L, Cesareni G (2012) The 4G10, pY20 and p-TYR-100 antibody specificity: profiling by peptide microarrays. New Biotechnology 29: 571-577  Tohgo A, Pierce KL, Choy EW, Lefkowitz RJ, Luttrell LM (2002) beta-Arrestin scaffolding of the ERK cascade enhances cytosolic ERK activity but inhibits ERK-mediated transcription following angiotensin AT1a receptor stimulation. J Biol Chem 277: 9429-9436  Torii S, Kusakabe M, Yamamoto T, Maekawa M, Nishida E (2004) Sef is a spatial regulator for Ras/MAP kinase signaling. Developmental cell 7: 33-44  Tsigelny I, Greenberg JP, Cox S, Nichols WL, Taylor SS, Ten Eyck LF (1999) 600 ps molecular dynamics reveals stable substructures and flexible hinge points in cAMP dependent protein kinase. Biopolymers 50: 513-524  Tyagi N, Anamika K, Srinivasan N (2010) A framework for classification of prokaryotic protein kinases. PLoS One 5: e10608  Uttamchandani M, Chan EW, Chen GY, Yao SQ (2003) Combinatorial peptide microarrays for the rapid determination of kinase specificity. Bioorg Med Chem Lett 13: 2997-3000  Vacic V, Iakoucheva LM, Radivojac P (2006) Two Sample Logo: a graphical representation of the differences between two sets of sequence alignments. Bioinformatics 22: 1536-1537  Vakiani E, Solit DB (2011) KRAS and BRAF: drug targets and predictive biomarkers. The Journal of Pathology 223: 219-229  VanMeter A, Signore M, Pierobon M, Espina V, Liotta LA, Petricoin EF, 3rd (2007) Reverse-phase protein microarrays: application to biomarker discovery and translational medicine. Expert Review of Molecular Diagnostics 7: 625-633   174  Vigil D, Cherfils J, Rossman KL, Der CJ (2010) Ras superfamily GEFs and GAPs: validated and tractable targets for cancer therapy? Nature Reviews Cancer 10: 842-857  Villanueva J, Vultur A, Lee JT, Somasundaram R, Fukunaga-Kalabis M, Cipolla AK, Wubbenhorst B, Xu X, Gimotty PA, Kee D, Santiago-Walker AE, Letrero R, D'Andrea K, Pushparajan A, Hayden JE, Brown KD, Laquerre S, McArthur GA, Sosman JA, Nathanson KL, Herlyn M (2010) Acquired resistance to BRAF inhibitors mediated by a RAF kinase switch in melanoma can be overcome by cotargeting MEK and IGF-1R/PI3K. Cancer Cell 18: 683-695  Vomastek T, Schaeffer HJ, Tarcsafalvi A, Smolkin ME, Bissonette EA, Weber MJ (2004) Modular construction of a signaling scaffold: MORG1 interacts with components of the ERK cascade and links ERK signaling to specific agonists. Proc Natl Acad Sci U S A 101: 6981-6986  Vossler MR, Yao H, York RD, Pan MG, Rim CS, Stork PJ (1997) cAMP Activates MAP Kinase and Elk-1through a B-Raf- and Rap1-Dependent Pathway. Cell 89: 73-82  Walsh C (2000) Molecular mechanisms that confer antibacterial drug resistance. Nature 406: 775-781  Walsh DA, Perkins JP, Krebs EG (1968) An adenosine 3',5'-monophosphate-dependant protein kinase from rabbit skeletal muscle. J Biol Chem 243: 3763-3765  Wang S, Huang X, Sun D, Xin X, Pan Q, Peng S, Liang Z, Luo C, Yang Y, Jiang H, Huang M, Chai W, Ding J, Geng M (2012) Extensive crosstalk between O-GlcNAcylation and phosphorylation regulates Akt signaling. PLoS One 7: e37427  Wellbrock C, Karasarides M, Marais R (2004) The RAF proteins take centre stage. Nature reviews Molecular Cell Biology 5: 875-885  Wingren C, Borrebaeck CA (2008) Antibody microarray analysis of directly labelled complex proteomes. Curr Opin Biotechnol 19: 55-61  Winkler DF, Campbell WD (2008) The spot technique: synthesis and screening of peptide macroarrays on cellulose membranes. Methods Mol Biol 494: 47-70  Wortzel I, Seger R (2011) The ERK Cascade: Distinct Functions within Various Subcellular Organelles. Genes and Cancer 2: 195-209  Xie F, Liu T, Qian WJ, Petyuk VA, Smith RD (2011) Liquid chromatography-mass spectrometry-based quantitative proteomics. J Biol Chem 286: 25443-25449  175   Xue L, Wang P, Cao P, Zhu JK, Tao WA (2014) Identification of Extracellular Signal-regulated Kinase 1 (ERK1) Direct Substrates using Stable Isotope Labeled Kinase Assay-Linked Phosphoproteomics. Mol Cell Proteomics 13: 3199-3210  Xue Y, Gao X, Cao J, Liu Z, Jin C, Wen L, Yao X, Ren J (2010) A summary of computational resources for protein phosphorylation. Curr Protein Pept Sci 11: 485-496  Yao Y, Li W, Wu J, Germann UA, Su MS, Kuida K, Boucher DM (2003) Extracellular signal-regulated kinase 2 is necessary for mesoderm differentiation. Proc Natl Acad Sci U S A 100: 12759-12764  Ye Y, Godzik A (2003) Flexible structure alignment by chaining aligned fragment pairs allowing twists. Bioinformatics 19 Suppl 2: ii246-255  Yee KL, Weaver VM, Hammer DA (2008) Integrin-mediated signalling through the MAP-kinase pathway. IET Systems Biology 2: 8-15  Yoon S, Seger R (2006) The extracellular signal-regulated kinase: multiple substrates regulate diverse cellular functions. Growth Factors 24: 21-44  Yu JS, Chen WJ, Ni MH, Chan WH, Yang SD (1998) Identification of the regulatory autophosphorylation site of autophosphorylation-dependent protein kinase (auto-kinase). Evidence that auto-kinase belongs to a member of the p21-activated kinase family. Biochem J 334 ( Pt 1): 121-131  Yung Y, Dolginov Y, Yao Z, Rubinfeld H, Michael D, Hanoch T, Roubini E, Lando Z, Zharhary D, Seger R (1997) Detection of ERK activation by a novel monoclonal antibody. FEBS Lett 408: 292-296  Zerweck J, Masch A, Schutkowski M (2009) Peptide microarrays for profiling of modification state-specific antibodies. Methods Mol Biol 524: 169-180  Zhang H, Pelech S (2012) Using protein microarrays to study phosphorylation-mediated signal transduction. Semin Cell Dev Biol 23: 872-882  Zhang H, Shi X, Hampong M, Blanis L, Pelech S (2001) Stress-induced inhibition of ERK1 and ERK2 by direct interaction with p38 MAP kinase. J Biol Chem 276: 6905-6908  Zhang W, Zhang J, Xu C, Li N, Liu H, Ma J, Zhu Y, Xie H (2012) LFQuant: a label-free fast  176  quantitative analysis tool for high-resolution LC-MS/MS proteomics data. Proteomics 12: 3475-3484  Zhang Y, Dong C (2007) Regulatory mechanisms of mitogen-activated kinase signaling. Cellular and Molecular Life Sciences : CMLS 64: 2771-2789  Zheng CF, Guan KL (1994) Activation of MEK family kinases requires phosphorylation of two conserved Ser/Thr residues. The EMBO Journal 13: 1123-1131  Zheng J, Knighton DR, ten Eyck LF, Karlsson R, Xuong N, Taylor SS, Sowadski JM (1993) Crystal structure of the catalytic subunit of cAMP-dependent protein kinase complexed with MgATP and peptide inhibitor. Biochemistry 32: 2154-2161  Zhou H, Watts JD, Aebersold R (2001) A systematic approach to the analysis of protein phosphorylation. Nat Biotechnol 19: 375-378  Zuccotto F, Ardini E, Casale E, Angiolini M (2010) Through the "gatekeeper door": exploiting the active kinase conformation. Journal of Medicinal Chemistry 53: 2681-2694    177  Appendices Appendix 1. Amino acid frequency in the human STK alignment  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42A 0.51 1.53 1.27 0.76 4.58 1.78 0.25 2.29 1.78 1.02 21.63 0.00 12.98 4.07 2.80 2.29 4.58 36.64 2.04 1.53 3.56 8.65 3.05 2.04 2.80 0.25 2.80 3.31 87.02 5.09 0.00 3.05 1.53 2.80 0.00 4.58 3.56 6.87 7.89 5.34 18.58 2.80G 1.27 1.53 1.78 13.49 4.33 0.25 0.25 87.79 2.80 94.15 12.98 1.78 67.18 1.27 0.25 0.00 1.78 18.07 1.02 0.76 2.04 12.72 5.85 39.44 2.80 0.00 0.25 0.51 0.00 0.51 0.00 0.25 0.25 0.00 0.00 8.91 3.05 0.00 3.82 1.02 1.27 0.51F 29.01 1.02 6.87 2.80 0.25 2.29 0.25 0.25 0.51 0.25 1.53 55.47 0.51 0.76 0.25 13.74 1.53 0.00 3.56 2.04 0.51 0.51 0.51 0.00 1.78 0.00 4.58 10.18 0.00 0.25 0.00 3.56 8.40 2.04 0.51 1.78 3.56 2.04 1.53 5.34 7.38 6.11I 5.85 4.07 13.49 9.92 2.29 4.83 39.95 0.00 1.27 0.00 0.51 