Characterisation of plasmodium falciparum pfnek3, an atypical MAP kinase activator

Two copies of Plasmodium falciparum MAPKs Pfmap1 and Pfmap2 have been identified and are believed to influence parasite proliferation.. Hence, the presence of alternative upstream MAPK

CHARACTERIZATION OF PLASMODIUM FALCIPARUM PFNEK3,

AN ATYPICAL ACTIVATOR OF A MAP KINASE

LYE YU MIN (B.Sc (Hons.), NUS)

M ASTER OF S CIENCE

N ATIONAL U NIVERSITY OF S INGAPORE

2007

me, without whom, there would not have been two journal publications with a position for my authorship

I am indebted to Maurice Chan, my mentor Maurice has always graciously taken extra steps to assist and advise me I am also grateful to Dr Doreen Tan for invaluable assistance during immunofluorescence analyses Appreciation goes to Jasmine and Madam Seah for many fruitful discussions and technical help It has been enriching working with Jason and Wenjie They have assisted me on countless occasions Huiyu is

an ardent supporter of this work and a fantastic listener too Chun Song has the ability to remain helpful and calm in the most testing situations for which I am thankful

I am indebted to Prof Christian Doerig (Univ of Glasgow) for providing the plasmids carrying Pfmap1 and Pfmap2 The pXJ40 plasmid was a gift from Prof E Manser (Centre for Molecular Medicine, Singapore) Technical expertise for analytical gel

filtration was sought from Mr Lam Kin Wai MR4 provided the Plasmodium falciparum

3D7 parasites contributed by Dr D.J Carucci Mr Robin Philp, Ms Ee Kim Huey and Mr Li Rong from Protein Analytics Lab (Biopolis Shared Facilities) were extremely accommodating with the interpretation of mass spectra

I would like to thank my family and friends for their understanding, encouragement, support and advice

CONTENTS

Acknowledgements……….…… i

Contents……….ii

Abstract……… vi

List of publications……… viii

List of tables……… ix

List of figures……… x

Abbreviations……… xii

Symbols……… xiii

1 Introduction 1

2 Literature review 5

2.1 The canonical MAPK cascade 5

2.2 The unusual kinome of the malaria parasite 6

2.3 The P falciparum genome encodes two MAPKs 9

2.4 Rodent Map2 is male!specific and necessary for the sexual cycle 10

2.5 MAPKKs upstream of Pfmap1 and !2 remain elusive 11

2.6 Current thoughts 12

3 Materials and Methods 14

3.1 Bio!computation 14

3.2 Use of Escherichia coli for cloning and expression 16

3.2.1 Growth and maintenance of E coli 16

3.2.2 Preparation of electrocompetent E coli 16

3.2.3 Transformation of E coli 17

3.3 Use of Saccharomyces cerevisiae for yeast!two!hybrid studies 18

3.3.1 Growth and maintenance of S cerevisiae 18

3.3.2 Preparation of competent S cerevisiae AH109 18

3.3.3 Yeast transformation 19

3.4 Mammalian cell culture 20

3.4.1 Growth and maintenance of mammalian cell lines 20

3.4.2 Mammalian cell transfection 21

3.5 Manipulation of the Pfnek3 gene 21

3.5.1 Pfnek3 gene isolation and amplification 21

3.5.2 Obtaining deletion!constructs for the purpose of Y2H screens 26

3.5.3 Generating mammalian!two!hybrid (M2H) fusion plasmid constructs 27

3.6 Conventional DNA ligation 28

3.7 Gateway™!based cloning 28

3.8 Site!directed mutagenesis of Pfnek3 and Pfmap2 30

3.9 DNA sequencing 35

3.10 Recombinant protein production 36

3.10.1 Harvesting cell!free extracts 36

3.10.2 Affinity purification of GST! and His!tagged proteins 37

3.11 Estimation of protein concentration by Bradford assay 37

3.12 SDS!PAGE 38

3.13 Mass spectrometric identification of proteins 39

3.14 Analytical gel filtration 40

3.15 Protein adsorption 41

3.16 Protein kinase assay 41

3.16.1 Validation of the ELISA!based kinase assay 42

3.17 Western blot 43

3.18 Antibody production 43

3.19 Immunofluorescence microscopy 45

3.20 Yeast!two!hybrid protein interaction studies 46

3.20.1 Principles 46

3.20.2 Testing the bait fusion protein for auto!activation and toxicity 47

3.20.3 Total cDNA library synthesis, amplification and fractionation 48

3.20.4 Constructing and screening a two!hybrid library 52

3.20.5 Prey plasmid rescue and identification by DNA sequencing 53

3.20.6 Re!testing the interaction by small!scale transformation 54

3.20.7 "!galactosidase reporter assay 54

3.21 Mammalian!two!hybrid protein interaction studies 55

3.21.1 Principles 55

3.21.2 M2H cell transfection 57

3.21.3 Normalized SEAP activity assay 59

4 Results and Discussion 62

4.1 Bio!computational identification of Pfnek3 62

4.2 Molecular cloning of FL! and TR!Pfnek3 67

4.3 Construction of plasmids for bacterial expression 69

4.4 Construction of plasmids for yeast!two!hybrid studies 70

4.5 Construction of plasmids for mammalian!two!hybrid studies 72

4.6 Generating kinase!inactive mutants of Pfnek3 and Pfmap2 74

4.7 Bacterial expression of recombinant proteins 75

4.8 Ensuring the expression of GST!Pfmap2 76

4.9 Kinase activity of recombinant Pfnek3 79

4.10 Pfnek3 directionally phosphorylates Pfmap2 82

4.11 Pfnek3 stimulates Pfmap2 kinase activity 84

4.12 Pfnek3 autophosphorylates in the presence of Pfmap2, active or inactive 86

4.13 Pfnek3 stimulates Pfmap2 but not Pfmap1 or hMAPK1 86

4.14 Determining Pfnek3 localization using a surrogate assay 88

4.15 Expression of endogenous Pfnek3 91

4.16 Total cDNA library construction for yeast!two!hybrid studies 93

4.17 Protein partners of Pfnek3 identified from yeast!two!hybrid 94

4.17.1 The advantages and limitations of Y2H assays 100

4.18 Mammalian!two!hybrid confirmation of protein interactions 106

4.19 Future challenges: unraveling the malarial MAPK cascade 109

4.19.1 Current strategies for kinase substrate identification 109

4.19.2 A dual!component strategy to decipher the phosphoproteome network of the

malaria parasite 115

4.19.2.1 The in vitro component 115

4.19.2.2 The in vivo component 119

4.20 Outlook 124

5 References 126

Appendices

Appendix I: Publications

Appendix II: Media, buffers and reagents

Appendix III: Vector maps

ABSTRACT The canonical mitogen!activated protein kinase (MAPK) signal cascade was

previously suggested to be atypical in the malaria parasite Two copies of Plasmodium

falciparum MAPKs (Pfmap1 and Pfmap2) have been identified and are believed to

influence parasite proliferation However, the regulators and substrates of malarial MAPKs have remained elusive Hence, the presence of alternative upstream MAPK kinases and substrates in the malaria parasite is a tantalizing research question

