STRUCTURAL AND BIOCHEMICAL STUDIES OF

PROTEINS INVOLVED IN POLYAMINE TRANSPORT

FROM Pseudomonas aeruginosa PAO1 AND mRNA DECAY

FROM Saccharomyces cerevisiae

WU DONGHUI

NATIONAL UNIVERSITY OF SINGAPORE

2010

STRUCTURAL AND BIOCHEMICAL STUDIES OF

PROTEINS INVOLVED IN POLYAMINE TRANSPORT

FROM Pseudomonas aeruginosa PAO1 AND mRNA DECAY

FROM Saccharomyces cerevisiae

WU DONGHUI

(B.Sc., Tongji University)

A THESIS SUBMITTED

FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF BIOLOGICAL SCIENCES

NATIONAL UNIVERSITY OF SINGAPORE

2010

To my family

I

Acknowledgements

First of all, I would like to express my sincere gratitude to my supervisor Dr. Haiwei

Song for giving me an opportunity of being a postgraduate in the field of X-ray

crystallography. Without his excellent guidance and encouragement, I could not complete

my thesis successfully. I would also like to thank Dr. Daiwen Yang as my co-supervisor

for his continuous support.

I am indebted to our collaborator Professor Lianhui Zhang and his lab members for the

contribution in section 3.1 and section 3.3 of my first project.

I would like to thank all current and former SHW lab members for their precious help

either in wet lab protein purification and crystallization or in dry lab discussions of

structure determination, especially former members Dr. Zhihong Cheng and Dr. Meipei

She. I want to sincerely thank Ms. Siew choo Lim and Dr. Shimin Jiang for their great

experimental help. Thanks also go to Dr. Portia Gloria Loh, Ms. Siew choo Lim and Mr.

Ting Feng Lai for critically proofreading the manuscript.

I would like to appreciate help from all administrative staffs in Department of

Biological Sciences (NUS) and IMCB for their supports. Thanks also go to the beamline

scientists at ESRF (France) for technical support and the DNA sequencing facility in

IMCB.

Finally, I am greatly indebted to the support from my family all these years, especially

from my wife and my lovely son for their understanding and encouragement.

II

Table of contents Acknowledgements………………………………………………………………………I

Table of contents…………………………………………………………………………II

List of figures……………………………………………………………………………VI

List of tables…………………………………………………………………………... .IX

List of abbreviations……………………………………………………………………IX

Part I Structural and biochemical studies of proteins involved in polyamine

transport from Pseudomonas aeruginosa PAO1

Summary…………………………………………………………………………………..1

Chapter 1 Introduction…………………………………………………………………...2

1.1 Polyamine metabolism..……………………………………………………………….3

1.1.1 Synthesis………………………………………………………………………….4

1.1.2 Degradation………………………………………………………………………7

1.1.3 Transport (uptake and secretion)………………………………………………....7

1.2 Roles of polyamine at the molecular level……………………………………………9

1.2.1 DNA regulation…………………………………………………………………...9

1.2.2. RNA regulation…………………………………………………………………10

1.2.3 Protein regulation………………………………………………………………..11

1.2.4 Post-translational modification………………………………………………….12

1.3 Roles of polyamine at the cellular level……………………………………………...13

1.3.1 Regulation of proliferation and differentiation…………………………………13

III

1.3.2 Regulation of apoptosis…………………………………………………………14

1.3.3 Regulation of cell cycle………………………………………………………...15

1.4 Polyamines in embryonic development……………………………………………...15

1.5 Polyamines and cancer...…………………………………………………………….16

1.6 Pseudomonas aeruginosa PAO1…………………………………………………….18

1.6.1 General property and clinical state……………………………………………...18

1.6.2 Virulence determinants…………………………………………………………19

1.6.3 Type III secretion system (T3SS)………………………………………………20

1.6.4 Effector proteins…………………………………………………………………23

1.6.5 Polyamine transport……………………………………………………………...26

1.7 Rationale and Aims..…………………………………………………………………27

Chapter 2 Materials and Methods………………………………………………………..29

2.1 Experimental materials………………………………………………………………29

2.2 Gene cloning and protein expression………………………………………………...34

2.3 Protein purification…………………………………………………………………..38

2.3.1 Protein expression in large-scale cell culture…………………………………….38

2.3.2 Protein purification procedures…………………………………………………..39

2.3.2.1 Cell lysis…………………………………………………………………..39

2.3.2.2 First GST affinity column chromatography……………………………….39

2.3.2.3 Second GST affinity column chromatography……………………………39

2.3.2.4 Gel filtration chromatography……………………………………………..40

2.3.2.5 Analysis of polyamines by RP-HPLC…………………………………….40

2.3.2.6 Preparation of ligand-free protein samples………………………………..41

IV

2.4 Circular Dichroism (CD)…………………………………………………………….41

2.5 Isothermal titration calorimetry (ITC)……………………………………………….41

2.6 Crystallization………………………………………………………………………..42

2.7 Data collection, structure determination and refinement…………………………….43

2.7.1 Data collection…………………………………………………………………...43

2.7.2 Structure determination and refinement…………………………………………43

2.8 Other methods for structure analysis………………………………………………...46

Chapter 3 Results………………………………………………………………………...47

3.1 spuD and spuE in the regulation of T3SS……………………………………………47

3.2 Protein purification……………………..……………………………………………49

3.3 spuD is a putrescine binding protein while spuE is a spermidine binding protein…..54

3.4 Crystallization, data collection and model refinement…………………………..…..57

3.5 Overall structure description…………………………………………………………61

3.6 Ligand binding induces conformational changes……………………………………65

3.7 Putrescine binding site in spuD and comparison with potF…………………………69

3.8 Spermidine binding site in spuE……………………………………………………..73

Chapter 4 Discussion…………………………………………………………………….76

4.1 Common features between spuD and spuE in ligand binding ………………………79

4.2 Differences between spuD and spuE in ligand binding……………………………...82

4.3 Possible ligand binding process in spuD and spuE…………………………………..84

4.4 polyamine and T3SS in Pseudomonas aeruginosa ………………………………….86

Part II In vitro assembly of decapping activator Lsm1p-7p-Pat1p eight-protein

complex and a preliminary structural study of Pat1p carboxyl terminal region

V

Summary…………………………………………………………………………………88

Chapter 1 Introduction…………………………………………………………………...89

1.1 Post transcriptional regulation……………………………………………………….89

1.1.1 5’-end processing………………………………………………………………...90

1.1.2 3’-end processing………………………………………………………………...90

1.1.3 Splicing…………………………………………………………………………..91

1.1.4 Export……………………………………………………………………………92

1.1.5 Translation……………………………………………………………………….92

1.2 mRNA decay…………………………………………………………………………93

1.2.1 Normal mRNA decay…………………………………………………………...94

1.2.2 mRNA surveillance pathways…………………………………………………..96

1.2.3 Decapping regulation…………………………………………………………..100

1.2.4 Pat1 in decapping, P-body formation and translation repression……………...102

1.2.5 Lsm proteins in pre-mRNA processing and mRNA decay……………………103

1.3 Rationale and aims……………………………………………………………….....107

Chapter 2 Methods……………………………………………………………………...108

2.1 Gene cloning, protein expression…………………………………………………...108

2.2 Protein purification………………………………………………………………...109

2.3 Native-PAGE………………………………………………………………………111

2.4 Reconstitution of Lsm2-3-Pat1-465c, Lsm1-7 and Lsm1-7-Pat1-465c……………112

2.5 Crystallization………………………………………………………………………112

2.6 Data collection……………………………………………………………………...113

2.7 Surface Plasmon Resonance (SPR)………………………………………………...113

VI

Chapter 3 Results and Discussion………………………………………………………113

3.1 Protein purification ………………………………………………………………...113

3.2 Crystallization and data collection ………………………………………………...117

3.3 Identification of interaction relationships among Pat1-465c and

seven Lsm proteins from Lsm1 to Lsm7…………………………………………..119

3.4 Reconstitution of Lsm2-3-Pat1-465c, Lsm1-7 and Lsm1-7-Pat1-465c ……………123

3.5 mRNA binding properties ………………………………………………………….126

3.6 Difficulties in structure determination of Pat1-465c and Pat1-473c……………….132

References………………………………………………………………………………134

Lists of figures

Figure I-1. Structures of polyamines……………………………………………………..4

Figure I-2. Polyamine synthesis and degradation pathways……………………………...6

Figure I-3. Polyamine transport in E. coli………………………………………………………8

Figure I-4. Virulence determinants in P. aeruginosa………………………………………..20

Figure I-5. Type III secretion system (T3SS)…………………………………………...22

Figure I-6. Schematic diagram of four effector proteins………………………………..23

Figure I-7. Roles of spuD and spuE on T3SS gene expression…………………………48

Figure I-8. Purification of spuD in complex with putrescine…………………………...49

Figure I-9. Purification of spuD-C236A in complex with putrescine…………………..50

Figure I-10. Purification of spuE in complex with spermidine…………………………50

Figure I-11. Purification of apo-spuD…………………………………………………...51

VII

Figure I-12. Purification of apo-spuE…………………………………………………...52

Figure I-13. Circular dichroism of spuD and spuE in apo- and ligand-bound forms…...53

Figure I-14. Analysis of spuD and spuE by RP-HPLC…………………………………54

Figure I-15. Analysis of polyamine binding of spuD and spuE using ITC……………..57

Figure I-16. Crystal pictures of spuD and spuE…………………………………………58

Figure I-17. Overall structures of apo and ligand-bound spuD and spuE………………63

Figure I-18. Superposition of Cα backbones from spuD, spuE and potF………………65

Figure I-19. Ligand biding induces conformational changes…………………………...67

Figure I-20. Stereo view and superposition of connector I and II from spuD and spuE in

both ligand-free and ligand-bound structures ……………………………..69

Figure I-21. Putrescine binding sites in spuD…………………………………………..72

Figure I-22. Side by side comparison of putrescine binding sites between spuD and

potF………………………………………………………………………...73

Figure I-23. Spermidine binding sites in spuE………………………………………….75

Figure I-24. Sequence alignment of polyamine binding proteins in bacteria…………...77

Figure I-25. Stereo view and superposition of amino acids from spuD and spuE for

ligand recognition………………………………………………………….81

Figure II -1. Post transcriptional regulation points of eukaryotic mRNA……………...90 Figure II -2. Normal mRNA decay pathways in eukaryotes…………………………...95

Figure II -3. The core components of mRNA decay and translation repression

machinery in P body……………………………………………………...96

Figure II -4. mRNA surveillance pathways……………………………………………..99

Figure II -5. Pat1 from Saccharomyces cerevisiae contains several conserved

VIII

regions……………………………………………………………………103

Figure II -6. Crystal structure of Lsm3 and models for Lsm1-7 and Lsm2-8…………106

Figure II -7. Purification of semet-Pat1-465c…………………………………………114

Figure II -8. Purification of semet-Pat1-473c…………………………………………115

Figure II -9. Purification of Lsm1……………………………………………………..115

Figure II -10. Purification of Lsm4……………………………………………………116

Figure II -11. Purification of Lsm2-3…………………………………………………116

Figure II -12. Purification of Lsm5-6-7……………………………………………….117

Figure II -13. Crystals of Pat1 C-regions alone and in complex with Lsm2-3………..118

Figure II -14. Interaction networks of Pat1-465c towards different Lsm proteins by

Native-PAGE……………………………………………………………121

Figure II -15. Interaction networks of Lsm proteins by Native-PAGE………………..122

Figure II -16. Reconstitution of Lsm2-3-Pat1-465c…………………………………...125

Figure II -17. Reconstitution of Lsm1-7………………………………………………125

Figure II -18. Reconstitution of Lsm1-7-Pat1-465c…………………………………...126

Figure II -19. Binding of Lsm1-7 complex and 5’ biotin-labeled poly (U)…………...127

Figure II -20. Binding of Lsm1-7-Pat1-465c complex and 5’ biotin-labeled poly

(U)……………………………………………………………………….128

Figure II -21. Binding of Lsm2-3 complex and 5’ biotin-labeled poly (U)…………...129

Figure II -22. Binding of Lsm2-3-Pat1-465c complex and 5’ biotin-labeled poly

(U)……………………………………………………………………….130

Figure II -23. Binding of Lsm1, 4, 5-6-7 and Pat1-465c towards 5’ biotin-labeled poly

(U)……………………………………………………………………….131

IX

Lists of tables

Table 1. Data collection and refinement statistics of spuD in apo- and

putrescine-bound forms. ……………………………………………………….50

Table 2. Data collection and refinement statistics of apo-spuE…………………………51

Table 3. Δφ and Δψ of connector I and II from spuD and spuE in ligand-free

and ligand-bound structures……………………………………………………66

Table 4. Summary of amino acids involved in aromatic stacking and hydrogen bonding

with ligand in spuD and spuE…………………………………………………...78

Table 5. Absorption edge data collection for semet-Pat1-465c and semet-Pat1-473c…119

Lists of abbreviations

ATP adenosine triphosphate

CD circular dichroism

DFMO α-difluoromethylornithine

DTT dithiothreitol

EDTA ethylene diamine tetra acetic acid

EG ethylene glycol

EJC exon junction complex

GST Glutathione-S-transferase

IPTG isopropyl β-D-thiogalactopyranoside

X

ITC isothermal titration calorimetry

kDa kilo Dalton

Lsm Sm-like

M molar

mg milligram

ml milliliter

mM millimolar

MPD 2-methyl-2, 4-pentanediol

mRNA messenger RNA

mRNP messenger ribonucleoprotein particle

NGD No-go mRNA decay

nm nanometer

NMD Nonsense-mediated decay

NSD Non-stop mRNA decay

OD optical density

ODC ornithine decarboxylase

PAGE poly-acrylamide gel electrophoresis

P-body processing body

PCR polymerase chain reaction

PDB Protein Data Bank

PEG polyethylene glycol

r.m.s.d root-mean-square-deviation

XI

RP-HPLC reverse phase-high pressure liquid chromatography

SAD single-wavelength anomalous dispersion

scPat1-465c S. cerevisiae Pat1p amino acids 465-796

scPat1-473c S. cerevisiae Pat1p amino acids 473-796

SDS sodium dodecyl sulfate

SeMet Seleno-L-methionine

SPR surface plasmon resonance

Tris Tris(hydroxymethyl)aminomethane

tRNA transfer RNA

T3SS type III secretion system

WT wild type

μl microliter

μM micromolar

1

This thesis comprises two parts. The first part addresses structural and biochemical studies of

proteins involved in polyamine transport in Pseudomonas aeruginosa PAO1. The second part

briefly describes an in vitro assembly of decapping activator Lsm1-7-Pat1-465c eight-protein

complex involved in mRNA decay and a preliminary structural study of the carboxyl

terminal domain of Pat1 which plays an important role in mRNA decay and translational

repression.

Part I Structural and biochemical studies of proteins involved in polyamine transport

from Pseudomonas aeruginosa PAO1

Summary

Polyamines (putrescine, spermidine and spermine) are naturally small polycationic molecules

and play important roles in DNA condensation, DNA-protein interaction, embryonic

development and cancer development. In Pseudomonas aeruginosa, spuDEFGH proteins are

supposed to be a spermidine transport ATP binding cassette (ABC) system, in which spuD

and spuE are the periplasmic spermidine binding proteins. Moreover, this system plays a role

in the regulation of type III secretion system (T3SS), one of the important virulence

determinants in Gram-negative bacteria. In this study, spuD is found to be a putrescine

binding protein and has no role in the T3SS gene expression. In contrast, spuE is a

spermidine binding protein and shows a distinct effect upon the T3SS gene expression.

Furthermore, we solved four structures including spuD in apo open form, spuD in putrescine-

bound closed form, spuE in apo open form and spuE in spermidine-bound closed form. This

is the first report in which both putresine and spermidine binding proteins in both apo and

ligand-bound forms are presented. Comparison of these four structures reveals novel insight

2

into the basis of polyamine binding specificity, as well as direct insight into the

conformational changes associated with ligand binding.

Chapter 1 introduction

Polyamines (putrescine, spermidine and spermine) are naturally occurring small linear

polycationic molecules. They are important regulators of a wide spectrum of processes at

both the molecular and cellular levels (Thomas and Thomas 2001). As regulators, they

themselves are also regulated tightly so as to adapt to dynamic cellular changes (Thomas and

Thomas 2001). Metabolic regulations of polyamines consist of synthesis, degradation and

transport (uptake and excretion) (Igarashi and Kashiwagi 1999). Polyamines are found to be

involved in cancers such as prostatic and colorectal cancer and infectious diseases such as

African sleeping sickness (Wallace 2003). Given their roles in disease, designing polyamine

inhibitors (analogues or derivatives) may have potential therapeutic effects in the treatment

of such diseases (Wallace 2003).

Recent study has implicated polyamines as a specific regulator of the type III secretion

system (T3SS), one of the essential virulence determinants in a Gram negative bacterium,

Pseudomonas aeruginosa (Zhou et al. 2007; Hauser 2009). Pseudomonas aeruginosa is a

serious human pathogen. It is resistant to most of the current antibiotics treatments and hence

poses a mortal threat to immuno-compromised patients. In this chapter, an overview of the

polyamine metabolic pathways in bacteria will be given, followed by a review of the current

knowledge of the physiological and pathological roles of polyamine and potential

applications of polyamine inhibitors in disease treatment. Finally, the implications of

polyamines in the virulence of Pseudomonas aeruginosa will be reviewed.

3

Section 1.1 Polyamine metabolism

Polyamines are small, natural molecules consisting of an aliphatic hydrocarbon backbone of

flexible length and interspersed positively charged amino groups at physiological pH.

Polyamines identified include putrescine (1,4-diaminobutane), spermidine ([N-(3-

aminopropyl) butane-1,4-diamine]) and spermine ([N,N′- bis(3-aminopropyl)butane-1,4-

diamine]) (Figure I-1). A polyamine is defined by the number of amine groups it contains.

As shown in Figure I-1, putrescine is divalent, spermidine is trivalent and spermine is

tetravalent. Putrescine and spermidine are found widely in both prokaryotes and eukaryotes,

while spermine is found exclusively in eukaryotes (Thomas and Thomas 2001). Regulation

of polyamine metabolism consists of synthesis, degradation and transport (uptake and

excretion) processes (Igarashi and Kashiwagi 1999). Polyamine synthesis and degradation

are well conserved in both prokaryotes and eukaryotes (Childs et al. 2003). However, while

much is known about polyamine transport in E. coli, most genes involved in eukaryotic

polyamine transport have not yet been identified. Hence in the following sections, discussion

of polyamine metabolism mainly refers to the studies carried out in E. coli.

4

Figure I-1. Structures of polyamines. The position of nitrogen and carbon atoms in

putrescine and spermidine is labeled for the convenience of structural analysis in the Results

and Discussion sections below.

1.1.1 Synthesis

In bacteria, putrescine and spermidine are the two most widely utilized polyamines within the

cells. In contrast, cadaverine is the least abundant among all natural polyamines. Cadaverine

is only synthesized under rare conditions such as in anaerobicity and low pH, or in the

absence of putrescine synthesis. It can be synthesized via catalysis by lysine decarboxylase

(CadA) from the precursor lysine when this residue is in abundance (Watson et al. 1992).

The synthesis pathway of spermine has not been identified to date, however, it seems that

exogenous spermine is sufficient to satisfy the cellular requirements and therefore bacteria do

not need a self synthesis system. Another interpretation might be that spermine plays a trivial

role in the metabolism of bacteria and is hence not needed in copious amounts.

5

In E. coli and many Pseudomonas species, two major putrescine synthesis pathways

(Figure I-2) have been described (Childs et al. 2003). In the first pathway, ornithine is the

precursor and catalyzed to putrescine by ornithine decarboxylase (ODC), which is encoded

by the speC gene. In the second pathway, arginine is the precursor and catalyzed via the

intermediate agmatine by SpeA (arginine decarboxylase) to the end product putrescine by

SpeB (agmatinase). In Pseudomonas aeruginosa, a third pathway of putrescine synthesis has

been identified in which agamatine is hydrolyzed to putrescine via N-carbamoylputrescine

(Nakada and Itoh 2003).

Spermidine synthesis involves the initial conversion of methione to S-adenosyl methionine

(SAM) by the enzyme methionine adenosyltransferase (SAM synthetase encoded by metK),

followed by conversion to decarboxylated SAM by SAM decarboxylase (speD). Finally,

putrescine and decarboxylated SAM are catalyzed as the substrates of spermidine synthase

(speE) to spermidine (Tabor et al. 1986).

6

Figure I-2. Polyamine synthesis and degradation pathways. Putrescine is synthesized via

two pathways. The first is the conversion of ornithine to putrescine by ornithine

decarboxylase (ODC), and the other is the conversion of arginine to the intermediate product

agmatine by arginine decarboxylase (ADC) and then to putrescine by agmatinase. S-

Adenosylmethione decarboxylase (SAMDC), spermidine (Spd) synthase and spermine (Spm)

synthase catalyze putrescine to spermidine and spermine, respectively. Spermine is converted

to spermidine and putrescine by spermine/spermidine N1-acetyltransterase (SSAT) and

polyamine oxidase (PAO). Figure is modified from the reference Childs et al. 2003.

7

1.1.2 Degradation

Polyamine degradation (Figure I-2) is considered as an interconversion cycle (Seiler 2004).

Two major enzymes have been implicated in the degradation of spermine, spermidine and

putrescine. Spermidine and spermine are first acetylated to N1-acetylspermidine (N1acSpd)

and N1-acetylspermine (N1acSpm) by spermidine/spermine-N1-acetyltransferase (SSAT).

Following acetylation, spermine and spermidine are catalyzed by polyamine oxidase (PAO)

to produce spermidine and putrescine, respectively. This is coupled to the production of

aldehyde and highly oxidative hydrogen peroxide (H2O2). As the generated spermidine and

putrescine can be reused, this cycle is then called an interconversion cycle.

1.1.3 Transport (Uptake and Secretion)

In E. coli, several polyamine transport systems have been identified (Figure I-3). One is the

spermidine preferential uptake system involving four proteins known as potABCD; the other

is a putrescine-specific uptake system composed of a potFGHI four-protein complex

(Kashiwagi et al. 1990; Furuchi et al. 1991). Both uptake systems belong to the ATP binding

cassette (ABC) transporter family (Higgins 1992), which comprises one periplasmic binding

protein (PBP) for substrate capture (potD or potF), two transmembrane proteins for channel

formation (potBC or potHI), and one membrane-associated ATPase (potA or potG) for

energy consumption coupled with polyamine uptake. Another polyamine transport system

has been identified (potE), which shows pH dependent activity (Kashiwagi et al. 2000). In

acidic medium it execretes putrescine and uptakes ornithine while in neutral medium it

uptakes putrescine (Kashiwagi et al. 2000). This pH-dependent polyamine transport activity

may imply that the transport systems in E. coli are highly sensitive and efficient in adjusting

to the changes in external environment. The spermidine excretion system has been identified

8

for E. coli when grown in neutral medium (Higashi et al. 2008). In this system, members of

the drug exporter family, MdtJI, are involved in the excretion activity.