1.02 0.25 5.60 3.05 2.80 1.53 1.53 6.62 1.27 5.60 1.53 0.00 0.51 0.76 0.00 8.65 3.05 1.02 29.26 0.00 24.94 31.30 0.76 0.25 1.53 3.31 4.83 3.05 2.04 12.72 7.12L 9.92 5.60 22.39 28.24 2.29 11.20 44.27 0.51 1.78 0.76 0.51 0.51 0.25 2.80 2.80 5.60 26.72 1.02 5.60 4.83 6.11 6.36 3.05 1.02 3.56 0.25 16.03 4.07 0.51 17.30 0.25 5.09 26.21 3.82 0.25 3.05 4.33 7.89 0.76 6.87 17.81 17.05M 0.25 1.02 2.04 0.76 1.02 1.02 1.27 0.00 0.25 0.00 0.00 0.25 0.00 0.76 0.51 2.80 1.27 1.53 1.27 1.78 1.27 0.25 0.25 0.25 0.00 1.02 3.05 3.56 1.02 11.45 0.00 3.05 7.12 0.51 0.00 0.51 1.53 1.53 2.04 3.05 1.02 4.83P 0.51 0.51 3.82 0.76 1.53 5.34 0.51 0.51 1.02 0.51 1.27 0.76 0.76 0.76 0.51 0.00 0.25 0.25 0.00 1.02 1.27 2.80 0.25 2.29 0.51 0.00 5.09 0.00 0.00 0.25 0.00 0.25 0.25 9.41 1.27 1.27 2.54 4.83 4.33 2.54 0.51 1.27V 3.05 4.58 11.45 7.63 1.78 20.87 8.40 0.00 3.05 0.25 2.04 0.76 0.00 13.74 85.75 5.85 2.54 19.34 6.62 3.31 7.89 3.31 0.76 0.76 4.33 0.00 9.16 46.82 6.62 28.24 0.00 17.56 15.78 1.27 0.25 4.33 6.36 10.43 3.31 3.56 10.94 2.54W 5.09 0.00 0.51 0.51 0.00 0.51 0.00 0.00 1.02 0.00 0.00 0.25 0.00 0.00 0.00 5.09 0.25 0.25 2.04 5.09 0.25 0.25 1.02 0.51 0.00 0.25 0.76 0.76 0.51 1.78 0.00 0.00 0.00 0.25 0.00 0.51 0.00 1.78 1.78 0.51 1.27 1.78C 0.25 2.29 1.27 1.02 0.51 1.02 0.51 0.51 1.27 0.25 0.76 0.51 0.25 1.27 1.27 7.63 0.25 14.50 2.29 1.78 1.78 0.25 0.25 0.51 1.27 0.00 0.25 2.04 0.00 2.80 1.02 3.05 0.76 1.53 0.00 0.00 1.27 1.53 1.27 3.31 2.54 1.78N 0.76 2.54 2.54 0.25 1.78 1.02 0.25 1.02 2.80 0.51 5.09 1.53 0.25 1.78 0.25 0.51 0.25 0.51 0.51 8.91 3.05 4.83 5.34 5.09 3.56 0.25 2.54 0.00 0.00 0.76 0.25 0.00 0.51 11.70 0.00 3.05 4.07 2.29 3.05 5.34 2.04 1.53Q 1.02 8.14 2.04 3.56 7.12 2.80 0.76 0.51 5.85 0.00 3.31 1.78 0.51 6.36 0.25 2.54 3.56 0.00 7.38 2.04 2.80 5.09 2.29 5.34 12.98 0.76 3.56 0.25 0.25 0.00 0.00 5.85 0.51 4.33 0.00 5.60 8.14 4.33 12.47 10.69 1.78 8.14S 0.51 5.85 1.78 3.31 4.07 1.02 0.51 3.56 12.72 0.25 24.17 3.82 12.21 5.09 0.76 2.29 4.33 1.53 2.29 5.09 1.78 7.63 10.43 3.56 2.04 0.25 0.25 0.00 0.25 0.25 0.00 1.53 1.27 9.41 0.00 7.63 6.62 4.83 6.11 8.40 2.04 1.53T 0.51 7.38 4.83 1.27 2.04 10.69 0.51 0.25 4.07 0.51 13.49 2.80 0.25 9.92 0.25 1.02 2.29 1.27 6.11 1.02 4.83 5.60 43.00 3.05 4.07 0.25 2.54 0.51 1.53 0.51 0.00 1.53 0.51 1.02 0.00 4.58 4.07 4.83 2.04 6.87 9.41 0.76Y 38.68 0.76 1.53 3.05 0.76 1.27 0.25 0.00 0.25 0.00 0.25 25.19 0.76 0.76 0.25 30.53 1.27 0.76 5.09 2.04 0.51 1.78 0.51 0.00 0.25 0.00 4.33 22.39 0.00 0.25 0.00 0.25 0.76 1.53 0.25 0.51 4.58 1.27 1.02 0.76 0.76 4.07D 0.76 9.92 3.82 5.60 6.87 1.78 0.00 0.25 2.29 0.76 0.25 0.76 0.76 2.04 0.51 0.51 0.25 0.00 1.27 18.58 1.02 7.63 7.89 4.58 4.58 0.00 3.82 0.51 0.76 0.00 0.00 0.51 0.25 12.98 0.00 9.41 13.23 3.56 3.82 5.09 1.27 1.27E 1.02 27.23 5.09 4.83 24.17 10.69 0.00 1.27 15.52 0.51 2.54 1.27 0.51 15.27 0.25 1.02 6.11 0.00 5.34 7.38 2.29 10.18 5.60 3.05 15.78 0.25 15.27 0.00 0.00 0.00 0.00 5.60 3.31 4.07 0.25 11.20 12.21 9.92 18.83 4.58 3.56 12.21H 0.00 2.54 1.53 2.29 1.02 0.76 0.25 0.00 2.54 0.00 2.29 0.76 0.25 1.27 0.00 2.04 0.51 0.25 1.02 18.32 6.62 3.56 0.25 2.54 3.05 0.25 1.78 0.76 0.00 0.25 0.00 1.53 0.51 1.27 0.00 2.04 2.54 3.31 2.29 6.87 0.76 2.54K 0.51 6.87 6.87 5.60 18.32 15.27 1.27 0.51 22.14 0.00 2.29 0.51 1.53 18.07 0.00 8.14 26.97 0.25 12.72 7.89 27.99 10.94 4.83 9.16 12.21 0.25 8.65 0.25 0.00 0.25 97.71 15.27 0.51 18.58 0.25 14.76 7.12 6.87 10.94 9.16 1.27 10.43R 0.51 6.62 5.09 4.33 15.27 5.60 0.51 0.76 17.05 0.25 5.09 0.25 0.76 6.62 0.25 5.34 13.74 2.29 27.23 5.34 18.83 6.11 3.31 5.85 8.65 0.51 6.36 0.51 0.25 0.76 0.76 7.12 0.25 12.72 0.51 9.92 4.58 15.27 8.40 7.89 2.80 11.45Space 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.78 0.00 0.25 0.00 0.00 0.00 0.00 0.00 0.00 1.53 10.43 15.01 87.53 0.25 0.51 0.25 0.00 0.00 0.00 0.00 0.00 7.63 4.83 3.31 1.78 1.27 0.76 0.25 0.25Insert 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.89 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 88.55 0.00 0.00 0.00 0.00 0.00 0.00 0.00Highest 38.68 27.23 22.39 28.24 24.17 20.87 44.27 87.79 22.14 94.15 24.17 55.47 67.18 18.07 85.75 30.53 26.97 36.64 27.23 18.58 27.99 12.72 43.00 39.44 15.78 1.02 16.03 46.82 87.02 29.26 97.71 24.94 31.30 18.58 1.27 14.76 13.23 15.27 18.83 10.69 18.58 17.05Y E L L E V L G K G S F G K V Y K A R H K X T G X X L V A I K I I K X K D R E Q L L43 44 45 4 47 48 49 5 51 52 53 54 55 5 7 58 59 60 61 62 63 64 6 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 4A 6 62 76 7 1 7 89 5 09 0 00 7 89 5 34 5 34 1 3 0 51 4 07 25 25 2 29 7 12 76 2 54 25 2 54 2 9 3 31 4 07 1 02 3 82 76 0 76 1 78 11 2 5 8 3 31 2 54 2 04 2 29 4 83 2 54 3 6 1 02 3.56 6.87 2. 4 3.31G 0 76 0 25 0 00 1 78 0 51 25 2 04 7 12 4 58 6 11 25 1 27 0 6 0 00 9 16 0 1 3 05 0 00 12 98 2 54 0 25 1 8 51 0 25 25 0 7 0 25 1 78 6 11 48 6 66 92 0 6 0 51 1 02 1 5 0 51 00 0.76 . 4.07 21.63F 1 78 1 78 25 3 5 1 3 25 2 29 6 11 25 1 53 1 02 13 23 51 51 1 53 15 78 8 91 0 0 0 25 4 07 5 60 8 65 9 4 5 85 15 01 00 15 27 1 78 0 51 25 51 25 1 53 2 1 5 09 6 36 3 05 0.00 2.29 3.56 .I 3 56 0 00 30 03 27 34 10 53 1 78 1 3 4 58 51 1 8 25 00 52 16 20 10 0 51 3 82 8 4 0 00 02 02 1 69 3 31 22 39 16 03 2 29 1 2 3 82 0 51 27 1 2 51 3 82 51 9 41 28 0 3.31 0.76 . .27L 4 07 0 00 1 21 6 87 17 05 38 4 7 38 10 69 38 68 1 3 0 25 0 76 2 04 13 49 11 96 3 3 48 09 12 72 00 2 04 2 4 4 27 3 82 46 82 6 62 18 32 0 00 25 19 17 56 3 05 2 54 1 27 1 78 7 5 12 98 1 78 22 90 41 73 5.8 3. 1 . 3.31M 2 4 0 00 29 1 3 4 58 1 98 1 02 4 07 3 0 0 00 0 25 25 1 02 76 1 27 4 33 2 29 0 00 0 51 1 78 3 31 1 02 9 92 0 76 4 1 0 00 1 78 17 56 51 76 0 51 76 5 09 6 11 1 78 1 53 58 1.78 1.78 3.31 0.25P 25 0 2 0 6 0 5 0 76 0 25 1 3 0 25 1 8 51 1 02 57 00 0 5 0 25 25 04 0 25 0 25 0 00 4 83 1 78 00 0 1 0 0 2 04 51 51 9 92 1 78 3 05 78 0 00 0 51 0 00 0 76 0.5 1.27 3. 5 4.8V 3 82 0 00 18 32 3 82 16 5 51 4 8 2 9 11 70 25 1 27 54 76 21 37 48 09 4 83 4 83 0 00 53 0 76 1 27 6 87 5 34 59 29 2 04 0 25 1 27 17 05 1 78 76 0 25 7 63 1 27 1 27 9 6 6 7 6.11 3.05 3.56 5.09W 0 00 00 0 00 0 51 00 0 00 00 51 00 0 00 3 05 25 1 27 0 00 1 02 0 0 0 76 0 25 11 45 25 0 00 2 00 00 00 0 00 1 7 00 1 7 0 51 0. 0 .00 .51 0.76C 1 27 0 00 0 51 1 02 1 02 1 53 3 82 4 83 2 54 1 02 1 8 0 76 76 1 27 1 02 2 29 0 00 0 25 3 5 1 27 13 23 0 00 2 80 76 0 00 1 02 18 83 76 1 2 25 76 1 02 1 78 0 25 0 76 0 7 1.27 1.78 3.82 .02N 10 94 0 00 1 27 5 5 2 04 25 3 5 6 36 0 76 11 96 4 8 0 5 53 18 1 51 6 11 0 76 51 0 00 8 14 6 62 2 29 2 04 0 0 0 25 1 02 0 5 76 25 8 91 5 9 0 76 6 1 0 51 3 56 3 05 0 00 0 25 5.34 8.40 6.87 4.58Q 6 62 0 00 78 10 43 2 80 7 38 7 63 1 3 12 21 0 76 2 29 1 02 0 51 11 20 25 2 04 0 25 2 80 4 83 0 1 0 00 1 53 1 78 2 29 25 1 53 3 5 2 04 1 3 4 33 7 63 0 1 27 9.92 7.38 8.40 .0S 3 82 0 51 1 78 8 4 1 78 0 5 7 12 5 60 5 60 6 11 6 36 2 29 2 04 2 4 1 2 5 3 1 02 1 7 0 00 63 4 83 51 3 31 0 7 0 5 1 7 1 27 1 02 76 12 98 3 82 4 58 23 66 1 02 4 83 7 38 1 27 1 02 9. 2 8.91 5.34 4.07T 5 85 0 00 0 76 3 1 1 02 2 04 1 3 4 07 2 54 2 9 0 0 1 02 1 0 7 38 8 65 0 6 1 02 0 00 9 41 4 1 53 82 1 02 0 76 12 2 2 04 00 76 5 09 1 02 5 34 6 11 0 51 4 33 7 1 0 25 0 51 4.07 3.56 5.6 4.33Y 1 2 1 78 3 31 51 51 25 2 8 76 51 25 76 4 83 0 6 1 27 2 9 13 74 2 1 51 7 38 3 82 30 79 25 0 25 1 27 51 39 44 1 78 25 1 53 1 2 10 18 3 2 12 98 3 31 2.04 1.53 4.33 1.02D 1 53 51 8 4 76 51 4 7 1 53 11 45 3 31 3 56 1 78 25 76 1 2 51 76 15 27 6 87 25 51 25 0 25 13 74 1 2 6 87 3 82 76 3 79 1 2 3 6 29 77 1 78 76 4.58 5. 9 3. 5 4.33E 6 87 9 0 1 02 7 12 0 00 00 6 11 8 65 1 78 5 8 0 25 10 3 1 02 1 2 0 51 4 3 0 25 3 05 0 00 7 3 8 40 0 51 0 76 0 00 0 51 0 0 73 79 0 76 1 11 9 4 58 1 27 19 59 0 1 5 34 1 45 2 04 1 78 7. 