To address this issue, available transcriptome datasets were scrutinized to

identify candidate Plasmodium MAPK regulators by comparing the transcriptional

profiles of numerous protein kinases As a result, a candidate kinase was identified to

possess transcriptional activities similar to both Pfmap1 and Pfmap2 This gene!of!

interest, named Pfnek3 (P falciparum NIMA!like kinase 3), is a homologue of the NIMA (Never in Mitosis, Aspergillus) mitotic kinase family

Immunofluorescence data indicated that endogenous Pfnek3 was expressed during late asexual to gametocyte stages Sequence analyses unveiled unusual kinase sequence motifs For instance, the lack of an ATP!binding glycine triad, a common

feature of many eukaryotic kinases, might have contributed to the absence of a strong in vitro kinase activity Moreover, the phylogenetic distance of Pfnek3 from mammalian

NIMA!kinases concurs with its preference for manganese, rather than magnesium, as a cofactor Recombinant Pfnek3 exists primarily as monomers, unlike the majority of NIMA kinases, which are dimeric

Recombinant Pfnek3 was able to phosphorylate and stimulate Pfmap2, a malarial MAPK known to be necessary for the completion of the male sexual cycle Contrastingly, this was not observed with two other MAPKs, Pfmap1 and human MAPK1, suggesting that the Pfnek3!Pfmap2 interaction may be specific for Pfmap2 regulation and possibly playing a role in gametocyte maturation, during which both genes have been reported to

be highly upregulated

Because protein interaction data is currently unavailable for Pfnek3, yeast!two!

hybrid experiments were performed using Pfnek3 as bait to screen a P falciparum total

cDNA library Out of about a hundred clones, five yielded potential interaction data However, only one interaction pair activated the "!galactosidase reporter gene when the interactions were re!tested As an attempt to confirm the interactions observed in the yeast system, mammalian!two!hybrid assays were employed Reporter gene activation was not detected, suggesting that one of the two systems, or the interaction, could be artifactual

To pursue the role of Pfmap2 and Pfnek3 in the sexual development of the malaria parasite, a dual!component phosphoproteomics strategy is discussed It would

be useful to first attempt an in vitro component to determine the substrates of Pfmap2 and Pfnek3, followed by an in vivo study to determine the total phosphoproteome of the malaria parasite Future studies could involve in vivo experiments that could illuminate

the functions and substrates of a kinase!of!interest, but only when gene!disrupted parasites or kinase!specific inhibitors become available

LIST OF PUBLICATIONS Journal publications:

[1] Low H, Lye YM, Sim TS (2007) Pfnek3 functions as an atypical MAPKK in Plasmodium

falciparum Biochem Biophys Res Commun 361(2):439!444

[2] Lye YM, Chan M and Sim TS (2006) Pfnek3: an atypical activator of a MAP kinase in

Plasmodium falciparum FEBS Letters 580: 6083!6092

[3] Lye YM, Chan M, Sim TS (2005) Endorsing functionality of Burkholderia pseudomallei

glyoxylate cycle genes as anti!persistence drug screens J Mol Cat B: Enzymatic 33: 51!

56

Poster abstracts:

[4] Lye YM, Lin W, Tay J, Low H, Chan M, Sim TS (2007) A survey of genes coding for N!

terminal regions in Plasmodium falciparum proteins affecting their heterologous

bacterial expression Research abstract in: Keystone symposia meeting: Cell signaling and Proteomics, Steamboat Springs, Colorado, USA Mar 22 – 27, 2007 Poster 214, p129

[5] Lye YM, Tan D, Chan M, Sim TS (2007) Tracking Plasmodium falciparum’s protein

network through heterologous interacting systems Research abstract in: Keystone symposia meeting: Cell signaling and Proteomics, Steamboat Springs, Colorado, USA Mar 22 – 27, 2007 Poster 312, p137

[6] Low H, Lye YM, Chan M, Sim TS (2006) Data mining for malarial kinases with bipartite

localization signals Research abstract in: 18th Annual Meeting of the Thai Society for Biotechnology, Bangkok, Thailand Nov 2 – 3, 2006 Poster VII!P!8, p214

[7] Lye YM, Chan M, Sim TS (2006) A proposed text!based data!mining tool for malaria

Research abstract in: 20th IUBMB International Congress of Biochemistry and Molecular Biology and 11th FAOBMB Congress, Kyoto, Japan Jun 18 – 23, 2006 Poster 3P!B!522, p21

[8] Chan M, Lye YM, Sim TS (2006) Validity of datamining the transcriptome data of

Plasmodium falciparum for procuring genes of its glycolytic and TCA pathways Research

abstract in: 20th IUBMB International Congress of Biochemistry and Molecular Biology and 11th FAOBMB Congress, Kyoto, Japan Jun 18 – 23, 2006 Poster 3P!B!525, p22

[9] Lye YM, Chan M, Sim TS (2004) NIMA!related protein kinases in Plasmodium falciparum

Research abstract in: 4th Combined Scientific Meeting, Singapore 8th NUS!NUH Annual Scientific Meeting, Singapore, Oct 7!8, 2004 Poster P!62, p 109

[10] Lye YM, Chan M, Sim TS (2004) Cloning and characterization of glyoxylate cycle genes

from Burkholderia pseudomallei Research abstract in: 1st Pacific!Rim International

Conference on Protein Science, Yokohama, Japan April 14 – 18, 2004 Poster P15/16!