Figure I-3. Polyamine transport systems in E. coli. Spermidine uptake system is

composed of potABCD proteins of which the periplasmic binding protein potD binds and

transfers spermdine through the membrane channel formed by potB and potC. This process is

coupled to ATP hydrolysis by the ATPase potA. Similarly, putrescine is transported by the

potFGHI system. Here, potF is the periplasmic putrescine-binding protein, while potH and

potI form the transmembrane channel proteins and potG hydrolyzes ATP during putrescine

transport. PotE shows pH-depenedent putrescine transport activity. Figure is adapted from

the reference Igarashi and Kashiwagi 1999.

Periplasmic binding proteins (PBP) are found in bacteria and act as receptors of a wide

spectrum of small molecule substrates for transport and chemotaxis (Tam and Saier 1993).

PBPs show high binding affinity and specificity to their ligands and are generally thought to

switch between open and closed forms upon ligand binding and ligand release. Free PBPs

show weak binding to their corresponding transmembrane channel proteins; however, the

binding affinity between periplasmic binding proteins and their respective transmembrane

protein increases when PBPs are loaded with their targets. Meanwhile, the interaction of

9

PBPs with their transmembrane channel proteins induces a conformational change of the

channel proteins in order to accept the ligands from the PBPs. This process is energetically-

driven by the ATPase of each ABC transporter.

Structures of potD in complex with spermidine and potF in complex with putrescine have

been reported (Sugiyama et al. 1996a; Sugiyama et al. 1996b; Vassylyev et al. 1998).

Recently, the structure of TppotD, a putrescine binding receptor from Treponema pallidum,

has been solved (Machius et al. 2007). In this structure, two 2-(N-morpholino) ethanesulfonic

acid molecules (Mes) that were used as a crystallization agent were found within the

receptor. One Mes molecule was observed to overlap with the putrescine binding site after

superposing with the structures of potD and potF.

Section 1.2 Roles of polyamine at the molecular level

1.2.1 DNA regulation

Positively charged polyamines (putrescine, spermidine and spermine) interact with the

phosphate groups of DNA via non-specific salt bridges (Ouameur and Tajmir-Riahi 2004).

These non-specific salt bridge interactions are in the form of nuclear aggregates of

polyamines (NAPs) in vivo (D'Agostino et al. 2005). There are three types of NAPs with

approximate molecular weights of 1000 Da, 4800 Da and 8000 Da, respectively. The

simplest formulas for 1000Da, 4800 Da and 8000 Da are Put-Ph-Spd-Ph-Spm, Put-Ph-Spd-

(Ph-Spm)2 and Put-Ph-Spd-(Ph-Spm)2, respectively (Put for putrescine, Spd for spermidine,

Spm for spermine, Ph for phosphate group). These three types of NAPs can be assembled in

vitro, which is in agreement with their in vivo function (Di Luccia et al. 2009). NAPs may

influence the behavior of DNA in several aspects: (1) protection of DNA from damaging

agents such as ionizing radiation and reactive oxygens (Khan et al. 1992; Spotheim-Maurizot

10

et al. 1995); (2) protection of DNA from exonuclease III and endonuclease cleavage

including that of DNAse I (Pingoud et al. 1984); (3) induction of DNA condensation and

hence genome packaging from around 1 meter in length to several micrometers in the

nucleus due to the shielding of more than 90% of DNA’s negative charges (Wilson and

Bloomfield 1979); (4) prevention of DNA fragmentation, which is often an onset of

apoptosis; (5) induction of DNA transition from the right handed B-form to the left handed

Z-form, wherein the Z form DNA is important in the regulation of the transcriptional

activities of certain genes (Rich and Zhang 2003); and (6) stabilization of DNA from heat

denaturation (Tabor 1961). As a result of these properties, derivatives of polyamines have

been commercially developed and used as carriers for the delivery of anti-sense

oligonucleotides into target cells to knock out specific genes (Bielinska et al. 1996).

1.2.2. RNA regulation

RNA comprises three types: tRNA, rRNA and mRNA. mRNA carries genetic information;

rRNA is a major component of the translational machinery ribosome; and tRNA contains

anticodon for basepairing to the triplet codon from mRNA while at the same time provides

amino acid for peptide elongation. Evidence shows that polyamines can regulate these three

types of RNAs. For example, purified tRNAs have been found to be associated with

spermidine and Mg2+ (Quigley et al. 1978). Both spermidine and Mg2+ are important in

modulating the conformation of tRNAs and their biological functions. Spermidine has been

found to facilitate the crystallization of tRNAs in ethanol and the presence of spermidine in

tRNAs has been confirmed from earlier crystallographic studies (Kim et al. 1971).

Polyamines are also associated with the ribosomes from bacteria and eukaryotes (Bachrach

2005). This suggests that they have a role in regulating the assembly of ribosomes, possibly

11

via an interaction with rRNAs, and hence protect RNA strand in the ribosome from

degradation, thereby enhance protein synthesis (Bachrach 2005). Besides association with

tRNA and ribosomes, polyamines are also found to modulate the stability of several specific

mRNAs in intestinal epithelial cells (Wang 2007). For example, polyamine depletion

stabilizes p53, junD and TGF β by increasing their half lives significantly. On the other hand,

polyamine supplementation restores their half lives to wild type levels. In addition to the

influence of polyamines upon mRNA stability, polyamines are found to induce a

translational frameshift of some mRNAs, one of which is antizyme (Matsufuji et al. 1995).

The sequence of antizyme mRNA contains a translational frameshift site and this site is

sensitively regulated by the concentration of cellular polyamines. When the cellular

concentration of polyamines exceeds a certain threshold, translation of frameshifting

antizyme is stimulated. Subsequently, antizyme binds to the rate-limiting polyamine

synthesis enzyme ornithine decarboxylase (ODC), induces the degradation of ODC and

hence decreases the rate of polyamine synthesis.

1.2.3 Protein regulation

Most of the proteins that are involved in DNA binding are transcription factors. Polyamines

have regulatory effects on many transcription factor-DNA interactions (Childs et al. 2003).

For instance, the interactions of estrogen receptor (ER) and its response element (ERE), NF-

κB and its response element NRE, Oct-1 and its DNA response element, and several viral

transcription factors ICP4, USF and TEF3 to their DNA response elements, are all regulated

by polyamines in a concentration-dependent manner. However, as polyamines can induce a

conformational change in DNA (reviewed in section 1.2.1), hence it may be possible that

12

regulation of DNA-protein interactions by polyamines might be via the regulation of the

DNA molecule itself.

On the other hand, polyamines can directly bind some proteins and modulate their

functions. For example, spermidine and spermine are found to bind the extracellular sites of

the N-methyl-D-aspartate (NMDA) receptor and to regulate the transport efficiency of

NMDA, which is an important neurological transmitter (Childs et al. 2003). Besides the

NMDA receptor, polyamines can regulate many other transmitter receptors through direct

interaction with these receptors (Childs et al. 2003).

1.2.4 Post-translational modification

Protein phosphorylation is a major post translational modification, which is carried out by

protein kinases. Polyamines have been found to regulate the phosphorylation of many of

these kinases, which are essential in the regulation of signal transduction and metabolic

pathways. For example, stimulation of tyrosine kinase activity in NIH3T3 cultured cell

requires polyamine synthesis (Auvinen et al. 1992). Phosphorylation of the kinase ERK1 and

ERK2 in leukemia L1210 cells and human breast epithelial cells is modulated by polyamines

(Bachrach 2005). The regulation of kinases involved in signal transduction or metabolism

may in turn modulate the expression of many downstream genes.

Besides non-covalent modification of many protein kinases, polyamines are also involved in

covalent protein modifications. eIF-5A is a ubiquitous eukaryotic translation initiation factor

that contains a unique amino acid, hypusine [N-epsilon-(4-aminobutyl) lysine]. Formation of

hypusine is a two-step enzymatic reaction. First, deoxyhypusine synthase catalyzes the

transfer of a 4-aminobutyl moiety from the spermidine to a specific lysine residue in the eIF-

5A precursor protein to form an intermediate amino acid, deoxyhypusine. Secondly, this

13

intermediate deoxyhypusine is hydroxylated by deoxyhypusine hydroxylase to form the

mature eIF-5A. eIF-5A is involved in cell proliferation and cell viability (Childs et al. 2003).

Hypusine modified eIF-5A is found to regulate the processing of certain RNAs (Pollard and

Malim 1998). Rev is the first RNA identified to be processed in a eIF-5A dependent manner.

Rev encodes a viral protein of HIV 1 and it shuttles between the nucleus and cytoplasm of

the host cell to mediate the translocation of the viral mRNAs of nonspliced or partially

spliced species across the nuclear envelope and hence regulate the gene expression of HIV. It

has a eIF-5A binding domain and it has been shown that binding of eIF-5A is essential for its

proper function (Ruhl et al. 1993). In addition to modulation of RNA processing, further

evidence points to a role in modulation of mRNA decay by hypusine-modified eIF-5A (Zuk

and Jacobson 1998). A single amino acid substitution in yeast eIF-5A shows a defect in

mRNA decay with accumulated uncapped mRNAs, suggesting the decay defect occurs

downstream of mRNA decapping.

Section 1.3 Roles of polyamine at the cellular level

As polyamines have been demonstrated to exert important roles in the regulation of all the

basic molecules, DNA, RNA and protein, roles of polyamines at the molecular level can be

reemphasized at the cellular levels. These roles span the fundamental cellular processes

including proliferation, differentiation, apoptosis, and cell cycle progression.

1.3.1 Regulation of proliferation and differentiation

Numerous studies show that polyamines are implicated in cell growth (Thomas and Thomas

2001). Overexpression of polyamines is often found in many malignant cells, while depletion

of polyamines causes cell growth arrest and may induce cell death. Addition of exogenous

polyamines can reverse this cell growth arrest. Many enzymes involved in polyamine

14

metabolism are essential in regulation of cell growth. For example, mutants of S-

Adenosylmethione decarboxylase (SAMDC), the key enzyme for spermidine and spermine

synthesis, cause growth arrest in Saccharomyces cerevisiae, Escherichia coli, mice embryos

and Leishmania donovani (Childs et al. 2003). Disruption of ODC, the key enzyme for

polyamine synthesis, causes Chinese hamster ovary cell death and failure of mice embryonic

development (Steglich and Scheffler 1982; Pendeville et al. 2001).

Besides the involvement of proliferative regulation, polyamines are also found to affect

differentiation. Inhibition of polyamine synthesis by DFMO, a specific ODC inhibitor,

prevents 3T3-L1 fibroblasts and L6 myoblasts from differentiation (Bethell and Pegg 1981;

Erwin et al. 1983). By contrast, addition of exogenous putrescine or spermidine restores the

differentiation of 3T3-L1 and L6 cell lines into adipocytes and muscle cells respectively. As

differentiation and proliferation are closely related, it is often difficult to distinguish the two

processes clearly. Therefore it is possible that involvement of polyamines in the regulation of

proliferation also intertwines with the regulation of cell differentiation.

1.3.2 Regulation of apoptosis

Polyamines have controversial roles in apoptosis. Some studies show that polyamines can

inhibit apoptosis. For example, spermine can prevent the apoptosis induced by ionomycin, a

Ca2+-stimulating agent produced in CD28- T-cells (Brune et al. 1991) and dexamethasone-

induced apoptosis in thymocytes can be blocked by the addition of polyamines (Desiderio et

al. 1995). In contrast, some studies show that polyamines can induce apoptosis. In a

particular study, overexpression of ODC, the key polyamine synthesis enzyme, can

accelerate apoptosis induced by withdrawal of IL-3 from murine myeloid cells and this

process is ODC dose-dependent (Packham and Cleveland 1994). Hence, addition of DFMO,

15

the inhibitor of ODC, can block ODC-induced apoptosis. Oxidation of spermidine and

spermine by amine oxidase can induce apoptosis. This apoptosis induction is due to the

production of highly oxidative H2O2 and highly reactive aminoaldehydes, which are strong

apoptosis inducers (Parchment and Pierce 1989; Ha et al. 1997).

1.3.3 Regulation of cell cycle

Polyamines have distinct distribution peaks during cell cycle progression (Heby et al. 1973).

The cell cycle is composed of four phases: G1, gap 1 or growh phase 1; S, the DNA synthesis

phase; G2, gap 2 or growth phase 2; and M, the mitotic phase. Cyclins and cyclin-dependent

kinases (CDKs) are the major cell cycle modulators. Six different cyclins (A, B1, D1, D2, D3

and E) have been identified and they function at different phases of cell cycle (Thomas and

Thomas 2001). In addition, cyclin-CDK is regulated by a group of molecules termed CDK

inhibitors. Evidence shows that polyamines are involved in cell cycle regulation based on the

following observations: (1) degradation of cyclin B1 requires the presence of polyamines and

inhibition of polyamine synthesis by DFMO leads to cell cycle retardation (Thomas and

Thomas 1994); (2) decrease of polyamines is correlated with the decrease of cyclin D1 and

cell cycle blockage in the G1 phase (Thomas, 1997); (3) p21, an CDK inhibitor, is

upregulated upon treatment with polyamine analogues, and growth arrest of melanoma cells

is induced (Kramer et al. 1999).

Section 1.4 Polyamines in embryonic development

Accumulating evidence shows that polyamines have important roles in the embryonic

development of commonly used model species. The earliest documented species for the

study of polyamine and embryonic development may be the sea urchin. Sea urchin egg

cleavage at the first, second and third stages is inhibited by α-methylornithine, an ODC

16

inhibitor (Kusunoki and Yasumasu 1978). This inhibition of egg cleavage is reversible upon

addition of ornithine or polyamines within certain ranges of concentrations. In the chick

embryo system, addition of DFMO, the ODC inhibitor, delays the development of erythroid

cells, heart beating tube and adipocyte-like cells (Heby and Emanuelsson 1981). This delay

may be associated with the disruption of nucleolar formation which is normally protected by

polyamines. Treatment of DFMO also causes pregnancy abortion in rat and rabbit (Reddy

and Rukmini 1981; O'Toole et al. 1989). In Caenorhabditis elegans, defective ODC causes

embryonic developmental retardation (MacRae et al. 1998). Addition of exogenous

polyamines can reverse this retardation.

Section 1.5 polyamines and cancer

Polyamines are essential for cell growth. However, overproduction of polyamines may

induce abnormal cell growth and lead to neoplastic transformation. Accumulation of

polyamines is often found in many cancerous tissues and body fluids of cancer patients.

Concentrations of polyamines in body fluids such as blood and urinary may be used as a

diagnostic tool for cancer and a therapeutic evaluation tool for drug treatment. Increased

ODC activity is observed in many common cancers, such as colorectal cancer, prostate

cancer, non-melanoma skin cancer, and breast cancer (Leveque et al. 2000; Milovic and

Turchanowa 2003; Schipper et al. 2003; Gilmour 2007). Most of these cancers show poor

prognostic outcomes despite extensive chemotherapy.

As such, blocking polyamine synthesis via ODC inhibitors will predictably exert a

therapeutic role during cancer treatment. A specific ODC inhibitor, DFMO, was earlier

proposed as a promising agent for cancer treatment through blocking polyamine synthesis.

However, clinical test data showed that the expected therapeutic effects in cancer were weak

17

or poor, possibly due to the compensatory effect from the polyamine uptake system.

Consistent with this view, polyamine uptake activity increased several folds after DMFO

treatment and hence overall polyamine contents of putrescine and spermidine in cancer cell

showed no obvious decrease compared to the cancer cell without DMFO treatment (Alhonen-

Hongisto et al. 1980; Bogle et al. 1994). Therefore, combinational therapy of blocking both

polyamine synthesis and its uptake system will have a more thorough effect of polyamine

depletion and hence promising response during cancer treatment. Recently, Burns and

coworkers (Burns et al. 2009) used this combinational strategy to achieve a very exciting

result upon treatment of several different cancer cell lines in a mouse cancer model. In their

study, a spermine conjugate, D-Lys (C16acyl)-Spm was found to block spermidine transport

in the nanomolar range and meanwhile cannot be taken up inside the cells. This suggests that

the conjugate is an efficient spermidine transport inhibitor. Single use of either the conjugate

or DMFO showed no growth inhibitory effect among several tested cancer cell lines,

including breast (MDA-MB-231), prostate (PC-3), melanoma (A375), and ovarian (SK-OV-

3). In contrast, combination of DMFO and the conjugate dramatically displayed several

orders of magnitude of cell growth inhibitory effect. More excitingly, the cell growth

inhibitory effect of combinational treatment in vitro was also observed in a transgenic skin

cancer mouse model with squamous cell carcinomas (SCC). 88% of SCCs showed “complete

or nearly complete remission”. This is the first study demonstrating a therapeutic effect of

polyamine depletion during cancer treatment.

All together, it is clear that polyamines play an important role in cancer cell growth. To

achieve polyamine depletion and therapeutic effects in cancers requires blockage of both

polyamine synthesis and uptake pathways.

18

Section 1.6 Pseudomonas aeruginosa PAO1

1.6.1 General property and clinical state

Pseudomonas aeruginosa (P. aeruginosa) is a Gram-negative bacterium with a large genome

(Visca et al. 2002). Although functions of many predicted open reading frames in P.

aeruginosa PAO1 genome are not clear, most identified genes have roles in the catabolism,

regulation and transport of small molecules (Visca et al. 2002). The large genome reflects its

flexibility in adaptation to diverse environmental conditions. P. aeruginosa is found nearly

everywhere, including water, soil, plants, and animals (Bonten et al. 1999). It is an important

opportunistic pathogen to patients in immuno-compromised states, such as organ-transplant

patients treated by immuno-suppressors, burn victims, cancer and neutropenia patients, HIV

1 patients, cystic fibrosis patients, and patients in intensive care units (Sadikot et al. 2005). It

can form different types of infections in patients with different conditions, such as acute and

severe infection in patients with ventilator-associated pneumonia (VAP), nosocomial

pneumonia and community-acquired pneumonia. It can also cause chronic infections in

patients with cystic fibrosis.

Unfortunately, P. aeruginosa is notoriously known as a multi-drug resistant bug and has

been found to develop resistance to nearly every antibiotic to which it is exposed, such as

fluoroquinolone, imipenem, ceftazidime, tobramycin, ciprofloxacin, polymyxin and

carbapenem (Talbot et al. 2006; Boucher et al. 2009). Therefore very limited drugs are

available for selection to fight this super-resistant bug. On the other hand, limited new drugs

are under development. Resistance of P. aeruginosa to carbapenem, one antibiotic agent in

clinical phase 2, has been recently reported (Talbot et al. 2006). Currently, the target sites of

nearly all available or under-development antibiotics largely fall into the following three

19

categories, cell wall synthesis inhibitor, protein synthesis inhibitor and topoisomerase

inhibitor (Talbot et al. 2006).

The resistance mechanisms include production of β-lactamases such as penicillinases,

cephalosporinases, and carbapenemases which digest antibiotics (Walsh et al. 2005),

activation of efflux systems to pump out antibiotics (Poole 2004), mutation of cell wall to

block the travelling of antibiotics before reaching their target sites (Gales et al. 2001), and

change of target structure (Schweizer 2003). Hence targeting efflux pump system or other

new sites may help to alleviate the increasing drug resistance in P. aeruginosa.

1.6.2 Virulence determinants

Infection severity is dictated by virulence determinants in P. aeruginosa and the host

immunological response (Sadikot et al. 2005; Driscoll et al. 2007). Virulence determinants in

P. aeruginosa include flagellum, pili, biofilm, efflux pump, type III secretion system (T3SS)

and other secreted factors such as exotoxin A, proteases and phenazines (Figure I-4). P.

aeruginosa owns a single flagellum that controls its movement and several pili that are

involved in attachment to injured epithelial cells such as in the respiratory tract, urinary tract

or surgery site. Mucoid polysaccharide alginate is responsible for obstruction of bacteria

clearance by the infected host and facilitation of biofilm formation. Biofilm is composed of a

biopolymer matrix that shields bacteria inside in the form of microcolonies and enhances

bacteria attachment to a surface. Quorum sensing is a cell-to-cell communication pathway

through small molecules termed autoinducers. Quorum sensing is a dynamic process for

bacteria to respond efficiently to their constantly changing external conditions. T3SS is a key

virulence determinant in P. aeruginosa and via the T3SS bacteria contact and inject several

20

effector proteins inside the host cell directly. It is thought that biofilm is responsible for

chronic infections whereas T3SS plays a major role in acute infections.

Figure I-4. Virulence determinants in P. aeruginosa. Figure is modified from the reference

Driscoll et al. 2007.

1.6.3 Type III secretion system (T3SS)

T3SS is a multi-protein complex in Gram-negative bacteria. This secretion system (Figure I-

5) is constituted of at least 42 genes which can be grouped into five categories: a needle

structure with a basal body across the inner membrane, the peptidoglycan layer, and the outer

membrane, which is associated with a cytoplasmic bulb and protrudes into the bacterial

extracellular surroundings; a translocation apparatus that dwells on the tip of the needle and

punches holes through the host cell plasma membrane into the cytoplasm; proteins that

regulate the process of secretion; chaperones that bind and enhance the delivery of their

cognate partners through the needle syringe; effector proteins that are injected into the host

cell through the needle and translocation apparatus (Enninga and Rosenshine 2009; Hauser

2009).

21

Of the 42 genes, a gene cluster of 36 genes, located within a chromosomal region of around

26 kb and encoded by five operons, has been identified. These 36 genes function in the

biogenesis and regulation of this T3SS complex. At least 6 other genes are scattered among

the chromosomal region which encode the four effector proteins, exoS, exoT, exoY and

exoU and two chaperones, spcS and spcU.

T3SS is regulated by environmental signals. Expression of T3SS can be induced by several

types of environmental signals, such as calcium in low concentration, contact with eukaryotic

host cells and high osmolarity (Yahr and Wolfgang 2006). ExsA, the master regulator of

T3SS and encoded by the exsCEBA operon, coordinates T3SS gene expression induced by

environmental signals, via binding to the promoter regions of known T3SS genes and

activating their expression. When T3SS is under non-inducible conditions, exsA is

sequestered by its repressor exsD while its activator exsC is inactivated by forming a

heterodimeric complex with exsE. In contrast, under inducible conditions, exsE is secreted

out of the bacterial cell through T3SS, which leaves exsC to interact with exsD and frees

exsA to activate T3SS gene expression.

22

Figure I-5. Type III secretion system. a) 42 genes are identified in the type III secretion

system. 36 genes are clustered together within a chromosomal region, encoded by five

operons. 6 others are distributed elsewhere. b) The type III secretion system can be dissected

into five components, needle complex, translocation apparatus, chaperones, regulatory

proteins and effector proteins. c) The type III secretion system is activated or inactivated via

23

regulation of the master regulator ExsA by three other regulatory proteins, ExsC, ExsE and

ExsD. Figure is adapted and modified from the reference Hauser 2009.

1.6.4 Effector proteins

As the four effector proteins (ExoS, ExoT, ExoY and ExoU) are directly translocated into the

host cell, understanding the cytotoxicity of these four effectors will be necessary to tackle the

pathogenesis of P. aeruginosa (Figure I-6).

Figure I-6. Schematic diagram of four effector proteins. S, secretion signal; CBD,

chaperone binding domain; MLD, membrane localization domain; GAP, GTPase activating

protein; ADPRT, ADP ribosyl transferase (ADPRT); PLA2, phospholipase A2; CF, cofactor

binding site. Figure is modified from the reference Hauser 2009.

ExoS

ExoS is a protein with 453 amino acids. It consists of several functional motifs (Zhang et al.