9 9.92 7.3 .63H 1.53 0.00 0.00 1.27 1.53 1.53 2.04 3.05 0.51 5.60 72.52 0.76 8.65 0.51 0.51 2.04 0.51 12.98 0.00 2.54 12.98 1.27 2.80 0.25 0.00 0.51 1.02 2.29 2.80 1.53 3.31 2.04 0.76 0.00 2.54 3.31 10.94 0.25 5.60 2.80 5.85 2.29K 8.65 0.00 8.65 12.98 3.56 0.00 22.65 10.94 1.53 17.30 0.25 6.62 0.76 0.25 1.78 16.03 0.00 6.87 0.00 9.41 13.74 1.02 0.76 0.51 0.00 0.25 0.51 2.29 0.25 3.31 3.82 1.02 0.00 0.00 7.12 7.38 2.80 1.78 15.78 13.23 8.40 11.96R 28.75 0.25 9.41 12.98 2.29 0.00 17.05 8.65 2.29 15.27 0.76 4.83 3.05 0.51 0.51 13.74 1.53 3.31 0.00 8.40 8.65 1.53 0.51 1.02 0.00 0.51 0.00 2.04 0.00 4.58 4.83 1.53 1.78 0.25 3.05 3.31 8.65 0.25 10.43 11.96 12.47 9.67Space 0.25 0.25 0.51 0.51 1.02 8.14 4.33 0.76 0.00 0.00 0.00 0.00 0.00 0.76 0.76 0.76 1.27 1.78 2.29 1.78 1.02 0.76 0.25 0.25 0.25 0.25 0.51 0.25 1.02 13.99 4.33 4.33 0.51 0.51 1.27 1.02 0.76 1.02 1.27 2.54 3.31 4.83Insert 0.00 0.00 0.00 0.00 0.00 29.26 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 97.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.51 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Highest 28.75 97.20 30.03 12.98 34.10 38.42 22.65 10.94 38.68 17.30 72.52 57.00 53.18 52.16 48.09 16.03 48.09 20.10 0.25 15.27 13.74 44.27 30.79 46.82 59.29 40.71 73.79 39.44 18.83 12.98 48.60 66.92 30.79 71.50 20.10 29.77 22.90 41.73 15.78 13.23 12.47 21.63R E I K I L K X L K H P N I V K L Y X X K L Y L V M E Y L S G G D L F D L L K K R G 178    85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126A 4.58 0.00 0.25 1.27 2.54 7.63 2.04 26.72 10.18 6.62 0.25 18.32 15.01 3.31 2.04 16.79 1.53 43.51 0.25 13.23 2.54 0.25 1.02 5.60 4.33 2.80 0.00 0.00 6.11 0.00 1.78 0.25 0.25 0.00 6.87 7.63 0.51 0.25 1.53 0.25 4.33 5.60G 6.62 1.02 0.25 1.02 0.76 3.31 0.51 0.76 2.54 1.27 1.02 4.33 2.04 3.82 0.00 2.04 1.78 44.02 0.00 3.56 0.00 1.53 0.00 2.80 2.80 32.06 0.76 0.25 0.51 0.00 5.34 0.00 0.25 0.00 5.09 6.11 0.25 0.25 0.00 0.00 6.11 5.09F 4.07 0.00 26.97 0.00 0.25 1.27 2.04 0.76 1.02 16.03 11.70 13.99 4.07 0.00 0.25 1.27 0.00 0.00 1.02 1.27 19.08 0.00 0.25 1.53 0.51 1.27 0.00 4.83 1.27 0.51 1.02 0.00 0.51 0.25 0.00 0.00 0.00 4.33 9.67 3.56 1.02 0.51I 1.78 1.02 6.36 0.76 2.04 0.76 3.82 20.10 4.58 1.78 19.08 9.41 2.80 0.00 40.46 6.87 2.04 0.76 12.98 2.04 0.25 10.69 0.00 0.51 1.78 0.51 0.51 55.47 38.68 0.00 0.00 0.00 25.19 0.00 0.00 0.25 0.00 46.56 2.54 18.32 0.76 0.51L 3.05 1.27 41.48 2.29 8.91 3.05 8.91 7.63 10.18 6.62 13.23 22.14 10.18 1.27 23.16 39.19 14.50 1.02 53.44 2.80 1.53 65.90 0.25 2.80 11.70 1.53 0.00 9.41 14.76 0.76 4.58 0.25 61.32 0.51 8.65 0.25 0.00 24.68 74.30 50.64 2.54 3.56M 1.27 0.00 7.63 1.02 3.05 1.78 4.07 1.53 1.02 2.04 3.82 9.41 1.27 1.27 7.12 1.53 1.27 1.02 8.91 1.27 0.51 3.82 0.00 0.25 5.34 1.27 0.00 1.27 2.54 0.00 0.25 0.00 1.53 0.00 0.51 0.00 0.00 2.04 6.11 3.05 1.78 1.53P 6.11 3.05 1.53 22.90 5.60 9.67 0.25 1.02 0.00 0.76 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.51 0.51 0.25 0.00 0.00 0.25 0.00 0.00 0.00 0.00 57.76 0.76 0.00 0.00 0.00 0.25 0.00 2.29V 5.09 0.25 5.09 0.25 2.54 3.05 6.62 20.36 6.36 4.83 11.45 5.09 4.07 0.25 16.28 17.56 1.53 1.78 19.85 1.02 1.27 3.56 1.02 0.25 1.02 0.76 0.00 19.85 30.53 0.00 0.00 0.00 9.92 0.00 1.02 0.25 0.25 18.32 5.09 14.50 5.85 1.78W 0.00 0.00 0.00 0.00 6.11 0.25 1.53 0.51 1.27 1.53 5.60 0.00 1.53 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.25 0.00 0.25 0.00 0.00 1.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.25 0.00 0.00 0.00 0.00 0.25 0.51 0.00C 1.53 0.00 0.76 0.51 0.51 0.25 1.53 1.02 3.05 0.51 3.05 5.60 2.04 2.29 1.02 6.11 3.31 1.02 0.25 0.76 0.25 12.21 0.00 1.27 2.80 2.04 0.00 1.27 1.02 0.51 2.80 0.00 0.00 0.76 2.54 0.25 0.51 2.29 0.00 2.04 1.78 0.25N 3.82 0.76 0.00 4.33 0.76 4.33 1.27 0.76 0.00 3.31 0.00 0.00 0.76 1.27 0.00 0.00 2.29 0.00 0.25 6.62 0.25 0.25 0.76 6.62 13.23 11.70 0.25 0.00 0.00 0.00 0.51 2.80 0.00 1.27 0.25 2.80 94.66 0.00 0.00 0.00 9.67 7.12Q 2.29 0.51 0.25 2.29 1.02 3.31 10.43 0.25 5.34 3.05 0.00 0.25 8.65 48.85 0.00 1.27 9.16 0.00 0.00 7.38 0.51 0.00 1.02 10.69 7.38 3.31 0.51 0.51 0.00 0.00 0.25 0.00 0.51 0.25 0.25 7.38 0.00 0.00 0.00 0.00 1.78 3.56S 5.85 0.25 0.76 20.87 2.04 7.89 1.53 3.82 8.91 9.67 1.27 4.07 2.54 4.33 0.25 3.82 15.78 2.80 0.25 5.34 2.54 0.25 0.25 31.30 2.29 3.31 0.76 0.25 1.27 0.00 0.51 1.53 0.00 1.53 13.74 13.23 1.02 0.25 0.00 0.25 8.40 15.27T 4.07 1.53 0.76 15.52 0.76 2.54 9.41 7.89 2.54 4.83 0.00 5.85 3.05 1.78 6.87 3.56 3.31 2.29 0.51 1.02 0.76 0.00 1.78 4.58 0.00 1.53 0.25 0.51 2.54 0.00 1.78 0.00 0.00 2.29 2.04 0.51 0.51 0.25 0.00 0.00 12.98 5.85Y 4.33 0.00 5.09 0.51 0.25 1.02 1.78 0.00 0.51 6.36 26.46 0.00 11.20 0.00 0.00 0.00 0.51 0.00 0.00 2.04 51.40 0.00 0.51 0.25 1.53 1.02 0.00 3.31 0.00 11.45 0.51 0.00 0.00 0.00 0.00 0.51 0.00 0.00 0.00 5.09 2.04 1.02D 3.31 0.25 0.76 11.45 2.29 12.21 7.89 0.00 0.25 4.07 0.00 0.00 1.02 7.63 0.00 0.00 4.83 0.00 0.25 14.25 0.76 0.00 0.00 6.62 1.53 4.07 0.00 0.00 0.00 0.00 0.25 93.38 0.00 0.00 1.02 15.01 0.00 0.00 0.00 0.00 29.26 11.70E 3.31 0.51 0.25 4.07 57.00 13.74 20.10 0.25 0.25 2.29 0.00 0.25 1.78 21.37 0.25 0.00 11.70 0.00 2.04 15.52 0.76 0.76 0.00 9.67 5.34 4.07 0.00 0.25 0.25 0.00 0.00 0.25 0.00 0.00 0.00 35.88 0.00 0.00 0.00 0.00 1.02 15.78H 3.31 1.02 0.25 1.27 1.27 4.83 6.36 0.00 0.76 4.58 0.25 0.51 4.33 0.51 1.78 0.00 2.80 0.00 0.00 6.11 15.27 0.25 89.57 2.29 10.94 5.60 0.25 0.25 0.00 86.26 0.51 0.25 0.25 1.02 0.00 2.29 0.51 0.25 0.25 0.25 1.53 5.34K 11.45 2.04 0.25 7.38 0.51 10.69 2.54 2.29 13.99 9.92 2.29 0.51 6.36 0.25 0.25 0.00 11.70 0.00 0.00 8.40 0.76 0.25 1.78 7.63 17.81 13.49 1.53 2.29 0.00 0.00 1.02 0.25 0.00 91.35 0.00 3.82 0.76 0.00 0.00 0.00 6.87 6.62R 10.43 2.04 1.02 2.04 1.53 8.40 7.38 4.07 26.97 9.67 0.00 0.00 17.05 1.27 0.25 0.00 11.96 1.78 0.00 7.38 1.27 0.25 1.53 5.34 9.16 8.14 0.51 0.00 0.25 0.00 78.63 0.76 0.00 0.76 0.00 2.80 0.51 0.00 0.00 1.02 1.27 5.85Space 13.74 53.94 0.25 0.25 0.25 0.00 0.00 0.25 0.25 0.25 0.25 0.25 0.25 0.51 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 83.21 0.25 0.25 0.25 0.25 0.25 0.25 0.00 0.00 0.25 0.51 0.51 0.51 0.51 0.51 0.76Insert 0.00 30.53 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 11.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Highest 11.45 3.05 41.48 22.90 57.00 13.74 20.10 26.72 26.97 16.03 26.46 22.14 17.05 48.85 40.46 39.19 15.78 44.02 53.44 15.52 51.40 65.90 89.57 31.30 17.81 32.06 1.53 55.47 38.68 86.26 78.63 93.38 61.32 91.35 57.76 35.88 94.66 46.56 74.30 50.64 29.26 15.78K X L S E E E A R F Y L R Q I L X G L E Y L H S K G X I I H R D L K P E N I L L D E1 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 1 5 156 157 158 1 9 16 16 162 163 16 165 166 167 168A 3.56 0. 0 4.8 4.07 .85 2.04 .00 27.99 . .00 .76 1.78 52.16 7.89 4.33 4. 7 4.58 6.36 5. 9 3. 5 2.29 1 .18 2.54 2.8 4.33 6.87 4.83 .51 5.85 8.91 .25 8.65 7 .99 .25 0.00 3.82 2.54 5.6 4.83 6.36 .76 4.58G 4.83 1.02 43.77 1. 7 0.76 0.2 0.25 11.9 0.7 0.25 9 .09 0. 0 3. 5 3.56 2.04 4. 8 8.14 8.91 3.99 1 .45 0.00 0.7 2.29 7.12 . 1. 8 7 .74 0.2 .25 6.87 . . 1.0 . 0. 0 .00 .76 1.53 30.28 3.82 1.78 20.10F .51 .25 3.56 .51 0.76 . 0.25 .00 0.00 82.19 0.0 14.76 .02 1. 2 5. 9 8.4 .76 1.78 1.27 2.8 1.78 2.54 2.04 .25 19.85 2.29 1.78 . 1. 2 4.33 1 .43 .51 . . .00 .51 6.36 1.53 .51 1. 7 .25 2.29I 0. 1 .25 2.04 3.3 31.04 .51 4 .7 9.9 0. 0 .00 0.2 4.33 0. 0 1.78 7.38 8. 4 5.60 .53 2. 9 2.04 . .54 1.53 0.51 4.58 9.92 1. 2 0. 5 4.58 1.02 0.51 8.65 .25 2.54 0.00 18.32 28. 0 1. 7 .25 3.31 0.00 1.02L 2.29 .25 4.83 3.82 2 .3 0.51 45.55 2.54 . 9.92 .76 45.55 .51 2.80 6.62 27.23 4.07 3.82 3.56 4.83 2.80 15. 7 3.05 1.27 2.47 4.07 .00 1.78 12.47 4. 7 . 13.23 .76 0.76 .00 16.03 39.44 2 .87 1.78 5.34 0.51 1.78M 1.02 .00 4.58 1.02 0. 