181, p 154

LIST OF TABLES

Table 2!1: Annotated NIMA family protein kinases from the malaria parasite 8

Table 3!1: Cell lines and organisms used in this study 15

Table 3!2: List of primers used in this study 24

Table 3!3: Recipe for PCR amplification of Pfnek3 (Protocol Pf1.1) 26

Table 3!4: T4 ligation recipe 29

Table 3!5: LR clonase™ recombination recipe using Gateway™ technology 29

Table 3!6: Recipe for site!directed mutagenesis 30

Table 3!7: Plasmids used in this study 31

Table 3!8: Fusion constructs used in this study 33

Table 3!9: Recipe for cDNA library preparation 51

Table 3!10: cDNA library thermal cycling parameters 51

Table 3!11: M2H transfection set!up 60

Table 3!12: Summary of kits used in the study 61

Table 4!1: Typical protein kinase domain features 65

Table 4!2: List of prey genes (identified from Y2H) sub!cloned into pVP16 74

Table 4!3 : List of peptide ions detected in a LC!MS/MS experiment identifying Pfmap2 78

Table 4!4: Representative raw data for Figure 4.12(C) Directional phosphorylation of Pfmap2 by Pfnek3 84

Table 4!5: Representative raw data for Figure 4.12(D) Kinase activity after co!incubation of kinases 84

Table 4!6: Representative raw data for Fig 4.14 88

Table 4!7: Representative raw data for Figure 4.16(A) 93

Table 4!8: Prey genes sub!cloned for M2H studies 99

Table 4!9: Advantages and disadvantages of yeast!two!hybrid screens 105

Table 4!10: Comparison of two!hybrid technology to other methods 105

Table 4!11: Representative raw data for Figure 4.20(A) Mammalian!two!hybrid protein interaction assay 109

Table 4!12: A list of phosphorylation site prediction web servers 119

LIST OF FIGURES

Figure 2.1: A typical eukaryotic MAP kinase pathway 6

Figure 3.1: Proposed strategy to study Pfnek3 as a potential, functional MAPKK 14

Figure 3.2: Schematic diagram depicting five deletion!constructs and the regions of the Pfnek3 gene that was PCR amplified for cloning into yeast bait vectors 27

Figure 3.4: Screening for protein!protein interactions with the Matchmaker Two!Hybrid System 46

Figure 3.5: A process flow chart for two!hybrid protein interaction screening 48

Figure 3.6: Schematic diagram of the SMART III system to generate cDNA with the SMART III and CDS III anchors 49

Figure 3.7: Reporter gene activation during a stable protein interaction event 57

Figure 3.8: Plasmids required for a M2H assay 58

Figure 4.1: Sequence alignment of Pfnek3 with other NEKs and comparison of domain architecture with the closest homologue 64

Figure 4.2: Phylogenetic analysis 66

Figure 4.3: Structural models of Pfnek3 and Pfnek1 66

Figure 4.4: Gel visualization of PCR products 68

Figure 4.5: Restriction digestion screening for FL! and TR!Pfnek3!pGEX recombinant constructs 70

Figure 4.6: Construction of yeast bait vectors fused with coding sequences derived from Pfnek3 71

Figure 4.7: Plasmid construction for M2H protein interaction studies 73

Figure 4.8: Generation of site!directed kinase!inactive mutants, GST!#Pfnek3 and GST! #Pfmap2 75

Figure 4.9: Recombinant expression of proteins 76

Figure 4.10: Pfmap2 amino acid sequence covered by mass spectrometry (~35%) 77

Figure 4.12: Pfnek3 phosphorylates Pfmap2 and stimulates its kinase activity 83

Figure 4.13: Confirmatory experiments demonstrating that increased MBP phosphorylation was due to Pfmap2 pre!activated with Pfnek3 85 Figure 4.14: Pfnek3 activates Pfmap2 but not hMAPK1 88 Figure 4.15: Cytoplasmic localization of Pfnek3 in HepG2 cells 91 Figure 4.16: Endogenous expression of Pfnek3 via immunofluorescence microscopy 92 Figure 4.18: Protein interactions of Pfnek3 97 Figure 4.19: Translation of prey fusion plasmid DNA sequences to determine the correctness of reading frame for interactions showing "!galactosidase activation 98 Figure 4.20: Mammalian!two!hybrid interactions 108 Figure 4.22: The KESTREL approach 112 Figure 4.23: Some phosphate sources used for the identification of kinase substrates 114 Figure 4.24: Schematic diagram of a protein chip!based global phosphoproteome analysis 115 Figure 4.26: The principles of SILAC 120

Figure 4.27: An outline for an organism level, proteome!scale in vivo identification of

protein kinase substrates when a specific inhibitor is unavailable 122 Figure 4.28: An integrated approach to study the phosphoproteome of the malaria parasite 123

ABBREVIATIONS

SYMBOLS

% (v/v) Milliliter per 100 milliliters

% (w/v) Gram per 100 milliliters

1 Introduction

Plasmodium falciparum is responsible for the most lethal form of human malaria

Currently, reports of widespread re!emergence of drug resistance exacerbate the problem (Martens and Hall, 2000) The development of effective vaccines has yet to produce significant success Therefore, a more detailed understanding of parasite development and growth may be vital to fortifying our molecular arsenal against the disease The malaria life cycle is appreciably complex (Leete and Rubin, 1996) Briefly, the sporozoites target the liver upon inoculation by the mosquito vector and develop into asexual forms that specifically infect red blood cells During the asexual intraerythrocytic stage, some parasites develop into gametocytes which are picked up by another feeding mosquito In the mosquito midgut, the gametocytes fuse and form zygotes that escape the midgut and transform into sporozoites that migrate to the vector’s salivary glands, ready to infect a new human host Surprisingly, the signaling mechanisms throughout the entire developmental process are poorly understood It has been established in other eukaryotic cells that molecular signaling is a key to a cell’s fate Therefore, it is reasonable to suggest that signal transduction control is essential to parasite growth A methodical approach to unveil these pathways may therefore lead to the identification of important signaling mediators