2007). The first 15 amino acids are thought to act as a secretion signal to direct ExoS into the

TTSS needle complex. Amino acids 16-51 are the potential binding site for SpcS, a

chaperone of exoS. Amino acids 52-72 are considered to be the membrane localization

domain (MLD) to allow the attachment of exoS to the plasma membrane of the host cells

24

after translocation. Amino acids 96-233, identified as a GTPase activating protein domain

(GAP), are found to affect the assembly of the host cell cytoskeleton through regulating its

substrates such as the small GTPase Rho and Rac. The disruption of the cytoskeleton

assembly within the host cell can lead to cell rounding due to cellular shape change. This

rounding is assumed to weaken the activity of phagocytosis from the host cell as reduced

internalization of P. aeruginosa within several cell types is observed. Amino acids 233-453

are identified as the ADP ribosyl transferase domain (ADPRT) and the ADPRT activity

requires the binding of a cofactor, the 14-3-3 protein (Ottmann et al. 2007). 14-3-3, which is

produced by the eukaryotic host cells, is necessary for exoS to trigger several adverse effects

on the host cell, such as induction of cell death, disruption of membrane trafficking,

endocytosis and DNA synthesis (Pederson and Barbieri 1998; Barbieri et al. 2001; Fraylick

et al. 2001; Rocha et al. 2003). Meanwhile, the recruitment of eukaryotic 14-3-3 as a cofactor

may prevent the ADP ribosyl activity of P. aeruginosa towards its own proteins resulting in

its self disruption. Clearly, domain dissection of ExoS shows it is a bifunctional toxin that

can affect the normal function of the host cell through distinct pathways.

ExoT

ExoT is a protein with 457 amino acids that shares 76% sequence identity with ExoS. ExoT

contains two domains: the amino-terminal GAP domain (amino acids 78-235) and the

carboxyl-terminal ADPRT (amino acids 235-457). The function of the GAP domain in exoT

seems to be quite similar to exoS with respect to the interference of cytoskeleton organization

and inhibition of phagocytosis and cell migration (Cowell et al. 2000; Garrity-Ryan et al.

2000). The GAP domain has also been observed to inhibit cytokinesis through inactivating

Rho GTPase (Shafikhani and Engel 2006). In contrast to the similar function shared by the

25

GAP domains of both ExoS and ExoT, the ADPRT domain in ExoT has different and limited

protein targets in the host cell. ExoT is found to specifically alter the activity of CT10

regulator of kinase (CRK) І and CRK II (Sun and Barbieri 2003), causing the inhibition of

cytokinesis and blockage of signaling to RAC1 (Deng et al. 2005; Shafikhani and Engel

2006). Like ExoS, ExoT is also a bifunctional toxin and can inhibit the phagocytosis and

cytokinesis of the host cell, and this inhibition facilitates the invasion of P. aeruginosa into

the host cell and hence its survival.

ExoY

ExoY is a protein with 378 amino acids. It has been identified as an adenylyl cyclase, with

two domains similar to the domains of other adenylyl cyclases such as edema factor and

CyaA (Yahr et al. 1998). These two domains act in concert to bind ATP. Injection of exoY

into the mammalian cell can lead to an intracellular increase of cAMP and the expression of

many different genes regulated by cAMP. Upregulation of these gene expressions can

interfere with the function of actin cytoskeleton assembly (Yahr et al. 1998) and increase the

endothelial permeability (Sayner et al. 2004). Therefore, it is conceivable that exoY, in

synergy with exoS and exoT, will make the host more vulnerable to the infections caused by

P. aeruginosa.

ExoU

ExoU is a 687-amino acid protein with strong phospholipase activity towards a wide range of

substrates, such as phospholipids, neutral lipids, and lysophospholipids (Sitkiewicz et al.

2007). It contains the binding site for chaperone spcU within residues 3-123 (Finck-

Barbancon et al. 1998), a phospholipase domain within residues 107-357 (Phillips et al.

2003; Sato et al. 2003), and a C terminal region with no recognizable domains but has been

26

reported to be important for killing the host cell (Sato et al. 2003). ExoU is highly cytotoxic

and can induce rapid host cell death. This rapid cell death induction may be closely

associated with the finding that once ExoU is translocated into the host cell, it will bind to the

membrane instantly, which might be through the recognition of its phospholipid substrate

within the membrane that leads to an increase of the permeability of the membrane (Rabin et

al. 2006). However, unlike cell death induced by ExoS and ExoT, this type of cell death is

characterized by an increased disruption of plasma membrane integrity, which more

resembles necrosis (Finck-Barbancon et al. 1997). As expected, it is found that up to 90%

cases of serious human P. aeruginosa infections can be attrituted to ExoU (Engel 2003).

Hence, blockage of exoU translocation into the host cell could have a therapeutic effect

against infections caused by P. aeruginosa.

Given the important pathogenic roles of TTSS and its effector proteins, identification of

TTSS modulators will become a hot research topic as the modulators of this system will have

the potential to become new drug targets. However, to date, only a limited number of TTSS

modulators have been reported.

1.6.5 Polyamine transport

Despite extensive studies of polyamine transport in E. coli, limited studies of polyamine

transport in P. aeruginosa have been reported. One gene cluster of spuABCDEFGH in P.

aeruginosa PAO1 was identified by Lu and coworkers (2002). In this report, the authors

proposed that spuDEFGH is a spermidine transport ABC system based on phenotypes

observed from gene knock-outs. Moreover, they found that polyamines could “induce

resistance to cationic peptide, aminoglycoside, and quinolone antibiotics in Pseudomonas

aeruginosa PAO1” (Kwon and Lu 2006). Recently, Zhou et al. (2007) found that the

27

polyamine transport pathway is closely associated with the expression of nearly all TTSS

genes. In this study, it was found that deletion of spuE or spuEFGH, which is the purported

polyamine transport system in P. aeruginosa, downregulated 34 of the 36 genes from the

clustered five operons by more than 2 folds, and the remaining two by around 1.6 folds.

Moreover, three TTSS effector genes, ExoT, ExoY, and ExoS were downregulated by 7.5-,

4.8- and 8.3-fold, respectively. Although exoU was not reported, this might be due to the

absence of ExoU in the experimental P. aeruginosa strain because it had been observed that

very few isolates contained all four effector proteins (Frank 1997). Thus, it is highly possible

that polyamine transport pathway also regulates ExoU activity for isolates that express exoU.

This global downregulation of TTSS expression indicates that polyamine transport system

could specifically regulate TTSS activity. Consequently, the polyamine transport system in

P. aeruginosa may become a new drug target against infections caused by this strain given

the link between polyamine with TTSS and antibiotic resistance.

Section 1.7 Rationale and aims

Limited studies are available to address polyamine transport in P. aeruginosa. Several

unresolved issues exist including:

• SpuD and spuE are supposed to be spermidine binding proteins in P. aeruginosa

based on the phenotypes of gene knock-outs, however, no direct evidence is available

to confirm that spuD and spuE can indeed bind polyamine.

• SpuE or spuEFGH are found to be involved in the TTSS gene regulation, however,

the role of spuD that plays in the regulation of TTSS gene expression has not been

defined.

28

• The mechanism for polyamine recognition is still not clear although the structures of

spermidine-bound potD and putrescine-bound potF have been solved. This

uncertainty may be largely due to the lack of ligand-free structures as the structures of

spermidine-bound potD and putrescine-bound potF are very similar.

• In many cases, a protein conformation will switch from an open to a closed form

upon ligand binding. However, it is not known if this can be applied to spuD and

spuE since their structures in both apo- and ligand-bound forms are not available.

The aims of this study were to:

• study the role of spuD in the regulation of T3SS gene expression.

• examine the binding between the protein (spuD and spuE) and the ligand (putrescine,

spermidine and spermine) by isothermal titration calorimetry (ITC) as this approach

can yield crucial thermodynamic information.

• solve the structures of spuD and spuE in both apo- and ligand-bound forms by X-ray

crystallography.

• delineate the mechanism of ligand binding specificity and affinity based on the

structures.

• study the ligand-induced conformational changes based on the structures.

The results of this study may have the significance in the following aspects:

• elucidation of the mechanisms for putrescine and spermidine recognition and

identification of the structural determinants for the ligand induced conformational

changes

• providing important preliminary information for the structure-based drug design

against the polyamine transport system in the P. aeruginosa

29

Chapter 2 Materials and Methods

2.1 Experimental materials

2.1.1 Strains E.coli BL21 star (DE3) Invitrogen

E.coli BL21 star (DE3) RIL Invitrogen

E.coli DH5α Invitrogen

2.1.2 Media LB 5 g Bacto-yeast extract, 10 g Bacto-trypton, 10 g NaCl (1L)

LB agar liquid LB medium supplemented with 1.5% Agar

2.1.3 Protein expression vector pGEX-6p-1 Amersham Biosciences 2.1.4. Oligonucleotides Oligonucleotides were ordered from Proligo and Research Biolabs.

A) Oligonucleotides for construct cloning (restriction and mutation sites are underlined)

BamHI-spuD-aa25-F

CGCGGATCCGCGGACAACAAGGTGCTGCACGTGTACAACTGG

EcoRI-spuD-aa367-R

CGCGAATTCTCACTTGCCGGACTTGATCTTGG

spuD-C236A-F

GGCCAACGGCAACATCGCCGTCGCCATCGGCTACTCC

spuD-C236A-R

GGAGTAGCCGATGGCGACGGCGATGTTGCCGTTGGCC

B) Oligonucleotides for reverse-transcription PCR (Work from ZLH’s lab)

30

exsC-F TATCGGTTTGCCTTCCCTGT

exsC-R TGATCGAGCAGATTGGCCAA

popB-F CGAGGGATGTCGAGAGTCTG

popB-R CGCCGACGATGATCGAAGCT

pscF-F GCAGATATTCAACCCCAACC

pscF-R GATCTTCTGCAGGATGCCTT

pscQ-F TGCAACTGGCCTTGCTGGAG

pscQ-R GTTGGCGAGGATGCGCACTT

16s rRNA gene was used as control.

2.1.5 Enzymes

Pfu DNA polymerase Stratagene

PreScission protease Amersham Biosciences

Restriction enzymes New England Biolabs

2.1.6 Kits

DNeasy Tissue Kit QIAGEN

RNeasy Mini Kit QIAGEN

ProSTAR Ultra HF RT-PCR system Stratagen

QIAprep spin Miniprep Kit QIAGEN

QIAquick Gel Extraction Kit QIAGEN

QIAquick PCR purification Kit QIAGEN

Coomassie Protein Assay Kit PIERCE

Site-Directed Mutagenesis Kit QIAGEN

2.1.7 Chromatography Columns

31

Glutathione Sepharose 4B

HiPrep Desalting Column

Hiload Superdex-75 26/60

Hiload Superdex-200 26/60

All Columns were purchased from GE Healthcare.

2.1.8 Crystallization kits and tools NeXtal DWBlock AmSO4 Suite QIAGEN

NeXtal DWBlock Anions Suite QIAGEN

NeXtal DWBlock Cations Suite QIAGEN

NeXtal DWBlock Classics Suite I-II QIAGEN

NeXtal DWBlock Classics L Suite QIAGEN

NeXtal DWBlock ComPAS Suite QIAGEN

NeXtal DWBlock Cryos Suite QIAGEN

NeXtal DWBlock JCSG Core Suite I- IV QIAGEN

NeXtal DWBlock JCSG+ Suite QIAGEN

NeXtal DWBlock MbClass Suite I-II QIAGEN

NeXtal DWBlock MPD Suite QIAGEN

NeXtal DWBlock Nucleix Suite QIAGEN

NeXtal DWBlock PACT Suite QIAGEN

NeXtal DWBlock PEGs Suite I-II QIAGEN

NeXtal DWBlock pHClear I-II Suite QIAGEN

NeXtal DWBlock ProComplex Suite QIAGEN

ADDITIVE SCREEN Hampton Research

32

Wizard I- II Emerald Biostructures

Crystallization tools Hampton Research

2.1.9 Buffers

DNA electrophoresis and Protein SDS-PAGE (Laemmli 1970)

DNA electrophoresis Buffer 40 mM Tris, 1.14% acetic acid, 1mM ethylene

diamine tetra acetic acid (EDTA)

SDS-PAGE running Buffer 25 mM Tris base, 192 mM glycine, 0.1% (w/v)

sodium dodecyl sulfate (SDS)

SDS-PAGE Staining Buffer 0.025% Coomassie Brilliant Blue R-250

30% Methanol, 10% Acetic Acid

SDS-PAGE Destaining Buffer 30% Methanol, 10% Acetic Acid

Protein purification

Cell Lysis Buffer 20 mM Tris-HCl, pH 8.0, 500 mM NaCl,

3mM dithiothreitol (DTT), Roche EDTA free

protease inhibitor tablet

Elution Buffer 20 mM Tris-HCl, pH 8.0, 500 mM NaCl,3mM

dithiothreitol (DTT), 20 mM reduced

Glutathione

Desalting Buffer 20 mM Tris-HCl, pH 8.0, 100 mM NaCl,3mM

dithiothreitol (DTT)

Gel Filtration Buffer 20 mM Tris-HCl, pH 8.0, 100 mM NaCl,3mM

dithiothreitol (DTT)

33

2.1.10 Chemicals

100 bp DNA ladders New England Biolabs

1 kb DNA ladders New England Biolabs

2-log DNA ladders New England Biolabs

Prestained protein markers New England Biolabs

IPTG Invitrogen

30% Acrylamide/Bis solution (29:1) Bio-Rad

2-Methyl-2, 4-pentanediol (MPD) Hampton Research

Polyethylene Glycol 400 Hampton Research

Polyethylene Glycol 3350 Hampton Research

Ethylene Glycol (EG) Hampton Research

Nuclease-free Water Ambion

All other chemicals were purchased from Sigma-Aldrich.

2.1.11 Instruments and Facilities

ND-1000 Spectrophotometer NanoDrop

Institute in-house DNA sequencing facility

ÄKTA HPLC system Amersham Biosciences

Jasco J-810 Circular Dichroism Spectrapolarimeter Jasco (ZLH lab)

Phoenix liquid handling robot Rigaku

In-house FRE X-ray generator Rigaku

Synchrotron beam lines

34

ID-23-1 ESRF, Grenoble France

2.2 Gene cloning and protein expression

2.2.1 Genomic DNA extraction from P. aeruginosa PAO1

Pseudomonas aeruginosa PAO1 strain was inoculated into the LB media for overnight

culture at 37°C. Cells were harvested and genomic DNA was extracted with DNeasy Tissue

Kit and precipitated in ethanol. DNA was air-dried and dissolved in buffer (10 mM Tris-HCl,

pH 7.5, 1 mM EDTA) in aliquots for storage at -20°C.

2.2.2 Gene knock-out and growth conditions (Done by ZLH’s lab)

P. aeruginosa PAO1 was used as a parental strain for generation of PAO1pClacZ reporter

strain (Zhou et al. 2007). Under the background of PAO1pClacZ, spuD and/or spuE gene(s)

were deleted either as single or double deletion using method as described by Dong et al.

2008. For ∆spuD, a 886-nt coding region (from 27 - 912 bp) was deleted; for ∆spuE, it was

the same construct as in Zhou et al., (2007); for double deletion of spuD and spuE, the same

region of ∆spuD was deleted in the background of ∆spuE.

Bacteria were routinely grown at 37°C in Luria-Bertani broth (LB) or LB medium

supplemented with the chelating reagent nitrilotiracetic acid (NTA) at a final concentration of

5 mM to induce T3SS gene expression.

2.2.3 Total RNA extraction from Pseudomonas aeruginosa PAO1

RNeasy Mini Kit was used to extract total RNA. Purified RNAs were treated with DNase I

and stored in RNase-free distilled water. The quantity and quality of RNA was examined by

electrophoresis and UV260/280 absorption (NanoDrop).

2.2.4 Reverse Transcription (Done by ZLH’s lab)

35

The reverse transcription reaction was carried out using Pseudomonas aeruginosa PAO1

total RNA with ProSTAR Ultra HF RT-PCR system (Stratagene). The reaction was heated

to 65°C for 5 minutes to inactivate the DNase I in the RNA sample prior to incubation at 42

°C for 30 minutes. First strand cDNA was stored in aliquots at -20°C.

2.2.5 β-Galactosidase assay (Done by ZLH’s lab)

β-Galactosidase activity was measured as described by Sambrook et al. (1987) by using P.

aeruginosa reporter strains based on PAO1pClacZ. The reporter strains were grown at 37°C

in LB or LB medium supplementing with NTA. Results are averages from four repeats and

given as Miller units of β-galactosidase activity per OD600.

2.2.6 PCR

General Reaction

Genomic DNA, 1st strand cDNA 1 μl Primers Forward and Reverse Primer each (10 μM) 1 μl Buffer 10X Pfu buffer 5 μl dNTPs 25 mM each of dATP, dCTP, dGTP, dTTP 1 μl Enzyme Pfu turbo DNA polymerase (2.5 U/ μl) 1 μl

Water Double-distilled water to volume of 50 μl

Site-directed mutagenesis PCR (QuikChange Site-Directed Mutagenesis Kit)

Template

Plasmid

1 μl

Primers

Forward and Reverse Primer each (10 μM)

1 μl

Buffer

10X reaction buffer

5 μl

dNTPs

40 mM each of dATP, dCTP, dGTP, dTTP

1 μl

Enzyme Pfu turbo DNA polymerase (2.5 U/ μl) 1 μl

36

Water

double-distilled water

Add volume to 50 μl

PCR Programs

General program

94°C 3 minutes, 1 cycle; 94°C 30 seconds, 55°C

45 seconds, 72°C 2 minutes, 35 cycles; 72 °C 10 minutes, 1 cycle

Sequencing

96°C 10 seconds, 50°C 5 seconds, 60°C 4 minutes, 25 cycles

Site directed Mutagenesis

95°C 30 seconds, 1 cycle; 95°C 30 seconds, 55°C 1 minute, 68°C 1 minute / kilo base of plasmid length, for 20 cycles

2.2.7 Purification of PCR products from agarose gel

PCR products were loaded and run on 1.2% agarose gel. The PCR fragments with correct

size were then excised and purified using QIAquick Gel Extraction Kit.

2.2.8 Restriction Enzyme digestion of PCR products and plasmid

Enzyme digestion reaction (volume 50 μl) Template

PCR fragments, plasmid vector

41.5 μl

Buffer

10X reaction buffer

5 μl

Enzyme

Enzyme 1

2 μl

Enzyme 2

2 μl

BSA

100X

0.5 μl

Digestion condition: 37°C for 3 hours.

37

Digested PCR products were purified using QIAquick PCR purification Kit, while digested

vectors were purified using QIAquick Gel Extraction Kit.

2.2.9 Ligation

Ligation reaction (volume 15 μl)

Ligation reaction was conducted at room termperature for 3 hours.

Buffer

10X reaction buffer

1.5 μl

Template

Digested PCR fragments

11.5 μl

Digested plasmid vector

1 μl

Enzyme

T4 ligase

1 μl

2.2.10 Transformation

5 μl of ligation products were added into 100 μl competent cell DH5α, which was put on ice.

After 5 minutes, the competent cell containing the ligation products were subjected to heat

shock at 42°C for 90 seconds and incubated in ice for a few minutes. After incubation, 400 μl

LB were added inside and gently rotated in shaker at 37°C for 45 minutes. 200 μl were

transferred to the agar plate containing 100 μg Ampicilin and spread evenly. The agar plate

was incubated at 37°C incubator for 12 hours.

2.2.11 Positive clone screening and plasmid preparation

Four clones were picked up and inoculated into 3 ml LB containing appropriate antibiotics to

grow overnight at 37°C shaker. Clones were screened by colony PCR. Plasmids from

positive colonies were extracted using QIAprep Spin Miniprep kit and submitted for

sequencing.

2.2.12 DNA sequencing

38

DNA sequencing was conducted in in-house DNA sequencing facility. DNA sequencing

components were described as below.

DNA sequencing (volume 20 μl)

Plasmid

2.5 μl

Forward or Reverse Primer each (10 μM)

1 μl

BigDye V3.1 mixture

8 μl

Distilled water

8.5 μl

2.2.13 Expression test The plasmids after sequencing confirmation were transformed into the expression

strain E.coli BL21 star (DE3). Two colonies from the expression strain were

inoculated into 3 ml LB with appropriate antibiotics and grew overnight at 37°C

shaker. 500 μl of cells were stored in 50% glycerol at -80°C. For expression test,

100μl of cells was transferred into 10 ml of fresh LB medium and cultured at 37°C.

When the optical density at OD600 reached about 0.5-0.6, 0.5 mM of Isopropyl-ß-

Dthiogalactopyranosid (IPTG) was added to induce gene expression for an

additional 3 hours at 37°C. 100μl of cells were pelleted and boiled in SDS-PAGE

loading buffer. The samples were separated by SDS-PAGE using a 12%

polyacrylamide gel.

2.3 Protein purification

Note: all steps of protein purification were done at 4°C. All samples were filtered with 0.22

µm filter before running on columns at FPLC systems.

2.3.1 Protein expression in large-scale cell culture The expression strains were inoculated into 100 ml of LB medium containing the

39

appropriate antibiotics at 37°C overnight. 60 ml of cell culture was added into 6 L of LB

medium and incubated in shaker with 200rpm at 37°C. When OD600 reached at 0.5-0.6,

protein expression was induced by addition of IPTG to a final concentration of 0.1 mM at

18°C overnight. The cell pellets were harvested by centrifugation and stored at -80°C for

protein purification.

2.3.2 Protein purification procedures 2.3.2.1 Cell lysis Frozen cell pellets from -80°C were re-suspended in the lysis buffer containing 1mg/ml Hen

egg Lysozyme. Loosened cells after Lysozyme digestion were disrupted by sonication with

10 cycles of 15-second on and 90-second off using a Soniprep 150 (SANYO). The resulting

crude extract was centrifuged for 1 hour at 18000 rpm (Beckman Coulter). The supernatant

was mixed with Glutathione Sepharose 4B beads (Amersham) equilibrated with the cell lysis

buffer.

2.3.2.2 First GST affinity column chromatography After incubation with the Glutathione Sepharose 4B beads for 1 hour at 4°C, the beads were

harvested by centrifugation at 1200rpm for 10 minutes. After extensive wash to remove non-

specific binding proteins to beads, GST-fusion proteins were eluted by 10-20 ml elution

buffer. Protein yields were determined by the Bradford method using the Coomassie Protein

Assay kit (Pierce) with BSA as a standard. PreScission protease (Amersham) was added at a

ratio of 50:1 (w/w) to cleave the GST tag off the fusion proteins. The reaction mixture was

gently rotated at 4°C overnight.

2.3.2.3 Second GST affinity column chromatography

40

After digestion by PreScission protease, samples were loaded onto a HiPrep desalting

column equilibrated by the desalting buffer. The desalted samples were passed through a

regenerated Glutathione Sepharose 4B column to remove the cleaved GST tag and non-

cleaved GST-fusion proteins. The flow through was checked by SDS-PAGE.