0 1.53 1.02 . 0.00 0.51 .51 3.82 0.2 0.76 3.56 5.60 2.29 2.29 .51 4.07 1. 3 6.11 2.80 .76 2.80 1.02 .00 .25 . 1. 2 .00 33.84 .00 . .00 5. 4 2.29 3. 5 1. 2 3.82 .02 2.04P 8.6 .00 3.31 1.78 1.27 0.25 0.00 0.00 0.00 0.25 0.51 0.51 0.00 0.00 1.78 0.25 .09 9.67 9.92 3.82 .0 1.78 1.53 4. 3 5.0 2.04 .25 .76 43.7 5. 9 .00 0.25 8.14 92.62 0.25 0.00 0.25 . 2.54 1.53 4.83 3.74V 0.51 0.00 . 8.14 3 .03 5.85 . 5.60 0.00 1.02 .25 7.38 0.76 6.62 5.60 8.40 5.60 3.31 1.27 6.6 .51 3. 5 .53 3.05 9.41 29.01 6.87 .00 4.33 1.27 0.00 8.65 . 1.02 0. 0 45.04 9.4 3.31 3. 5 2.8 1.02 1.27W 0.51 . 0. 0 . .76 0. 0 0.00 . .00 0.51 0.25 .76 0. 5 . 1.27 1. 2 .00 1.27 1. 2 0.51 0.0 1.27 2.04 .00 0.76 0.25 . 0.5 . 0. 3 15.0 . . . . .76 .76 .25 . 1. 2 0. 0 0. 0C 0.76 .00 3.5 .53 4.58 1.78 0.00 1 .18 .00 0. 0 1. 2 1.53 10.69 3.56 1. 2 1.53 1.78 .51 .00 1.78 .00 2.29 .25 1.0 0.76 25.19 .76 1.78 .25 1.78 . 1.27 1.27 .51 0. 0 0.76 1.78 .25 2.8 1.27 0.00 0.25N 11.20 .25 3.31 7.89 0. 0.00 .0 .00 .51 0. 5 .76 1. 7 .02 4. 3 1.02 2.54 5.34 3.82 6.87 .60 .25 2.29 6.87 3.82 1.27 0.76 .25 2.04 .51 4.83 . .00 .25 . 2.29 0.51 . 8.65 5.85 4. 7 0.51 2.29Q 5.85 0.51 2.04 6.8 0.00 2.29 0.25 .25 .00 0.00 .00 .76 0.25 2.04 12. 2 1.27 4.07 5. 9 2.54 .36 0.76 1.02 2.29 4.58 4.07 0.76 3.56 .00 1.27 3.05 .00 .76 .00 0.51 .25 1.78 .25 9.41 7.63 9.92 0.76 3.82S 4.07 1. 2 5.34 5. 4 .00 .00 0.00 9.16 0.51 0.51 1.02 9.4 25.45 5.8 2. 9 3.82 .96 1 .43 1 .18 7.38 . 4.07 12.47 18.58 5.34 1.02 .78 13.9 1. 2 3. 5 0.25 1.53 16. 3 0.2 .25 0.25 .51 7.12 7.38 1 .69 1.02 4.83T 3.82 1.78 2.80 8.91 .02 0.76 0.51 17.56 0.00 0.51 .25 2.29 1. 2 9.16 2.54 2. 4 7.89 5.60 8.14 . .02 7.63 17.05 33.59 1.78 3.82 1.27 73. 3 .76 2. 4 0.51 3.05 .25 .25 . 1.53 .00 4. 7 1. 2 5.34 3.56 3.31Y 1. 8 0.00 1. 7 1.53 1.53 1.53 . . . 3.31 .00 1.02 .51 .76 3. 5 6.62 .76 .76 1.27 1. 2 .25 7.38 5. 9 .25 9.92 2. 4 2. 9 .51 2.04 12. 1 69. 1 .76 .00 . .51 1. 7 2.29 .25 3.3 1.78 .00 7.63D 15.27 1. 3. 5 3.56 . . . . 95.42 . 1. 2 .51 1. 2 .25 2.29 2. 4 7.38 12.47 9.67 4.33 .25 2. 4 4.33 4.33 2.54 1.27 0.51 1.53 1. 2 5.34 .25 .25 .25 .25 3.31 .51 .25 4.33 4.33 4.58 . 4.33E 0.69 0.51 3.82 8.14 0. 0.00 .0 0.00 1.2 0. 0 0.51 1.02 0.2 1.78 15.27 3.05 6.62 6.11 8.1 0.43 2.29 2.8 3.82 5.85 3. 2 1. 2 .02 0.51 2.29 15.01 0.51 0.00 0.00 0.51 92.62 0.25 0.25 10.43 8.14 1 .43 1. 7 6.87H 3.82 0.25 1.53 18.58 0.51 2.04 0.00 0.25 0.51 0.25 0.25 1.78 0.25 1.27 4.07 2.04 4.83 2.80 4.58 3.31 0.25 2.04 3.82 2.54 0.76 1.27 0.00 0.25 0.51 2.80 1.27 0.00 0.51 0.00 0.00 0.51 3.56 1.27 0.51 2.29 0.76 3.05K 7.63 0.25 3.05 5.60 0.25 71.76 0.25 0.00 0.25 0.00 0.25 0.51 0.51 23.92 9.67 4.58 6.87 7.12 4.58 5.34 3.31 5.09 15.52 2.04 2.54 1.78 0.76 0.76 4.33 0.25 0.00 0.76 0.00 0.25 0.00 0.00 0.25 5.85 4.33 12.72 3.31 4.33R 9.41 0.76 1.78 6.87 0.76 8.14 0.00 0.00 0.25 0.25 0.00 0.51 0.25 21.88 7.63 1.78 5.09 5.34 3.56 4.58 1.27 10.18 6.87 1.78 5.60 2.54 1.27 0.51 13.23 5.85 0.51 17.56 0.25 0.00 0.51 2.80 0.25 10.43 9.67 6.11 2.54 4.58Space 3.31 65.39 0.76 0.76 0.51 0.51 0.76 0.76 0.25 0.25 0.51 0.51 0.76 0.76 0.76 1.02 1.27 1.02 1.53 3.31 18.07 9.67 2.29 1.53 1.27 1.27 1.02 0.76 0.51 0.76 0.51 0.25 0.00 0.00 0.00 0.00 0.25 0.25 0.76 1.53 52.16 7.89Insert 0.00 25.95 0.00 0.00 0.00 0.00 0.00 3.31 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 60.31 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 23.92 0.00Highest 15.27 1.78 43.77 18.58 31.04 71.76 45.55 27.99 95.42 82.19 91.09 45.55 52.16 23.92 15.27 27.23 11.96 12.47 13.99 11.45 3.31 15.27 17.05 33.59 19.85 29.01 70.74 73.03 43.77 15.01 69.21 33.84 70.99 92.62 92.62 45.04 39.44 20.87 30.28 12.72 4.83 20.10D X G H I K L A D F G L A K Q L S X X X X L T T F V G T P E Y M A P E V L L G K X G 179    Frequency (%) of the 20 amino acids at each position in the human STK alignment was calculated and presented in this table. The amino acid with highest frequency was used to generate the STK consensus sequence.  169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210A 3.05 2.29 3.56 20.10 17.81 0.51 0.76 0.00 24.43 4.07 4.58 2.29 0.51 10.18 9.67 3.05 1.53 18.58 8.91 2.29 4.07 2.54 3.82 1.27 5.34 9.67 4.83 1.53 6.11 6.11 7.12 8.14 4.33 5.60 2.29 2.04 7.38 0.00 3.56 3.82 4.07 4.07G 1.27 27.23 1.27 2.80 1.78 0.00 0.25 0.00 4.33 0.00 92.37 0.00 0.00 0.00 2.54 0.25 0.00 1.02 2.54 66.16 2.29 1.78 0.00 1.27 3.82 30.53 4.33 4.58 4.58 1.53 2.04 0.51 1.53 4.83 3.82 0.76 3.05 0.00 5.85 8.14 1.78 2.29F 4.07 1.78 6.36 0.51 0.51 0.00 0.51 7.63 0.00 12.72 0.51 0.00 0.76 8.91 10.94 0.51 4.33 4.58 2.54 0.51 4.58 3.31 1.02 60.05 7.89 0.00 1.78 1.53 0.51 1.78 2.04 1.27 9.92 8.91 4.33 1.53 4.58 0.00 3.05 8.91 4.33 4.83I 4.33 0.00 2.54 0.00 8.65 0.00 29.52 0.00 0.51 8.40 0.00 19.59 35.37 9.41 10.43 7.38 6.11 5.34 1.78 0.76 2.29 4.07 3.56 1.53 0.76 2.80 4.07 0.76 2.80 7.38 3.05 5.09 9.41 3.56 4.07 43.00 7.63 0.00 6.87 2.29 2.29 4.07L 2.54 1.53 4.58 2.29 3.31 0.00 10.43 0.25 0.00 45.04 0.00 5.60 20.10 43.51 4.58 0.76 39.69 45.55 3.82 5.85 3.31 11.20 12.47 6.36 5.85 0.51 7.12 4.07 6.87 9.41 5.85 20.87 23.41 10.43 12.21 5.34 16.54 0.25 4.07 9.16 10.69 14.25M 0.51 0.25 0.51 0.51 0.25 0.00 15.78 0.25 0.00 5.34 0.25 2.80 6.36 13.74 2.29 1.78 37.66 3.31 0.76 0.25 1.02 2.54 0.51 1.02 1.27 0.25 2.54 0.25 1.53 7.89 0.76 3.31 4.33 3.31 4.58 4.33 8.40 0.00 3.82 1.02 1.78 1.78P 1.78 1.27 11.20 9.16 0.76 0.00 0.00 0.25 0.25 0.00 0.51 0.00 0.00 0.25 0.00 0.00 0.00 0.00 0.00 0.76 1.27 34.61 68.70 1.78 12.47 2.80 4.58 2.80 14.76 5.60 3.05 2.29 3.82 1.78 0.76 0.51 3.05 0.00 11.20 14.50 11.20 12.72V 3.56 0.76 3.31 1.27 34.86 0.00 22.39 0.00 0.25 13.49 0.00 35.62 18.07 4.83 2.80 4.07 4.58 10.18 8.91 0.25 3.05 4.33 2.29 1.78 2.04 1.53 4.33 1.53 4.07 5.60 3.05 5.60 3.31 2.80 1.53 12.72 11.20 0.00 4.07 4.83 4.33 7.89W 1.27 0.00 0.51 0.25 0.00 0.00 14.50 77.35 0.00 0.25 0.00 0.00 0.00 0.51 8.91 0.00 0.76 0.25 0.25 0.00 1.53 0.51 0.25 6.36 1.53 0.00 0.76 0.25 0.25 1.02 0.00 0.25 1.53 2.04 0.00 0.25 0.25 0.00 0.00 1.53 2.04 1.27C 0.76 1.27 1.02 0.51 9.67 0.00 1.27 0.00 1.78 3.56 0.25 26.21 3.82 1.78 1.02 1.78 2.04 2.54 8.40 0.51 5.34 0.76 0.51 0.00 2.04 1.02 2.29 0.76 1.27 0.00 0.76 1.27 1.27 0.25 2.04 0.25 2.29 0.00 0.76 1.02 0.25 1.27N 0.51 9.92 2.29 0.76 0.00 0.00 0.00 0.00 0.25 0.00 0.00 0.00 0.51 0.00 0.25 0.51 0.25 1.53 1.53 1.27 1.78 2.29 0.00 1.27 1.78 6.62 5.34 13.23 4.83 3.56 3.56 0.76 1.53 5.34 6.36 1.78 1.53 0.25 4.33 2.54 4.33 2.04Q 3.56 0.25 6.62 4.33 0.51 0.25 0.00 0.25 0.51 0.00 0.51 0.25 0.00 0.00 1.78 2.04 0.00 0.76 2.04 1.27 9.92 1.02 0.00 1.02 7.63 2.04 3.82 4.58 4.07 6.62 8.91 16.28 4.33 6.62 7.63 3.05 3.05 0.51 2.80 2.29 1.78 2.54S 3.56 19.59 4.07 13.49 12.47 0.25 0.76 0.00 65.65 0.51 0.51 0.51 1.27 1.02 0.51 1.27 1.02 0.51 12.47 2.04 8.65 9.41 1.27 0.25 5.34 6.36 5.60 15.78 5.34 6.11 2.29 2.29 2.04 3.82 4.07 2.29 5.34 0.76 7.12 6.87 6.11 9.67T 0.51 12.47 8.14 3.82 4.33 0.51 1.27 0.00 0.00 2.29 0.00 2.54 12.72 4.07 0.51 5.60 0.25 1.53 25.45 1.53 4.83 6.11 1.53 0.51 2.29 4.07 3.56 8.14 2.54 2.29 5.09 13.23 1.27 2.29 1.78 2.29 3.05 0.25 4.58 3.82 6.36 3.56Y 55.98 1.78 4.07 0.25 1.02 0.00 1.78 10.69 0.00 3.56 0.00 2.29 0.00 0.76 42.75 3.05 1.02 2.04 4.07 0.25 6.36 4.33 0.76 9.92 8.65 0.00 2.29 0.51 2.04 3.82 1.27 0.76 10.94 6.11 4.33 3.31 1.53 0.25 1.27 3.56 1.53 5.09D 3.05 17.30 0.51 0.76 2.29 97.46 0.00 0.00 0.25 0.25 0.00 0.25 0.00 0.00 0.00 1.78 0.25 0.00 1.78 0.51 2.80 0.76 0.51 1.53 9.92 14.76 13.23 16.03 10.18 5.85 13.99 4.83 3.05 2.54 3.31 4.58 2.54 0.25 5.60 6.11 8.91 2.54E 1.02 0.51 9.67 8.40 0.00 0.76 0.51 2.54 0.00 0.25 0.00 0.00 0.00 0.25 0.51 62.60 0.00 0.00 5.09 1.53 11.70 2.29 0.25 0.51 