There is a growing interest in understanding the role of protein kinase pathways

in the malaria parasite Among eukaryotes, protein kinases belonging to the mitogen! activated protein kinase superfamily (MAPK, also called ERK, extracellularly!regulated

kinase) are currently among the best understood MAPKs are believed to be highly

conserved among eukaryotes and are central to the transduction of extracellular

mitogenic stimuli down a cascade of ATP!dependent protein kinases Two copies of P falciparum MAPKs (Pfmap1 and Pfmap2) have been identified so far (Graeser et al.,

1997; Dorin et al., 1999) Both MAPKs share a peptide sequence identity of 41% in their

catalytic domain Phosphorylation of a threonine!tyrosine (TXY) sequence motif by an upstream kinase is usually needed for the activation of classical MAPKs The TXY motif is completely conserved in Pfmap1 as TDY (PlasmoDB identifier: PF14_0294), and is altered

to TSH in Pfmap2 (PlasmoDB identifier: PF11_0147) Pbmap2, the P berghei counterpart

of Pfmap2 was demonstrated to regulate male gametogenesis via gene!disruption

studies as well as in sex!specific proteomic analyses (Khan et al., 2005; Rangarajan et al.,

2005)

Numerous attempts have been made to identify candidate kinases upstream of Pfmap1 and Pfmap2 For example, it has previously been suggested that Pfnek1, a NIMA!

family kinase is an upstream regulator (i.e MAPKK, also called MEK, MAPK/ERK Kinase)

of Pfmap2 (Dorin et al., 2001) Unfortunately, in vivo activity of Pfnek1 was not

established and recombinant Pfnek1 did not stimulate Pfmap1 A second protein kinase,

P falciparum Protein Kinase 7 (PfPK7), with a MEK!like motif was reported (Dorin et al.,

2005) Interestingly, its sequence similarity to MEKs is limited to the C!terminal domain

while its N!terminal region bears similarity to PKA (protein kinase A) In view of the

above, and the fact that the malaria ‘kinome’ (genome!scale analyses of kinase!encoding

sequences) reveals the lack of other MEK!coding DNA sequences (Ward et al., 2004),

Dorin et al (2005) suggested that a regular three!step MAP kinase cascade is possibly

non!existent in the malaria parasite Thus, the existence of unusual signaling mediators

that can regulate plasmodial MAPKs remains an intriguing question of Plasmodium

biology

The advent of the PlasmoDB repository (www.PlasmoDB.org) has eased our search for potential signaling mediators Capitalizing on the transcriptome data offered

in PlasmoDB (Bozdech et al., 2003; Le Roch et al., 2004), it was possible to identify a

range of annotated gene products with expression profiles well!correlated with Pfmap1

and/or Pfmap2 The result of data!mining revealed five Plasmodium genes clustered as

sequences with homology to the NIMA (Never in Mitosis, Aspergillus) protein kinase

family (Ward et al., 2004) Of these five genes, three were revealedby microarray data to

be expressed predominantly in gametocytes (Bozdech et al., 2003; Le Roch et al., 2004) NIMA, the founding member of the NEK kinase family was described in Aspergillus nidulans and shown to be required for G2/M transition (Osmani et al., 1988) Thus far,

other NIMA homologues have been identified in many eukaryotes, and there is growing evidence that NIMA!family kinases (NEKs) play the role of cell cycle regulators (O’Connell

et al., 2003)

We were intrigued by a sequence identified as PFL0080c (hereafter referred to as

Pfnek3 following Ward et al., 2004), because it has an asexual intraerythrocytic

expression profile akin to that of Pfmap1 whilst also being highly up!regulated during the

gametocyte stage, where Pfmap2 is specifically expressed (Dorin et al., 1999)

Preliminary studies suggested that Pfmap2 and Pfnek3 act synergistically to

phosphorylate substrate proteins in vitro (Lye et al., 2006) To understand the

relationship between the kinases, the objectives of this study are as follows:

1 To identify the mode of interaction between Pfnek3 and Pfmap2

2 To verify the directionality of phosphorylation between Pfnek3 and Pfmap2

3 To determine the localization pattern of endogenous Pfnek3

4 To reveal the protein interaction partners of Pfnek3

2 Literature review

2.1 The canonical MAPK cascade

All living cells face the need for stimuli perception and mitotic signal transduction during cell cycle progression Strategies for resolving this necessity have therefore evolved very early, as evident by the high level of conservedness of classical signal

transduction pathways in all eukaryotic cells One well!conserved pathway is the MAPK (mitogen!activated protein kinase) pathway and MAPK family members have been

identified in all eukaryotes investigated so far, from unicellular organisms to mammals and plants (Figure 2.1) Reviewed in Garrington and Johnson (1999), the MAPKs, also

called ERKs (extracellularly regulated kinases), are central to the adaptive responses of

eukaryotic cells to a wide range of stimuli Phosphorylation at both the Thr and Tyr

residues of the conserved MAPK activation motif (TXY) by a specific MAPK kinase (MAPKK; also called MEK, for MAPK/ERK kinase) is necessary for MAPK activation Correspondingly, another upstream kinase, MAPKKK or MEKK, often associated with

membrane receptor tyrosine kinases, are in turn responsible for the phosphorylation and

activation of MEKs

Figure 2.1: A typical eukaryotic MAP kinase pathway

The pathway involves the sensing of extracellular stimuli which are transduced through a series

of phosphotransfer cascades via GTP exchangers and GTPases which culminate in a kinase cascade resulting in the activation of MAPK1/2 which in turn phosphorylate its target proteins, many of which are phosphorylation!activated (or –deactivated) transcription factors Figure constructed from textual description in Garrington and Johnson (1999)

2.2 The unusual kinome of the malaria parasite

The sequencing of the P falciparum genome has led to a plethora of opportunities

for genome!wide analyses of broad gene groupings, and this has been done with protein

kinases and termed the kinome (Ward et al., 2004) The malarial kinome has been

reported to be evolutionarily!divergent from most eukaryotes and this provides new opportunities for the identification of novel drug targets that are parasite!specific and

thus less likely to present host toxicity issues With the aim of identifying and classifying

all protein kinases in the malaria parasite, Ward et al (2004) used a variety of

bioinformatics tools to identify 65 malarial kinase sequences and constructed a phylogenetic tree to position these sequences relative to the seven established

eukaryotic protein kinase groups The main features of the tree were: (1) several malarial sequences did not cluster within any of the known protein kinase groups; (2) the highest number of malarial protein kinases fall within the CMGC group, which is a collective term for cyclin!dependent kinases (CDKs), mitogen activated protein kinases (MAPKs), glycogen synthase kinases (GSKs), and CDK!like kinases (CLKs), whose members are usually involved in the control of cell proliferation; and (3) no malarial protein kinases

clustered with the tyrosine kinase group, pointing to the possible absence of a typical MAPK cascade in the parasite