2.3.2.4 Gel filtration chromatography After the second GST affinity chromatography, the flow through was loaded onto gel

filtration column Superdex-75 or Superdex-200 (Amersham) pre-equilibrated in the gel

filtration buffers. The peak fractions were assessed using SDS-PAGE and the corresponding

peak fractions with high purity were pooled together. The amount of proteins was estimated

by the Bradford method and then concentrated in vivaspin 20 concentrators (5,000 MWCO,

Vivascience) to 30 mg/ml.

2.3.2.5 Analysis of polyamines by reverse phase-high pressure liquid chromatography

(RP-HPLC) (Conducted by ZLH lab)

Sample Preparation In this study, 75μl perchloric acid was added to 25μl protein solution to

extract potential polyamines. Half of the mixture (50μl) was mixed with 150μl saturated

sodium bicarbonate and 50μl dansyl chloride in acetone (9 mg/ml) (Kabra et al. 1986). The

sample was incubated at 55°C for 1 hour. After the sample was cooled down to room

temperature, 500μl toluene was added to extract the polyamines’ derivatives. 100μl of the

toluene solution was mixed with 100μl methanol and filtered through 0.2 μm filter to prepare

samples for RP-HPLC analysis.

RP-HPLC analysis RP-HPLC analysis was performed using a Waters 2695 system and

monitored by a 996 PDA and a 2475 fluorescence detector. Separation was performed on a

reverse-phase column (Phenomenex, Luna 5μ C18 250*4.6 mm) with gradient program and

41

peaks were monitored by PDA detector with λ = 350 nm and fluorescence detector with λ =

520 nm. The gradient progam was (MeOH in H2O): 55% (0 minute), 55%-65% (0-7

minutes), 65% (7-14 minutes), 65%-70% (14-18 minutes),70%-75% (18-20 minutes), 75%-

80% (20-22 minutes), 80%-100% (22-27 minutes), 100% (27-35minutes), 100%-55% (35-36

minutes), 55% (36-40 minutes).

2.3.2.6 Preparation of ligand-free protein samples

Purified and concentrated spuD and spuE were dissolved into 8M urea and loaded into

desalting column (Amersham Biosciences) equilibrated by 8M urea. After desalting, spuD

and spuE were loaded into gel filtration column (Amersham Biosciences) equilibrated by the

gel filtration buffer for on column refolding. The flow rate of gel filtration was set at 2

ml/min. The final concentration for crystallization of ligand-free spuD (hereinafter termed as

apo-spuD) and ligand-free spuE (hereinafter termed as apo-spuE) was 30 mg/ml and

20mg/ml respectively. RP-HPLC analysis showed that the ligands for both spuD and SpuE

have been successfully removed (see below).

2.4 Circular Dichroism

The circular dichroism (CD) spectra were measured on a Jasco J-810 Circular Dichroism

spectra-polarimeter at the wavelength ranging from 195 to 260 nm at a resolution of 0.1 nm.

SpuD and spuE in apo- and ligand-bound states were exchanged into 25 mM Na/K phosphate

pH 7.5 at the concentration of 25 μM and were transferred to a cell of 0.1cm path length.

Three measurements were averaged for each sample and the molar ellipticity was calculated.

2.5 Isothermal titration calorimetry (ITC)

Isothermal titration calorimetry (ITC) measurements of protein (spuD and spuE) and ligand

(putrescine and spermidine) were performed at 22°C in a VP-ITC microcalorimeter

42

(MicroCal Inc.) while ITC measurements of protein (spuD and spuE) and spermine were

conducted in a MicroCal iTC200 (MicroCal Inc.). Protein samples were exchanged by

desalting to a solution containing 20mM Tris pH 7.4 and 100mM NaCl. The protein

concentration (spuD and spuE) in cell was 20 μM, while the ligand concentration in syringe

was 100–200 μM. In the VP-ITC microcalorimeter, titrations began with one 2 µl injection

and subsequent twenty-eight 10 µl injections. In the MicroCal iTC200, titrations initiated

from one 0.5 µl injection and followed by nineteen 2 µl injections. MicoOrigin 7.0 program

was used for data analysis. Based on the solved ligand-bound structures of spuD and spuE

(see below), one-site binding model was used for data fitting to obtain the parameters of Kd,

ΔH, and stoichiometry (N).

2.6 Crystallization

Crystallization screening was performed using the sitting-drop vapor-diffusion method at

15°C by mixing 200nl protein solution with 200nl reagent solution in 96-well plates using

Phoenix liquid handling robot. Initial crystallization hits were optimized using hanging drop

vapor diffusion method and seeding in combination with additive screening (Hampton).

All crystals grew in 1 ml of reservoir solution. For the apo-spuD, 1 µl of protein sample was

mixed with 1 µl of 0.1 M Mes, 25% polyethylene glycol (PEG) 3350, pH 6.5. For the apo-

spuE, 1 µl of protein sample was mixed with 1 µl of 0.1 M Tris, 0.2 M MgCl2, 25% PEG

3350, pH 8.5. For the putrescine-bound spuD in wild type and C236A mutant, 1 µl of protein

sample was mixed with 1 µl of 0.1 M sodium acetate, 10 mM ZnCl2, 28% PEG 3350, pH 5.0.

All crystals grew to 0.1 * 0.02 * 0.02 mm3 within one week.

Prior to freezing into liquid nitrogen, all crystals were transferred in serial steps to the

corresponding cryoprotectant conditions. Crystals from apo-spuD were transferred to 0.1 M

43

Mes, 28% PEG 3350, 10% glycerol, pH 6.5. Crystals from apo-spuE were transferred to 0.1

M Tris, 0.2 M MgCl2, 28% PEG 3350, 10% glycerol, pH 8.5. Crystals from spuD in complex

with putrescine were transferred to the mother liquor containing 10% 2-Methyl-2, 4-

pentanediol (MPD), pH 5.0.

2.7 Data collection, structure determination and refinement

To study the mechanism of ligand recognition and ligand binding affinity of spuD and spuE,

four structures, apo-spuD, putrescine-bound spuD, apo-spuE and spermidine-bound spuE,

were solved (spermidine-bound spuE was solved by Siew choo Lim in our lab). In order

to study the effect of disulfide bond upon ligand recognition, the structure of putrescine-

bound spuD with C236A mutation was also solved.

2.7.1 Data collection

Data for spuD in complex with putrescine were collected at the beamline ID23-1 (European

Synchrotron Radiation Facility). Data for apo-spuD and apo-spuE were collected at local

Rigaku Cu-Kα X-ray machine with a Rigaku R-axis IV++ imaging-plate detector. Data for

apo-spuD, apo-spuE, putrescine-bound spuD in wild type and C236A mutation were

integrated with the program of Mosflm (CCP4 package) (1994). All data were scaled and

merged with the program of Scala (CCP4 package) (1994).

2.7.2 Structure determination and refinement

SpuD consists of 367 amino acids with a 24-amino acid secretion signal peptide, while spuE

is a 365-amino acid protein with a 25- amino acid secretion signal peptide. Regions for the

crystallization of spuD and spuE were amino acids 25-367 and 26-365, respectively.

Molecular replacement using PHASER in CCP4 package (1994) was carried out to find

44

initial solutions for apo-spuD, putrescine-bound spuD, putrescine-bound spuD with C236A

and apo-spuE.

Crystals of spuD in apo and putrescine-bound forms belonged to P212121 space group.

Three molecules were located in one asymmetric unit (ASU) of apo-spuD while two

molecules were included in one ASU of putrescine-bound spuD (wild type and C236A

mutation). To solve the putrescine-bound spuD, one whole chain of potF in E. coli was used

as the search model (PDB code 1A99). The searches yielded two clear solutions

corresponding to the two molecules in the ASU. Each molecule contained two structural

domains and the electron density map of putrescine in each of the two molecules was clearly

visible between the two structural domains. Model building in COOT (Emsley and Cowtan

2004) and model refinement via restrained refinement in REFMAC5 (Murshudov et al. 1997)

and simulated annealing in CNS (Brunger et al. 1998), were conducted iteratively. Water

molecules were added using CNS (Brunger et al. 1998) where chemically reasonable.

Criteria of water pick up include peaks are above 2.5 sigma units, and the minimum and

maximum hydrogen bonding distance between water and oxygen or nitrogen atom is 2.0 Å

and 3.2 Å, respectively. The final model had an R-work of 21.1%, an R-free of 25.5% for the

putrescine-bound wild-type spuD and an R-work of 20.4%, an R-free of 23.7% for the

putrescine-bound spuD C236A (shown in Table 1).

To solve apo-spuD, the two individual domains from the putrescine-bound wild-type spuD,

the amino-terminal (N) domain including residues 26-132 and 276-322, and the carboxyl-

terminal (C) domain including residues 133-275 and 342-366 were used as the search models

as one whole chain from the putrescine-bound spuD as the search model failed to yield a

reliable solution. Three clear solutions corresponding to each of the two individual domains

45

were identified. Combination of these solutions yielded three molecules in one ASU. The

linker region of residues 323-341 in each molecule was confidently added based on the omit

maps using coefficients 2Fo-Fc and Fo-Fc. Model refinement and water addition were

similar to that of putrescine-bound spuD. An R-work of 21.9% and an R-free of 28.8% were

obtained for the final model.

Apo-spuE was crystallized in P21 space group. Two molecules were located in one ASU.

Similar to the solution of apo-spuD, two individual domains from spermidine-bound spuE,

the N domain including residues 30-131 and 273-329, and the C domain including residues

132-272 and 340-360 were used as the search models. Two clear solutions corresponding to

each of the two individual domains were identified. Two molecules were built through the

combination of these solutions. The linker region 330-339 in each molecule was traced

clearly. After refinement and water addition, the final model had an R-work of 22% and an

R-free of 27% (Table 2).

46

All final structures were geometrically checked using PROCHECK (Laskowski et al. 1993).

In the final models, some residues were missing due to poor densities. They were Ala-25,

Gly-366 and Lys-367 in each chain of apo-spuD, Ala-25 and Lys-367 in each chain of

putrescine-bound spuD and putrescine-bound spuD with C236A, residues 26-27 and 363-365

in each chain of apo-spuE, and residues 26-28, and 362-365 in each chain of spermidine-

bound spuE. In addition, residues 71-73 in chain A of spermidine-bound spuE were also

disordered. Hence chain B in spermidine-bound spuE was more complete than chain A.

Superposition showed that the root-mean-square-deviation (r.m.s.d.) of all Cα atoms among

the two chains of putrescine-bound spuD, putrescine-bound spuD with C236A, apo-spuE and

spermidine-bound spuE was 0.4 Å, 0.4 Å, 0.3 Å and 0.2 Å, respectively while the average of

r.m.s.d of all Cα atoms among the three chains of apo-spuD was 0.77 Å (with range 0.6 Å-

1.0 Å). For the convenience of description and discussion, chain A from apo-spuD,

putrescine-bound spuD and apo-spuE and chain B from spermidine-bound spuE were used

below.

2.8 Other methods for structure analysis

Structures were superimposed using Top3D (CCP4 1994) and all structural figures were

prepared using PyMol (http://www.pymol.org). Dali server was used for r.m.s.d. calculation

(Holm et al. 2008). DynDom was used to identify domain boundary and hinge regions

(Hayward and Berendsen 1998). Lsqkab (CCP4 package) was used for assessment of

orientational changes between the N domain and C domain of ligand-free and ligand-bound

structures when the N domain in both structures was superimposed as a reference.

47

Chapter 3 Results

3.1 spuD and spuE in the regulation of T3SS (Done by ZLH).

SpuDEFGH was supposed to be a spermidine transport ABC system in which spuDE are

the periplasmic spermidine binding proteins (Lu et al. 2002). Zhou and coworkers (Zhou

et al. 2007) found that knock out of spuE or spuEFGH in P. aeruginosa PAO1

significantly down-regulates most of the T3SS gene expression. However the effect of

spuD upon T3SS gene expression was not tested. In order to examine the effect of spuD

upon T3SS gene expression, a reporter system was used (Zhou et al. 2007) in which the

LacZ reporter gene was fused to the promoter region of exsCBEA and integrated into the

attB site of the P. aeruginosa PAO1 strain. Three knock-outs, spuD, spuE and spuDE

were created and LacZ activity was monitored against exsA knock-out and wild type

PAO1, the negative and positive controls respectively. The results (Figure I-7a) showed

that the LacZ activity in spuD knock-out exhibited no obvious change compared to the

activity in wild type PAO1. In contrast, the LacZ activity in spuE knock-out significantly

decreased to the equivalent level of exsA knock-out. The LacZ activity in the double

knock-out spuDE was equivalent to the single spuE knock-out. These results indicate that

spuD has no regulatory role in the T3SS gene expression while spuE demonstrates a

distinct role in the regulation of T3SS.

To further test our results, quantitative reverse transcription (RT)-PCR was used. Four

genes: exsC, popB, pscF and pscQ were randomly chosen from four operons of T3SS. In

agreement with our results from the LacZ reporter system, spuD knock-out showed no

obvious effect on the expression of these four genes, while spuE knock-out significantly

down-regulated the expression of these genes (Figure I-7c). The effect of spuE knock-

48

out upon T3SS gene regulation further confirmed the findings by Zhou and coworkers

(Zhou et al. 2007). All together, these data indicate that spuD is different from spuE in

the regulation of T3SS genes in P. aeruginosa PAO1.

Figure I-7. Roles of spuD and spuE on T3SS gene expression. a) Expression of LacZ gene

under the promoter of exsCBAE in single or double mutants of ∆spuD and ∆spuE. β-

Galactosidase activity was measured in each mutant grown in LB or LB medium

supplementing with NTA. Parent strain of PAO1pClacZ was used as positive control and

exsA mutant was used as negative control. The results were averages from four repeats and

given as Miller units of β-galactosidase activity per OD600. b) Locations of genes in operons

for RT-PCR analysis were shown in red and bold. c) RT-PCR analysis of the expression of

T3SS genes in ∆spuD, ∆spuE and ∆spuDE mutants.

49

3.2 Protein purification

After the first and second GST affinity column chromatography, samples (spuE, spuD of

wild type and C236A mutant) were loaded onto gel filtration column for further purification.

Gel filtration profiles and SDS-PAGE for the peaks of these samples were shown in Figure

I-8, Figure I-9, Figure I-10.

Figure I-8. Purification of spuD in complex with putrescine. a) spuD in complex with

putrescine is eluted as monomer (peak position 180.80 ml) from the gel filtration column S75

26/60. b) SDS-PAGE shows the purity of protein from the peak fraction.

50

Figure I-9. Purification of spuD-C236A mutant in complex with putrescine. a) spuD-

C236A in complex with putrescine is eluted as monomer (peak position 245.4 ml) from the

gel filtration column S200 26/60. b) SDS-PAGE shows the purity of protein from the peak

fraction.

Figure I-10. Purification of spuE in complex with spermidine. a) spuE in complex with

spermidine is eluted as two peaks from the gel filtration column S75 26/60. The first peak

51

corresponds to the void volume at 117.1 ml, while the second peak (177 ml) is the monomer

position of spuE in complex with spermidine. b) SDS-PAGE shows the purity of protein

from the second peak.

Apo-spuD and apo-spuE were prepared as described at methods section 2.3.2.6. Gel filtration

profiles and purity of the samples checked with SDS-PAGE were shown in Figure I-11 and

Figure I-12. The majority of eluted spuD corresponded to the position of ligand-bound spuD

(see comparison with Figure I-8). In contrast, only a small portion of spuE after refolding

eluted at the corresponding position of the spuE-spermidine as in Figure I-10, while most of

the spuE was found in the void volume (Figure I-12). This indicates that the majority of

spuD and a small fraction of spuE were refolded properly.

Figure I-11. Purification of apo-spuD. a) ligand-free spuD is eluted as monomer (peak

position 177.80 ml) from the gel filtration column S75 26/60. b) SDS-PAGE shows the

purity of protein from the peak fraction.

52

Figure I-12. Purification of apo-spuE. a) ligand-free spuE is eluted as two peaks (the first

peak at 117.3 ml for the void volume and the second peak at 177.1 ml for the monomer spuE)

from the gel filtration column S75 26/60. b) SDS-PAGE shows the purity of protein from the

second peak fraction.

In order to test if spuD and spuE after urea treatment are well folded, circular dichroism

(CD) analysis was used to compare the spectra of spuD and spuE before and after urea

treatment. As shown in Figure I-13, the CD spectra of spuD in apo- and putrescine-bound

states showed quite similar pattern. Also, the CD spectra of spuE in apo- and spermidine-

bound states were close to each other. In agreement with the elution behavior observed from

the gel filtration chromatography after on column refolding, these data further support that

spuD and spuE were refolded properly after urea denaturation.

53

Figure I-13. Comparison of CD spectra of spuD and spuE in apo- and ligand-bound

states. a) Overlay of CD spectra between ligand-free and putrescine-bound spuD. b) Overlay

of CD spectra between ligand-free and spermidine-bound spuE.

54

3.3 spuD is a putrescine binding protein while spuE is a spermidine binding protein.

To determine the polyamine binding property of spuD and spuE, we expressed and

purified spuD and spuE, followed by using ITC to test their binding affinity towards

putrescine, spermidine and spermine. To our surprise, spuD and spuE showed no binding

towards all these tested polyamines (data not shown) which is in conflict with the

previous report (Lu et al. 2002) that spuD and spuE are supposed to be spermidine

binding proteins. One possibility is that spuD and spuE have already been loaded with

some ligands during the process of protein purification and the ligand pre-loading blocks

spuD and spuE to bind any additional ligands and therefore causes false negative results.

In order to test this possibility, RP-HPLC was used and we found that spuD was clearly

preloaded by putrescine and spuE by spermidine (Figure I-14).

Figure I-14. Analysis of spuD and spuE with RP-HPLC. Four polyamine standards

were used which include putrescine (1), cadaverine (2), spermidine (3) and spermine (4)

55

as labeled in the picture. The arrow indicates the presence and position of ligand released

from spuD and spuE before urea denaturation.

To assess the ligand binding affinity, we prepared ligand-free spuD and spuE (see

procedures in method for sample preparation), and repeated the ITC experiment. ITC data

(Figure I-15) showed that SpuD bound to putrescine with high affinity (Kd = 3 nM) and to

spermidine with low affinity (Kd = 6.8 μM). In contrast, spuE showed high binding affinity to

spermidine (Kd = 14.3 nM) and no detectable binding to putrescine. Both proteins showed no

binding to spermine. Despite the structural similarity between putrescine and spermidine,

spuD and spuE can still recognize their respective ligand efficiently. These results underscore

that spuD and spuE bind their each ligand with high specificity and high affinity and also

argue that spuD may be a periplasmic putrescine binding protein while spuE is a periplasmic

spermidine binding protein.

56

57

Figure I-15. Analysis of polyamine binding of spuD and spuE using ITC. A time course of

power supply of each injection was shown in the upper panel of each part. In the lower panel

of each part, the generated heats from each injection were shown in circles and fitted by one

site model in continuous line. a) Binding between spuD and putrescine. b) Binding between

spuE and spermidine. c) Binding between spuD and spermidine. d) Binding between spuE

and putrescine. e) Binding between spuD and spermine. f) Binding between spuE and

spermine.

3.4 Crystallization, data collection and model refinement

Crystals of spuD and spuE, optimized by streak seeding, are shown in Figure I-16.

58

Figure I-16. Crystal pictures of spuD and spuE. a) ligand-free spuD. b) spuD in complex

with putrescine. c) spuD-C236A in complex with putrescine. d) ligand-free spuE.

Data collection and model refinement of apo-spuD, spuD-putrescine and spuD-C236A-

putrescine is shown in Table 1. Data collection and model refinement of apo-spuE is shown

in Table 2.

59

Table 1. Data collection and refinement statistics of spuD in apo- and putrescine-bound

forms.

Data collection apo spuD spuD-putrescine spuD-C236A-putrescine

Wavelength (Å) 1.5418 0.97988 0.97988 Resolution (Å) 2.97(3.045-2.97) 2.0(2.052-2) 2.0(2.052-2) Space group P212121 P212121 P212121 Cell parameters

a/b/c (Å) 47.8/109.0/228.5 65.9/82.9/128.3 66.0/83.0/128.2 α/β/γ (°) 90/90/90 90/90/90 90/90/90 Total reflections 76892 217313 569273 Unique reflections (N) 21850 44053 48071 I/σ 8.8 (2.3) 5.9 (5.6) 9.0 (6.3) Completeness (%) 85.4 (80.1) 92 (63.5) 99.4 (98.8) Rmerge (%)a 7.3 (32.9) 7.4 (12.2) 5.7 (11.6) Refinement statistics

Data range (Å) 20-2.97 69.67-2 69.67-2 Used reflections (N) 20613 41789 45588 Nonhydrogen atoms 8157 5656 5668 Water 133 284 298 Rwork/Rfree (%)b 21.9/28.8 21.1/25.5 20.4/23.7 R.m.s. deviation

Bond length (Å) 0.007 0.01 0.007 Bond angle (deg) 1.207 1.184 1.025 Protein B factors (Å2) 40.0 17.6 13.8 Water B factors (Å2) 32.4 20.2 17.6 Ligand B factors (Å2) --- 10.5 7.2 Ramachandran plot (%)c

Most favored region 82.3 92.5 93.1 Additionally allowed region 17 7.5 6.9 Generously allowed region 0.7 0 0

60

Table 2. Data collection and refinement statistics of apo-spuE.

Data collection apo spuE Wavelength (Å) 1.5418 Resolution (Å) 2.3(2.359-2.3) Space group P21 Cell parameters

a/b/c (Å) 64.3/48.8/114.3 α/β/γ (°) 90/103.2/90 Total reflections (N) 205552 Unique reflections (N) 30163 I/σ 7.1 (4.2) Completeness (%) 96.8 (74.5) Rmerge (%)a 7.4 (17.2) Refinement statistics

Data range (Å) 20-2.3 Used reflections (N) 28592 Nonhydrogen atoms 5441 Water 205 Rwork/Rfree (%)b 22.0/27.0 R.m.s. deviation

Bond length (Å) 0.011 Bond angle (deg) 1.381 Protein B factors (Å2) 30.8 Water B factors (Å2) 30.8 Ramachandran plot (%)c

Most favored region 91.7 Additionally allowed region 8.1 Generously allowed region 0.2

Values in the highest resolution shell are shown in parentheses.

a Rmerge = Σ|Ij-<I>|/ΣIj, where Ij is the intensity of an individual reflection, and <I> is the

average intensity of that reflection.

b Rwork =Σ||Fo| - |Fc|| / Σ|Fc|, where Fo denotes the observed structure amplitude,

and Fc denotes the structure factor amplitude calculated from the model. Rfree is

61

calculated with 5.0% of randomly chosen reflections omitted from the refinement (Brunger

1992).

c Fractions of residues in most favoured/allowed/generously allowed/disallowed regions of

the Ramachandran plot were calculated according to PROCHECK (Laskowski et al. 1993).