7.12 10.18 14.50 6.87 15.52 6.62 19.34 5.09 4.07 12.72 4.83 2.80 4.58 0.76 8.65 4.58 8.91 7.63H 7.38 0.00 1.53 1.02 0.51 0.00 0.00 0.51 0.00 0.00 0.00 1.02 0.00 0.25 0.00 1.02 0.00 1.02 1.53 1.53 1.78 1.78 0.76 1.02 4.58 3.05 5.34 6.62 3.56 2.04 2.29 2.04 1.02 2.04 2.04 1.02 0.51 0.25 1.78 3.56 3.56 1.27K 0.00 1.02 17.30 22.14 0.51 0.00 0.00 0.00 0.51 0.00 0.00 0.25 0.25 0.00 0.00 1.78 0.00 0.51 1.53 1.53 12.21 2.29 0.25 0.76 3.31 2.04 4.07 6.36 4.83 9.67 6.36 3.31 3.56 6.62 18.83 4.07 6.87 1.27 9.41 5.09 7.89 4.83R 0.76 0.51 10.69 7.38 0.51 0.00 0.00 0.00 1.02 0.00 0.25 0.25 0.00 0.00 0.00 0.25 0.00 0.25 6.11 10.18 10.69 3.05 0.51 0.76 5.09 0.51 4.33 2.80 3.31 6.11 7.89 1.78 4.33 7.38 10.18 3.05 5.60 0.51 6.62 3.05 5.34 4.33Space 0.51 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.51 0.51 0.51 0.51 0.51 0.51 1.02 0.51 1.02 1.02 1.02 1.02 1.27 1.27 1.02 1.02 1.02 1.27 1.02 1.02 1.02 1.02 1.02 1.02 22.39 4.58 3.31 2.54 2.04Insert 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 72.26 0.00 0.00 0.00 0.00Highest 55.98 27.23 17.30 22.14 34.86 97.46 29.52 77.35 65.65 45.04 92.37 35.62 35.37 43.51 42.75 62.60 39.69 45.55 25.45 66.16 12.21 34.61 68.70 60.05 12.47 30.53 14.50 16.03 15.52 9.67 19.34 20.87 23.41 12.72 18.83 43.00 16.54 1.27 11.20 14.50 11.20 14.25Y S K K V D I W S L G V I L Y E L L T G K P P F X G E S E L E L L X K I L X X P X L211 212 213 214 215 216 217 21 219 22 221 222 223 224 225 226 2 7 228 229 2 0 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247A 3.82 3.82 3.05 4.58 4.33 5.60 3.56 7.63 5.60 34.86 7.89 5.09 0.51 3.56 3.31 5.60 1.78 2.29 5.85 3.56 1.27 6.62 12.21 7.12 0.51 5.6 .53 34.61 10.94 3.56 19.34 2.29 8.40 1.27 3.82 1.02 1.27G 3.56 1.78 3.31 3.56 5.09 1.78 3. 5 .04 5.09 3.31 1.27 3.05 .51 0.76 1.78 6. 6 1.27 0.25 1.02 1. 2 1.53 1.02 1.53 .76 0.00 2.54 .02 0.76 4.07 2.0 1.78 0.25 2.04 0. 0 1.78 1.53 0.76F 3.82 8.40 6.87 2.29 4.07 6.36 0.51 0.76 0.00 15.27 3.56 0.25 21.63 1.53 .25 1.02 5.34 4.83 0.51 4.83 0.76 1.27 1.53 1.53 0.00 2.54 0.25 5.8 3.31 0.25 3.05 1.53 0.76 0.76 0.25 34.61 33.33I 0.51 10.43 1.02 1.53 1.78 12.98 2.29 0.51 1.27 2.80 6.87 1.02 7.38 7.74 3.56 1.53 2.80 5.09 2.8 3.82 1.7 0.25 .02 0.51 0.00 14.25 .25 11.7 2.54 0.51 20.61 3.31 0.76 0.76 3.82 .53 12.21L 7.38 9. 2 5.85 3.05 4.58 18.32 4. 7 2.8 6.62 16.28 7.89 3.56 55.47 34.61 .60 4.83 32.82 65.39 4.33 9.41 3.05 2.29 3.56 3.31 0.00 27.99 .02 9.16 6.36 3.31 22. 4 55.98 4.83 1.53 5.09 9.67 23.41M 1.02 5. 9 0.51 0.51 0.25 4.33 0.76 . 1.02 4.07 0.51 0.76 .78 8.91 1.53 2.80 22.14 3.05 2.54 2.80 0.00 2.29 1.27 0.76 0.00 3.82 0.51 1.27 0.51 0.76 0.76 6.11 .76 0.76 1.78 0.76 3.82P 16.28 8.65 28.2 12.21 4.58 3.31 11.20 21.12 5.34 1.27 0.25 1.27 0.51 2.29 0.25 1. 7 0.25 0.25 0.0 6.87 0.76 65.65 0.25 0.76 0.00 32.32 2.04 4.58 2.54 0.51 1.27 1.78 0.00 0.00 48.0 0.00 0.76V 3.56 5.85 3.05 2.80 4.33 11.70 1.53 5.09 4.33 6.87 5.60 1.02 2.54 13.23 1.02 1.27 1.02 1.02 10.69 10.43 0.76 2.80 0.25 3.05 0.25 1. .25 9.9 1.27 1.02 11.20 2.54 1.53 1.02 2.2 4.33 12.21W 2.54 1.27 9.92 0.00 1.78 5.09 0.25 0.00 0.25 0.0 0.51 0.0 .00 0.00 .25 1.27 0.00 10.18 0.0 1.53 0.00 0.25 0.51 0.25 0.51 0.51 0.00 0.51 0. 0 0.00 0.00 0.00 .25 0.00 1.02 24.43 0.25C 0.51 1.27 0.0 .27 0.76 5.09 1.53 1.27 1.02 7.89 1.78 1.78 .25 2.54 .76 2. 4 25.95 2.29 2.29 2.29 1.27 0.00 0.51 2.04 0.00 1.02 .02 2.54 0.76 0.51 3.82 1.53 .02 0.76 1.02 0.25 0.25N 2.29 3.82 3.05 1.27 7.38 .25 3.31 2.29 3.05 0.0 1.27 5.85 .76 0.25 4.33 4.07 0.25 0.00 4.83 0. 0 21.63 1.02 2.80 3.31 0.00 0.00 1.02 0.25 4.33 1.53 0.76 0.25 7.89 1.78 1.02 0. 0 1.53Q 4.33 3.31 3.56 .82 2.29 1.27 0.25 4.58 4.58 1.02 6.36 5.34 .00 0.76 7.12 9.16 0.25 0.00 12.98 5.85 2.04 3.31 5.85 12.21 0.00 0.00 0.25 0.51 5.09 20.10 1.27 3.56 1 .52 3.56 3.05 1.53 1.02S 4.83 4.33 8.91 12.72 8.40 2.29 45. 4 9.67 7.89 0.76 3.31 12.21 .51 0.00 1 .72 6.36 0.51 .53 4.33 1.27 3.3 2.29 12.98 4.83 0.00 0.76 21.63 4.3 6.62 3.05 2.29 1.02 .65 3.05 8.40 2.29 0.00T 3.56 2.80 1.53 5.34 6.87 3.82 5.34 3.82 2.29 1.02 2.54 1.27 1.27 1.27 3.82 1. 2 1.27 .76 11.7 2.80 1.53 0.25 5.34 2.04 0.25 0. 27.99 4.33 3.82 2.80 3.82 1.27 .31 2.04 1.53 . 8 0.76Y 1.53 8.65 1.78 1.53 2.80 4.83 1.27 1. 7 1.27 0.76 1. 2 0.0 1.02 0.00 0.51 0.76 0.00 .51 1.27 7.89 0.0 2.04 1.53 0.51 0.00 3. 5 0.00 2.04 2.04 0.5 0.51 0.51 .76 3.05 0.51 12.72 1.78D 2.80 3.56 4.58 12.47 5.85 3.05 7.12 7.38 9.16 0.51 0. 0 35.62 2.54 0.00 5.85 2.54 0.25 .00 4.83 0.76 43.26 0.76 7.63 6.87 0. 0 0.51 2.29 0.76 8.91 16.03 0.76 .25 5.34 6.62 2.8 .51 0.76E 1 .78 6.62 3.82 9.16 8.40 3.56 3.05 14.76 27.99 1.78 5.09 8. 1 1.27 0.00 10.69 7.89 0.2 0.25 10.94 3.05 9.41 1.02 7.8 1 .47 0.00 0.51 1.02 3.05 21.37 3.33 0.76 0.76 8.65 2.04 5.09 1.27 1.78H 2.04 2.29 2.04 4.07 7.89 1.02 1.78 2.04 2.29 0.25 3.31 3.05 0.25 0.51 2.80 3.56 0.00 0.51 4.07 2.54 1.27 1.53 5.09 2.80 0.51 0.51 0.76 0.00 2.54 1.27 0.00 0.25 1.78 65.14 0.51 0.25 0.76K 8.91 2.80 3.82 10.69 9.67 2.80 2.04 6.62 6.62 0.51 24.43 4.58 0.25 0.00 16.03 18.83 0.76 0.76 7.38 18.58 4.33 1.78 11.70 27.48 1.78 0.51 0.51 2.04 8.65 6.11 2.80 11.45 16.28 2.04 4.33 0.51 1.02R 9.16 2.54 3.82 6.11 5.60 1.78 1.27 5.60 2.80 0.00 15.01 2.80 0.25 0.76 16.79 16.54 2.54 0.25 6.87 10.43 1.53 3.05 5.85 6.87 95.67 1.02 0.25 0.76 3.31 1.78 2.54 4.58 10.43 3.05 3.31 0.51 1.53Space 1.78 2.80 1.27 1.02 3.31 0.76 0.76 0.76 1.53 0.76 1.53 1.27 1.27 1.27 1.02 1.27 0.51 0.76 0.76 0.25 0.51 0.51 0.76 0.51 0.51 0.51 3.56 1.02 1.02 1.02 0.51 0.76 1.02 0.76 0.51 0.51 0.76Insert 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.27 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 32.82 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Highest 16.28 10.43 28.24 12.72 9.67 18.32 45.04 21.12 27.99 16.28 24.43 35.62 55.47 34.61 16.79 18.83 32.82 65.39 12.98 18.58 43.26 65.65 17.81 27.48 95.67 32.32 27.99 11.70 21.37 33.33 22.14 55.98 16.28 65.14 48.09 34.61 33.33E I P X X L S P E A K D L L R K L L X K D P X K R P T A E E L L K H P F F 180  Appendix 2. Kinase substrate consensus sequences from literature    181  Appendix 3. Protocol of the kinase substrate microarray technique Reagents: 1. Microarray Blocking Buffer: 1% bovine serum albumin (BSA) in 100 mM HEPES, pH 7.5. 2. Protein kinase buffer: 5 mM MOPS pH 7.2, 2.5 mM β-glycerol-phosphate, 5 mM MgCl2, 1 mM EGTA, 0.4 mM EDTA, 0.5 mM dithiothreitol (DTT). 3. Tris-buffered saline (TBS): 50 mM Tris-HCl, 150 mM NaCl, pH 7.5. 4. Tris-buffered saline with Tween (TBST): 0.05% Tween-20 in TBS.  5. ATP stock solution: 10 mM ATP in distilled H2O. 6. 0.5% Sodium dodecyl sulphate (SDS) in distilled H2O. 7. Pro-Q Diamond phosphoprotein/phosphopeptide microarray stain and destain solution (Life Technologies).  Procedures: 1. Block the slide in a microarray reaction tray with the Microarray Blocking Buffer for 1 hour at room temperature with agitation. 2. Wash the slide with TBST for 3 times, 2 min each time. 3. Take out the slide from TBST using forceps, and allow excess liquid to slip off. Clip on to the slide a microarray incubation chamber. 4. Make up a proper volume of kinase reaction mix by diluting purified active protein kinase in  182  protein kinase buffer, and supplement with 100 μM ATP. Load the solution into the incubation chamber. 5. Incubate the reactions for 2 hours at 30 ℃ in a shaking incubator. Keep the microarray in a humidity chamber if necessary.  6. Carefully remove the incubation chamber. Wash the slide with 0.5% SDS twice, 5 min each time. 7. Wash the slide with TBST for 7 times, 5 min each time to remove SDS. 8. Block the slide in the Microarray Blocking Buffer again for 1 hour at room temperature. 