A novel family of 20 kinase sequences was identified and called the “FIKK” group,

on the basis of a conserved “FIKK” amino acid motif The FIKK family seems restricted to

the Apicomplexan protozoan, with 20 members in P falciparum Many of the malarial

FIKK kinases are believed to contain protein export signals that transport the FIKK

Figure 2.2: The export of a GFP! tagged FIKK kinase from transgenic parasites into the host erythrocyte membrane

Parasite nuclei were confirmed with DAPI staining (blue) Figure source:

Nunes et al (2007)

protein kinase to the parasitized human cell (Schneider and Mercereau!Puijalon, 2005)

Recently, transgenic parasites expressing GFP!tagged FIKK kinases allowed the detection of exported FIKK kinases at the erythrocyte cytoskeleton (Figure 2.2) Moreover, the FIKK kinases co!immunoprecipitated from red cells were enzymatically active, suggesting yet unknown roles of these kinases in the remodeling of the erythrocyte during an infection Hence, these findings emphasize the need to study the function and localization of malarial protein kinases so as to validate them as candidates

(2001) Pfnek2 Serine/threonine!protein kinase Nek1,

Pfnek3 Serine/threonine!protein kinase

Nek1, putative (this study) PFL0080c Apicoplast Lye et al (2006)

Pfnek4 Serine/threonine protein kinase 2,

Khan et al

(2005);

Reininger et al

(2005)

Note: 1 Nomenclature follows Ward et al (2004)

2.3 The P falciparum genome encodes two MAPKs

During investigations of molecular mechanisms possibly regulating parasite

multiplication and development, two plasmodial MAPKs (Pfmap1 and Pfmap2) have

been identified (Doerig et al., 1996; Dorin et al., 1999) Both malarial MAPKs were

originally described as members of the MAPK1/2 subfamily, whose members are normally involved in the regulation of cell proliferation in response to external stimuli

Pfmap1 is expressed during erythrocytic schizogony and in gametocytes (Graeser

et al., 1997), while Pfmap2 mRNA and protein are detectable only in the latter stage

(Dorin et al., 1999) Pfmap1 contains the conserved peptide motif comprising the amino

acids, TXY, as an activation site Mammalian MAPK1 is activated by phosphorylation on the threonine and tyrosine residues of the TXY motif by an upstream protein kinase (i.e

MAPKK) Although the TXY motif is similarly present on the Plasmodium MAPK1 (Pfmap1), it is not yet clear which Plasmodium kinase is capable of activating Pfmap1 Intriguingly, the requirement of a MAPK1 homologue for parasite survival, development

and proliferation has been demonstrated in Trypanosoma brucei (Muller et al., 2002) The case for Plasmodium MAPKs remains to be tested

Surprisingly, in Pfmap2, the TXY activation site is substituted by an atypical TSH

motif Site!directed mutagenesis showed that both the Thr (T290) and the His (H292)

residues in this motif are important for kinase activity of recombinant Pfmap2 (Dorin et

al., 1999) Recent mass spectrometric studies indicate that the phosphorylation on T290

is crucial for the activation of Pfmap2 (Low et al., 2007) suggesting that the regulation of

Pfmap2 activity may differ from that of typical MAPKs The only other example of such a

divergent MAPK activation site is found in the free!living protozoan Tetrahymena sp (Nakashima et al., 1999)

In a pioneering study to understand the sex!specific proteomes of Plasmodium parasites, P berghei, the rodent malaria parasite, was transfected with GFP reporter

constructs under the control sex!specific promoters which enabled the differentiation

and sorting of male and female gametocytes (Khan et al., 2005) A comparative

proteomic analysis revealed the presence of sex!specific proteins, among which, Pbmap2, the rodent homologue of Pfmap2 was demonstrated to be specific to, and necessary for, male gametocytogenesis

In another independent study, Pbmap2!deficient parasites were impaired in

sexual cycle completion (Rangarajan et al., 2005) These studies suggest the necessity of

Pbmap2 to the malaria parasite and it follows that the use of a rodent malaria model for searching upstream kinases and candidate drugs capable of disrupting their phosphorylation appears relevant for human malaria

Figure 2.3: Pbmap2 knock!out male gametocytes of the rodent malaria parasite,

P berghei, were disrupted in the ability to

exflagellate

Abbreviations: M, male gametocytes; F,

female gametocytes (Image: Khan et al., 2005)

2.5 MAPKKs upstream of Pfmap1 and !2 remain elusive

Before the advent of the Plasmodium genome resource (www.PlasmoDB.org), P falciparum MAPKK homologues were identified using PCR with degenerate primers, an

approach which led to the identification of genes from various families of eukaryotic

protein kinases (Dorin et al., 1999) When genome data became available, Dorin et al.,

(2001) isolated Pfnek1 (P falciparum NIMA!like kinase 1), a protein kinase exhibiting

maximal homology to members of the NIMA family of protein kinases (also called NIMA! like kinases, NEKs), but possessing a putative protein motif for activation (SMAHS) reminiscent of the conserved SMANS activation site found in mammalian

MAPKK1/MAPKK2 enzymes Bacterially!expressed Pfnek1 can phosphorylate

recombinant Pfmap2 in vitro Unfortunately, in vivo functionality of Pfnek1 still awaits

formal demonstration Pfnek1 has no in vitro effect on Pfmap1 or on mammalian ERK2

(also called MAPK2)

PfPK7 is another plasmodial protein kinase that encodes the ‘most MAPKK!like

enzyme’ in the Plasmodium genome based on sequence similarity (Dorin et al., 2005) However, PfPK7 was unable to phosphorylate the two P falciparum MAPK homologues

in vitro, and was insensitive to PKA and MAPKK inhibitors Together with the absence of

a typical MAPKK activation site, this suggests that PfPK7 is not a MAPKK orthologue,

although this enzyme is the most ‘MAPKK!like’ enzyme encoded in the P falciparum genome Consequently, Dorin et al (2005) suggested that the classical three!component