3.5 Overall structure description

Structures of apo and putrescine-bound spuD belong to α/β type of proteins and consist of

two similar globular domains, amino-terminal domain (N domain) and carboxyl-terminal

domain (C domain), with 3 domain hinge segments (Figure I-17a, b and c). Each domain is

composed of a centeral strands-mixed β sheet flanked by α helices at both sides. N domain

contains amino acids 26-132 and 276-322 while C domain comprises amino acids 135-272

and 342-366. The three domain hinge segments are segment I with amino acids 133-134,

segment II with amino acids 273-275 and segment III with amino acids 323-341. Segment I

and segment II are from a part of two anti-parallelled β strands and sandwiched by the N and

C domain with low solvent accessibility. In contrast, segment III is composed of two short α

helices and located at the surface of the two domains with high solvent accessibility. This

suggests that segment III is more flexible than segment I and II. One disulfide bond between

Cys-173 and Cys-236 was identified in both apo and putrescine-bound spuD. This disulfide

bond is located at the beginning of two adjacent β strands within the C domain, which may

stabilize the conformation of C domain. However, r.m.s.d. between the wild type and the

mutant C236A of spuD is 0.13 Å. This shows that the disulfide bond is dispensable in

maintaining the conformation of C domain and in ligand binding specificity although Cys-

171 and Cys-214 in spuE (the equivalents to Cys-173 and Cys-236 in spuD) fail to form a

disulfide bond.

62

Similar to the overall shape of spuD, spuE in apo and spermidine-bound structures are also

composed of two globular domains, the N domain and the C domain (Figure I-17d and e)

and three hinge segments between the two domains. Each domain consists of a centeral β

sheet of mixed strands which are surrounded by α helices at both sides. The N domain

comprises amino acids 30-131 and 273-329 while the C domain includes amino acids 134-

270 and 340-360. Three hinge segments are segment I with amino acids 132-133, segment II

with amino acids 271-273 and segment III with amino acids 330-339. Segment I and II are

from two anti-paralleled β strands and link the N domain and C domain closely. Segment III

is composed of one short α helix and located at the surface of the two domains. R.m.s.d. of

all Cα atoms between the apo-spuD and apo-spuE is 1.5 Å while r.m.s.d. between putrescine-

bound spuD and spermidine-bound spuE is 0.9 Å (Figure I-18a and b). Although the overall

shape of spuD and spuE in apo and ligand-bound forms looks quite similar, close

examination showed that ligand-bound spuD and spuE are more compact than apo-spuD and

apo-spuE. This indicates that ligand binding induces a conformational change.

63

Figure I-17. Overall structures of apo- and ligand-bound spuD and spuE. The amino

terminal domain (N domain) is shown in pale green; the carboxyl terminal domain (C

64

domain) is shown in pale yellow; the hinge regions are shown in light magenta. Ligand

(putrescine and spermidine) is shown in stick model; the carbon atoms are orange, the

nitrogen atoms are blue. Disulfide bond is shown in black line. a) apo spuD. b) wild type

spuD in complex with putrescine. c) C236A spuD in complex with putrescine. d) apo spuE.

e) spuE in complex with spermidine.

PotD and potF are polyamine receptors located in the E. coli periplasm. PotD shows

preference to spermidine (Kd = 3.2 μM) and low affinity to putrescine (Kd = 100 μM)

(Kashiwagi et al. 1993) , whereas potF only shows putrescine specific binding (Kd = 2.0 μM)

(Vassylyev et al. 1998). Structure of potF in complex with putrescine is α/β type and consists

of two globular domains with three linker segments between the two domains (Vassylyev et

al. 1998). R.m.s.d. of all Cα atoms between putrescine-bound spuD and putrescine-potF

(PDB code 1A99 chain A) was 0.97 Å (Figure I-18c). This indicates the overall structures of

spuD and potF are similar.

65

Figure I-18. Superposition of Cα backbones from apo spuD (green), apo spuE (violet),

spuD in complex with putrescine (red), spuE in complex with spermidine (cyan), potF

in complex with putrescine (slate blue). Ligand (putrescine and spermidine) is shown as a

stick model. a) apo spuD and apo spuE. b) spuD in complex with putrescine and spuE in

complex with spermidine. c) spuD in complex with putrescine and potF in complex with

putrescine.

3.6 Ligand binding induces conformational changes

It is believed that ligand induced protein conformational change, in most cases, if not all,

switches between open and closed forms. The structures from apo and ligand-bound spuD

66

and spuE support this view. In spuD, one domain rotates 45.9° with respect to the other

domain when compared between the apo and putrescine-bound structures. In spuE, the

rotational angle of one domain with respect to the other domain is 38.8° between the apo and

spermidine-bound structures (Figure I-19a and b). In contrast, the two domains in spuD and

spuE show no translation upon ligand binding. Although, in spuD and spuE, rotation between

the two domains upon ligand binding is obvious, however each of the two domains is quite

rigid. The rigidity of the two domains after rotation can be reflected from the r.m.s.d. of 0.65

Å between the N domains of apo and putrescine-bound spuD, and the r.m.s.d. of 0.90 Å

between the C domains of apo and putrescine-bound spuD. Similarly, the r.m.s.d. between

the N domains of apo and spermidine-bound spuE is 0.82 Å, and the r.m.s.d. for the C

domains of apo and spermidine-bound spuE is 0.96 Å. The hinge region can be clearly

identified through the superposition of apo and ligand-bound structures though single

structure may give a rough estimation of the boundary of hinge region.

Spermidine-bound potD is closed by 43.6° compared with the apo-spuE open form, which

is 4.8° more closed as compared to that of spermidine-bound spuE. Similarly, Putrescine-

bound potF is closed by 44.9° when compared with the apo-spuD open form, which is very

close to the rotational change of the two domains (45.9°) identified in the putrescine-bound

spuD.

67

Figure I-19. Stereoview of Cα backbone superpositions from apo spuD (green), spuD in

complex with putrescine (red), apo spuE (violet), spuE in complex with spermidine

68

(cyan). Putrescine and spermidine are shown in stick models. The N domain in both spuD

and spuE is fixed as the reference domain. a) Cα superposition of apo spuD and spuD in

complex with putrescine. b) Cα superposition of apo spuE and spuE in complex with

spermidine.

In addition to the global conformational changes induced upon ligand binding, examination

of individual amino acid reveals that local changes occur largely within the three hinge

segments. Connector I and II may be active in the ligand induced hinge motion, while

connector III seems to be passive in the hinge motion as it is located at the surface of the two

globular domains and distant from the ligand binding site. Changes of dihedral angle of

ligand-free and ligand-bound spuD and spuE were shown in table 3.

Table 3. Δφ and Δψ of connector I and II from spuD and spuE in ligand-free and

ligand-bound structures.

Superposition of connector I and II from pairs of spuD and spuE in apo and ligand-bound

forms shows different shifts among these corresponding hinge amino acids (Figure I-20).

These differences may reflect the underlying ligand binding specificity for spuD and spuE.

69

Figure I-20. Stereo view and superposition of connector I and II from spuD and spuE in

both ligand-free and ligand-bound structures. Amino acids are shown in stick models.

Oxygen is shown in red; nitrogen is shown in blue. Carbon atoms in spuE are shown in

green; carbon atoms in spuD are shown in cyan; carbon atoms in putrescine and spermidine

are shown in orange. a) apo spuD and apo spuE. b) putrescine-bound spuD and spermidine-

bound spuE.

3.7 Putrescine binding site in spuD and comparison with potF

70

Putrescine is buried within a pocket formed by the N domain and C domain in spuD (Figure

I-17b). The ligand is fastened in the pocket through hydrogen bonding and van der Waal

interactions with amino acids from the N domain, C domain and hinge region (Figure I-21).

In the inner side of the pocket, the N2 of putrescine establishes hydrogen bonds to the OD2

of Asp-275 (2.78 Å) and water 4(S) (2.89 Å). This water is tethered by the OG of Ser-83

(2.60 Å) from the N domain and the OE1 of Glu-183 (2.50 Å) from the C domain. At the

other side (entry side) of the pocket, the main chain carboxyl oxygen atoms from Ser-36

(2.84 Å) and Asp-37 (3.32 Å) in the N domain and the OD1 of Asp-244 (2.73 Å) in the C

domain form hydrogen bonds with the N1 of putrescine directly. Water 51(S) makes

hydrogen bond (2.85 Å) with the N1 and this water is tightened by the O and OG of Ser-36.

Although water 32(S) forms weak hydrogen bond with the N1 (3.5 Å), it establishes a strong

hydrogen bond (2.70 Å) with the 51(S), and it is further held in position by the O of Ser-223

(2.70 Å) and the OD2 of Asp-244 (2.80 Å) from the C domain. These hydrogen bonds

converge to lock the two ends of putrescine in position.

Besides the hydrogen bond interactions between putrescine and spuD, the hydrocarbon

backbones of putrescine establish extensive van der Waal interactions with five aromatic

acids from spuD, Trp-35, Tyr-38 and Phe-311 from the N domain, Tyr-241 from the C

domain and Phe-273 from the linker region connector II (Figure I-21). All these polar and

non-polar mediated interactions together contribute to the high binding affinity of spuD

towards putrescine.

Comparison of the putrescine interaction between spuD and potF (PDB accession code

1A99 chain A) shows that most hydrogen bonding and van der Waals are well conserved.

Close examination of putrescine binding reveals a few variations (Figure I-22). Firstly, in

71

spuD, the OE1 of Glu-183 establishes a strong hydrogen bond with the N2 of the putrescine

through water 4(S) bridging. In contrast, the OE2 of Glu-185 in potF, the equivalent to Glu-

183 in spuD, makes a strong hydrogen bond with the N2 of putrescine through water

bridging. On the other hand, the OE2 of Glu-183 in spuD is fastened by the OG of Ser-180

and the OH of Tyr-241, whereas the OE1 of 185E in potF is tethered by the OG1 of Thr-138

and the NE1 of Trp-244 (the equivalent to Tyr-241 in spuD). The observation that the

different rotamers of the equivalent amino acid are involved in the ligand binding indicates

that ligand binding is determined not only by the direct amino acids but also by the adjacent

interaction surroundings of these amino acids. Secondly, the position of water 4(S) in spuD

and the corresponding water in potF shows a shift compared to the position of the other two

water 32(S) and 51(S) with their corresponding water in potF. This shift may be due to the

different interaction surrounding of Glu-183 in spuD and Glu-185 in potF. Of note, these

variations between spuD and potF are all from the N2 region (the inner side of the pocket)

while the N1 region is more conserved (the entry side of the pocket).

72

Figure I-21. Putrescine binding sites in spuD. Ligand and amino acids involved in the

ligand binding are shown as stick models. Water is shown as sphere. Oxygen and nitrogen

are shown in red and blue respectively. The N domain, C domain and hinge regions are

shown in the same color as in Figure I-17. Amino acids involved in ligand binding are

shown in the same color as their domain distributions. Carbon atoms in putrescine are shown

in orange. Hydrogen bond is shown in black dashed line.

73

Figure I-22. Side by side comparison of putrescine binding sites between spuD and potF.

Ligand and amino acids involved in the ligand binding are shown as stick models. Water is

shown in sphere. Oxygen and nitrogen are shown in red and blue respectively. SpuD is

shown in green while potF is shown in cyan. Carbon atoms in putrescine are shown in

orange.

3.8 Spermidine binding site in spuE

Similar to the spuD in complex with putrescine, spermidine is engulfed within a cleft formed

by the N and C domains in spuE (Figure I-17e). Amino acids from the N domain, C domain

and hinge region are involved in the ligand binding.

74

N1, N2 and N3 of spermidine are found to establish hydrogen bonding interactions with the

amino acids from the N domain, C domain and hinge region (Figure I-23). N1, like the N1

of putrescine, refers to the entry site of the pocket, while N2 is located in the middle of the

pocket and N3, the most inner side. The amino acids for hydrogen bonding with the N3

include Asp-180, Glu-181, Asn-269 and Ser-343. Of note, these four amino acids are all from

the C domain. The OE2 of Glu-181 (2.78 Å) and the ND2 of Asn-269 (2.80 Å) make direct

hydrogen bonds with the N3 whereas the OD2 of Asp-180 (2.88 Å) and the OG of Ser-343

(2.80 Å) establish hydrogen bonds with the N3 through water 119 (S) bridging (2.99 Å). In

the middle of the pocket, the OD2 of Asp-273 from the hinge region connector II makes a

hydrogen bond with the N2 of spermidine. In the entry site of the pocket, the O of Thr-35

(2.94 Å) from the N domain, the OD1 (2.88 Å) and OD2 (3.13 Å) of Asp-242 from the C

domain form direct hydrogen bonding with the N1 of spermidine, while the OG1 of Thr-35

(2.65 Å), the OD2 of Asp-36 (3.16 Å) and the OD2 of Asp-242 (3.24 Å) establish hydrogen

bonds with the N1 of spermidine through water 94(S) bridging (3.11 Å).

In addition to the hydrogen bonding of the three amino groups from spermidine to spuE, the

hydrocarbon chains of spermidine are found to establish van der Waal interactions with four

aromatic residues of spuE which are Trp-34, Tyr-37 from the N domain, Tyr- 239 from the C

domain, and Trp-271 from the hinge region connector II (Figure I-23).

It is noted that more hydrogen bonding interactions are established within the two ends (N1

and N3), while the middle portion (N2) only has one hydrogen bonding interaction. This

might be a strategy of interaction between proteins and their linear ligands, with the two ends

more emphasized and the middle portion less emphasized.

75

Figure I-23. Spermidine binding sites in spuE. Ligand and amino acids involved in the

ligand binding are shown as stick models. Water is shown as sphere. Oxygen and nitrogen

are shown in red and blue respectively. The N domain, C domain and hinge regions are

shown in the same color as in Figure I-17. Amino acids involved in ligand binding are

shown in the same color as their domain distributions. Carbon atoms in spermidine are

shown in orange. Hydrogen bond is shown in black dashed line.

76

Chapter 4 Discussion

Although the overall structures of spuD and spuE either in apo-or ligand-bound forms are

quite similar, which can be reflected from their low r.m.s.d. and 59% sequence identity

(Figure I-24), however, spuD and spuE exhibit sharply different ligand binding preferences.

As the N1 and N2 of spermidine are corresponding to those of putrescine, we predict that the

overlapping region between putrescine and spermidine may reflect a common feature for

putrescine and spermidine binding, i.e. between the N1 and N2 while the key factors in

determining ligand specificity of spuD and spuE possibly stem from the non-overlapping

region, i.e. between the N2 and N3 of spermidine.

To test our prediction, we compare the four structures, apo-spuD, putrescine-bound spuD,

apo-spuE and spermidine-bound spuE. Amino acids involved in aromatic stacking and

hydrogen bonding with ligand in spuD and spuE are summarized in table 4. Several common

features are found within the overlapping region of putrescine and spermidine.

77

Figure I-24. Sequence alignments of polyamine binding proteins in bacteria. From top to

bottom: spuD from Pseudomonas aeruginosa, spuE from Pseudomonas aeruginosa, potF

from E.coli, potD from E.coli. The sequence alignment was obtained by CLUSTALW

(Thompson et al. 1994), and the figure was generated by ESPrpit (Gouet et al. 1999). Strictly

conserved amino acids were shown with white letters in red, amino acids with similartity

scores > 0.7 were in red letters and amino acids with similarity scores < 0.7 were shown in

black letters. Cysteines involved in disulfide bond were labelled as digit 1 in green. The

78

secondary structural elements of spuD wild type in complex with putrescine were generated

by DSSP (Kabsch and Sander 1983) and labeled above the sequences.

Table 4. Summary of amino acids involved in aromatic stacking and hydrogen bonding

with ligand in spuD and spuE.

Nitrogen atoms of ligand in hydrogen bonding interaction

spuD spuE

N1 S36 T35 S36 via water T35 via water D37 D36 via water D244 D242 D244 via water D242 via water S223 via water s221

N2 D275 D273 S83 via water s82 E183 via water e181

N3 e183 E181 g271 N269 t182 D180 via water f345 S343 via water

Aromatic residues in van der Waals

spuD spuE

N domain W35 W34 Y38 Y37 F311 Y309

Hinge region F273 W271 C domain Y241 Y239

Amino acids involved in interactions with ligand in spuD and spuE are shown in upper case

while structurally equivalent amino acids but not involved in interactions with ligand are

shown in lower case.

79

4.1 Common features between spuD and spuE in ligand binding

First, residues from the hinge region, Phe-273 and Asp-275 in spuD and their equivalents

Trp-271 and Asp-273 in spuE interact with their respective ligand by van der Waals and

hydrogen bonding. These interactions highlight the role of hinge region played upon ligand

binding and subsequently conformational change around the hinge axis. Second, residue Tyr-

38 from the N domain of spuD forms a strong hydrogen bond with the residue Asp-275

located at the hinge, in both apo (2.71 Å) and putrescine-bound form (2.66 Å) (Fig. 6A and

6B). Similarly, Tyr-37 from the N domain of spuE establishes a hydrogen bond to the hinge

residue Asp-273 in both apo (2.50 Å) and spermidine-bound forms (2.57 Å). The hydrogen

bonding interaction between domain and hinge region may be important in transducing the

rotational signal from the hinge region to the domain and then coordinating domain

movement upon ligand binding. In contrast, no amino acid from the C domain in both spuD

and spuE is found to make hydrogen bonding to the hinge amino acid. This may indicate that

the N domain interacts with the ligand earlier than the C domain due to the close link

between the N domain and the hinge amino acid. Third, we notice that the distance between

the CG of Tyr-38 and the CG of Asp-275 in apo-spuD (6.09 Å) and putrescine-bound spuD

(6.30 Å) is close to the distance between the N1 and N2 of putrescine (5.95 Å) (Figure I-25a

and b). Similarly, the distance between the CG of Tyr-37 and the CG of Asp-273 in apo-spuE

(6.37 Å) and spermidine-bound spuD (6.37 Å) coincides well to the distance between the N1

and N2 of spermidine (6.13 Å) (Figure I-25a and b). Thus it seems that Tyr-38/Asp-275 in

spuD and Tyr-37/Asp-273 in spuE act as a ruler to identify the region between the N1 and

N2 of putrescine and spermidine. In this ruler, Asp-275D in spuD and Asp-273D in spuE

form a strong hydrogen bond to the N2 of putrescine and spermidine respectively, while Tyr-

80

38 in spuD and Tyr-37 in spuE establish van der Waal interactions to hold the corresponding

hydrocarbon atoms of putrescine and spermidine in place. Moreover, this correlation between

the corresponding Tyr/Asp and the N1 and N2 of putrescine and spermidine is also observed

in potD and potF (PDB code 1A99, 1POT, 1POY). Although the TpPotD is in complex with

Mes (PDB code 2V84), the distance between the CG of Tyr-17 and the CG of Asp-240

(equivalent to Tyr-38 and Asp-275 in spuD respectively) is 6.27 Å which is close to the

distance from the corresponding amino acid residues in spuD, spuE, potD and potF. This

further supports our proposal that the Tyr/Asp forms a ruler to recognize the region between

the N1 and N2 of putrescine and spermidine. And this ruler may also explain the observation

that spermidine binds to spuE in a unidirectional manner is due to the asymmetrical

distribution of nitrogen atoms (N1, N2 and N3) in spermidine.

81

Figure I-25. Stereo view and superposition of amino acids from spuD and spuE to show

the similarities and differences in ligand recognition. Amino acids are shown as stick

models. Water is shown as sphere. Oxygen is shown in red; nitrogen is shown in blue.

Carbon atoms in spuD are shown in cyan; carbon atoms in spuE are shown in green; carbon

atoms in putrescine and spermidine are shown in orange. Distance between the N1 and N2 of

82

putrescine and between the CG of 38Y and CG of 275D in spuD is shown in dashed lines. a)

apo-spuD and apo-spuE. b) putrescine-bound spuD and spermidine-bound spuE.

All together, these common features may reflect a conserved mechanism for the recognition

of linear ligands such as polyamines. As suggested by Vassylyev and coworkers (Vassylyev

et al. 1998), the N domain moves earlier than the C domain upon ligand binding based on the

observation that more amino acids from the N domain in potF are involved in ligand binding.

Based on this assumption, we further propose that polyamine binding proteins may undergo

three steps of conformational change upon ligand binding. Polyamine is first recognized by

the hinge amino acids and the N domain then moves towards the ligand due to the hydrogen

bonding connection between the N domain and the hinge amino acid. The rotational change

of the hinge region and movement of the N domain during this process induces the shift of

the C domain towards the ligand, burying it between the two domains. Also, we suppose that

the interaction between the N domain and the hinge amino acids may be the reason to address

that more amino acids from the N domain are involved in the ligand binding.

4.2 Differences between spuD and spuE in ligand binding

Despite the common features shared by spuD and spuE in ligand binding, our ITC studies

clearly reflect that the two proteins have different ligand binding specificities. Structural

comparison of ligand-free and ligand-bound forms between spuD and spuE reveals several

notable differences within the non-overlapping region of putrescine and spermidine, i.e.

between the N2 and N3 of spermidine. First, the OD1 of the hinge amino acid Asp-275 in

apo-spuD establishes a hydrogen bond to the OG of Ser-83 (2.57 Å), and this hydrogen bond

is maintained in the form of putrescine-bound spuD (2.58 Å). In contrast, Asp-273 in apo-

83

spuE (the equivalent to Asp-275 in spuD) is 4.57 Å away from Ser-82 (the equivalent to Ser-

83 in spuD), but a hydrogen bond (2.71 Å) between Asp-273 and Ser-82 is induced upon

spermidine binding (Figure I-25a and b). Second, besides hydrogen bonding interaction

between Ser-83 and Asp-275 in spuD, Ser-83 also hydrogen bonds to the Asn-63 in both apo

(3.03 Å) and putrescine-bound (2.91 Å) spuD. However this hydrogen bond is only observed

between Ser-82 and Asn-62 (the equivalents to Ser-83 and Asn-63 in spuD respectively) in

spermidine-bound spuE (2.77 Å) while not in apo-spuE (distance 4.52 Å) (Figure I-25a and

b). Third, in the apo-spuD, the OE2 of Glu-183 establish two strong hydrogen bonds to the

OG of Ser-180 (2.48 Å) and the OH of Tyr-241 (2.87 Å) (Figure I-25a). These two strong

hydrogen bonds are also observed in the Ser-180 (2.76 Å) and Tyr-241 (2.63 Å) upon

putrescine binding (Figure I-25b). In contrast, the OE2 of Glu-181 in the apo-spuE (the

equivalent to Glu-183 in spuD) fails to establish hydrogen bonding interactions with the Ser-

178 and Tyr-239 (the equivalents to Ser-180 and Tyr-241 in spuD respectively) (Figure I-

25a); however a hydrogen bond is observed between the OE1 of Glu-181 and the ND2 of

Asn-135 in both apo-spuE and spermidine-bound spuE while the hydrophobic Ile-136 of

spuD, the equivalent to Asn-135 in spuE, fails to establish hydrogen bond interaction to Glu-

183 in both apo-spuD and putrescine-bound spuD. Fourth, Phe-273 in spuD is replaced by

Trp-271 in spuE to establish van der Waals to their respective ligand. Fifth, Asn-269, Asp-

180 and Ser-343 are found to hydrogen bond to spermidine either directly or through water

bridging. In contrast, Gly-271, Thr-182 and Phe-345, their equivalents in spuD, fail to

establish hydrogen bonding with putrescine.

It is noted that three amino acids, Ser-83, Glu-183 and Phe-273 in spuD and their

corresponding amino acids Ser-82, Glu-181 and Trp-271 in spuE are involved in most of

84

these differences. As these residues are located in the non-overlapping region, we propose

that they are the determining factor of ligand specificity via interaction at the region occupied

between the N2 and N3 of spermidine.

