Crystal structure of the catalytic and
ubiquitin-associated domains of
the protein kinase MARK2 / PAR-1
from Rattus norvegicus (Berkenhout, 1769)
Thesis submitted to the Department of Biology, Faculty of Mathematics, Informatics and Natural Sciences of the University of Hamburg in partial
fulfillment of the requirements for the degree of Ph.D.
By
Saravanan Panneerselvam
from India
Hamburg 2006
1 Introduction 1
1.1 The protein kinase superfamily 1
1.2 Classification of the protein kinase superfamily 1
1.3 Identification of MAP/microtubule affinity regulating kinase (MARK) 3
1.3.1 Homologues of MARK 6
1.3.2 Activation of MARK 9
1.3.3 Inhibition of MARK 9
1.3.4 Proteins interacting with MARK 10
1.4 Ubiquitin binding domains 11
1.4.1 The ubiquitin-associated domain (UBA) 11
1.4.2 Structure of the UBA domain 12
1.4.3 The UBA domain of the MARK 13
1.5 Aim of the work 14
2 Materials and methods 15
2.1 Materials 2.1.1 Chemicals 15 2.1.2 Enzymes 15 2.1.3 Bacterial strains 16 2.1.4 Cloning vectors 17 2.1.5 Expression vectors 17 2.1.6 Media 17 2.1.7 Crystallization 2.1.7.1 Crystallization supplies and tools 18
2.1.7.2 Crystallization solutions 19
2.1.8 Equipment and accessories 19
2.2 Molecular biology and microbiological methods 2.2.1 Culture and storage of E. coli strains 20
2.2.2 Transformation of E. coli strains 20
2.2.3 Isolation of plasmid DNA 21
2.2.5 Ligation reaction 21
2.2.6 Restriction analysis of DNA 22
2.2.7 DNA sequencing 22
2.2.8 Mutagenesis of DNA 23
2.2.9 Gene cloning using the Invitrogen Gateway technology 24
2.3 Protein methods 2.3.1 SDS-Polyacrylamide gel electrophoresis (SDS-PAGE) 29
2.3.2 Protein expression and purification 30
2.3.3 Cell lysis and solubility test 30
2.3.4 Chromatography 31
2.3.4.1 Ni-NTA affinity chromatography 31
2.3.4.2 GST affinity chromatography 31
2.3.4.3 Anion and cation exchange chromatography 32
2.3.4.4 Gel filtration chromatography 32
2.3.5 Determination of concentration of proteins 33
2.3.6 Concentrating the protein solution 33
2.3.7 Protein kinase assay 33
2.3.8 Limited proteolysis 34
2.3.9 N-terminal sequencing 34
2.3.10 Mass spectrometry 34
2.3.11 Detailed protocols on purification of MARK2 wild type and mutant proteins 35 2.3.11.1 Expression 35 2.3.11.2 Purification 35 2.3.11.3 Selenomethionine labelling 36 2.4 Crystallographic methods 2.4.1 Crystallization techniques 36 2.4.2 Cryoprotection of crystals 38 2.4.3 Data collection 38
3 Results 41
3.1 Cloning of different constructs of MARK2 41
3.2 Identification of structurally folded part of MARK2 42
3.2.1 Limited proteolysis 42
3.2.2 N-terminal sequencing and mass spectrometry analysis 43
3.3 Cloning of the stable fragment of MARK2 44
3.4 Expression and purification 45
3.5 Kinase activities of the purified protein 47
3.6 Preparation of selenomethionine labelled protein 48
3.7 Crystallization 48
3.8 Data collection and structure determination 50
3.8.1 Molecular replacement 53
3.8.2 Heavy atom derivatives of MARK2 crystals 53
3.8.3 Model building and refinement 54
3.8.4 Crystals of selenomethionine labelled MARK2 55
3.9 Overall structure of the MARK2 catalytic and UBA domains 56
3.10 Structure of the catalytic domain 56
3.10.1 Conformation of the activation loop 58
3.10.2 Intermolecular disulfide bridge 59
3.10.3 Dimerization 60
3.11 Structure of the UBA domain 61
3.12 UBA linker and common docking domain for kinase activators 66
4 Discussion 68
4.1 Activation segment 70
4.2 Activation loop of MARK2 interferes with substrate binding 70
4.3 Catalytic cleft and nucleotide binding site 73
4.4 Conformation of the catalytic loop 74
4.5 Dimerization 75
4.6 C-terminal extension of the kinase core 77
5 Summary 81 6 References 83 7 Appendix 96 7.1 Abbreviations 96 7.2 List of figures 98 7.3 List of tables 100 7.4 Oligonucleotides 100 7.5 Purification buffers 102 7.6 List of Coordinates 103 8 Acknowledgements 104 9 Curriculum vitae 105 10 Declaration 107
1 Introduction
1.1 The protein kinase superfamily
Protein kinases comprise one of the largest families of proteins in eukaryotic organisms. They catalyze the phosphorylation of cellular proteins on serine, threonine or tyrosine residues in order to alter their functional properties. They are hence central to cellular signaling networks that co-ordinate various activities like metabolism, stress response, transcription, translation, DNA replication and cell cycle control, development of organs, neuronal signaling and apoptosis. Improper functioning of these enzymes is often manifested in various human diseases and has been implicated in several types of cancers.
There are at least 518 protein kinases identified in the human genome, which is nearly 1.7% of all human genes (Manning et al., 2002). All the protein kinases have a strong sequence similarity in their catalytic domains. Despite their sequence similarity, protein kinases have different substrate specificities, mechanisms of regulation, modes of action, etc. The phylogenetic tree (Fig. 1.1) depicts the relationships between members of the superfamily of human protein kinases.
1.2 Classification of the protein kinase superfamily
Protein kinases are classified on the basis of aminoacid sequence similarity in the catalytic domain. Generally, the protein kinase superfamily is divided into 9 major groups, 90 families and 145 subfamilies (Hanks and Hunter, 1995, Manning et al., 2002).
The major groups are as follows:
AGC group - includes the cyclic-nucleotide-dependent protein kinase
families, βARK and ribosomal S6 kinase families
CaMK group - includes the families of protein kinases regulated by
Ca2+/calmodulin, the Snf1/AMPK families and other
Fig. 1.1: Phylogenetic tree of the complete superfamily of human protein kinases.
Most protein kinases belong to a single superfamily of enzymes whose catalytic domains are related in sequence and structure. The main diagram illustrates the similarity between the protein sequences of these catalytic domains. The inset diagram shows trees for seven atypical protein kinase families (Manning et al., 2002). The MARK subfamily is highlighted with a red circle.
PTK group - conventional protein tyrosine kinases
CMGC group - includes the CDK, MAPK, GSK3 and CLK protein
kinase families
STE group - includes homologues of Ste7/MAP2K, Ste11/MAP3K
and Ste20/MAP4K protein kinases
CK1 group - includes the CK1, TTBK and VRK protein kinase
families
TKL group - Tyrosine kinases like kinases, includes kinase families
that resemble both tyrosine and serine/threonine kinases like MLK, LISK, IRAK and STRK families.
RGC group - kinases that are similar in domain sequences to tyrosine
kinases
OPK group - other protein kinases that are not falling into major
groups
In addition to these major groups of kinases, there were some other proteins reported to have protein kinase activity but lack sequence similarity with other protein kinases. These proteins have been classified as atypical protein kinases.
1.3 Identification of MAP/microtubule affinity regulating kinase (MARK)
Microtubules (MTs) serve as tracks for cellular transport, and regulate cell shape and polarity. Rapid transitions between stable and dynamic forms of MTs are central to these processes. This dynamic instability is regulated by a number of cellular factors, including the structural MT-associated proteins (MAPs), which in turn are regulated by phosphorylation. Tau is a microtubule-associated protein prominent in the brain, particularly in the axonal compartment of neurons, where it helps to stabilize microtubules. Microtubules in turn serve as tracks for the intracellular transport of vesicles and organelles, for providing stability of axons, and for growth cone advance. The tau-microtubule interaction is regulated by phosphorylation, especially at the KXGS motifs in the repeat domain of tau which represents the core of the microtubule-binding domain (Biernat et al., 1993). The same domain also forms the core of the abnormal tau aggregates ("paired helical filaments", PHF) in Alzheimer's disease. Both functions, microtubule binding and PHF assembly, are efficiently suppressed when tau
is phosphorylated by MAP/microtubule affinity regulating kinases. Thus, excess activation of MARK in cells leads to microtubule breakdown because they are not properly stabilized. A selection of protein kinases that are phosphorylating tau at different sites are shown in Fig. 1.2.