MAPK pathway may be absent in the malaria parasite

2.6 Current thoughts

Plasmodium falciparum causes 350 to 500 million clinical episodes of malaria

occur every year (World Malaria Report, 2005) The impact of this parasite on the socio!economic development of affected countries is considerable as a result of the spread of drug resistance Although the malaria parasite possesses only approximately 6000 genes, its life cycle is surprisingly complex, consisting of a succession of developmental stages staggering both the human and mosquito hosts In order to complete its life cycle, the malaria parasite needs to sense changes in its environment and to provide rapid and adequate adaptive responses, such as stimulation or inhibition of the cell division

machinery In contrast, the genome of Trichomonas vaginalis, a human vaginal

pathogen, encodes nearly 10 times more genes but displays a much simpler life cycle

Figure 2.4: Transcription levels of Pfmap2 and Pfnek3 at various intraerythrocytic stages

Both kinases are predominantly expressed in the schizont and gametocyte stages (Bozdech et al., 2003; Le Roch et al., 2004) However, in proteomic studies, native Pfmap2 could only be specifically detected in male gametocytes (Khan et al., 2005)

(Carlton et al., 2007), suggesting that the malaria parasite possesses: (1) a low!

redundancy genome and/or (2) atypical gene products that contain multiple domains or

singular domains with multiple functions, thereby enabling a single gene to take part in multiple physiological pathways The existence of a large number of unusual malarial genes, that encode protein domains that do not possess homology to other eukaryotic genomes, points in the direction of such a possibility

The cell signaling pathways of the malaria parasite is currently believed to deviate from classical eukaryotic models Therefore, the evolutionarily!divergent protein kinases

that possibly regulate the Plasmodium cell cycle progression become attractive drug

targets In particular, the MAPKs are believed to play central roles in the adaptive response of eukaryotic cells to a wide range of stimuli However, the current understanding of molecular regulatory mechanisms on Pfmap1 and Pfmap2 is insufficient to explain the complexity of the parasite’s elusive reproductive pathways Intriguingly, the classical eukaryotic MAPK pathway appears to be absent in the malaria

parasite To this day, the in vivo upstream regulators of Plasmodium MAPKs have

remained enigmatic

With this in mind, there is a compelling imperative to characterize potential regulators of the previously identified plasmodial MAP kinases As part of an earlier study, the synergistic kinase activity of Pfmap2 and Pfnek3 has been demonstrated To take this lead further, the relationship between the kinases would be further dissected and the interaction partners of Pfnek3 identified in an attempt to decipher the malarial MAPK signaling pathway

3 Materials and Methods

3.1 Bio!computation

Multiple sequence alignments were carried out using the ClustalW program

(Thompson et al., 1994) Analysis of the protein sequences was performed using the

software packages at the ExPASy molecular biology server (www.expasy.org) PlasmoDB (www.PlasmoDB.org) was the major source of sequence annotation and transcriptome data Alignment dendrograms were viewed with the TreeView program (http://taxonomy.zoology.gla.ac.uk/rod/treeview.html)

Figure 3.1: Proposed strategy to study Pfnek3 as a potential, functional MAPKK.

RT-PCR and PCR amplification to obtain Pfnek3 (full-length and various truncations)

Parasitic mRNA and DNA

extraction

Ligate into cloning vectors

Sub-clone into destination vectors (GST/6xHis/Two- hybrid bait)

Activity assays (ELISA)

Recombinant protein production expression

Databank (PlasmoDB)

mining

Kinase activation assays

Immunofluorescence microscopy

Two-hybrid protein interaction studies Site-direct mutagenesis

Table 3!1: Cell lines and organisms used in this study

E coli

BL21 (DE3)

Expression host with T7 RNA polymerase gene placed under lacUV5 promoter control

Novagen, USA

E coli TOP10 A general purpose cloning host

E coli ccd!survival

A lethality!resistant strain required for the maintenance of native pDEST series Gateway destination vectors

Invitrogen, USA

P falciparum 3D7 strain Genome sequencing strain MR4, USA

P falciparum Tan strain A clinical isolate from Singapore

National University Hospital, Singapore

S cerevisiae AH109 strain For screening protein interactions in

yeast!two!hybrid experiments

Clontech, USA

3.2 Use of Escherichia coli for cloning and expression

3.2.1 Growth and maintenance of E coli

Escherichia coli cultures were grown in autoclaved Luria!Bertani (LB) medium as

well as LB agar Petri dishes containing an addition of 1.5% (w/v) agar Cultures were grown at 37oC with shaking at 250 RPM When necessary, antibiotics such as kanamycin (50µg/ml), chloramphenicol (40µg/ml) and/or ampicillin (100µg/ml) were added to

culture media Stocks of E coli strains were preserved at !80oC in 10% (w/v) glycerol The

strains of E coli used in this study are listed in Table 3!1

3.2.2 Preparation of electrocompetent E coli

The preparation of electrocompetent cells with the highest competence requires

the cells to be in the early to mid!logarithmic growth phase A single well!isolated E coli

colony was inoculated into 5 ml of sterile LB broth and incubated overnight at 37oC with shaking at 250 RPM Two ml of the overnight culture was then added to 100 ml of sterile

LB broth This is incubated at 37oC with shaking at 250 RPM until mid!log phase (A600 0.6!0.8)

Following that, the culture was pelleted by centrifuging for 10 min at 7000 RPM

at 4°C Cells are washed once in 50 ml of ice!cold 10% (w/v) glycerol, pelleted again, followed by supernatant removal and re!suspension by short vortexing The re!suspended cells were then divided into 50 !l aliquots The cells can now be used for electro!transformation or frozen at !80°C for subsequent use The minimum acceptable

3.2.3 Transformation of E coli

The ligation reaction (10 out of a 20 µl reaction) was diluted with 100 µl of sterile water and the mixture added to 50 µl of competent cells in an ice!chilled electroporation cuvette with an electrode gap of 0.2 cm (Bio!Rad) The Bio!Rad Gene Pulser was adjusted to 2.5 kV and an electric pulse rendered across the cuvette Immediately, one

ml of sterile LB broth was added to the mixture to recover the cells at 37°C for one hour with shaking at 250 RPM