4.3 Possible ligand binding process in spuD and spuE

Based on the identified contrast and common features, the ligand binding in spuD and spuE

may undergo the following processes. The putrescine first interacts with the Asp-275 in the

hinge region of spuD, and the strong hydrogen bonding between Asp-275 and Ser-83,

induces the OH of Ser-83 to shift to a position to coordinate a water molecule with the help

of Glu-183 (Figure I-25b). Besides establishing a strong hydrogen bond to the N2 of

putrescine, this water determines the depth of the putrescine binding pocket and blocks the

way of the ligand to slide deeper. The importance of this water can be addressed by the

observation that disruption of this water by the mutation of Ser-85 (equivalent to Ser-83 in

spuD) to alanine or Glu-185 (equivalent to Glu-183 in spuD) to glutamine significantly

decreases the binding affinity of potF to putrescine and putrescine uptake activities in E.coli

(Vassylyev et al. 1998). Meanwhile, the hydrogen bond between Tyr-38 and Asp-275

induces other amino acids from the N domain to move in place. Rotation of the hinge region

and movement of the N domain induce the amino acids other than Glu-183 from the C

domain to shift to the proper position. Together, putrescine is locked in place through

hydrogen bonding and van der Waal interactions with the two domains and the hinge region.

On the other hand, the hydrogen bonding interaction between Asp-275 and Ser-83 and

subsequent recruitment of water through the cooperation with Glu-183 upon ligand binding

clearly prevents the binding of larger ligand such as spermidine and spermine. The distance

of water molecule W(4S) in putrescine-bound spuD is 1.4 Å, 1.6 Å and 1.9 Å to the carbon

85

atoms C5, C6 and C7 of the spermidine, respectively; the distance between the OE1 of Glu-

183 in putrescine-bound spuD and the N3 of spermidine is 1.2 Å (shown in Figure I-25b).

And presence of Gly-271 and Phe-345 in spuD (the equivalents of Asn-269 and Ser-343 in

spuE) clearly fails to hydrogen bond to spermidine. Moreover, the substitution of Trp-271 in

spuE by Phe-273 in spuD weakens the van der Waal interactions to spermidine due to the

lack of one more ring in the side chain of Phe-273. All these differences may explain the

reason of relatively low binding affinity (Kd = 6.8 μM) of spuD towards spermidine.

In the apo spuE, the lack of hydrogen bonding between Asp-273 and Ser-82, and different

side chain orientation of 181E disrupts the subsequent recruitment of the water molecule in

the right position that will greatly weaken the binding ability of spuE towards putrescine.

Moreover, Trp-271 in spuE has a larger hydrophobic side chain than its equivalent Phe-273

in spuD. Substitution of Phe-273 by Trp in spuD may disfavor the presence of water

molecule W(4S) in position. This is supported by sequence alignment of spuD, spuE, potF

and potD (Figure I-24). As shown in the alignment, potF with high binding affinity towards

putrescine contains Phe-276 (the equivalent of Phe-273 in spuD), while potD with high

binding affinity towards spermidine has Trp-255 (the equivalent of Trp-271 in spuE). This

highlights that recruitment of the water to suitable position to hydrogen bond the N2 of

putrescine in spuD requires the proper conformation of Ser-83 and Glu-183 and presence of

Phe-273, but not Trp at that position. However the different conformation of Ser-82 and Glu-

181 and presence of Trp-271 in spuE, compared to their equivalents in spuD, act together to

disfavor the presence of water, which is very important in fastening putrescine in position.

These differences explain our ITC analysis that spuE has no detectable binding towards

putrescine. The absence of water molecule in spuE due to the unique conformations of Ser-

86

82 and Glu-181 and presence of Trp-271 allows the N2 and N3 part of spermidine to occupy

the pocket. Moreover, unlike the side chain of Glu-183 in spuD, the side chain of Glu-181 in

spuE is oriented such that it allows the N2 and N3 part of spermidine to move inside the

pocket without blockage. Furthermore, Glu-181 strengthens spermidine binding by

establishing a strong hydrogen bond with the N3 of spermidine (Figure I-25b). The

importance of Glu-181 of spuE in spermidine binding can be addressed from the previous

mutation study that E171Q of potD in E. coli (the equivalent to Glu-181 in spuE)

significantly decreases the binding affinity and uptake activity of potD towards spermidine

(Kashiwagi et al. 1996). At the same time, the interaction of Tyr-37 from the N domain and

Asp-273 from the hinge region induces the shift of other amino acids in the N domain to

establish interactions with the N1 of spermidine. Subsequently, the amino acids other than

Glu-181 from the C domain shift to hold spermidine tightly inside the pocket.

4.4 polyamine and T3SS in Pseudomonas aeruginosa

T3SS is an important virulence determinant for Gram-negative bacteria pathogens.

Identification of its key regulators and then design of new antibiotics targeting the key

regulators may become a promising alternative for the treatment of associated infectious

diseases. It is clear that spermidine transport system is such a key T3SS regulator, while

putrescine transport appears to have no role in the regulation of T3SS based on the effects of

spuD knock-out in our study. However, we cannot rule out the possibility that putrescine

transport is also involved in the regulation of T3SS. The negative result from our analysis

might indicate that other unidentified putrescine transport system exists in P. aeruginosa

PAO1 and hence compensates the potential effect of spuD knock-out. Consistent with this

view, we notice that several possible polyamine transport genes are in the Pseudomonas

87

genome database v2 (http://www.pseudomonas.com) with gene identification number

PA1410, PA2711 and PA2592 respectively. Although their function is not confirmed, it

seems that P. aeruginosa PAO1 owns multiple polyamine transport genes and the gene

multiplicity underscores the important roles of polyamine involved in the metabolism of P.

aeruginosa PAO1 and possibly in its virulence.

In summary, spuD is found to be a putrescine binding protein and has no effect on the

T3SS gene expression. In contrast, spuE acts as a spermidine binding protein and shows a

significant role in the T3SS gene expression. Structural comparison of the apo- and ligand-

bound spuD and spuE provides novel insight into the mechanism of ligand binding

specificity, as well as a common mechanism of polyamine binding protein upon ligand

induced conformational change. We suggest that Ser-83, Glu-183 and Phe-273 in spuD and

their corresponding amino acids Ser-82, Glu-181 and Trp-271 in spuE play important roles in

determining their respective ligand binding specificity. These findings deepen our knowledge

of ligand recognition and polyamine transport in P. aeruginosa and may set a solid

foundation for later drug design against polyamine transport in P. aeruginosa as polyamines

have close links with the antibiotics resistances of this strain (Kwon and Lu 2006). However,

it will be intriguing if spuD and spuE share the common spuFGH for membrane transport. As

previous studies of polyamine transport in E. coli showed that the major putrescine and

spermidine ABC systems are distinct (Kashiwagi et al. 1990; Furuchi et al. 1991). Further

studies will be needed to address the polyamine transport system of P. aeruginosa in detail.

88

Part II In vitro assembly of decapping activator Lsm1p-7p-Pat1p complex and a

preliminary structural study of Pat1p carboxyl terminal region

Summary

Decapping is a critical and rate-limiting step in the 5’-3’ mRNA decay pathway. It is

regulated by several decapping activators. One important decapping activator is Lsm1p-7p-

Pat1p, an eight-protein complex. This complex prefers deadenylated mRNAs destined for

degradation over polyadenylated mRNAs for translation. However, it is not clear how this

eight-protein complex is assembled and how it interacts with mRNA. In this part, we aim to

decipher the interaction networks within the complex and study the mRNA binding

properties of this complex and roles of each individual component in the process of mRNA

binding. Through Native-PAGE, we identified the interaction networks between Pat1 and the

Lsm proteins. We found that the carboxyl terminal region (C-region) of Pat1 alone is

sufficient to interact with the Lsm1-7 heptameric complex and successfully assembled the

eight-protein complex in vitro. This sets a firm foundation for future structural studies. We

also tested the mRNA binding properties of the eight-protein complex and each individual or

subcomplex component in mRNA binding. Our study revealed that Lsm2 and Lsm3 are

important not only for facilitating binding between the Pat1 C-region and Lsm1-7 protein

complex but also for Lsm1-7-Pat1-mediated mRNA binding. These findings have not been

reported previously. In addition, we crystallized and performed a preliminary structural study

of the C-region of Pat1.

89

Chapter 1 Introduction

1.1 Post transcriptional regulation

Post transcriptional regulation of gene expression is a key aspect in eukaryotes. Regulation

includes several key steps: mRNA processing (5’ capping, splicing and 3’ poly (A) tail

formation), transport, surveillance, translation and degradation (Figure II -1). Regulation on

these key steps effects the control of quality and quantity of mRNA, which consists of a pool

of non-translated and translated transcripts. mRNA decay and mRNA translation are often in

a competitive relationship (Jacobson and Peltz 1996; Coller and Parker 2004). mRNA decay

is coupled with translation repression, while mRNA translation is coupled with the repression

of mRNA decay.

90

Figure II -1. Post transcriptional regulation points of eukaryotic mRNA. Pre-mRNA is

transcribed from DNA by RNA polymerase II (RNAP II). Pre-mRNA is processed to mature

mRNA by 5’-end capping, 3’-end cleavage, polyadenylation and splicing. After processing,

mature mRNAs are assembled into messenger ribonucleoprotein particles (mRNPs) and then

exported from the nucleus to the cytoplasm via the nuclear pore complex (NPC). In the

cytoplasm, mRNA undergoes a quality check for any aberrant transcripts. Non-sense

mediated mRNA decay (NMD) is activated when a pre-stop codon is harbored within a

transcript. After the quality check, the mRNA is translated into protein or degraded based on

the dynamic balance between translational pool and degradative pool. Figure is modified

from the reference Aguilera 2005.

1.1.1 5’-end processing

The nascent mRNA with a length of 22-25 nucleotides is capped with a 7-methyl guanosine

at its 5’-end by cap enzymes (Moteki and Price 2002). Cap enzymes in yeast include Cet1-

Ceg1 with triphosphatase and guanylyltransferase activities, and Abd1 with guanine-N7-

methyltransferase activity (Aguilera 2005). Capped nascent pre-mRNA is recognized by the

nuclear cap binding protein complex (CBC). The interaction between the cap structure and

CBC complex is important for subsequent pre-mRNA processing (Das et al. 2003).

1.1.2 3’-end processing

After transcription, most pre-mRNAs in eukaryotes have to be modified at the 3’-end by

cleavage and polyadenylation. 3’-end processing requires both cis-elements in the pre-

mRNAs and trans-factors. Cis-elements in mammals include three primary sequence

fragments that determine the polyadenylation and cleavage site. The three cis-elements are

91

the conserved AAUAAA polyadenylation signal (PAS), the cleavage site and the less

conserved downstream elements (DSE). These cis-elements are recognized by the cleavage

and polyadenylation specificity factors (CPSF) and poly(A) polymerase (PAP). After correct

positioning of the cis-elements by these trans-factors, the poly(A) tail will be synthesized to

around 200-300 bases in length. Polyadenylation at the 3’-end have important roles in the

regulation of transcription, splicing, translation and mRNA stability (Mandel et al. 2008).

1.1.3 Splicing

A characteristic feature in eukaryotes is that most genes contain introns. There are several

advantages of harboring introns inside genes. Firstly, transcription of one pre-mRNA via

alternative splicing may generate more mRNA transcripts and hence provide more specific

and subtle functions within different cell types. Secondly, mutations inside introns do not

alter the final mRNA transcript, thus maintaining the stability and function of the synthesized

protein. Splicing of pre-mRNAs is carried out in the spliceosome, which is conserved from

yeast to human (Mandel et al. 2008). Spliceosome is a highly dynamic and protein-rich

machine with around 170 proteins (Wahl et al. 2009). The spliceosome comprises five small

ribonucleoprotein particles (snRNPs), U1, U2, U4, U5 and U6. Seven Sm proteins (B, D1,

D2, D3, E, F and G) are found to be involved in the biogenesis of U1, U2, U4 and U5

snRNPs through the interaction with a conserved sequence present in these snRNPs (Branlant

et al. 1982). These snRNPs are assembled into a series of stepwise discrete complexes. The

early (E) complex is initiated by the interaction of the U1 snRNP with the 5’ splice site.

Following the E complex formation, the A complex (or pre-spliceosome) is formed by the

association of the U2 snRNA–branch-region duplex. After recruitment of the B complex

(U4-U5-U6 snRNPs) to the A complex, this bigger complex is rearranged to yield the mature

92

spliceosome C complex (U1-U2-U4-U5-U6). The C complex functions in splice site

recognition, intron excision and exon ligation. Concomitant with splicing, the exon junction

complex (EJC) is recruited to the newly processed mRNAs at the region 24-nucelotide

upstream of the exon junction sites. The EJC functions in a series of cellular processes,

including mRNA export from the nucleus to the cytoplasm, subcellular localization,

translation and quality control (Tange et al. 2004; Giorgi and Moore 2007).

1.1.4 Export

After processing, the pre-mRNAs turn into mature mRNAs. The mature mRNAs are

assembled into mRNPs and associated with the transcription export complex (TREX). TREX

is composed of the THO complex and several export factors such as Sub2, Yra1, ALY.

Through ALY of TREX, two major export receptors, the heterodimer Nxf1-Nxt1 is recruited

to the mRNPs. Via the Nxf1-Nxt1 pathway, mRNPs are docked to NPC and then transported

across the nucleus and released into the cytoplasm (Carmody and Wente 2009).

1.1.5 Translation

Following export into the cytoplasm, mRNAs are recruited into the ribosomes for protein

synthesis. The ribosomes are highly conserved across all species from prokaryotes to

eukaryotes. The ribosomes (70S in bacteria and 80S in eukaryotes) are composed of two

subunits: the small subunit (30S in bacteria and 40S in eukaryotes) and the large subunit (50S

in bacteria and 60S in eukaryotes). The small subunit functions in the correct base pairing of

the triplet codons in the mRNA with the anticodons of the corresponding tRNAs, and

decodes the genomic information into the peptide sequence information. The large subunit

contains the peptidyl-transferase center and functions in peptide bond formation. Three tRNA

binding sites have been identified and they cooperate closely for peptide synthesis. These

93

three tRNA binding sites are the aminoacyl-tRNA binding site (A site), peptidyl-tRNA

binding site (P site), and deacylated tRNA binding site (E site) (Steitz 2008).

Translation contains four stages: initiation, elongation, termination and ribosome recycling.

In eukaryotes, the 40S small subunit with Met-tRNA and initiation factors eIFs 1, 1A, 2, 3

and 5 is recruited to the 5’ capped mRNA via the cap binding complex eIF4F (composed of

eIF4A, eIF4E and eIF4G). Then the 40S subunit with these initiation factors scans along the

mRNA until a start codon AUG is encountered. Through base pairing, the anticodon Met-

tRNA within the 40S subunit is anchored to the start codon and then triggers the recruitment

of the large subunit 60S to assemble into a whole ribosome 80S for translation initiation and

elongation. Translation terminates when a stop codon in the A site is encountered. Eukaryotic

class I release factor 1 (eRF1) in conjunction with class II release factor eRF3 hydrolyze the

ester-linked peptide in the peptidyl-tRNA of the P site and the newly synthesized peptide is

released from ribosome. After peptide release, the whole ribosome is disassembled into the

two distinct subunits for a subsequent round of translation (Steitz 2008; Sonenberg and

Hinnebusch 2009). Translation control is very important in maintaining cellular homeostasis

and it mainly occur at the initiation stage (Sonenberg and Hinnebusch 2009).

1.2 mRNA decay

After the functions of mRNAs are fulfilled, they enter the degradation pathway. mRNA

decay is an important post-transcriptional checkpoint in the aspect of quantitative and

qualitative control of transcripts. The importance of quantitative control of mRNAs is to

maintain a dynamic balance between the pools of non-translated and translated transcripts. In

contrast, aberrant mRNAs, generated during the process of mRNA synthesis either from

transient mistakes or from inherited genetic mutations, may interfere with the functions of

94

normal transcripts and cause any associated disorders. Hence, efficient elimination of these

aberrant transcripts requires quality surveillance systems.

1.2.1 Normal mRNA decay

In eukaryotes, two major pathways of normal mRNA decay have been identified (Parker and

Song 2004), namely the 5’-3’ pathway and the 3’-5’ pathway (Figure II -2). Both pathways

are initiated by a common step, poly(A) tail shortening (deadenylation) by a Ccr4/Pop2/Not

deadenylase complex (Meyer et al. 2004; Parker and Song 2004). In the 5’-3’ pathway, after

deadenylation, the transcripts are subject to decapping by the decapping enzyme

(Dcp1/Dcp2), and then further degraded by the 5’-3’ exonuclease Xrn1 (Decker and Parker

1993). Alternatively, in the 3’-5’ pathway, following deadenylation, the transcripts are

degraded by the exosome, a cytoplasmic protein complex with 3’-5’ exonuclease activity

(Anderson and Parker 1998). The remaining oligonucleotide cap structure after exosome

cleavage is digested by the DcpS scavenger decapping enzyme (Liu et al. 2002).

95

Figure II -2. Normal mRNA decay pathways in eukaryotes. Figure is modified from the

reference Coller and Parker 2004.

mRNA decay can occur in cytoplasmic loci referred to as processing bodies (P bodies or

DCP bodies or GW bodies), which are conserved in eukaryotes (Sheth and Parker 2003). P

bodies are visible granules when viewed under a light microscope and dynamic sites for non-

translational mRNA transcripts either for decay or for storage (Parker and Sheth 2007).

Although the exact components of P bodies are not clear, components identified to date can

be roughly categorized into three classes (Parker and Sheth 2007). First, a large multi-subunit

core of proteins forms the general translation repression/decay machinery. This core is

conserved from yeast to mammals. The core is composed of the decapping hetero-enzyme

Dcp1/Dcp2, the decapping activators Pat1p, Lsm1p-7p complex, Dhh1p, Edc1p, Edc2p and

Edc3p, the 5’-3’ exonuclease Xrn1p, and the deadenylase complex CCR4/POP2/NOT

96

(Shown in Figure II -3) (Parker and Sheth 2007). Second, mRNAs under translation

repression or decay are found in P bodies (Teixeira et al. 2005). P bodies disrupt when

treated with RNase A (Teixeira et al. 2005). This is strong evidence to show that the

assembly of P bodies requires the existence of RNAs and it suggests that the size and number

of P bodies may be associated with the quantity of non-translated mRNAs within P bodies.

Consistent with this view, the number and size of P bodies increase when a non-translating

mRNA fragment is overexpressed in yeast (Teixeira et al. 2005). Third, some specific

components restricted to particular organisms or mRNAs are found in P bodies. For example,

the nonsense-mediated mRNA decay factors Upf1p, Upf2p and Upf3p have been found in P

bodies of certain Saccharomyces cerevisiae mutants (Sheth and Parker 2006). On the other

hand, some proteins that are involved in microRNA silencing have been detected in

metazoan P bodies, but not in Saccharomyces cerevisiae P bodies (Jakymiw et al. 2005; Liu

et al. 2005).

Figure II -3. The core components of mRNA decay and translation repression

machinery in P body. Figure is modified from the reference Coller and Parker 2004.

1.2.2 mRNA surveillance pathways

97

Abnormal mRNAs may occur during the process of mRNA synthesis. These aberrant

transcripts could compromise the functions of normal transcripts. Three types of mRNA

surveillance systems have been identified to minimize these abnormalities, including the

nonsense-mediated mRNA decay (NMD), non-stop mRNA decay (NSD), and no-go mRNA

decay (NGD) (Clement and Lykke-Andersen 2006; Isken and Maquat 2007) (Figure II -4).

NMD

NMD is a translation-dependent surveillance system that detects mRNAs containing a

premature translation termination codon (PTC) and triggers elimination of such aberrant

transcripts. Synthesis of C-terminal truncated proteins may exert a dominant-negative effect

by competing with the functions of respective full length proteins (Holbrook et al. 2004). It

has been observed that around one third of the known human diseases associated with

mutations result from the production of non-sense mRNAs (Holbrook et al. 2004). This

observation stresses the pathological role of these aberrant transcripts. Therefore,

understanding the mechanism of NMD may bring in a promising therapeutic strategy to deal

with these diseases (Welch et al. 2007).

In the NMD, the key issue for cells to make a decision is how to distinguish normal

translation termination codon from pre-mature stop codon. In mammal cells, two NMD

models have been proposed, namely, an EJC-dependent premature termination and a faux 3’-

untranslated region (UTR) mediated premature termination (Wen and Brogna 2008).

Normally, an EJC is located within 20-24 nucleotides upstream of the exon-exon boundary.

During the process of translation, the EJC is replaced by ribosome translation machinery.

However, when a stop codon located 50-55 nucleotides upstream of the last exon-exon

boundary is encountered, this stop codon is defined as a PTC and translation machinery fails

98

to displace the EJC. Subsequently, NMD is triggered through the recruitment of NMD

factors to the EJC (Maquat 2004). Recently, several observations challenge this EJC-

dependent premature termination model. For example, NMD could occur even in the absence

of 3’ most intron and the efficiency of NMD is affected by the length of 3’ UTR (Buhler et

al. 2006). Another example is that a NMD sensitive β-globin mRNA shows resistance to the

NMD by artificially modifying the relative position between the poly(A) binding protein

(PABP) and a downstream PTC (Silva et al. 2008). These observations indicate an alternative

NMD model in mammals, namely the faux 3’-untranslated region (UTR) mediated premature

termination, which has been observed in S. cerevisiae (Amrani et al. 2004).

Exact NMD trans-factors vary among species, however, three core factors, up-frame shift 1

(Upf1), Upf2 and Upf3, are highly conserved across species and form a surveillance complex

with the Upf2 sandwiched by the Upf1 and Upf3. This complex can physically interact with

the ribosome and EJC through Upf1 and Upf3 respectively (Wen and Brogna 2008).

Moreover, this complex has been found to interact directly or co-immnopurifiy with the

general mRNA decay components such as DCP2, Xrn1, exosome, and deadenylase PARN

(He and Jacobson 1995; Lykke-Andersen 2002; Lejeune et al. 2003). These interactions with

the general mRNA decay machinery could direct non-sense mRNAs to decay either in a

deadenylation-independent manner or through deadenylation and 3’-5’ decay pathway

(Mitchell and Tollervey 2000; Lejeune and Maquat 2005). In contrast to the mRNA decay

from the both ends, the decay of non-sense mRNAs in Drosophila occurs through an

endonucleolytic cleavage near the PTC region, followed by exonucleolytically rapid

degradation from both directions (Gatfield and Izaurralde 2004).

99

Figure II -4. mRNA surveillance pathways. a) The mRNA with a premature stop codon is

recognized by NMD factors and degraded throught the exonucleolytic pathways from both

ends. b) The mRNA without stop codon is degraded by the exosome from the 3’ end. c)

The mRNA with a stem-loop secondary structure hinders translation elongation. This

hindrance is recognized by the no-go decay factors Dom34/Hbs1 and cleaved by an

unidentified endonucleose. This figure is adapted from Clement and Lykke-Andersen 2006.

NSD

Non-stop mRNAs have been studied in S. cerevisiae and human. Non-stop mRNAs cause

ribosome stalled at the 3’ end and leave an empty A site in the ribosome. This stall may

trigger the NSD through the recruitment of exosome-associated Ski7 factor to the A site and

100

thus degrade the aberrant transcripts by the exosome in 3’-5’ direction (van Hoof et al. 2002).