Fig. 1.2: Bar diagram of human tau and the sites phosphorylated by different protein kinases. Tau is a microtubule-associated protein, and htau40 is the longest
isoform of the splice variants which contains additional N-terminal inserts (I1 and I2). It has an N-terminal projection domain and a C-terminal microtubule binding domain in which the KXGS repeats are located. Tau protein can be phosphorylated by many different kinases. The SP/TP motifs are the main targets for proline directed kinases such as GSK3β, CDK5 and MAP kinase. The main targets of PKA are S214 and to a lesser extent KXGS repeats. The main targets of MARKs are KXGS repeats; the phosphorylation on S262 particularly by MARKs detaches tau from microtubules which in turn causes increased microtubule dynamics and tau aggregation (Biernat et al., 2001; Drewes et al., 1997).
A second function of tau is its interference with motor proteins moving along microtubules; this function is also fine-tuned in axons by MARK (Mandelkow et al., 2004). Using the phosphorylation of tau as a readout, MARK was purified and cloned from brain tissue (Drewes et al., 1997).
PKA
MAPK
GSK3ß
CDK5
C 3 4 441 I1 N I2 P1 P2 1 2 1 TP 153 181 TP 175 TP SP 205 SP 199 SP 202 212 TP 217 TP 231TP 235SP SP 396 422SP SP 404 S214 S262 S293 S324 S356 KXGSMARK
SP 46 TP69 TP 50 TP 111 Projection domainMicrotuble binding domain
Repeat region
PKA
MAPK
GSK3ß
CDK5
C 3 4 441 441 I1 N I2 P1 P2 1 2 1 1 TP 153 181 TP 175 TP SP 205 SP 199 SP 202 SP 205 SP 199 SP 202 212 TP 217 TP 231TP 235SP SP 396 422SP SP 404 SP 396 422SP SP 404 S214 S262 S293 S324 S356 KXGSMARK
SP 46 TP69 TP 50 TP 111 SP 46 TP69 TP 50 TP 111 Projection domainMicrotuble binding domain
Compared to other kinases, MARK is a relatively large protein (~720-790 aa) which contains several domains: an N-terminal leader sequence, a typical kinase catalytic domain, an ubiquitin-associated domain (UBA), a spacer and a tail domain containing the KA1 (kinase-associated) motif characteristic for the family of kinases ending with the ELKL motif (Fig. 1.3). Four isoforms of MARK (1-4) were found in mammals, encoded by different genes, with additional splicing variants (Drewes, 2004; Drewes et al., 1997).
Fig. 1.3: Domain organization of human MARK2. Residue numbers refer to the
longest isoform of human MARK2 (Swiss-Prot entry Q7KZI7). CD, common docking domain; UBA - ubiquitin associated domain; KA1, kinase associated domain 1 and the tail domain ends with ELKL motif, typical for this family of kinases. T208 phosphorylation by MARKK or LKB1 is necessary for MARK activation (Timm et al., 2003; Lizcano et al., 2004) and T596 phosphorylation by aPKC leads to binding of 14-3-3 proteins and negatively regulates the MARK kinase activity (Hurov et al., 2004).
MARKK
LKB1 aPKC
MARKK
1.3.1 Homologues of MARK
MARK kinases are conserved from yeast to human and share a similar primary structural organization. A sequence alignment of MARKs and their homologous proteins is shown in Fig. 1.4. In table 1.1, genes encoding the MARK/PAR-1/KIN1 subfamily in animal and yeast species are summarized.
Organism Genes Mammals MARK1 MARK2 MARK3/C-TAK1/p78 MARK4/MARKL1 pEG3/MELK/MPK38 D. melanogaster PAR-1 C. elegans PAR-1 S. cerevisiae KIN1 KIN2 S. pombe KIN1
Table 1.1: Genes encoding MARK/PAR-1/KIN1 kinases in animal and yeast species (Tassan et al., 2004)
PAR-1: PAR-1 kinase shows a very high homology to MARK. PAR-1 kinase was
initially identified in C. elegans and later in D. melanogaster. In both organisms this kinase plays a major role in the anterior-posterior (A/P) axis formation and cell polarity. Many of the identified PAR-1 substrates are involved in cell polarization and oogenesis (Kemphues, 2000; Pellettieri and Seydoux, 2002).
KIN1: The yeast kinases KIN1 and KIN2 also share a striking homology to MARK
kinases. These kinases belong to the SNF1 kinase family of the Ca2+ /calmodulin-dependent kinase II (CaMK II) group. In S. pombe it has been shown that KIN1 kinases are involved in the formation of cell morphology (Levin and Bishop, 1990)
Members of the MARK/PAR-1 family occur in most organisms examined so far. Analysis of the human genome showed that there are four members of the human MARK family which belong to the class of CaMK II kinases (Fig. 1.1; Hanks and Hunter, 1995; Manning et al., 2002). They occur in a variety of cell types and presumably serve a variety of functions, depending on isoform or localization. Table 1.2 summarizes the different isoforms of mammalian MARKs and their different substrates.
Table 1.2: Isoforms, localization and substrates of mammalian MARK kinases
(Drewes et al., 2004; Riechmann et al., 2004)
Tau, MAP2, MAP4 High in brain, glioma,
testis MARKL1, hPAR-1d MARK4 PTPH1,Cdc25C, KSR1, Plakophilin2, Dishevelled Highest in brain and
pancreas EMK2; KP78; hPAR-1a; C-TAK1 MARK3
Tau, MAP2, MAP4 Dcx, Oskar, Raf, Exuperantia Similar to MARK1 EMK1; hPAR-1b MARK2
Tau, MAP2, MAP4 High in brain, spleen,
skeletal, muscle, pancreas, kidney and heart
EMK3; hPAR-1c MARK1 Substrates Expression Pattern Synonyms Isoform
Tau, MAP2, MAP4 High in brain, glioma,
testis MARKL1, hPAR-1d MARK4 PTPH1,Cdc25C, KSR1, Plakophilin2, Dishevelled Highest in brain and
pancreas EMK2; KP78; hPAR-1a; C-TAK1 MARK3
Tau, MAP2, MAP4 Dcx, Oskar, Raf, Exuperantia Similar to MARK1 EMK1; hPAR-1b MARK2
Tau, MAP2, MAP4 High in brain, spleen,
skeletal, muscle, pancreas, kidney and heart
EMK3; hPAR-1c MARK1 Substrates Expression Pattern Synonyms Isoform
Fig. 1.4: Multiple sequence alignment of MARK/PAR-1 kinase family members.
The sequence numbering is that of human MARK2 kinase. Colour coding: Invariant residues white on red background; conservatively substituted residues red. The sequences are highly similar in the catalytic domain. Sequences used here were from: human (Q7KZI7), rat (O08679) for MARK, C. elegans for PAR-1 (Q17346) and S.
cerevisiae for KIN1 (P13185). Swiss-Prot data base accession numbers are given in
1.3.2 Activation of MARK
Protein kinases are normally regulated by phosphorylation in their "activation loop" which controls the access of the substrate to the catalytic centre (Huse and Kuriyan, 2002; Johnson et al., 1996). Like other kinases, MARK family members can be activated by phosphorylation in the activation loop. This can be achieved by the protein kinase MARKK (MARK-kinase) which phosphorylates T208 in MARK2 (Timm et al., 2003). This kinase was also found in the context of activation of MEKs and named TAO-1 (Hutchison et al., 1998). MARK can also be phosphorylated at T208 and activated by LKB1 which plays a role in tumor suppression (Lizcano et al., 2004).