For heat shock transformation of commercial E coli TOP10 competent cells,

regular ligation mixes or TOPO™ ligase!free cloning reactions were mixed with ice!thawed cells for 2 min on ice Transformation was achieved by heat shock for 30 sec in a 42°C water bath followed by icing for 2 min Sterile SOC medium (250 µl) was immediately added to the mixture followed by one hour of recovery at 37°C with shaking

at 250 RPM

Selection for transformed colonies was achieved by spreading the transformation mixtures onto LB agar containing relevant antibiotics, depending on the antibiotic resistance marker encoded on the plasmid The plates were incubated overnight at 37oC and well!isolated colonies were picked the following day

3.3 Use of Saccharomyces cerevisiae for yeast!two!hybrid studies

3.3.1 Growth and maintenance of S cerevisiae

Yeast strains were stored in YPD or an auxotrophic!selection SD medium with 25% (w/v) glycerol at –80°C To prepare glycerol stocks, isolated colonies from an agar plate were re!suspended in 200–500 µl of YPD or the appropriate SD medium in a sterile 1.5 ml microfuge tube Sterile 50% (w/v) glycerol was added to a final concentration of 25% Frozen stocks were revived by streaking onto YPD or a selective SD drop!out agar and incubated at 30°C On YPD agar, colonies would generally take at least 4!5 days to appear To prepare liquid overnight cultures, fresh (< 2 mth old, 2!4 mm diameter) colonies from a working stock plate were inoculated at one colony per 5 ml of broth

3.3.2 Preparation of competent S cerevisiae AH109

A flask containing 30 ml of YPD was inoculated with several young (< 1 mth) colonies that are 2–3 mm in diameter Care was taken to disperse cell clumps by smearing the wire loop carrying the inoculums against the wall of the flask in a circular motion The flask was incubated at 30°C overnight for 16–18 h with shaking at 250 RPM

to stationary phase (OD600 > 1.5) The culture was transferred to another flask containing

300 ml of sterile YPD and incubated at 30°C for 3 h with shaking (250 RPM) achieving an

OD600 of 0.4–0.6 Cell cultures were transferred to six 50 ml Falcon tubes and centrifuged

at 1,000 gravitational!forces (RCF) for 5 min at room temperature (20–22°C) The supernatants were discarded and the cell pellets thoroughly re!suspended in sterile

1,000 RCF for 5 min at room temperature The supernatant was removed and the pellet

re!suspended in 1.5 ml of freshly prepared, sterile 1x TE/lithium acetate (LiAc) mix

3.3.3 Yeast transformation

The desired plasmid (0.1 µg) was freshly mixed with 0.1 mg of herring testes carrier DNA Freshly!made yeast competent cells (0.1 ml) were added to each tube and vortexed 0.6 ml of sterile PEG/LiAc solution was added to each tube and vortexed for 10 sec Cells were incubated at 30°C for 30 min with shaking at 200 RPM, after which, 70 µl

of DMSO was added Cells were then mixed by gentle inversion and heat shocked for 15 min in a 42°C water bath Cells were chilled on ice for 1–2 min for recovery followed by centrifugation for 5 sec at 14,000 RPM at room temperature After removal of the supernatant, cells were re!suspended in 0.5 ml of sterile 1X TE buffer and 100 µl from each sample were spread on the relevant SD!dropout agar plates that selected for cells expressing the corresponding auxotrophic markers For co!transformation samples, three controls plates (!Leu, !Trp and !Leu/!Trp) were needed to ensure comparable transformation efficiencies of both bait and prey plasmids Plates were incubated at 30°C until colonies appear (at least 2–4 days)

3.4 Mammalian cell culture

3.4.1 Growth and maintenance of mammalian cell lines

HepG2, a human liver cell line, was purchased from ATCC (USA) MCF7, a human breast cell line, was a gift from Prof Bay BH (National University of Singapore) Both cell lines were maintained in T25 flasks (Nunc, Denmark) using Dulbecco's Modified Eagle's Medium (DMEM; Sigma, USA) supplemented with 10% (v/v) fetal calf serum (FCS; Hyclone, USA) For maintaining HepG2 cells, 1X cell!culture grade pen!strep (Sigma, USA) was added to the medium Flask caps were loosened for ventilation while incubated at 37°C and 5% CO2 Approximately 0.5!1.0 million cells were split into a fresh flask each time 80!100% confluence was achieved

To prepare frozen stocks, cells in 2!4 day old flasks (with 80!100% confluence) were dislodged from the flask walls using 1.5 ml of 1X cell!culture grade trypsin (Sigma, USA) with 3!5 min of incubation at ambient temperatures Thereafter, fresh culture medium was added to quench the trypsin, followed by vigorous pipeting and transferred

to sterile 50 ml Falcon tubes The cells were pelleted by centrifugation at 1500 RPM for 5 min at 20°C The supernatant was decanted and the pellet resuspended with 1.5 ml of freezing medium, consisting of DMEM supplemented with 10% (v/v) FCS and 10% (v/v) DMSO (dimethyl sulfoxide; Sigma, USA) Thereafter, the cells were transferred to cryo!vials (Nunc, Denmark), sealed in Parafilm™, and floated in an isopropanol bath (~25°C), which was subsequently transferred to a !80°C freezer

To revive stocks, cryo!vials were thawed quickly in a 37°C water bath and their contents transferred into 50 ml Falcon tubes containing 5 ml of culture medium The Falcon tubes were then centrifuged to pellet the cells The supernatants were decanted and replaced with 7.5 ml fresh cell culture media After resuspension of the cells, the contents were transferred to a sterile T25 flask and incubated at 37°C with 5% CO2

3.4.2 Mammalian cell transfection

Prior to transfection, 6!well plates were seeded with about 1 million cells per well using antibiotic!free DMEM supplemented with 10% (v/v) FCS and allowed to grow overnight achieving approx 75!90 % confluence The cells were rinsed once with sterile PBS and immersed with 2 ml of the serum!reduced medium, OptiMEM™ (Invitrogen, USA) The wells were topped up with 500 µl of a mixture containing 10 µl of the Lipofectamine 2000™ reagent (Invitrogen, USA) and 4!20 µg of the desired plasmid(s)