9. Stain the slide with Pro-Q Diamond stain for 1 hour at room temperature in dark. 10. Destain with the Pro-Q Diamond destain solution for 3 times, 15 min each time. 11. Rinse the slide in H2O once. Wash in H2O for 15 min. 12. Dry the slide using N2 flow.  13. Scan in a microarray scanner at a wavelength of 532 nm or 543 nm. 14. Use microarray image analysis software to archive quantitative data and generate a consensus peptide sequence.   183  Appendix 4. Specificities of 214 human protein kinases tested on kinase substrate peptide microarrays   184    185    186    187    188     189  Appendix 5. Specificity of ERK1-pT207 antibody   Figure A5. Specificity of the ERK1-pT207 antibody. The ERK1- pT207 antibody was tested on a peptide dot blot (A) and a Western blot of ERK1 recombinant protein (B). Three purified peptides with ERK1 Y204 and T207 phosphorylated in different combinations were spotted on nitro-cellulose membrane (2 ng/spot). In A, it is shown that the pT207 antibody reacted with peptide #1 and #3 (T207 phosphorylated), but not with peptide #2 (only Y204 phosphorylated). In B, wild type, kinase-dead and T207A mutant ERK1 were phosphorylated by MEK1 and subjected to Western blotting analysis. The pT207 antibody was able to detect increases signal in WT with MEK1 stimulation without cross-reacting with KD or T207 mutants. These results indicated the pT207 antibody was specific for ERK1-T207 phosphorylation.    190  Appendix 6. S74 near Subdomain II is an autophosphorylation site The data from the mass spectrometry (MS) analyses of ERK1 also identified S74 as a novel phosphorylation site in wild-type, but not catalytically-dead preparations of the kinase. Phospho-specific antibody against this serine site was raised and tested with recombinant ERK1-WT and KD proteins, which were prepared under the same conditions with the MS samples. Results from Western blotting analyses indicated a similar phosphorylation level at S74 in ERK1-WT with and without incubation with MEK1, but no phosphorylation in ERK1-KD (Figure A6 A). Thus, S74 was autophosphorylated, and the phosphorylation was likely to be independent from phosphorylation of the TEY sites by MEK1. Preliminary results from A431, Jurkat and NIH-3T3 cells with different treatments also showed that S74 was phosphorylated in vivo (Figure A6 B). This serine is located two residues apart on the C-terminal side from the ATP-binding Kinase Subdomain II (AXK motif). Phosphosites in human c-TAK1/MARK3 kinase near the AXK motif were identified as inhibitory sites (Bachmann et al., 2004). Considering the addition of a phosphate at the serine may induce interaction with the positively charged lysine and disturb the orientation of Subdomain II, it was suspected that the phosphorylation of S74 might be inhibitory. ERK1-S74E mutant was generated and expressed as recombinant protein. Based on in vitro kinase assay with MBP, this phospho-mimicking mutant showed comparable  191  activity with ERK1-WT (Figure A6 C). This observation did not support the hypothesis that S74 phosphorylation inhibited the kinase activity of ERK1. The functional role, if any, of this phosphosite is still unclear.  Figure A6. Autophosphorylation of ERK1 S74 site. (A) The phosphorylation of S74 was examined in GST-ERK1 WT and KD with or without MEK1 activation. (B) Phosphorylation of S74 in A431, Jurkat and NIH-3T3 cells under various treatment conditions (PPi, phosphatase inhibitors; Stauro., staurosporine). (C) A mutant that mimics the phosphorylation of S74 (S74E) was created and expressed as a recombinant protein in bacteria. The samples were subsequently subjected to SDS-PAGE and Western blotting with phosphosite-specific antibodies or the pan-expression ERK-CT antibody, and the immunoblots from the region of the migration of the ERK1 are shown. The mutant showed comparable phosphotransferase activity towards the substrate bovine myelin basic protein (pMBP) with ERK1-WT in vitro.  192  Appendix 7. Generation and evaluation of a new polyclonal generic phosphotyrosine antibody  A7.1 Rationale Protein phosphorylation on tyrosine residues plays important regulatory roles under physiological and pathological conditions (Hunter, 1989). Tyrosine phosphorylation in response to extracellular stimulations mediates downstream signalling by controlling enzymatic activities and protein-protein interactions in a reversible manner (Pawson and Scott, 2005). Alternations in protein-tyrosine phosphorylation and protein-tyrosine kinase activities have been implicated in various human diseases including cancer (Blume-Jensen and Hunter, 2001; Machida et al., 2003). Since their first development in the 1980s (Glenney et al., 1988), phosphotyrosine (pY) specific monoclonal antibodies have been widely used to identify, purify and quantify tyrosine phosphorylation. Development of mass spectrometry techniques has allowed sensitive protein phosphorylation analysis at the proteome level. Considering the low abundance and stoichiometry of phosphoproteins, especially at phosphotyrosine sites, phospho-specific antibodies are still important tools in the enrichment step. However, a recent comprehensive comparison of these monoclonal antibodies using high throughput phosphopeptide microarray approach indicated low coverage rates and sequence preferences of all three generic pY antibodies tested, which  193  included 4G10, PY20 and p-Tyr-100 (PY100) (Tinti et al., 2012). In this chapter, a novel method to produce generic phosphotyrosine antibody is described. Polyclonal antibodies were raised in a large number of rabbits using a large number of diverse phosphopeptides that featured multiple phosphosites. Parallel testing of this polyclonal antibody preparation, termed PY-K, and the well-known and used monoclonal pY antibodies mentioned above was performed, and the findings demonstrate significant improvements in the performance of PY-K relative to these other widely used pY antibodies.  A7.2 Production of the generic phosphotyrosine antibody PY-K Peptides with a length of 7 to 20 amino acids that featured up to 7 confirmed physiological phosphorylation sites (including serine, threonine and tyrosine phosphosites) located on the activation loops of more than 210 of the human protein kinases were synthesized with an additional C-terminal beta-alanine (as spacer) and a cysteine residue (for conjugation). The syntheses were carried out according to Fmoc solid-phase synthesis strategy (Fields and Noble, 1990). The peptide products were confirmed by mass spectrometry and purified to a final purity between 70% and 98%. Following conjugation with keyhole limpet hemocyanin (KLH), antigenic peptides were divided into 28 distinct pools with equal mass ratios. Each peptide mixture was separately injected into two New Zealand White rabbits for immunization based on an extended 4-month protocol (Pacific   194       Figure A7.1. Schema for preparation of generic phosphotyrosine antibody PY-K. The PY-K antibody was originally raised in over 50 rabbits injected with 28 distinct pools of phosphopeptides selected from over 200 human protein kinase sequences. After 4 months of repeated immunizations, the sera of the rabbits were pooled and subjected to ammonium sulphate precipitation. The immunoglobulin fraction was dissolved in PBS and affinity-purified with phosphotyrosine-agarose columns.    