A further level of regulation lies in the association with other proteins and domains. By immunofluorescence, MARK2 has a vesicular distribution, but the target proteins (such as tau) are often cytosolic. A change in localization may be achieved via the non-catalytic domains of MARK, and indeed the deletion of the spacer domain causes a mislocalization in D. melanogaster (Vaccari et al., 2005).
1.3.3 Inhibition of MARK
A notable feature of MARK isolated from brain tissue is its double phosphorylation in the activation loop (at T208 and S212 in MARK2, (Drewes et al., 1997)). While phosphorylation of T208, the target site of MARKK or LKB1, activates the kinase, phosphorylation of S212 is probably inhibitory, but the responsible kinase is unknown at present. Mutation of this serine to glutamic acid or alanine abolishes the kinase activity (Timm et al., 2003). By comparing the MARK sequence with the closely related kinase, CHK1, it appears that this residue is involved in aligning the catalytic residue in the proper position during the phosphotransfer reaction (Chen et al., 2000). Moreover the oncogenic serine/threonine kinase Pim-1 interacts with MARK3 (C-TAK1) and phosphorylates the kinase domain of MARK which inhibits the MARK activity. The residue phosphorylated by Pim-1 is yet to be identified (Bachmann et al., 2004).
A recent study by Matenia et al. reveals that the MARK kinase activity can be inhibited also by some other mechanism. The PAK5 kinase, a member of the mammalian p21 activated kinases family inhibits MARK activity towards tau protein. This inhibition is mainly based on the protein-protein interaction between the MARK and PAK5 catalytic domains, rather than by phosphorylation (Matenia et al., 2005).
1.3.4 Proteins interacting with MARK
MARK interacts with the cytosolic scaffold protein 14-3-3 (alias Par-5) during D.
melanogaster cell polarity and development (Benton et al., 2002; Macara, 2004a;
Macara, 2004b) and 14-3-3 is also involved in many cellular functions and localization of various proteins. 14-3-3 frequently interacts with phosphorylated proteins and recruits them to different cellular compartments. Remarkably, binding between MARK and 14-3-3 occurs without the usual requirement of a phosphorylated peptide. Indeed, MARK does not bind to the canonical binding groove of 14-3-3 but to its C-terminus (Benton et al., 2002). It appears also that MARK phosphorylates other partners which then bind to 14-3-3, such as Cdc25C, KSR1, plakophilin, or Raf-1 (Benton et al., 2002; Muller et al., 2003). This suggests that MARK regulates 14-3-3 activity towards its binding partners.
On the other hand MARK2 is phosphorylated by aPKC at T596, a conserved residue in all MARK isoforms. This phosphorylation increases its 14-3-3 binding activity and negatively regulates the kinase activity and its plasma membrane localisation (Hurov et al., 2004). Moreover, proteomic analysis of MARK shows its interaction with different proteins involved in cytoskeleton organization (Brajenovic et al., 2004).
Recently, it has been shown that the yeast homologues of MARK, the KIN1 and KIN2 kinases, interact with t-SNARE, Sec9 and the Lgl homologue Sro7, proteins which are involved in the final stage of exocytosis. It has also been shown that the conserved 42 amino acids at the carboxy terminal KA1 domain (Kinase associated domain 1) interact with the kinase catalytic domain and/or N-terminus and leads to kinase auto-inhibition (Elbert et al., 2005).
1.4 Ubiquitin binding domains
Ubiquitin is a regulatory protein which takes part in numerous biological processes, including targeted protein degradation, endocytic sorting, transcriptional control, intracellular localization and retroviral virion budding. This small 76 aminoacid protein is present in all eukaryotes and is highly conserved from yeast to humans. Nine ubiquitin binding domains are identified so far in different proteins (Hicke et al., 2005): CUE - coupling ubiquitin to endoplasmic reticulum degradation
UIM - ubiquitin interacting motif
NZF - Npl4 zinc finger motif
UBA - ubiquitin associated domain
UEV - ubiquitin conjugating Enzyme Variant
GAT - Gga and Tom1 domain
GLUE - GRAM-like ubiquitin-binding in Eap45 PAZ (ZnF-UBP) - polyubiquitin-associated zinc finger
VHS - Vps27, HRS, STAM
All of these domains are found in various proteins and tend to have a role in ubiquitin dependent pathways. Among these nine ubiquitin binding domains, the UBA domain is one of the best characterized domains from different proteins.
1.4.1 The ubiquitin associated domain
The UBA domain is a commonly occurring sequence motif of ~45 amino acids which was initially identified in proteins involved in ubiquitin/proteasome pathways, and later found in diverse proteins involved in the DNA excision-repair and cell signaling via protein kinases (Mueller and Feigon, 2002). The UBA domain was the first ubiquitin binding motif to be described. This domain was identified through sequence database searches as a moderately conserved ~45 residue sequence found in a variety of proteins (Hofmann and Bucher, 1996). So far, 127 UBA domains from 107 human proteins have been identified, most of them are implicated in the ubiquitin-proteasome degradation machinery (Chen and Madura, 2002).
UBA domains have been shown to bind mono-, di-, tri-, and tetra-ubiquitin in vitro but appear to bind to polyubiquitin with a higher affinity. It is thought that polyubiquitinated proteins represent the true in vivo binding substrates. Some UBA domains appear to homo- and heterodimerize and to bind other proteins like Vpr protein of human immunodeficiency virus type 1 (Dieckmann et al., 1998).
As well as having different affinities towards mono- and polyubiquitins, UBA domains are specific for the linkage of polyubiquitins. According to the ubiquitin linkage specificity, the UBA domains are divided into four classes (Raasi et al., 2005):
Class 1 UBA domains selective for lysine-48 linked polyubiquitin chains Class 2 UBA domains selective for lysine-63 linked polyubiquitin chains Class 3 UBA domains not binding to polyubiquitin chains
Class 4 UBA domains binding to any polyubiquitin chains.
p62 is a novel cellular protein which was initially identified in humans as a phospho tyrosine independent ligand of the src homology 2 (SH2) domain of p56lck, a member of the c-src family of cytoplasmic tyrosine kinases. In addition to the SH2 domain, p62 possesses several structural motifs, including a ubiquitin associated (UBA) domain that is capable of binding ubiquitin noncovalently. It has been shown that the UBA domain of p62 binds to various proteins that are involved in neurodegenerative disorders such as Alzheimer’s disease (Pridgeon et al., 2003). The important interacting proteins include, myelin basic protein, 14-3-3 zeta isoform, syntaxin binding protein, FK506 binding protein 14, homeobox protein Meis2, transketolase, heat shock cognate hsp70, reelin isoform b, CaMKII, Unc51 like kinase II and nuclear receptor co-repressor 1.
1.4.2 Structure of the UBA domain
There are several UBA domain structures from various proteins solved mainly by the NMR method. Interestingly, all these proteins share very low sequence similarity in their UBA domains, but form a very similar three helical fold structure (Fig. 1.5). Most of these structures have a conserved large hydrophobic surface patch which has been predicted to play a role in protein-protein interactions (Mueller et al., 2002).
Fig. 1.5: Structure of the UBA domain from HHR23A. (PDB code: 1IFY; Mueller et
al., 2002) a: Ribbon representation of UBA domain structure, coloured according to the secondary structure elements. b: Surface representation of the UBA domain using the following colour coding: red, acidic residues Glu and Asp; blue, basic residues Arg and Lys; orange, polar residues Asn, Gln, His, Ser and Thr; white, hydrophobic residues Ala, Gly, Phe, Ile, Pro, Met, Leu, Tyr and Val. The major accessible residues on the hydrophobic surface, Met173, Gly174, Tyr175, Leu199 and Ile202, are marked. The surface area of the hydrophobic surface patch is about 470 Å2, which corresponds to ~17% of the total surface area of about 2830 Å2. This Figure was prepared with Deep View Swiss-PDB Viewer (Guex and Peitsch, 1997) and POVray for Windows (Persistence of Vision Pty. Ltd. (2004), Persistence of Vision Raytracer Version 3.5, retrieved from http://www.povray.org/).