3.5.1 Pfnek3 gene isolation and amplification

Parasites were grown according to Haynes et al (1976) The isolation and amplification of the Pfnek3 gene was described in Lye et al (2006) DNA and RNA were

extracted from erythrocyte!stage parasites using the QIAamp™ DNA Blood Mini Kit and RNeasy™ Mini Kit, respectively, following manufacturer’s instructions (both from Qiagen, USA)

PCR was performed using a pair of primers, OL771 and OL768 For easy

reference, all primers used in this study are listed in Table 3!2 Pfu DNA polymerase

(Promega, USA) was used for the amplification of this and other malarial genes in this laboratory (Chan and Sim, 2004), utilizing thermal cycling parameters comprising one min of denaturation at 94°C, followed by 35 repeats of the following: 94°C for one min, 45°C for two min and 68°C for three min An extension step for 10 min at 68°C was carried out, followed by a final hold at 4°C The reaction recipe is tabulated in Table 3!3

As evident in other studies (Dyson et al., 2004), the presence of N!terminal

hydrophobic regions in a recombinant protein has a propensity to interfere with

subsequent heterologous expression in E coli Therefore, in anticipation of this possibility, the truncated form of the Pfnek3 gene encoding a peptide lacking the first 59

amino acids was cloned using another forward primer, OL767 and the reverse primer OL768

In order to verify the transcription of Pfnek3 in P falciparum, the Sensiscript™

two!step reverse transcription kit (Qiagen, USA) or the Transcriptor™ First Strand cDNA Synthesis Kit (Roche, USA) was employed Poly!A+ mRNA!enriched templates purified from erythrocyte!stage parasites were subjected to 3 h of reverse transcription at 37°C with OL768 for first strand synthesis prior to thermal cycling as described above

Modern cloning techniques have been hastened with the introduction of Gateway technology (Invitrogen, USA) Because this technology involves genetic recombination rather than DNA ligation for sub!cloning, it was no longer necessary to

include restriction enzyme sites in PCR primers To amplify the Pfnek3 gene for

Gateway cloning, the following PCR primers were used instead: OL1106 which was

engineered to contain the directional vector insertion sequence (CACCATGGGA) and OL1107, a reverse primer

All PCR and RT!PCR products were separated on 0.8% (w/v) agarose gels and the

relevant DNA bands (FL! and TR!Pfnek3 = 1044 and 867 bp, respectively) were excised

using a scalpel and purified using the DNA Gel Extraction kit, following manufacturer’s

instructions (Qiagen, USA) The blunt!end Pfnek3 PCR products were cloned into two

different cloning vectors, (a) pCR!Blunt II!TOPO™ for conventional cloning or (b) the

pENTR/D!TOPO™ directional cloning vector compatible with Gateway technology (both

from Invitrogen, USA) Recombinant constructs were chemically!transformed into E coli

TOP10 cells (Invitrogen, USA) Successful transformants were selected on LB agar containing 50µg/ml of filter!sterilized kanamycin (Sigma, USA) Positive transformants were verified by restriction digestion for the regular pCR!TOPO™ vectors or DNA sequencing for the pENTR/D!TOPO™ Gateway™ vectors DNA inserts on both vector types were authenticated by sequencing The recombinant full!length and truncated

Pfnek3 were thereafter named FL!Pfnek3 and TR!Pfnek3, respectively

Table 3!2: List of primers used in this study

DNA sequencing

of TOPO™ and D!

TOPO™ plasmid inserts

Invitrogen

OL48 5’!TAATACGACTCACTATAGGG!3’ Sequences plasmids with

OL74 5’!TATGCTAGTTATTGCTCAG!3’ Sequences plasmids

from the T7 terminator DNA Sequencing Sigma!Proligo OL674

Pfnek3 reverse primer

(EcoRI site underlined;

reverse complemented termination signal in

Pfnek3 reverse primer

(SalI site underlined;

reverse complemented termination signal in

bold)

DNA and RT!PCR amplification of

Pfnek3, for pGEX

Pfmap1 reverse primer

Pfmap1 reverse primer

with XhoI site underlined Amplify Pfmap1 Sigma!Proligo

OL877

5’!

GTCGACTTAAACATATTTATGTTT!

3’

Pfmap1 reverse primer

with XhoI site underlined (for first 312 aa)

Amplify Pfmap1 Sigma!Proligo

inactive Pfnek3 mutant

TR!Pfnek3 forward primer 1st base

OL1107 5’! TTGAACCGTTATACATGAT!3’ Pfnek3 reverse primer (no

stop codon)

For D!TOPO cloning

inactive Pfmap2 mutant

Forward primer for FL!

Pfnek3 (EcoRI site underlined)

Forward primer for TR!

Pfnek3 (EcoRI site underlined)

Table 3!3: Recipe for PCR amplification of Pfnek3 (Protocol Pf1.1)

DNA template (double stranded DNA) 1.0 (approximately 0.5 ! 0.8 µg)

10X polymerase buffer (Promega, USA) 2.5

3.5.2 Obtaining deletion!constructs for the purpose of Y2H screens

Yeast!two!hybrid (Y2H) protein interaction screens make use of the expression of reporter genes, in this case, "! and #!galactosidase, under the control of transcriptional activators encoded by the bait and prey vectors The theoretical relevance of this system

is contested by the omnipresence of false positives and/or non!expression of heterologous proteins Excessive background yeast growths (false positives) are due to auto!activation of the reporter genes by the DNA!BD (binding domain) and/or DNA!AD (activation domain) fusion proteins In order to mitigate these possibilities, deletion of certain portions of bait and/or prey proteins may be required to eliminate unwanted transcriptional activity before two!hybrid screen results could be meaningfully

interpreted (Bartel et al., 1993) In this study, a total of five constructs of different lengths were amplified by PCR from a P falciparum genomic DNA template (Figure 3.2)

PCR products were gel!purified and cloned into the polylinker region of the yeast bait vector (pGBKT7) encoding a DNA!BD fusion of the Pfnek3 protein (methods described in Section 3.6) This strategy was designed to manage attrition owing to auto!activation of