195  Immunology). Sera of the two rabbits from each group were combined and immunoglobulins were precipitated with 50% saturated ammonium sulphate (SAS). A pY-peptide antibody library was created by pooling equal amounts of all 28 SAS fractions together, which was then subject to affinity purification on phosphotyrosine-agarose columns (Figure A7.1). The methods for purification and concentration of the antibody were described in Chapter 2.  A7.3 Phospho-specificity of the pY antibodies on KinexTM phosphopeptide microarray The KinexTM phosphopeptide microarray (Kinexus) features 500 human physiological phosphopeptides that were selected based on their high rates of evolutionary conservation, with scores calculated with sequences from 20 diverse species. Each peptide contains 15 amino acids with a phospho-serine, threonine or tyrosine in the middle of the sequence, and was printed in triplicate on the microarray slides. The immunoreactivity of PY-K antibody was tested in parallel with 4G10, PY20 and PY100 on the same slide. Alexa-546 dye-labelled secondary antibodies were used to detect antibody binding. Based on the normalized data, all four generic antibodies were able to detect about 90% of the 184 pY peptides with intensity higher than 25% of the signal median. Overall, PY-K provided stronger detection of phosphotyrosines than the monoclonal antibodies that 196   Figure A7.2. Immunoreactivities of generic phosphotyrosine antibodies for 500 phosphopeptides on Kinex™ phosphopeptide microarray. Data for immunoreactive peptide signal intensity were normalized to the median value for each tested antibody for each type of phosphorylated peptide on the microarray: phosphotyrosine (A), phosphothreonine (B), and phosphoserine (C). In each data set, the peptide signals for each antibody and type of phosphorylated peptide were sorted in order of the strongest signal first. The final concentrations of antibodies used for probing each grid of phosphopeptides on the microarray were ~2 µg/mL PY-K, ~5 µg/mL 4G10, ~5 µg/mL PY-20, and ~1 µg/mL PY-100.  197  usually require higher working concentrations (Figure A7.2, Panel A). Most serine- and threonine-phosphorylated peptides were recognized to about 0.2-0.8% of the signal intensity as recorded with most tyrosine-phosphorylated peptides on the microarray. However, the 4G10 antibody had stronger cross-reactivity to phosphothreonine than PY-K, PY-20 and PY-100 (Figure A7.2, Panel B). As for serine-phosphorylated peptides, PY-K showed slightly higher immunoreactivity for a few of the peptides than observed with the other antibodies. But the maximum signal intensity of phosphoserine was only 4% comparing to the signal of phosphotyrosine detected by the PY-K antibody (Figure A7.2, Panel C).  A7.4 Differential recognition specificities of pY antibodies To further evaluate the selectivity for the pY sequence context of each generic antibody, the annotated phosphosites-Tyr-phosphatase peptide microarray developed by Jerini Peptide Technologies was employed to assess binding of more than 6,000 distinct human protein-tyrosine phosphosites. The same incubation conditions and antibody concentrations were used for this chip as with the KinexTM phosphopeptide microarrays. The resultant image of the subarray that was probed with PY-K is presented in Figure A7.3. Overall, PY-K was able to detect most of the pY peptides with stronger immunoreactivity than the other tested phosphotyrosine antibodies (Figure A7.4). After normalizing the signal intensity to an equal median value, a subset of positive peptides that  198  showed signal intensity at least one standard deviation higher than the median was selected for each generic antibody. Of all of the 6099 pY peptides on the JPT chip, there were 1746 positive peptides that were picked up by at least one of the four generic antibodies (Figure A7.5). Among these peptides, PY-K detected 1316 (75.4%), 4G10 detected 1168 (66.9%), PY-20 detected 1193 (68.3%), and PY-100 detected 1307 (74.9%). Although PY-K and PY-100 had similar coverage, the median signal intensity of PY-K was about 10 times higher than PY-100 before the normalization scalar was applied. Thus, PY-K antibody was able to detect more pY peptides with stronger immunoreactivity comparing to the monoclonal antibodies. Interestingly, more than 25% (466/1746) of the positive pY peptides were uniquely recognized by only one of these antibodies. The PY-K antibody actually detected 243 pY peptides that were not recognized by any of the other three antibodies, which indicated differences in sequence context specificity of the generic antibodies.  To further characterize the selectivity of these antibodies, the Two Sample Logo web tool (Vacic et al., 2006) was used to visualize the positive and negative factors based on two sets of input data. For each antibody tested, the positive dataset was generated by peptides with signal intensity at least one standard deviation higher than the signal median, whereas the negative dataset included peptides with signal lower than 25% of the median value. As shown in Figure A7.6, all four antibodies positively selected for   199         Figure A7.3. JPT phosphopeptide chip probed with PY-K antibody. One of the three representative fields of the JPT phosphopeptide chip that features 6099 human phosphotyrosine peptides after incubated with PY-K (2 µg/mL) for 2 hours. The binding of primary antibody was detected by Alexa-546-labelled donkey anti-rabbit IgG secondary antibody.      200        Figure A7.4. Immunoreactivities of generic phosphotyrosine antibodies for 6099 phosphotyrosine peptides on JPT phosphopeptide chip. Data for immunoreactive peptide signal intensity were normalized to the median value for each tested antibody. In each data set, the peptide signals for each antibody and type of phosphorylated peptide were sorted in order of the strongest signal first. The final concentrations of antibodies used for probing each grid of phosphopeptides on the microarray were ~2 µg/mL PY-K, ~5 µg/mL 4G10, ~5 µg/mL PY-20, and ~1 µg/mL PY-100.    201           Figure A7.5. Venn diagram of tyrosine-phosphorylated peptides detected by four generic phosphotyrosine antibodies. This diagram was generated from the 1746 most immunoreactive phosphotyrosine peptides on JPT phosphopeptide microarray. Phosphopeptides showed immunoreactivities that were at least one standard deviation higher than the median value determined for each tested antibody were considered. Phosphopeptides that are strongly recognized by two or more antibodies are shown in overlapping circles. Of the 1746 strongly immunoreactive phosphopeptides, PY-K detected 1316 (75.4%), 4G10 detected 1,168 (66.9%), PY-20 detected 1,193 (68.3%) and PY-100 detected 1,307 (74.9%).   202   Figure A7.6. Logo diagram for amino acid specificity preferences of generic phosphotyrosine-specific antibodies. The amino acid residue recognition preferences of the flanking sequences for four generic phosphotyrosine-specific antibodies tested on 6099 distinct tyrosine-phosphorylated peptides printed on the JPT phosphopeptide microarray. The top peptides were selected using the cut-off criteria as described in the text for classifying the positive and negative data sets. The visualization graphs were generated with the Two Sample Logos algorithm.  203  leucine at -1 and +1 positions, and proline at +3 position, which was consistent with previous results published by other groups (Tinti et al., 2012; Zerweck et al., 2009). It is evident that the PY-20 and PY-100 displayed remarkably similar specificities. Acidic residues were negative determinants for all three monoclonal antibodies at most positions, whereas PY-K antibody positively selected for acidic amino acids at -1 and -2 positions. Significantly, aspartic acids and glutamic acids commonly precede the tyrosine phosphoacceptor sites in physiological phosphosites.  