1.4.3 The UBA domain of MARK
Interestingly, MARK kinases are one of the sub-families of protein kinases containing a UBA domain downstream to their catalytic domain. The function of the UBA domain in the MARK kinases remains unclear. The presence of a UBA domain adjacent to the catalytic domain suggests potential interactions with diverse proteins involved in the ubiquitin proteasome pathway, DNA repair, or cell signaling (Brajenovic et al., 2004; Hofmann and Bucher, 1996).
Y175 M173 G174 I202 L199 C N a b α3 α2 α1 Y175 M173 G174 I202 L199 C N a b Y175 M173 G174 I202 L199 C N Y175 M173 G174 I202 L199 Y175 M173 G174 I202 L199 C N C N a b α3 α2 α1
1.5 Aim of the work
MAP/microtubule affinity regulating kinases (MARK) are a family of protein serine/threonine kinases which have been identified by their ability to phosphorylate the microtubule-associated proteins tau, MAP2 and MAP4. Phosphorylation of the neuronal MAP tau on serine262 dramatically reduces its MT binding capacity and leads to the formation of neurofibrillary tangles, which is a hallmark of Alzheimer’s disease. The homologues of MARK kinases in C. elegans and D. melanogaster play important roles in embryonic polarity and cell cytoskeleton regulations. All these functions make MARK an important drug target for Alzheimer’s disease. It has been proven that regulating a protein kinase activity is a helpful approach to treat many diseases like cancer etc. In order to develop an inhibitor specific to MARK, there is a need for a high resolution structure of MARK.
The aim of this study was therefore to elucidate the X-ray crystal structure of MARK kinase. Generally, protein kinases can attain different conformations (active, inactive and semi-active) depending upon their phosphorylation status, cellular localization and availability of partner molecules. This makes the analysis of the regulation of protein kinases a complex task. To understand this, the crystal structures in different conformations have to be determined. All protein kinase structures determined so far show a similar fold in their catalytic domains, but they are highly specific regarding their substrates. While the biochemical studies reveal many different substrates for MARK kinase, the reason for substrate specificity remains to be answered. The MARK kinases have different interacting partners, but for most of the interacting proteins the mode of interaction is yet to be identified. The goal of this study was to express, purify, crystallize and to solve the structure of MARK kinase in various conformations.
2 Materials and methods 2.1 Materials 2.1.1 Chemicals
All chemicals used were of the highest purity available (ACS grade) and were purchased from the following companies:
Amersham Pharmacia Biotech AppliChem
Acros Organics Fluka
Kodak Merck
New England Biolabs Qiagen
Sigma
In addition, the chemicals for crystallization were purchased from the following companies as pre-formulated screens or separate reagents:
Hampton Research Jena Biosciences Molecular Dimensions 2.1.2 Enzymes
All restriction enzymes used for DNA engineering and the T4 DNA ligases were purchased from the companies New England Biolabs, United States Biochemical and Stratagene. The BP-clonase and LR-clonase enzyme mixtures were purchased from Invitrogen.
2.1.3 Bacterial strains
Strain Genotype Features
XL2-Blue recA1 endA1 gyrA96 thi-1 hsdR17 supE44 ReIA1 lac[F’ proAB laclqZΔM15 Tn10 (Tetr) Amy Camr]α
Host for cloning and plasmid propagation (Stratagene). DH5α Library Efficiency F-φ80lacZ ΔM15 Δ(lacZYA-argF)U169 recA1endA1 hsdR17 (rkmk+)phoA supE44 thi-1gyr
A96 relA1 λ
Host for cloning and propagation of Gateway vectors (Invitrogen).
Library Efficiency DB3.1
F- gyrA462 endA1 (sr1-recA)
mcrB mrr hsdS20(rB-, mB-)
supE44 ara-14 galK2 lacY1
proA2 rpsL20(SmR) xyl-5 - leu
mtl1
The DB3.1 E. coli strain is resistant to ccdB effects and can support the propagation of plasmids containing the ccdB gene (Invitrogen).
BL21-AI F- ompT hsdS(rΒ -, mB- ) dcm
araB:T7RNAP-tetA
The T7 RNA polymerase gene is contained in the araB locus of the araBAD operon, allowing the regulation of the expression of the T7 RNA polymerase by the sugars L-arabinose and glucose. Glucose represses basal expression. The strain is suitable for high yield expression from T7 based expression vector (Invitrogen).
B834 F- ompT hsdS6(rB- mB-) gal dcm met
Methionine auxotropic cell strain, used for selenomethionine labelling (Novagen).
Table 2.1: Cell strains and feature list. The first column lists the names of the E. coli
strains used for cloning, vector propagation and protein expression; the second column contains the genotype and the third some remarks about the purpose of use and their features.
Table 2.1 shows the bacterial strains used in this study: XL2-Blue (Bullock, 1987) and library efficiency DH5α (Hanahan, 1983) were used as cloning hosts, BL21 (DE3) (Studier et al., 1990), library efficiency DB3.1 (Bernard and Couturier, 1992), BL21 AI (Studier, 1986, Lee et al., 1987) and B834 (DE3) (Wood, 1966) were used for protein expression.
2.1.4 Cloning Vectors
Most inserts used in this study were amplified by PCR and cloned into the pDONR201 vector of the recombination-based Gateway cloning system (2.2.9). All vectors used in this study are shown in Table 2.2.
2.1.5 Expression Vectors
The expression vectors used in this study for the high-yield expression of recombinant proteins carry cloned inserts under the control of the T7 promoter. Thus, only E. coli strains engineered to express the T7 RNA polymerase upon IPTG induction can be used for expression, e.g. BL21 (DE3). The Gateway vectors used in this study are of 2 types: the donor vector pDONR201 and the destination vectors pDEST15/17 (Table 2.2). The BL21 AI strain that was used for expression of Gateway expression vectors is arabinose inducible (araBAD promoter).
Table 2.2: Summary of the vectors used in this study.
2.1.6 Media
Antibiotics: All antibiotic solutions were dissolved either in water or ethanol
depending on the solubility of the corresponding antibiotic. The water soluble antibiotics were sterilized by filtering through a 0.22 µM sterile filter. The stock solutions were prepared with the following concentrations: Ampicillin-15 mg/ml in ddH2O, Kanamycin-25 mg/ml in ddH2O, Chloramphenicol-34 mg/ml in ethanol and
Carbenicillin-15 mg/ml in ddH2O.
Vector Expression System Features
pDONR201 Used for generation of
entry clones
KanR
pDEST15 E. coli AmpR, N-terminal GST tag
pDEST17 E. coli AmpR, N-terminal His6 tag
pET16b E. coli AmpR, N-terminal His10 tag
Luria-Bertani (LB) medium: 1% (w/v) bacto-tryptone, 0.5% (w/v) bacto-yeast
extract, 1% (w/v) NaCl; sterilized by autoclaving. Purchased from Life Technologies as dried powder.
LB agar plates: LB medium and 1.5 % (w/v) bacteriological agar. Sterilized by
autoclaving. Plates were poured when the temperature reached ~50°C.
SOB medium: 2% (w/v) bacto-tryptone, 0.5% (w/v) bacto-yeast extract, 0.5 % (w/v)
NaCl. A solution of KCl was added to a final concentration of 25 mM. The pH was adjusted to 7 with NaOH and the solution sterilized by autoclaving. Before use a sterile solution of MgCl2 was added to a final concentration of 0.1 M.
SOC medium: SOB medium supplemented with 1.8 % glucose.
M9 Minimal Medium: one litre of 5X stock of M9 medium was prepared by
dissolving 30g Na2HPO4, 15g KH2PO4, 5g NH4Cl and 2.5g NaCl in distilled water and
sterilized by autoclaving. To make one liter of M9 medium, 200 ml of 5X M9 salts, 1 ml of 1 M MgSO4, and 10 ml of 40% glucose were mixed and diluted up to one litre
with sterile water.
2.1.7 Crystallization
2.1.7.1 Crystallization supplies and tools
Crystallization supplies and tools including crystallization plates, siliconised cover slides (round and square slides of different thickness), sealing tape, forceps and tools for crystal manipulation were purchased from Hampton Research.