This process is independent of deadenylation as the lack of stop codon may disrupt a normal

environment of 3’ UTR and interfere with the formation of poly(A) tail. Alternatively, NSD

can be initiated in 5’-3’ pathway when Ski7 is absent (Isken and Maquat 2007). The

physiological significance of NSD may include two aspects: to eliminate the dead-end

transcripts and to release the stalled ribosome for synthesis of other mRNAs (Isken and

Maquat 2007).

NGD

NGD has only been studied in S. cerevisiae. It was identified through an artificial insertion of

stem-loop structure into mRNAs, resulting in the blockage of translation elongation (Doma

and Parker 2006). This blockage triggers NGD through endonucleolytic cleavage of mRNAs

near the position where translation elongation hinders. The resulting free ends of mRNAs are

further degraded by the exosome and Xrn1 from the directions of 3’ and 5’ respectively. Two

proteins, Dom34 and Hbs1, are found to play an important role in NGD. Dom34 is a

paralogue of eRF1, which structurally resembles a tRNA molecule while Hbs1 is related to

eRF3 and belongs to the translational GTPase subfamily. It is thought that Dom34 can

recognize the stalled ribosome and thus trigger the endonucleolytic cleavage of the no-go

mRNAs. However the identity of the endonuclease is still not known.

In summary, the three mRNA surveillance systems are all closely associated with the

translation processes. When translation stalls abnormally either at the elongation or the

termination stage, mRNA surveillance systems will be activated to eliminate these aberrant

transcripts through the recruitment of the general mRNA decay machinery.

1.2.3 Decapping regulation

101

Decapping is a rate-limiting step in 5’-3’ mRNA decay. Two broad decapping regulators

(decapping activators and decapping inhibitors) have been identified. Decapping inhibitors

comprise the poly (A) tail binding protein (Pab1p) (Caponigro and Parker 1995) and cap

binding protein eIF-4E (Schwartz and Parker 1999; 2000). The mechanism of decapping

inhibition by Pab1p might be through two possible ways. First, Pab1p interacts directly with

eIF-4G, a component of the cap binding complex eIF-4F (Tarun and Sachs 1996). eIF-4F can

recruit the 40S ribosome subunit to the mRNA through the interaction of its subunit eIF-4G

with eIF3, a ribosome bound initiator factor. Studies show that the cap binding complex can

inhibit decapping (Schwartz and Parker 1999; 2000). Therefore, the inhibition of decapping

by Pab1p can be through its interaction with the cap binding complex eIF-4F. Second, Pab1p

interacts with the cap structure directly in vitro and therefore may block the decapping

activity of Dcp1/Dcp2 by this interaction (Khanna and Kiledjian 2004). Decapping

activators contain Pat1p, Lsm1p-7p, Dhh1, and Edc1-3 proteins (Schwartz et al. 2003; Coller

and Parker 2005).

Dhh1 is a DEAD-box RNA helicase that is conserved across species. It is required for

translation repression and stimulation of decapping in vivo (Coller and Parker 2005). Edc1

and Edc2 are basic proteins with low sequence identity with each other. They were found to

enhance the efficiency of decapping in vitro by direct binding to the mRNA substrate (Coller

and Parker 2004). Edc3 contains a Sm-like domain at the N-terminus, a central FDF domain

and a C-terminal YjeF-N domain. It enhances decapping and P-body formation, possibly

through the dimerization of YjeF-N domain (Ling et al. 2008). Besides being a general

mRNA decay factor, Edc3 can also modulate the specific Rps28B mRNA decapping (Coller

102

and Parker 2004). Details of Pat1p and Lsm1p-7p in decapping activation are reviewed

below.

1.2.4 Pat1 in decapping, P-body formation and translation repression

Pat1 is an important protein functioning in the processes of mRNA decapping, P body

formation and translation repression (Pilkington and Parker 2008). Pat1 is conserved in

Saccharomyces cerevisiae, Drosophila melanogaster, and mammalian cells (Eulalio et al.

2007; Scheller et al. 2007). The interaction partners of Pat1 in Saccharomyces cerevisiae

include Dcp1, Dcp2, Lsm1-7, Dhh1 and Xrn1 (Coller and Parker 2004). Deletion of Pat1p in

yeast shows a significant reduction in translation repression under glucose deprivation

(Coller and Parker 2005) and a significant defect in decapping (Tharun et al. 2000).

Overexpression of Pat1 in Saccharomyces cerevisiae enhances P-body assembly (Coller and

Parker 2005). In Saccharomyces cerevisiae, Pat1 is a 796-amino acid protein (Figure II -5).

Distinct functional domains in Pat1 have been identified based on deletion studies

(Pilkington and Parker 2008). The region of amino acids 254-422 determines the decapping

activity of Pat1 and deletion of this region has a similar effect to that of deletion of full length

Pat1 on decapping efficiency. The region of amino acids 422-763 is involved in promoting

Pat1’s accumulation in P bodies, translational repression, and association with Lsm1 and

recruitment of Lsm1 to P bodies. Furthermore, this region is responsible for cell growth

inhibition when overexpressed. In addition, Pat1 is an RNA-binding protein, in which the

regions of amino acids 254-422 and 422-763 interact independently with poly (U).

Moreover, the poly (U) binding property seems to be specific as the interaction can not be

competitively inhibited by an excess of poly (C).

103

Figure II -5. Pat1 from Saccharomyces cerevisiae contains several conserved regions.

Figure is modified from the reference Pilkington and Parker 2008.

1.2.5 Lsm proteins in pre-mRNA processing and mRNA decay

Lsm proteins (Sm like proteins with molecular weights between 8-28 kDa) are a conserved

family of proteins characterized by a “Sm motif” identified in Sm proteins (Hermann et al.

1995; Seraphin 1995). The Sm motif is composed of an amino-terminal α helix, followed by

five anti-parallel β strands (Kambach et al. 1999). This motif forms two domains, namely the

Sm domain I and Sm domain II. The Sm domain I contains α helix and a β sheet (β1-3) while

the Sm domain II comprises β strands β4-5. The two domains are generally connected by a

long loop region. Eight Lsm proteins (Lsm1-8) are conserved in eukaryotes, and two distinct

heptameric complexes, known as Lsm1-7 and Lsm 2-8, are formed (Salgado-Garrido et al.

1999; Bouveret et al. 2000; Tharun et al. 2000). Lsm1-7 complex is localized in the

cytoplasm and functions as mRNA decapping activator (Salgado-Garrido et al. 1999;

Bouveret et al. 2000) but is not involved in splicing (Bouveret et al. 2000). In contrast, Lsm2-

8 is localized in the nucleus, binds to U6 snRNA, forms the core of U6 related snRNP, and

functions in pre-mRNA splicing (Salgado-Garrido et al. 1999). Meanwhile, Lsm2-8 is also

involved in the degradation of pre-mRNAs and mRNAs restricted in the nucleus (Kufel et al.

2004), although it is not involved in mRNA decay in the cytoplasm (Tharun et al. 2000).

104

Lsm1-7 is found to physically interact with several mRNA decay factors from the 5’-3’

decay pathway based on two-hybrid interaction and coimmunoprecipitation studies. These

factors include Pat1, Dhh1, the decapping enzyme Dcp1, Dcp2, the 5’-3’ exonuclease Xrn1,

and Upf1 as well as mRNA (Fromont-Racine et al. 1997; Bouveret et al. 2000; Tharun et al.

2000; Coller and Parker 2004).

Chowdhury et al. (2007) found that Lsm1-7-Pat1 complex can bind to mRNA near its 3’-

end, and the presence of a stretch of U residues near the 3’-end of mRNA enhances this

complex binding. More importantly, this complex interacts strongly with oligoadenylated

mRNAs when compared with polyadenylated mRNAs. This property distinguishes

oligoadenylated mRNA destined for degradation from polyadenylated mRNA destined for

translation. This may be the basis of the in vivo function of the complex as a decapping

activator and translation repressor. However, when synthetic homo-oligoribonucleotides (20-

mers) were used to test the RNA binding of this complex, only oligo(U) was observed to

have a detectable binding with this complex. On one hand, this finding supports the general

binding preference of Sm-like protein complexes for U-rich sequences (Achsel et al. 1999;

Achsel et al. 2001; Toro et al. 2001; Schumacher et al. 2002), and, on the other, failure of

interaction with oligo(A) indicates that the stronger interaction with oligoadenylated mRNAs

over that with polyadenylated mRNAs not only depends on the presence of oligo(A) tail but

also other adjacent sequences such as a stretch of U residues or yet unidentified sequences. In

addition to the well-known role of Lsm1-7-Pat1 complex as an activator in mRNA decay,

Lsm1-7 proteins also play a role in the stabilization of certain mRNAs. It has been known for

about 20 years that the 5’-end of many orthopoxvirus mRNAs contains a unique

nontemplated poly(A) tract of 5 to 40 residues, although the function of this poly(A) tract is

105

not clear (Patel and Pickup 1987; Schwer et al. 1987; Ahn and Moss 1989; Ink and Pickup

1990). Recently, Bergman et al. (2007) found that the presence of this 5’ poly(A) tract is able

to stabilize mRNAs by repressing mRNA decay in both pathways, i.e. 5’-3’ and 3’-5’.

Moreover, Lsm1-7 complex is identified to be a trans-factor that specifically associates with

and stabilizes the mRNAs containing a 5’ poly(A) tract of 10 to 21 in length. It is proposed

that Lsm complex may bind to the 5’ and 3’ ends of such mRNAs simultaneously and

therefore inhibit mRNA decays from the both ends. However, it seems that binding of 5’

poly(A) tract by the Lsm1-7 complex is not direct as addition of excess poly(A) failed to

inhibit this binding. This observation is similar to that made by Chowdhury et al. (2007),

which underscores that binding of either 3’ oligo(A) tail or 5’ poly(A) tract by Lsm1-7

complex requires other sequences yet unidentified.

Recently, the crystal structure of Lsm3 from Saccharomyces cerevisiae was reported

(Naidoo et al. 2008). Lsm3 forms an octamer ring structure with an α-helix face at one end

and a loop face at the other end (Figure II -6a, 6b and 6c). Each monomer is characterized

by the typical “Sm motif” with an α helix at the N terminal, followed by five anti-paralleled β

strands (Figure II -6d). Based on the available interaction networks and structure of Lsm3,

an updated model for the organization of Lsm1-7 and Lsm2-8 heptameric ring is shown in

Figure II -6e and 6f.

106

Figure II -6. Crystal structure of Lsm3 and models for Lsm1-7 and Lsm2-8. Helix is

shown in blue; loop region between β3 and β4 is shown in magenta; other regions are shown

in green. a) Lsm octameric ring viewed from the helix face. b) Lsm octameric ring viewed

107

from the loop face. c) A side view of Lsm octameric ring. d) Structure of Lsm3 monomer. e)

Model for Lsm1-7 viewed from the helix face, colored by electrostatic potential distribution

(red for negative charge and less than -0.5 kT/e, blue for positive charge and more than 0.5

kT/e. f) Model for Lsm2-8. Orientation and color are identical to Lsm1-7. Figure a-d was

generated from PDB code 3BW1 and e-f was adapted from the reference Naidoo et al., 2008.

1.3 Rationale and aims

Structural study of Lsm1-7-Pat1 may provide important insight into the mechanisms of

mRNA recognition and decapping activation by this complex. The primary challenge in the

structural study of this eight-protein complex is that sufficient quantity and purity of the

complex have to be achieved. Currently, there is no protocol of successful assembly of the

complex with sufficient quantity and purity suitable for structural study, although through a

tandem affinity strategy, the Lsm1-7-Pat1 complex and Lsm1-7-Pat1-Xrn1 complexes have

been purified from Saccharomyces cerevisiae (Bouveret et al. 2000; Chowdhury et al. 2007).

However, in both studies, Pat1 shows several bands indicating protein degradation during the

process of purification. Moreover, yields of the recovered complexes are low due to the

intrinsic expression under the native promoter. The low yields make the structural study of

this complex difficult. Evidence has shown that the C-region of Pat1 may be the region that

interacts with Lsm1-7 complex (Pilkington and Parker 2008). Although it is clear that Lsm1-

7 and Pat1 can form an eight-protein complex, the subunits in the Lsm1-7 complex that are

responsible for Pat1 interaction are not known. Also the role that each protein plays in the

process of RNA binding is unclear though it has been established that Lsm1-7-Pat1 complex

can bind mRNA near the 3’ end containing a U-rich region (Chowdhury et al. 2007).

The aims of this study were to:

108

• reassemble the Lsm1-7-Pat1 complex of good quality and sufficient quantity in vitro.

• delineate the interaction networks within the Lsm1-7-Pat1 eight-protein complex

• explore the individual roles of the Lsm1-7-Pat1 complex in RNA binding

• solve the structure of Pat1 C-region by X-ray crystallography.

Chapter 2 Methods

Most methods and materials for protein purification and crystallization are similar to the

methods described in Part I unless otherwise stated.

2.1 Gene cloning, protein expression Primer lists for Saccharomyces cerevisiae Pat1 (short for Pat1 hereafter) (restriction site

is underlined)

BamHI-Pat1-aa465-F CGCGGATCCGGTGCCACAAATTTGAACAAATCTGGG BamHI-Pat1-aa473-F CGCGGATCCGGGGGCAAAAAATTCATTCTTG XhoI-Pat1-aa796-R CGCCTCGAGTTACTTTAGTTCTGATATTTCACC Primer lists for Saccharomyces cerevisiae Lsm (short for Lsm hereafter) (Done by Dr.

Shimin Jiang in our lab and restriction site is underlined)

Lsm1-BamHI-F: CGGGATCCATGTCTGCAAATAGCAAG Lsm1-XhoI-R: CCGCTCGAGTTAGTACATGTCAGATTT Lsm4-BamHI-F: CGGGATCCATGCTACCTTTATATCTT Lsm4-XhoI-aa117-R: CCGCTCGAGTTAAAATTCGACCTTTTG

109

Lsm2-BamHI-F: CGGGATCCGATGCTTTTCTTCTCC Lsm2-HindIII-R: CCCAAGCTTATTTTCTTTCAGTCAT Lsm3-NdeI-F: GGAATTCCATATGGAGACACCTTTGGAT Lsm3-KpnI-R: GGGGTACCTTATATCTCCACTGCGCC Lsm5-BamHI-F: CGGGATCCGATGAGTCTACCGGAGATT Lsm5-HindIII-R: CCCAAGCTTACAACGCCTCCGTAGG Lsm6-NdeI-F: GGAATTCCATATGTCCGGAAAAGCTTCT Lsm6-KpnI-R: GGGGTACCCTATATTTTTTGTTCACT Lsm7-BamHI-F: CGGGATCCGATGCATCAGCAACACTCCCAAAGGAAAAAATTCGAAGGCCC Lsm7-HindIII-R: CCCAAGCTTCTATTTTTGCATATATAG Cloning vector lists pGex-6p-1 Pat1 constructs, Lsm1, Lsm4 pETDuet-1 Lsm2, 3, 5, 6 pACYCDuet-1 Lsm7

Expression strains BL 21 (DE3) RIL for all Pat1 constructs. BL 21 (DE3) for all Lsm constructs. SDS-PAGE

As the molecular weights of Lsm2, 3, 4, 5, 6 and 7 are similar, in order to achieve good

visualization on an SDS-PAGE gel, we used a commercial NuPAGE 4-12% Bis-Tris gel

(Invitrogen) according to the manufacturer’s instructions.

2.2 Protein purification

110

Buffer lists for all constructs using pGex-6p-1 as the cloning vector including Pat1

constructs and Lsm1 and 4:

Cell Lysis Buffer

20 mM Tris-HCl, pH 8.0, 500 mM NaCl,

2mM dithiothreitol (DTT), Roche EDTA free

protease inhibitor tablet

Elution Buffer 20 mM Tris-HCl, pH 8.0, 500 mM

NaCl,2mM dithiothreitol (DTT), 20 mM

reduced Glutathione

Desalting Buffer 20 mM Tris-HCl, pH 8.0, 100 mM

NaCl,2mM dithiothreitol (DTT)

Gel Filtration Buffer 20 mM Tris-HCl, pH 8.0, 100 mM

NaCl,2mM dithiothreitol (DTT)

Buffer lists for all constructs using Duet-1 as the cloning vector including Lsm2-3 and

Lsm5-6-7.

Cell Lysis Buffer 10 mM HEPES, pH 7.6, 200 mM NaCl,

Roche EDTA free protease inhibitor tablet

Wash Buffer 10 mM HEPES, pH 7.6, 200 mM NaCl, 5mM

imidazole

Elution Buffer 10 mM HEPES, pH 7.6, 200 mM NaCl,

250mM imidazole

111

Gel Filtration Buffer 10 mM HEPES, pH 7.6, 100 mM NaCl, 2mM

dithiothreitol (DTT)

Briefly, GST-fused proteins were purified following cell lysis, first GST affinity column

chromatography, elution by reduced glutathione, overnight cleavage by PreScission protease,

desalting, and second GST affinity column chromatography to remove the cleaved GST tags

and any uncleaved GST-fusion proteins. Target proteins were further purified by gel

filtration chromatography using appropriate buffers as listed above. Both Pat1-465c and

Pat1-473c were concentrated to 10 mg/ml. The final concentrations of Lsm1 and Lsm4 were

4 mg/ml. SeMet-incorporated Pat1-465c and Pat1-473c constructs were expressed in a

minimal medium containing 20 mg/l seleno-L-methionine (SeMet), and purified in the same

way as that of native Pat1 proteins except 10 mM DTT was included at every step of

purification.

Duet-1 expressed proteins were purified using TALON Resin (CLONTECH). Cells were

sonicated and cell pellets were removed through centrifugation. Supernatants were mixed

with TALON Resin for 1 hour at 4°C. The resins were harvested and washed extensively

using wash buffer. Final target proteins were eluted with elution buffer and further purified

by gel filtration chromatography. Final concentrations of protein were 20 mg/ml for Lsm2-3

and 15 mg/ml for Lsm5-6-7.

2.3 Native-PAGE

Gel preparation and running was similar to SDS-PAGE except SDS was not included in the

gel and running buffer. Samples of different combinations at equal molar ratio were

incubated at room temperature for 30 min before running in the gels except the preparation of

Lsm1-7-Pat1-465c complex, in which Lsm1, Lsm4, Lsm2-3 and Lsm5-6-7 were incubated

112

together for 15 min before addition of Pat1-465c for another incubation of 15 min.The gels

were run at room temperature at constant current 12 mA. The percentage of stacking gel and

running gel was 5% and 8% respectively. The amount of protein loaded into the lanes of each

gel was comparable.

2.4 Reconstitution of Lsm2-3-Pat1-465c, Lsm1-7 and Lsm1-7-Pat1-465c

As for the reconstitution of trimeric Lsm2-3-Pat1-465c, equimolar amounts of Lsm2-3 and

Pat1-465c were mixed and incubated at room temperature for 20 minutes. As for the

reconstitution of Lsm1-7 and Lsm1-7-Pat1-465c, individual protein (Lsm1, Lsm4, and Pat1-

465c) or subcomplexes (Lsm2-3 and Lsm5-6-7) were mixed in equal molar in 8M urea for 20

minutes at room temperature. High urea concentration was supposed to disrupt any potential

ring formation ability of Lsm2-3 and Lsm5-6-7 and enhance the assembly of Lsm1-7. After

incubation, these mixtures were loaded into gel filtration column S200 26/60 (Amersham) for

purification which was preequilibrated in the same buffer as that for purification of Lsm2-3

and Lsm5-6-7.

2.5 Crystallization

Crystal screening was carried out using a Phoenix liquid handling robot (Rigaku). Initial

crystallization hits were optimized using the hanging drop vapor diffusion method and

seeding in combination with additive screening (Hampton).

All crystals grew under 1 ml of reservoir solution. For the semet-Pat1-465c, 1 µl of protein

sample was mixed with 1 µl of 0.1 M Tris HCl, 10 mM DTT, 10% MPD, 20% polyethylene

glycol (PEG) 3350, pH 7.4. For the semet-Pat1-473c, 1 µl of protein sample was mixed with

1 µl of 0.085 M Hepes Na, 16% PEG 3350, 15% glycerol, 6% MPD, pH 7.5. For the Lsm2-

113

3-Pat1-465c, 1 µl of protein sample was mixed with 1 µl of 0.05 M Hepes Na, 6% PEG

3350, pH 7.0.

2.6 Data collection

Se absorption edge data of Pat1-465c and Pat1-473c were collected at beamline ID23-1

(European Synchrotron Radiation Facility). As crystals decayed quickly after Se absorption

edge data set, inflection and remote data set were not processed.

2.7 Surface Plasmon Resonance (SPR)

SPR was carried out on a Biacore 2000 instrument (Biacore). The 5’ end biotin labeled

ssRNAs 15mer polyU purchased from DHARMACON was attached to a streptavidin-coated

sensor chip (Biacore). A buffer of 10 mM HEPES, pH7.6, 100 mM NaCl, 2mM Mgcl2 was

flowed across the chip until the baseline was stable. The 5’ end biotin-labeled polyU was

then attached to flow cell 2 by injecting 20 μl of 100nM RNA in 0.3 M NaCl at a flow rate of

5 μl/min. After immobilization, flow cell 2 and reference flow cell 1 were blocked with 100

μl of 1 mg/ml biotin at 5 μl/min. A binding buffer of 10 mM HEPES, pH7.6, 100 mM NaCl,

2 mM MgCl2, 2 mM DTT was flowed across flow cells 1 and 2. Proteins were dialyzed into

the binding buffer prior to injection. 60 μl of protein with a serial concentration gradient of

50-1000 nM was injected across the chip at 30 μl/min. Dissocation was followed with 3-

minute interval. Regeneration was done after each cycle of association and dissociation by

injection of 30 μl 1M NaCl at 30 μl/min. The data were fitted to a 1:1 binding mode with

mass transfer using BIAevaluation 3.1.

Chapter 3 Results and Discussion

3.1 Protein purification

114

After the first and second GST affinity column chromatography, Pat1-465c, Pat1-473c,

Lsm1, and Lsm4 were further purified using gel filtration chromatography. The gel filtration

profiles and protein purity checked with NuPAGE are shown in Figure II -7, Figure II -8,

Figure II -9 and Figure II -10 for Pat1-465c, Pat1-473c, Lsm1, and Lsm4, respectively.

Figure II -7. Purification of semet-Pat1-465c. a) Elution profile of gel filtration

chromatography from S200 26/60. b) Peak fraction at 228.6 ml was checked by NuPAGE.

115

Figure II -8. Purification of semet-Pat1-473c. a) Elution profile of gel filtration

chromatography from S200 26/60. b) Peak fraction at 232.9 ml was checked by NuPAGE.

Figure II -9. Purification of Lsm1. a) Elution profile of gel filtration chromatography from

S200 26/60. b) Peak fraction at 229 ml was checked by NuPAGE.

116

Figure II -10. Purification of Lsm4. a) Elution profile of gel filtration chromatography

from S200 26/60. b) Peak fraction at 244.5 ml was checked by NuPAGE.