To investigate the physiological relevance of binding specificities of the generic antibodies, the sequences of 34,004 confirmed human phosphotyrosine sites from PhosphoNET database (http://www.phosphonet.ca/) were aligned, and the frequencies of each amino acid at each position surrounding the phosphotyrosine were calculated (Figure A7.7). It is evident that the amino acid residues for leucine, serine, glutamic acid, lysine, glycine, arginine, alanine and aspartic acid most commonly flank phosphotyrosine sites in human proteins. It is important that generic phosphotyrosine antibodies will immunoreact favourably with tyrosine phosphosites that are bordered by these residues to permit detection of physiological phosphorylation sites. For every generic antibody tested, a frequency matrix was generated by subtracting the background frequency of each amino acid at each position based on all 6099 peptide sequences from the binding frequency calculated based on the positive dataset. Percent differences of each antibody from the expected background are presented in Figure A7.8.  204  The Frobenius norm of each matrix was calculated to represent the differential selectivity of each antibody comparing to the background. The data indicated that PY-K had the lowest overall bias, which is 10.1% lower than PY-100, 11.3% lower than 4G10, and 14.8% lower than PY-20. All four antibodies showed positive selectivity for leucine residues at most positions, which was consistent with the high frequencies of leucine residues surrounding physiological phosphotyrosine sites. However, PY-K had much lower negative bias towards aspartic acid and glutamic acid residues, which are also commonly found near phosphotyrosines in peptides, than the three tested monoclonal antibodies. Thus, it is concluded that the polyclonal PY-K antibody exhibited overall weaker sequence context selectivity and stronger affinity towards physiological phosphotyrosine sites when compared to the monoclonal pY antibodies tested.  A7.5 Phosphotyrosine detection of the generic antibodies in Western blotting The differential specificities observed from phosphopeptide microarrays analyses with the various pY antibodies might also be reflected in detectable differences in less sensitive methods such as Western blotting. The human epidermoid carcinoma cell line A431 is widely used in cell signalling and phosphorylation studies due to the overexpression of EGF receptor (EGFR) (Giard et al., 1973). To test the detection of phosphotyrosine- containing protein bands on Western blots, A431 cells were treated with  205  phosphatase inhibitors, or stimulated by EGF to induce protein-tyrosine phosphorylation. Four sets of cell lysate samples from the same experiment were resolved by SDS-PAGE. The concentration of each generic antibody used was consistent with the microarray experiment. As shown in Figure A7.9, all four generic antibodies yielded a strong signal at around 170 kDa, which was likely to be tyrosine phosphorylated EGFRs, with PTP inhibitors treatment or EGF stimulation. The overall band patterns were similar with slightly different relative intensities. The PY-K antibody displayed higher sensitivity than the other monoclonal antibodies. None of the antibodies showed any phosphotyrosine signals in serum-starved cell lysate (Figure A7.9, Lane 5), indicating low cross-reactivity with unphosphorylated proteins.   A7.6 Discussion Reversible protein-tyrosine phosphorylation plays critical regulatory roles under diverse physiological and pathological conditions, which has made generic phosphotyrosine antibodies valuable tools in cell signalling research. A recent study that compared the specificities of three widely-used monoclonal generic pY antibodies using high-density phosphopeptide chip technology (Tinti et al., 2012) has raised concerns about strong sequence context selectivities and low overall coverage rates. These issues can potentially introduce significant bias in phosphotyrosine site identification and quantitation.   206            Figure A7.7. Frequency of amino acids surrounding human phosphotyrosine sites. The percent frequency of each of the 20 common amino acids bordering the phosphotyrosine residues in 34,004 experimentally confirmed human phosphosites was determined using data from PhosphoNET (http://www.phosphonet.ca/).     207    Figure A7.8. Sequence context specificities of generic phosphotyrosine antibodies. The positive data set for each antibody based on the JPT phosphopeptide chip was used to calculate a matrix with percent frequency of each amino acid at each position. The specificity matrices were then generated by subtracting the background frequency of all 6099 phosphopeptide on the chip from the positive percent frequency of each antibody tested. All of the peptide were tyrosine-phosphorylated, so the background frequency for the phosphoacceptor site was zero. Red-coloured cells correspond to amino acids that occur in higher frequency than expected by random and are favoured, whereas blue-coloured cells identify residues that are unfavourable to the antibody. The Frobenius norm of each matrix was calculated to represent the strength of selectivity of each antibody (19.6 for PY-K, 22.1 for 4G10, 23.0 for PY-20, and 21.8 for PY-100).  208   Figure A7.9. Immunoreactivity of generic phosphotyrosine-specific antibodies with A431 cell lysates. Selectivity of generic phosphotyrosine antibodies towards protein-tyrosine phosphorylation induced by treatments of various phosphatase inhibitors and EGF in A431 human cervical carcinoma cells. Lane 1: 0.025% DMSO for 30 min; Lane 2: combination of the PTP inhibitors (25 mM phenylarsine oxide/PAO and 50 mM Na3VO4) for 30 min; Lane 3: STP inhibitor (30 mM NaF) for 30 min; Lane 4: combination of STP inhibitor (30 mM NaF) and PTP inhibitor (50 mM Na3VO4) for 30 min; Lane 5: serum-starved control; Lane 6: serum-starved cells with 100 ng/mL EGF for 5 min. The amount of lysate protein loaded in each lane was 25 µg.  209  In this study, my colleagues and I applied a novel strategy by generating a pool of polyclonal antibodies from a large set of physiological phosphotyrosine site sequences. The evaluation was carried out using the same JPT microarray as well as a phosphopeptide chip with phosphoserine, phosphothreonine, and phosphotyrosine peptides to test the cross-reactivities. Results from multiple tests all indicated that the PY-K polyclonal antibody was able to detect phosphotyrosine with higher sensitivity and less bias than the monoclonal pY antibodies 4G10, PY-20 and PY-100. The specificity of PY-K was also shown to be most similar to the profile of experimentally confirmed human phosphotyrosine sites, revealing more physiological relevant selectivity for this polyclonal antibody. The PY-K antibody can serve as a powerful tool to identify, quantify, and enrich phosphotyrosine-containing proteins. This study also proved the concept of using “pooled” polyclonal antibodies to detect generic phosphorylation. In the future, the same strategies can be applied with larger sets of antigenic phosphotyrosine peptides to develop even better generic phosphotyrosine antibodies, as well as phosphoserine- and phosphothreonine-specific antibodies.  

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0167195/manifest

Comment

Related Items