2.1.7.2 Crystallization solutions
Crystallization screens purchased from Hampton Research:
Crystal screen 1 Additive screen 1 Crystal screen 2 Additive screen 2 PEG/Ion screen Additive screen 3 Grid screen Na-Malonate Crystal screen Index
Grid screen PEG6000 Grid screen Ammonium sulfate SaltRx screen Crystal screen Lite
Crystallization screens purchased from Jena Biosciences:
JB High throughput Screen I and II
2.1.8 Equipment and accessories Centrifuges:
Cold centrifuge J2-21 M/E Beckman
Ultracentrifuge Beckman
Table centrifuge 5402 Eppendorf
Table centrifuge 5415C Eppendorf
Rotors:
JA-10, JA-20 Beckman
45Ti Beckman
Äkta purification system and corresponding accessories:
Mono-S column HR10/10 Pharmacia
HiLoadTM 16/60 SuperdexTM 200 Pharmacia
Ni-NTA Superflow Qiagen
GST-sepharose beads Amersham Biosciences
Other equipments and accessories:
French press cell Aminco
Gel dryer model 583 Bio-Rad
Incubator shaker model Innova 4330 New Brunswick Scientific UV/visible spectrophotometer Ultraspec1000 Pharmacia
Dynamic light scattering Firma Dierks and Partner
Scintillation counter Tricarb 1900 CA, Packard
2.2 Molecular biology and microbiological methods 2.2.1 Culture and storage of E. coli
Bacteria cells were grown on LB agar plates or in liquid LB medium at 37°C unless stated otherwise. For positive selection, media and plates were supplemented with the appropriate antibiotics in the following concentrations: 50 μg/ml ampicillin, 50 μg/ml carbenicillin, 25 μg/ml kanamycin. All media and manipulation tools were sterilized by autoclaving, or if heat-labile, by filtration through a 0.22 μm filter. For permanent storage at –80oC, cell strains were flash-frozen in liquid nitrogen after mixing with glycerol for cryoprotection. BL21 and XL2 blue strains were preserved in 30% (v/v) glycerol in LB; DH5α strain in 50% (v/v) glycerol in LB.
2.2.2 Transformation of E. coli strains
E. coli cells competent for transformation were either purchased from the companies as
listed in table 2.1 or prepared in the laboratory. XL2-Blue, DH5α library efficiency and BL21 AI cells were transformed by the heat-shock method: 20-100 ng of DNA was added to an aliquot of competent cells (~20-50 μl) previously thawed on ice for approximately 10 minutes. The mixture of DNA and competent cells was incubated on ice for 30 minutes. After the heat-shock at 42oC for 30 seconds, 200-400 μl of SOC medium was added and the cells were incubated at 37°C for 1 hour with shaking. Finally, 100-200 μl cells were plated on a selective medium plate and incubated overnight at 37oC.
2.2.3 Isolation of plasmid DNA
All plasmid mini-preparations were carried out with the Invisorb Spin Plasmid Mini Kit (Invitek) following the user manual. All midi-preparations were carried out with the Nucleobond AX Kit (Macherey-Nagel) according to the manual.
2.2.4 Determination of DNA concentration and purity
The concentration and the degree of purity of double stranded plasmid DNA was determined based on the Beer-Lambert Law by measuring the absorbance at 260 nm and 280 nm:
A260=ε260 c l and A260x 50= μg/ml (when l= 1 cm)
A260 is the absorbance at 260 nm, ε260 is the molar absorbtion coefficient, c is the molar
concentration and l is the optical path. For a protein-free and RNA-free solution of DNA the ratio of A260/ A280 should be close to 2. Protein contaminants would decrease
this ratio, whereas RNA contamination would increase it. To further estimate the concentration and purity of DNA preparations agarose gel electrophoresis was carried out.
2.2.5 Ligation reaction
The components below were mixed in a 500µl Eppendorf tube. The mix was incubated at 16°C overnight. The molar ratio between the digested vector and the digested insert was around 1:5. 10x Buffer 1µl Ligase (5U/µl) 1µl Digested vector 200 ng Digested insert 200 ng H2O x µl Total volume 10 µl
2.2.6 Restriction digestion analysis of DNA
DNA samples were analyzed with the use of restriction digestion enzymes (New England Biolabs). In each case the sample was mixed with the desired restriction enzyme/s and the proper reaction buffer (New England Biolabs) in a final volume of 20 μl. The mixture was incubated for 60 min at 37° C. The digested DNA was loaded immediately onto an agarose gel to check the result of the restriction digestion analysis.
2.2.7 DNA sequencing
DNA sequencing reactions were performed to confirm the sequence of a construct and the existence of mutations, especially after PCR amplification steps. The reactions were performed using fluorescent dye labelling and the Sanger Method (Sanger et al., 1977) in a Robocycler Gradient 96 PCR machine. The protocol for the temperature cycle reaction was:
Terminator ready reaction mix 8 μl
dsDNA 500 ng Primer (10pmol/ μl) 1 μl
ddH2O to a final volume of 20 μl
The PCR program for sequencing was: 1. Denaturation 96°C 10 sec
2. Annealing 45°C 5 sec X 30 cycles 3. Elongation 60°C 4 min
After the reaction, the DNA was precipitated by using the Pellet Paint NF Precipitant (Novagen). To the 20 μl reaction mixture, 1 μl Pellet Paint NF Co-Precipitant and 80 μl of 75% ethanol were added. The contents were mixed by vortexing, and centrifuged at 13 krpm for 10 min at RT. Then the DNA pellet was washed with 250 µl of 70% ethanol to remove any trace of salts in the DNA pellet and centrifuged at 13krpm, for 10 min at RT. The pellet was then air dried and resuspended in 30 μl of HPLC-grade dd H2O.
The ABI PRISM 310 Genetic Analyser (PE Applied Biosystems) was used to sequence the DNA. The sequencing results were analyzed with the VectorNTI software (Informax).
2.2.8 Mutagenesis of DNA
All mutations described in this thesis were created by site-directed mutagenesis, which was performed using the Quick Change Site-Directed Mutagenesis Kit (Stratagene). The method utilizes the Pfu Ultra High Fidelity polymerase to replicate the parental plasmid by using two synthetic oligonucleotide primers containing the desired mutation.
The primers, each complementary to opposite strands of the vector, are extended during temperature cycling by the DNA polymerase. Temperature cycling generates copies of the plasmid by linear amplification, incorporating the mutation of interest. The temperature cycling reaction was performed as follows:
10x Pfu Ultra High Fidelity buffer 2 μl ds DNA template (25ng/ μl) 5 μl
dNTPs (2.5 mM) 2 μl
Primer 1 (0.5 pmoles/μl) 1 μl Primer 2 (0.5 pmoles/μl) 1 μl Pfu polymerase (2.5 U/μl) 0,5 μl ddH2O to a final volume of 20 μl
The temperature cycling program used was:
Step Time Temperature Cycles
Initial denaturation 30 seconds 95°C 1 Denaturation 30 seconds 95°C Annealing 1 minute 55 – 58°C Extension 12 minutes 68°C 16
The primer annealing temperature was calculated according to the melting temperature (Tm) of the primers and the extension time was calculated according to the length of
the plasmid. Next, a treatment with DpnI endonuclease was carried out to digest the parental methylated DNA template, allowing the selection of the newly synthesized DNA containing the mutation. 2-4 μl of this reaction mixture was used to transform XL2-Blue cells. Alternatively, some of the mutants were created by cutting and ligating the mutated DNA fragment from the full length mutant clones.
2.2.9 Gene cloning using the Invitrogen Gateway cloning technology
The Gateway cloning technology is a method that enables rapid cloning of a gene in various expression systems (Invitrogen). Gateway uses a well-characterized lambda phage site-specific recombination system, thus restriction enzymes and ligases are not required in any step. Two reactions, ‘BP-reaction’ and ‘LR-reaction’, constitute the Gatewaycloning technology.
Reaction Reaction sites Product Product Structure
BP reaction attB x attP Entry clone attL1-gene-attL2
LR reaction attL x attR Expression clone attB1-gene-attB2
Table 2.3: Summary of reactions and nomenclature of the Gateway cloning technology
First, an entry clone is generated from a PCR product that spans the attB recombination sequences (BP reaction, Fig.2.1). Once a positive clone is verified and sequenced the second step is to transfer the gene of interest to a variety of expression vectors (LR reaction; Fig.2.1), featuring different tags and/or different expression systems (e.g. E.
coli, insect cells). The ccdB gene interferes with the E. coli DNA gyrase. Thus, every
cell that takes up an unreacted vector that still carries the ccdB gene or a by-product, will fail to grow.
Generation of MARK2 constructs with the Gateway technology
The first step to enter the system was to obtain the proper PCR product with the required attB recombination sequences. In this study, this was achieved in two steps, with a set of primers specific for each construct and a set of primers for the completion of the attB sites. The resulting DNA sequence would be attB1 – TEV cleavage site –
fragment of interest – attB2.
Fig. 2.1: Overview of the Gateway cloning technology.
The encoding sequence of the TEV cleavage site is incorporated in order to be able to remove the tag after purification of the relevant protein for crystallization purposes. The primers were designed as follows:
Primers to amplify the ORF:
attB1-Tev-f (forward primer)
5’-AAAAAGCAGGCTTC GAAAACCTGTATTTTCAGGGC- coding sequence-3’
attB1 (12 bases)- 2 bases for frame - TEV cleavage site attB2-stop-r (reverse primer)
5’-AGAAAGCTGGGTCTTA – coding sequence-3’ AttB2 (13 bases) - stop codon
BP reaction to generate an entry clone
LR reaction to generate an expression clone BP reaction to generate an entry clone BP reaction to generate an entry clone
LR reaction to generate an expression clone LR reaction to generate an expression clone
Adapter primers to generate complete attB sites:
AttB1f (forward primer)
5’-GGGGACAACTTTGTACAAAAAAGCAGGCT-3’ AttB2r (reverse primer)
5’-GGGGACCACTTTGTACAAGAAAGCTGGGT-3’
(The underlined sequences show the parts that overlap in both gene specific and adaptor primers)
PCR reactions
The first PCR step was done with primers that do not have very long non-specific overhangs in order to obtain a high success rate in amplification. The attB sites that are needed, were generated with a second PCR step using the adapter primers.
1st PCR
Template DNA 200 ng
10x reaction buffer 5 μl
Forward primer 10 pmoles
Reverse primer 10 pmoles
dNTPs Mix 5 mM
H2O up to 50 µl
The PCR program used was as follows:
Step Time Temperature Cycles
Initial denaturation
1 minute 95°C 1x
Denaturation 15 seconds 95°C
Annealing 30 seconds Depends on each
primer
Extension 30 seconds 68°C
10x
First PCR product 10µl 10x reaction buffer 5 μl Forward primer (attB1f) 40 pmoles Reverse primer (attB2r) 40 pmoles
dNTPs Mix 5 mM
H2O up to 50 µl
The cycling parameters for the second PCR were:
Step Time Temperature Cycles
Initial denaturation 1 minute 95°C 1x Denaturation 15 seconds 95°C Annealing 30 seconds 45°C Extension 30 seconds 68°C 5x Denaturation 15 seconds 95°C Annealing 30 seconds 55°C Extension 30 seconds 68°C 20x
The PCR products were treated with DpnI in order to digest the template plasmid. PCR product 50 µl
10x DpnI buffer 5 µl
DpnI 10 units
30 minutes at 37oC (digestion reaction) 15 minutes 65 oC (DpnI heat inactivation)
The PCR products were then analyzed by agarose gel electrophoresis and purified by using the pellet paint protocol (refer to section 2.2.7).
BP reaction:
Purified PCR product 200 ng pDONR201 (150 ng/μl) 2.5 μl 5x BP reaction buffer 5 μl BP-clonase enzyme mix 2 μl
ddH2O up to 25 μl
The mixture was incubated at 25oC overnight and then 2 μl of proteinase-K was added to stop the reaction. Next, 1-3 μl were used to transform DH5α library efficiency E.
coli cells. The clones were analyzed by setting up a double digestion with specific
enzymes. Positive clones were then sequenced (section 2.2.7) using the primers attL1-f (proximal to attL1) and attL2-r (proximal to attL2) and gene specific primers (Appendix). When a positive clone was confirmed for correct sequence, it was used for setting up an LR reaction in order to generate an expression clone.
LR reaction: LR buffer 5x 4 μl Destination vector (150 ng/ μl) 3 μl Entry clone (150 ng/ μl) 1 μl Topoisomerase I 0.5 μl LR clonase mix 2 μl ddH2O up to 20 μl
The topoisomerase was added to relax the entry clone plasmid as a supercoiled state of entry vectors was often observed in agarose gel electrophoresis. The supercoiled state of DNA lowers the efficiency of LR clonase mix.
The mixture was incubated at 25oC overnight and then 2 μl of proteinase K was added to stop the reaction. 1-2 μl of this DNA product was transformed into either DH5α library efficiency competent cells for plasmid propagation, or BL21 AI cells for expression of the recombinant protein.
2.3 Protein methods
2.3.1 SDS-Polyacrylamide gel electrophoresis (SDS-PAGE)
SDS-PAGE was performed in the lab following a modified protocol (Matsudaira and Burgess, 1978; Laemmli, 1970). The stacking gel was 4% acrylamide and the separating gel was 10% or 17% (Table 2.4). Protein samples were diluted 1:1 with 2x SDS-PAGE loading buffer (Laemmli, 1970), and heated for 2 min at 95°C. Electrophoresis was carried out at 150V and maximal 35mA in SDS-PAGE running buffer (25 mM Tris-HCI, 190 mM Glycine, 0.1% (w/v) SDS). The gels were then stained in a 0.1% (w/v) solution of Coomassie brilliant blue R-250, 45% (v/v) methanol and 9% (v/v) acetic acid for 20 min on a shaking platform. Next, the gels were destained in an intensive destaining solution (50 % (v/v) methanol, 10% (v/v) acetic acid) for 20 min and for a minimum of 1 hour in a normal destaining solution (5% (v/v) methanol, 7.5% (v/v) acetic acid).
Molecular weight marker proteins (Biofermentas) were:
Protein name Molecular Weight (in kDa)
ß-Galactosidase 116
Bovine serum albumin 66.2
Ovalbumin 45.0
Lactate-dehydrogenase 35.0 Restriction endonuclease Bsp98I 25.0
Lactoglobulin 18.0
Table 2.4: Solutions for preparing SDS-PAGE (volumes are in ml) 2.3.2 Protein expression and purification
The following strategy was applied for every new expression construct: First, 5 ml LB supplemented with the appropriate antibiotics was inoculated with the desired expression strain and grown at 37°C overnight. This pre-culture was used to inoculate a new 100 ml LB culture, which was left to grow at 37°C, shaking at 280 rpm until the optical density was reached to 0.6. A sample of 1 ml was centrifuged and kept as a un-induced control and the rest was un-induced with IPTG to final concentration of 0.5 mM or with arabinose to a final concentration of 0.2 %, depending on the E. coli cell strain used for expression. The cultures were left to grow at lower temperature 25-30°C for overnight. Cells were then harvested by centrifugation at 8 krpm for 5 min (Eppendorf 5810R) and resuspended in lysis buffer.
2.3.3 Cell lysis and solubility test
Cells were lysed using a French press (Aminco). First, cells were thoroughly resuspended in lysis buffer (2-4 ml of lysis buffer/100ml culture) and then they were transferred to a French press cell. Application of 20,000 PSI in two rounds ensured the lysis. Lysates were kept on ice and centrifuged at 14 krpm for 20 min at 4°C. A sample
Separating gel Stacking gel
Components 10% 17% 4% 40 % Acrylamide / Bis-acrylamide (37.5:1) 15.0 25.6 5.4 Tris HCl (1M pH 8.8) 22.0 22.0 - Tris HCl (0.25 M pH 6.8) - - 27.0 10%SDS 0.6 0.6 0.54 TEMED 0.12 0.12 0.108 10% APS 0.065 0.065 0.15 H2O 22.0 11.4 20.9
of the supernatant in which all soluble proteins were present and the pellet resuspended in lysis buffer were loaded onto an SDS gel together with samples before and after induction. In this way expression and solubility of the desired protein was tested in a first approach. For large-scale production of proteins, cultures of 2-6 litres were grown as described above.
2.3.4 Chromatography
Purification of proteins was performed by fast performance liquid chromatography (FPLC) using Äkta purifier and Äkta explorer machines (Pharmacia).
2.3.4.1 Ni-NTA affinity chromatography
Immobilised metal affinity chromatography (IMAC) makes use of the binding properties of metals towards proteins for purification purposes; nickel-nitriloacetic (Ni-NTA) resin (QIAGEN) contains chelated nickel, which is able to specifically bind to stretches of polyhistidine in proteins. Most expression systems include a tag of six histidines either at the N- or at the C-terminus or on both.
The resin was cast on a Pharmacia self-packed XK26 column or a batch protocol was performed with the use of Biorad columns. The material was rinsed with ddH2O to
remove the 20% (v/v) ethanol preservative and equilibrated with the appropriate loading buffer. The buffers should not contain DTT or other reducing agents in high concentrations, as this might strip the nickel from the resin. Once the column was equilibrated, the sample was passed twice over the column. After loading, the column was washed with loading buffer (~10-20 column volumes) and finally, the protein was eluted with 3 volumes of elution buffer. All buffers are listed in the Appendix.
2.3.4.2 GST affinity chromatography
Glutathione Sepharose (Pharmacia) is an agarose material coupled with glutathione, which is frequently used for purification of GST tagged proteins. Glutathione Sepharose resin was self-packed in a column (Pharmacia) or prepacked 5 ml GST Hi-Trap columns (Pharmacia) were used. The column was rinsed with 10 volumes of
ddH20 and 10 volumes of loading buffer. The sample was loaded onto the column with
a low flow rate of 0.2 ml/min and passed twice over the column. Next, the column was washed with 7-10 volumes of loading buffer. Elution was achieved with 4 volumes of GST elution buffer. Eluate fractions were pooled together and kept at 4oC for further processing.
2.3.4.3 Anion and cation exchange chromatography
Protein separation by ion exchange chromatography depends on the reversible adsorption of charged molecules to an immobilised ion exchange group of opposite charge. Varying conditions such as ionic strength and pH can control these interactions. To ensure electrostatic binding, the total ionic strength needs to be low. Generally, 100mM NaCl was included into the anion exchange buffers. The columns for anion exchange chromatography (AIEX) and cation exchange chromatography (CIEX) were MonoQ HR 10/10 and MonoS HR 10/10 (Pharmacia) respectively. After equilibration with 5 column volumes of the AIEX/CIEX buffer A, the sample was loaded and the washing step followed with 5-7 column volumes of AIEX/CIEX buffer A. The elution was performed with a linear gradient of AIEX/CIEX buffer B in two steps: first from 0 to 60% in 5-8 column volumes and then to 100% in 1-2 column volumes. Eluted fractions containing the protein of interest were pooled together.
2.3.4.4 Gel filtration chromatography
On a gel filtration (or size exclusion) column the molecules are separated according to differences in their sizes. The concentrated protein solution is injected into a 1 ml loop with a injection needle (Pharmacia). Small molecules which can diffuse into the pores of the gel beads are delayed in their passage through the column in contrast to the larger molecules, which cannot diffuse into the gel beads. The larger molecules thus leave the column first, followed by the smaller ones in order of their sizes. Gel filtration was performed with a Superdex G-200 HR 16/60 column (Pharmacia).
2.3.5 Determination of the protein concentration
Protein concentrations were estimated by using the Bradford method (Bradford, 1976). The assay is based on the observation that the absorbance maximum for an acidic solution of Coomassie Brilliant Blue G-250 shifts from 465 nm to 595 nm when binding to the protein occurs. Bovine serum albumin solution is used as a standard. 1µg to 5 µg of BSA solution in 10µl were used to calculate a standard curve. The test proteins were measured in two different dilutions. To these samples 200µl of Bradford reagent (Bio Rad) were added and mixed. The absorbance at 595 nm was measured in a Microtitre plate reader (BioLynx 2.2) and the protein concentrations were calculated from the standard curve.
2.3.6 Concentrating the protein solution
The protein solutions were concentrated using the Amicon (Millipore) device. In this device, a membrane with a molecular weight cut-off smaller than the protein of interest is placed at the bottom of a cell, which is filled with the protein solution. The cell is then placed in a 50 ml Falcon tube and centrifuged at 3000g. In this way, the flow through contains only lower molecular weight components, while the protein is concentrated in the chamber.
2.3.7 Protein kinase assay
Kinase activities were assayed in 50 mM Tris-HCl pH 7.5, 5 mM MgCl2, 2 mM
EGTA, 0.5 mM PMSF, 0.5 mM DTT and 0.5 mM benzamidine for 30 minutes at 30°C. Final concentration of [32P] ATP (3.7*107 MBq/mol; Amersham Biosciences) and substrate peptide were 100 μM. The substrate peptide derived from the first repeat of tau protein containing S262 in the KXGS motif (TR1-peptide NVKSKIGSTENLK, Drewes et al. 1997). Reactions were stopped by addition of half the volume of 30 %(w/v) TCA. After centrifugation, the supernatant was applied to phosphocellulose-paperdiscs, washed with phosphoric acid (0.1 M), dried by air and radioactivity was measured in a scintillation counter (Tricarb 1900 CA, Packard).
2.3.8 Limited proteolysis
Limited proteolysis is one of the novel techniques to identify the structurally folded part of the protein (Fontana et al., 2004). In a number of studies it has been demonstrated that the sites of limited proteolysis along the polypeptide chain of a protein are characterized by enhanced backbone flexibility, implying that proteolytic probes can pinpoint the sites of local unfolding in a protein chain. Limited proteolysis was used to analyze the partly folded (molten globule) states of several proteins, such as apomyoglobin, lactalbumin, calcium-binding lysozymes, cytochrome C and human growth hormone.
The N-terminal GST tagged MARK2 construct (GST-MARK2 [1-364]) was subjected to
limited proteolysis with different proteases like trypsin, chymotrypsin, GluC, AspN, and thermolysin in a ratio of 200 to 1. 100 µg of purified MARK protein was mixed with 0.5 µg of corresponding proteases in the protease reaction buffer (Appendix). This reaction mixture was incubated at 37ºC and samples were taken at different time points (10, 30, 60, 120 min). The reactions were quenched by adding 1mM PMSF and boiling the samples with 2x SDS loading buffer. As a control, a sample was incubated at 37ºC for 120 minutes to check for the heat stability of the protein. The results were analyzed by SDS-PAGE and further with N-terminal amino acid sequencing and mass spectrometry.
2.3.9 N-terminal aminoacid sequencing
N-terminal sequencing for MARK2 stable fragment was performed with Procise-cLC (ABI-Perkin Elmer) protein Sequencer. The MARK2 protein (GST-MARK2 [1-364])
was digested with trypsin in a ratio of 200 to 1 and the stable fragment was purified by gel filtration chromatography. This purified protein was applied on to a PVDF membrane and subjected to 5 cycles of sequencing reactions.
2.3.10 Mass spectrometry
Mass spectrometry analysis was performed with a SELDI Mass-Spectrometer PBS-I (Ciphergen, USA). 2µl of the protein sample were mixed with equal amount of matrix
(Sinapinic Acid in 50% acetonitrile and 0.5% trifluoric acid). From this mixture, 3µl were loaded on to a H4 Protein Chip (Ciphergen, USA) in two times and each time the chip was dried for ~15 minutes in the hot air oven. Mass spectrometry analysis was used to find the molecular weights of stable fragment of MARK2 and selenomethionine labelled proteins.
2.3.11 Detailed protocols on purification and crystallization of MARK2 wild type and mutant proteins
2.3.11.1 Expression
Proteins (wild type and mutant proteins) were expressed in E. coli strain BL21 AI (Invitrogen). Cells were induced overnight at 24°C by adding arabinose to a final concentration of 0.2% at OD600 ≈ 0.6. Cells from a litre culture was harvested by
centrifugation and resuspended in 40 ml lysis buffer (50mM Hepes pH 7.2, 300 mM NaCl, 5% glycerol) and supplemented with one tablet of EDTA-free complete protease inhibitor cocktail (Roche). Resuspended cells were lysed by passing two times through a French press cell (Aminco).
2.3.11.2 Purification
The expressed proteins were purified in four steps: Ni-NTA affinity chromatography, TEV protease cleavage to remove the His-tag, ion exchange chromatography and gel filtration chromatography. Clarified lysate were applied to a self-packed Ni-NTA affinity column of 5ml Ni-NTA beads (Qiagen). The protein was eluted with a 0 to 1000 mM gradient of imidazole in buffer A (50 mM Hepes pH 7.2, 300 mM NaCl and 5% glycerol). Pure protein fractions were pooled and mixed with purified TEV protease in a ratio of 1:20 to cleave off the His-tag. The protein mixture was dialyzed overnight against buffer B (50 mM Hepes pH 7.2, 200 mM NaCl, 5% glycerol, 1 mM EGTA, 1 mM DTT). Around 1 ml Ni-NTA beads were added to the dialyzed protein to remove the His-tagged TEV protease and the uncleaved MARK2 protein. After one hour, the Ni-NTA beads were removed and the salt concentration was reduced to 100 mM by dilution with salt free buffer (50 mM Hepes pH 7.2 and 5% glycerol). The
proteins were further purified by MonoS cation exchange chromatography (MonoS HR 10/10 column, Amersham Biosciences) using a 100 to 1000 mM NaCl gradient . Pooled fractions were concentrated using Ultrafree-30 concentrators and applied to a gel filtration column (Hiload 16/60 Superdex G200, Amersham Biosciences) equilibrated with 50 mM Bis-Tris pH 6.5 (H2SO4), 250 mM NaCl, 5% glycerol). The
pure protein fractions were pooled and concentrated to ~20 mg/ml. The concentrated protein solution was then aliquoted in 0.2 ml PCR tubes and shock frozen in liquid nitrogen.
2.3.11.3 Selenomethionine labelling
Selenomethionine labelled protein of the double mutant MARK2 was prepared by expression in methionine auxotrophic E. coli strain B834 (Invitrogen) (Table 2.1) using M9 minimal medium, supplemented with all amino acids except methionine that was substituted by selenomethionine (Acros Organics) (40 mg per litre of medium). The purification procedure was essentially the same as for the unlabelled proteins. The percentage of incorporation of selenomethionine was estimated by mass spectrometry.
2.4 Crystallographic methods 2.4.1 Crystallization techniques
A variety of methods exist to crystallize biological macromolecules. All of them aim to bring the protein solution into a supersaturated state. Among them, vapour diffusion is the most widely used method that also has been used in this study. A droplet containing the protein solution to be crystallized, buffer, crystallizing agent and additives is equilibrated against a reservoir containing a solution of crystallizing agent at a higher concentration than in the droplet. Equilibration proceeds by diffusion of the volatile species (water or organic solvent) until the vapour pressure in the droplet is equal to the one in the reservoir. If equilibration occurs by water exchange (from the drop to the reservoir), it leads to a decreasing volume of the droplet. This method was used in this study in two variations, either as hanging or as sitting drops (Fig. 2.2).
Crystallization solutions and supplies are described in the material section 2.1.7. Prior to each crystallization experiment, the highly concentrated protein solution (10-20 mg/ml) was centrifuged in an Eppendorf tabletop centrifuge at 13 krpm for 10 minutes at 4oC, in order to separate precipitates. First screens were set up in Crystal Quick 96 well sitting drop plates (Hampton Research) by mixing 1 µl of protein with 1µl of reservoir solution. During the optimization trials, the hanging drop method was preferred.
Fig. 2.2: Set up of crystallization trials using: a) the sitting or b) the hanging drop
vapour diffusion technique.
For each trial 2 μl of protein solution was mixed with 2 μl of reservoir. The plates were sealed with tape in the case of sitting drop and with a siliconised cover slip in the case of hanging drop trials and kept at 20°C and/or at 4°C. The plates were examined each day during the first week and 2 times a week during the first 2 months. The initial screens were performed by the use of the commercial screens described in section 2.1.7.2.
Occasionally, the technique of seeding crystals was performed in an attempt to improve the size and quality of micro crystals, irregular-shaped crystals or to avoid excessive nucleation. First, a drop with existing micro crystals was transferred to an Eppendorf tube, diluted with the proper buffer and a small plastic ball was added. Vortexing this solution resulted in breaking the nuclei/crystals and made it possible to transfer a small portion of them by a horse tail hair to new drops of lower concentration of protein and/or precipitant agent. This technique can lead to a slower growth thus yielding bigger crystals.
a b
2.4.2 Cryoprotection of crystals
The high intensity of synchrotron radiation can lead to radiation damage of the protein crystals. The interaction between the beam and the crystal generates thermal disorder and may even break bonds within the protein. A common method to prevent this radiation damage is freezing the crystals in a stream of cold nitrogen and collect the diffraction data at low temperature (Garman and Schneider, 1997). But the high solvent content (~50%) of protein crystals can lead to ice crystal formation during freezing and these ice crystals will destroy the crystal integrity and disturb the protein diffraction. This problem can be overcome by soaking the crystal in a cryoprotectant solution which maintains the crystal quality and prevents the formation of ice crystals (Garman and Schneider, 1997). MARK2 crystals were soaked in different cryoprotecting agents like glycerol, MPD and ethylene glycol and tested for the diffraction quality.
2.4.3 Data collection
X-ray diffraction data were collected using synchrotron radiation at the beamline of the X13 Consortium for Protein Crystallography at HASYLAB (DESY, Hamburg). X13 is a monochromatic, fixed-wavelength beamline, set at 0.802 Å and other characteristic features of this beam line are given below:
Beamline X13
Institute DESY, EMBL and University of Hamburg, Hamburg
Wavelength 0.802Å
Optics Triangular cut Si (111) monochromator crystal
Mirror continuous bent Rh-coated focusing mirror
Detector MAR300 CCD
2.4.4 Methods of Phase determination
The determination of the three-dimensional structure of macromolecules using X-ray crystal diffraction techniques requires the measurement of amplitudes and the calculation of phases for each diffraction point. Amplitudes |F (h,k,l)| can be directly measured from diffracting crystals, phases α (h,k,l) have to be determined indirectly. Thus, methods were developed to calculate phases. These include molecular replacement (MR), multiple isomorphous replacement (MIR) and multi wavelength anomalous dispersion (MAD) method.
Molecular Replacement (MR)
The molecular replacement method makes use of a known three-dimensional structure of a homologous protein as an appropriate starting model to provide initial phases for the unknown structure. Protein kinases normally have a similar fold in their catalytic domain. Therefore, the crystal structures of other protein kinases can be used as a search model for solving the MARK2 crystal structure.
Multiple Isomorphous Replacement (MIR)
Isomorphous replacement requires the introduction of high atomic number elements (heavy atoms), such as mercury, platinum, uranium, and so forth, into the macromolecule without disrupting its structure of packing in the crystal. Thus, a perfect isomorphous derivative is one in which the only change between it and the native crystal is the incorporation of one or more heavy atoms. This is commonly done by soaking crystals of native molecules in a solution containing the desired heavy atom. 1 mM stock solutions of various heavy atom salts were prepared with the cryoprotectant solution, and the MARK2 crystals were transferred to these solutions, soaked for few hours, and tested for heavy atom incorporation.
Multiple-wavelength anomalous dispersion (MAD)
Multiple-wavelength anomalous dispersion method utilizes the property of heavy atoms to absorb X-rays of specific wavelength which causes the anomalous scattering or anomalous dispersion. One common way to use MAD is to introduce selenomethionine (SeMet) in place of methionine residues in a protein. The selenium
atoms (which replace the sulfur atoms) have a strong anomalous signal at wavelengths that can be obtained from synchrotron X-ray sources. Selenomethionine labelled protein of the MARK2 was prepared by expression in methionine auxotrophic E. coli strain B834 (Table 2.1) and crystals were grown by using the wild type protein crystallization conditions.