Duet-1 expressed proteins of Lsm2-3 and Lsm5-6-7 were purified using TALON Resin

affinity chromatography and followed by gel filtration chromatography. The gel filtration

profiles and protein purity checked with NuPAGE are shown in Figure II -11 and Figure II

-12 for Lsm2-3 and Lsm5-6-7 respectively.

117

Figure II -11. Purification of Lsm2-3. a) Elution profile of gel filtration chromatography

from S200 26/60. b) Peak fraction at 218.9 ml was checked by NuPAGE.

Figure II -12. Purification of Lsm5-6-7. a) Elution profile of gel filtration chromatography

from S200 26/60. b) Peak fraction at 222.4 ml was checked by NuPAGE.

3.2 Crystallization and data collection

Crystals of semet-Pat1-465c, semet-Pat1-473c and Lsm2-3-Pat1-465c are shown in Figure

II -13.

118

Figure II -13. Crystals of Pat1 C-regions alone and in complex with Lsm2-3. a) Semet-

Pat1-465c crystals. b) Semet-Pat1-473c crystals. c) Native Lsm2-3-Pat1-465c crystals.

Data collection statistics for these crystals are listed in Table 4.

119

Table 5. Se absorption edge collection for semet Pat1-465c and semet Pat1-473c

Data collection Pat1-465c Pat1-473c

Derivative SeMet SeMet

Wavelength (Å) 0.9795 0.9792

Resolution (Å) 2.86(2.95-2.86) 2.82(2.91-2.82)

Space group P212121 P21

Cell parameters a/b/c (Å) 53.9/91.3/131.2 36.9/176.6/174.4

α/β/γ (°) 90/90/90 90/90.08/90 Unique reflections (N) 19716 68572

I/σ 4.4 (2.0) 5.6(2.4)

Completeness (%) 99.9 (99.9) 100 (100)

Rmerge (%)a 7.1 (37.1) 8.3 (30.8)

Values in the highest resolution shell are shown in parentheses.

a Rmerge = Σ|Ij-<I>|/ΣIj, where Ij is the intensity of an individual reflection, and <I> is the

average intensity of that reflection.

3.3 Identification of interaction networks among Pat1-465c and seven Lsm proteins

from Lsm1 to Lsm7

It is known that Pat1, possibly through the C-region of Pat1, can interact with Lsm1-7

complex (Pilkington and Parker 2008); however it is not known which subunits of this

complex interact with Pat1. To address this question, we used Native-PAGE to test the

different combinations of Pat1-465c with all Lsm proteins (Figure II -14). Figure II -14

shows that Pat1-465c has no interactions with Lsm4 (see lane 3 and 4) and Lsm5-6-7 (see

lane 7 and 8). In contrast, a thick band appeared after mixing Pat1-465c with Lsm2-3 (see

lane 5 and lane 6), which indicates that Pat1-465c binds to Lsm2-3 strongly. In accordance

120

with this observation, the amount of Lsm2-3 in lane 6 decreased when compared to the

amount of Lsm2-3 in lane 5, which further confirms the interaction between Pat1-465c and

Lsm2-3. However, it is not clear if Pat1-465c interacts with Lsm1 as the positions of the two

proteins on the gel were very close (see lane 1 and lane 2). But it seems that the band density

in lane 2 corresponding to the position of Lsm1 turned a little denser compared to the band

density of Lsm1 in lane 1. Thereby it is possible that Pat1-465c also interacts with Lsm1

although the interaction strength may not be as strong as the interaction between Pat1-465c

and Lsm2-3. Lsm1, Lsm4, Lsm2-3 and Lsm5-6-7 were mixed at equal molar ratio and

incubated at room temperature for 30 minutes and then loaded in lane 9. Of note, two major

bands emerged, the lower band is likely from the abundant Lsm2-3 3 (compared to lane 5

and lane 6 for the position of Lsm2-3) and the upper band seems to correspond to the

position of Lsm4 (compared to lane 3 and lane 4 for the position of Lsm4). However a shift

has occurred for Lsm 1 and Lsm5-6-7 in lane 9 when compared with lane 1/lane 2 for Lsm1

and lane 7/lane 8 for Lsm5-6-7, based on the shift, it is possible that the upper band in lane 9

may not be Lsm4 alone but be the Lsm1-7 complex. When Pat1-465c was added into the

mixture of lane 9, the upper band and the lower band in lane 9 completely shifted up to form

a sharp band just below the Pat1-465c band (lane 10). This indicates that the upper band in

lane 9 is not the Lsm4 alone but may contain the complex of Lsm1-7 as Lsm4 alone cannot

bind to Pat1-465c (see lane 3 and 4). Also, the lower band in lane 9 should correspond to

Lsm2-3 as Lsm2-3 can interact with Pat1-465c strongly (see lane 5 and lane 6). Therefore,

the sharp band in lane 10 may contain two components, the Lsm1-7-Pat1-465c eight-protein

complex and Lsm2-3-Pat1-465c three-protein complex.

121

Figure II -14. Interaction networks between Pat1-465c and different Lsm proteins

evaluated by Native-PAGE. Samples in each lane are shown at the left side of the gel.

Taken together, these results indicated that Pat1-465c interacts with Lsm1-7 efficiently and

this interaction is mainly through the subunits of Lsm2 and Lsm3.

To test the model of Lsm1-7 in Figure II -6, different combinations of Lsm proteins were

incubated and run on Native-PAGE as shown in Figure II -15.

122

Figure II -15. Interaction networks of Lsm proteins by Native-PAGE. Proteins in each

lane were labeled at the top left of each gel.

In Figure II -15a, two new bands appeared after mixing Lsm1 with Lsm2-3 as labeled with

red arrows although the density of the two bands was weak. This indicates that Lsm1 can

interact with Lsm2-3 but with low affinity. Lsm1 showed no interactions with Lsm4 and

Lsm5-6-7 as indicated in Figure II -15b and Figure II -15c. Also Lsm2-3 failed to interact

with Lsm4 (shown in Figure II -15d). In sharp contrast, a strong band appeared as labeled

123

with the red arrow in Figure II -15e, which indicates that Lsm2-3 interacts with Lsm5-6-7

with high affinity. However, it is not clear if Lsm4 can interact with Lsm5-6-7 as the two

proteins stack closely on the gel, making the results hard to interpret (shown in Figure II -

15f).

Taken together, our data are in agreement with the model provided by Naidoo et al. (2008).

It seems that this heptameric Lsm1-7 ring complex is mainly contributed by the interaction

between Lsm2-3 and Lsm5-6-7. The weak interaction between Lsm1 and Lsm2-3 and the

lack of interaction between Lsm1 and Lsm4 indicate that incorporation of Lsm1 and sclms4

into the heptameric ring structure is dependent on Lsm2-3 and Lsm5-6-7.

3.4 Reconstitution of Lsm2-3-Pat1-465c, Lsm1-7 and Lsm1-7-Pat1-465c

Reconstitution of trimeric complex containing Lsm2-3 and Pat1-465c is shown in Figure II -

16. A peak shift on the gel filtration was observed after the formation of the trimeric complex

(eluted at 193.4 ml) compared to the peak position of Pat1-465c (228.6 ml from Figure II -7)

and Lsm2-3 (218.9 ml from Figure II -11).

Recently, reconstitution of human Lsm1-7 have been reported by Zaric and coworkers

(2005). Through analytical ultracentrifugation and electron microscopy, they found that

human Lsm2-3 and Lsm5-6-7 are a ring-shaped and oligomeric structure in which Lsm2-3 is

an octameric molecule [Lsm2-3]4 and Lsm5-6-7 a hexameric one [Lsm5-6-7]2. In order to

disrupt the ring structure of the subcomplexes (Lsm2-3 and Lsm5-6-7) and reconstitute the

human Lsm1-7 heptameric complex, they took a strategy with several steps as described

below. Individual protein (Lsm1 and Lsm4) or subcomplexes (Lsm2-3 and Lsm5-6-7) were

first incubated in 4M urea and 1M NaCl in a “semi-denaturaing” state for 2 h at 37 °C and

then mixed in equimolar amounts for 2-5 h, and the sample was dialyzed overnight against

124

buffer with decreasingly salt concentration. Finally, the reconstituted human Lsm1-7

complex was purified by gel filtration and anion exchange chromatography. Unlike their

time-consuming strategy, we took a fast but efficient approach to reconstitute Lsm1-7 and

Lsm1-7-Pat1-465c. Instead of using the semi-denaturing of samples in 4M urea, a thorough

denaturing of samples were conducted in 8M urea, followed by direct on-column refolding

and reconstitution of the Lsm1-7 and Lsm1-7-Pat1-465c simultaneously. This strategy has

been successfully used for the sample preparation and structural solution of apo-spuD and

apo-spuE in the first part of my thesis.

Reconstitution of Lsm1-7 seven-protein complex is shown in Figure II -17. A noticeable

peak shift occurred when compared the peak position (208.9 ml in Figure II -17) with the

peak positions of Lsm1 (229 ml from Figure II -9), Lsm4 (244.5 ml from Figure II -10),

Lsm2-3 (218.9 ml from Figure II -11), and Lsm5-6-7 (224.4 ml from Figure II -12).

Reconstitution of Lsm1-7-Pat1-465c eight-protein complex is shown in Figure II -18. The

peak position of Lsm1-7 shifted forward from 208.9 ml (Figure II -17) to 201 ml (Figure II

-18) after incorporation of Pat1-465c. The NuPAGE gel clearly showed eight bands

corresponding to the eight proteins (Figure II -18).

Together, these shifts observed during the process of reconstitution also confirm our Native-

PAGE analysis of the interaction networks among the eight proteins.

125

Figure II -16. Reconstitution of Lsm2-3-Pat1-465c. a) Elution profile of gel filtration

chromatography from S200 26/60. b) Peak fraction at 193.4 ml was checked by NuPAGE.

Figure II -17. Reconstitution of Lsm1-7 complex. a) Elution profile of gel filtration

chromatography from S200 26/60. b) Peak fraction at 208.9 ml was checked by NuPAGE.

126

Figure II -18. Reconstitution of Lsm1-7-Pat1-465c complex. a) Elution profile of gel

filtration chromatography from S200 26/60. b) Peak fraction at 201 ml was checked by

NuPAGE.

3.5 mRNA binding properties

Although it is known that Lsm1-7-Pat1 can bind with oligo (U) (Chowdhury et al. 2007), it is

not known how each of the eight proteins participates in this binding. To this end, SPR was

employed to test the binding of Lsm proteins to 5’ biotin-labelled 15-mer poly (U). As shown

in Figure II -19, Lsm1-7 seven-protein complex binds to poly (U) with a dissociation

constant (Kd) of 41.3 nM, while the binding affinity of Lsm1-7+Pat1-465c complex towards

poly (U) (Figure II -20) decreased to around half that of the binding affinity between Lsm1-

7 and poly (U) (Kd = 86 nM).

127

Figure II -19. Binding of Lsm1-7 complex and 5’ biotin-labelled poly (U).

128

Figure II -20. Binding of Lsm1-7-Pat1-465c complex and 5’ biotin-labelled poly (U).

129

Lsm2-3 showed high binding affinity to poly (U) with a Kd of 21.2 nM (Figure II -21).

Similarly, the trimeric complex of Lsm2-3-Pat1-465c displayed reduced binding affinity

towards poly (U) (Kd = 51.5 nM) (Figure II -22).

Figure II -21. Binding of Lsm2-3 complex and 5’ biotin-labelled poly (U).

130

Figure II -22. Binding of Lsm2-3-Pat1-465c complex and 5’ biotin-labelled poly (U).

It seems clear that Pat1-465c decreases the binding between Lsm1-7 or Lsm2-3 and poly

(U) to approximately half level. Sensorgram of Pat1-465c binding revealed a fast-association

and fast-dissociation pattern, which may indicate that the binding mainly occurs through

electrostatic interactions between the positively charged surface residues of Pat1-465c and

negatively charged phosphate groups in poly (U) (Figure II -23). In contrast, Lsm1, Lsm4

and Lsm5-6-7 failed to interact with poly (U) (Figure II -23).

131

Figure II -23. Binding of Lsm1, 4, 5-6-7 and Pat1-465c towards 5’ biotin-labelled poly

(U).

Together, it seems that the RNA binding property of Lsm1-7 can be largely attributed to

Lsm2 and Lsm3, while Pat1-465c has a negative regulatory effect on RNA binding. Two

conserved and heptameric lsm protein complexes have been identified: theLsm1-7 complex

which is located in the cytoplasm and involved in mRNA decay, and the Lsm2-8 complex

localized within the nucleus and involved in pre-mRNA splicing. Lsm2-8 is not involved in

cytoplasmic mRNA decay, but plays a role in the decapping and degradation of pre-mRNAs

and nucleus-restricted mRNAs (Kufel et al. 2004). As the two heptameric Lsm protein

132

complexes have close associations with RNAs, their common components, namely, Lsm2, 3,

4, 5, 6 and 7, may be the core components for RNA interactions, while the unique

components in each of the two complexes, Lsm1 in the Lsm1-7 complex and Lsm8 in the

Lsm2-8 complex, may determine the functional sites of the complexes, whether in the

cytoplasm or in the nucleus. From our SPR data, Lsm2 and Lsm3 might be the major factors

for RNA binding in both Lsm1-7 and Lsm2-8 complexes. Currently, more efforts have been

employed to determine the role of Lsm1 in the cytoplasmic mRNA decay (Tharun et al.

2005; Chowdhury and Tharun 2008; 2009), while few report has addressed the roles of these

common components in mRNA decay. All together, the preliminary results in this study

suggest that Lsm2 and Lsm3 might play a more important but yet unidentified role in the

process of mRNA decay apart from being merely scaffolding ring-structure proteins. In

support of this hypothesis, Pat1-465c interacts with Lsm1-7 complex mainly through the

binding of Lsm2 and Lsm3.

3.7 Difficulties in structure determination of Pat1-465c and Pat1-473c

The C-region of Pat1 bears low sequence identity to other proteins, thereby hampering efforts

to solve the structure using the molecular replacement method.

Crystals of the C-region of Pat1 were grown with seleno-methionine substitution. Seleno-

methionine-substituted Pat1-422c was purified, crystallized and data was collected. However,

the highest resolution achieved was 3.6 Å and the selenium heavy atom sites could not be

identified clearly (Data not shown). The diffraction quality of this crystal did not improve

although extensive optimizations including additive screening, seeding, dehydration and

annealing were performed. Based on secondary structure predictions, a shorter construct

(Pat1-465c) was made. Seleno-methionine-substituted Pat1-465c crystals appeared after a

133

few hours. Although, a resolution of 2.5 Å was reached, the Pat1-465c crystals showed

twinning and selenium heavy atom sites failed to be traced out. A further shorter construct

was made (Pat1-473c), deleting eight residues at the N-terminal of Pat1-465c, however this

construct similarly yielded twinning problems. Primary sequence information showed that

Pat1-473c contains a high percentage of lysine residues (27 Lys out of 323 residues or 8%).

We suspect that the positively charged Lys residues located on the surface of Pat1-473 might

be the determining factor for twinning formation through salt bridge interaction with other

negatively charged surface amino acids. To solve the structure of the C-terminal region of

Pat1, the future work is to (1) systematically mutate the Lys residues in this region of Pat1,

(2) make more constructs within this region, and (3) study its homologues from other species

such as Homo sapiens or Schizosaccharomyces pombe. The future work also goes to obtain

diffraction quality crystal of Lsm2-3-Pat1-465c.Our ultimate goal is to solve the structure of

Lsm1-7-Pat1-465c complex in the presence and absence of poly (U).

134

References

Achsel, T., Brahms, H., Kastner, B., Bachi, A., Wilm, M., and Luhrmann, R. 1999. A doughnut-shaped heteromer of human Sm-like proteins binds to the 3'-end of U6 snRNA, thereby facilitating U4/U6 duplex formation in vitro. EMBO J 18: 5789-5802.

Achsel, T., Stark, H., and Luhrmann, R. 2001. The Sm domain is an ancient RNA-binding motif with oligo(U) specificity. Proc Natl Acad Sci U S A 98: 3685-3689.

Aguilera, A. 2005. Cotranscriptional mRNP assembly: from the DNA to the nuclear pore. Curr Opin Cell Biol 17: 242-250.

Ahn, B.Y., and Moss, B. 1989. Capped poly(A) leaders of variable lengths at the 5' ends of vaccinia virus late mRNAs. J Virol 63: 226-232.

Alhonen-Hongisto, L., Seppanen, P., and Janne, J. 1980. Intracellular putrescine and spermidine deprivation induces increased uptake of the natural polyamines and methylglyoxal bis(guanylhydrazone). Biochem J 192: 941-945.

Amrani, N., Ganesan, R., Kervestin, S., Mangus, D.A., Ghosh, S., and Jacobson, A. 2004. A faux 3'-UTR promotes aberrant termination and triggers nonsense-mediated mRNA decay. Nature 432: 112-118.

Anderson, J.S., and Parker, R.P. 1998. The 3' to 5' degradation of yeast mRNAs is a general mechanism for mRNA turnover that requires the SKI2 DEVH box protein and 3' to 5' exonucleases of the exosome complex. EMBO J 17: 1497-1506.

Auvinen, M., Paasinen, A., Andersson, L.C., and Holtta, E. 1992. Ornithine decarboxylase activity is critical for cell transformation. Nature 360: 355-358.

Bachrach, U. 2005. Naturally occurring polyamines: interaction with macromolecules. Curr Protein Pept Sci 6: 559-566.

Barbieri, A.M., Sha, Q., Bette-Bobillo, P., Stahl, P.D., and Vidal, M. 2001. ADP-ribosylation of Rab5 by ExoS of Pseudomonas aeruginosa affects endocytosis. Infect Immun 69: 5329-5334.

Bergman, N., Moraes, K.C., Anderson, J.R., Zaric, B., Kambach, C., Schneider, R.J., Wilusz, C.J., and Wilusz, J. 2007. Lsm proteins bind and stabilize RNAs containing 5' poly(A) tracts. Nat Struct Mol Biol 14: 824-831.

Bethell, D.R., and Pegg, A.E. 1981. Polyamines are needed for the differentiation of 3T3-L1 fibroblasts into adipose cells. Biochem Biophys Res Commun 102: 272-278.

Bielinska, A., Kukowska-Latallo, J.F., Johnson, J., Tomalia, D.A., and Baker, J.R., Jr. 1996. Regulation of in vitro gene expression using antisense oligonucleotides or antisense expression plasmids transfected using starburst PAMAM dendrimers. Nucleic Acids Res 24: 2176-2182.

Bogle, R.G., Mann, G.E., Pearson, J.D., and Morgan, D.M. 1994. Endothelial polyamine uptake: selective stimulation by L-arginine deprivation or polyamine depletion. Am J Physiol 266: C776-783.

Bonten, M.J., Bergmans, D.C., Speijer, H., and Stobberingh, E.E. 1999. Characteristics of polyclonal endemicity of Pseudomonas aeruginosa colonization in intensive care units. Implications for infection control. Am J Respir Crit Care Med 160: 1212-1219.

135

Boucher, H.W., Talbot, G.H., Bradley, J.S., Edwards, J.E., Gilbert, D., Rice, L.B., Scheld, M., Spellberg, B., and Bartlett, J. 2009. Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America. Clin Infect Dis 48: 1-12.

Bouveret, E., Rigaut, G., Shevchenko, A., Wilm, M., and Seraphin, B. 2000. A Sm-like protein complex that participates in mRNA degradation. EMBO J 19: 1661-1671.

Branlant, C., Krol, A., Ebel, J.P., Lazar, E., Haendler, B., and Jacob, M. 1982. U2 RNA shares a structural domain with U1, U4, and U5 RNAs. EMBO J 1: 1259-1265.

Brune, B., Hartzell, P., Nicotera, P., and Orrenius, S. 1991. Spermine prevents endonuclease activation and apoptosis in thymocytes. Exp Cell Res 195: 323-329.

Brunger, A.T. 1992. Free R value: a novel statistical quantity for assessing the accuracy of crystal structures. Nature 355: 472-475.

Brunger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P., Grosse-Kunstleve, R.W., Jiang, J.S., Kuszewski, J., Nilges, M., Pannu, N.S., et al. 1998. Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D Biol Crystallogr 54: 905-921.

Buhler, M., Steiner, S., Mohn, F., Paillusson, A., and Muhlemann, O. 2006. EJC-independent degradation of nonsense immunoglobulin-mu mRNA depends on 3' UTR length. Nat Struct Mol Biol 13: 462-464.

Burns, M.R., Graminski, G.F., Weeks, R.S., Chen, Y., and O'Brien, T.G. 2009. Lipophilic lysine-spermine conjugates are potent polyamine transport inhibitors for use in combination with a polyamine biosynthesis inhibitor. J Med Chem 52: 1983-1993.

Caponigro, G., and Parker, R. 1995. Multiple functions for the poly(A)-binding protein in mRNA decapping and deadenylation in yeast. Genes Dev 9: 2421-2432.

Carmody, S.R., and Wente, S.R. 2009. mRNA nuclear export at a glance. J Cell Sci 122: 1933-1937.

CCP4. 1994. The CCP4 suite: programs for protein crystallography. Acta Crystallogr D Biol Crystallogr 50: 760-763.

Childs, A.C., Mehta, D.J., and Gerner, E.W. 2003. Polyamine-dependent gene expression. Cell Mol Life Sci 60: 1394-1406.

Chowdhury, A., Mukhopadhyay, J., and Tharun, S. 2007. The decapping activator Lsm1p-7p-Pat1p complex has the intrinsic ability to distinguish between oligoadenylated and polyadenylated RNAs. RNA 13: 998-1016.

Chowdhury, A., and Tharun, S. 2008. lsm1 mutations impairing the ability of the Lsm1p-7p-Pat1p complex to preferentially bind to oligoadenylated RNA affect mRNA decay in vivo. RNA 14: 2149-2158.

Chowdhury, A., and Tharun, S. 2009. Activation of decapping involves binding of the mRNA and facilitation of the post-binding steps by the Lsm1-7-Pat1 complex. RNA 15: 1837-1848.

Clement, S.L., and Lykke-Andersen, J. 2006. No mercy for messages that mess with the ribosome. Nat Struct Mol Biol 13: 299-301.

Coller, J., and Parker, R. 2004. Eukaryotic mRNA decapping. Annu Rev Biochem 73: 861-890.

136

Coller, J., and Parker, R. 2005. General translational repression by activators of mRNA decapping. Cell 122: 875-886.

Cowell, B.A., Chen, D.Y., Frank, D.W., Vallis, A.J., and Fleiszig, S.M. 2000. ExoT of cytotoxic Pseudomonas aeruginosa prevents uptake by corneal epithelial cells. Infect Immun 68: 403-406.

D'Agostino, L., di Pietro, M., and Di Luccia, A. 2005. Nuclear aggregates of polyamines are supramolecular structures that play a crucial role in genomic DNA protection and conformation. FEBS J 272: 3777-3787.

Das, B., Butler, J.S., and Sherman, F. 2003. Degradation of normal mRNA in the nucleus of Saccharomyces cerevisiae. Mol Cell Biol 23: 5502-5515.

Decker, C.J., and Parker, R. 1993. A turnover pathway for both stable and unstable mRNAs in yeast: evidence for a requirement for deadenylation. Genes Dev 7: 1632-1643.

Deng, Q., Sun, J., and Barbieri, J.T. 2005. Uncoupling Crk signal transduction by Pseudomonas exoenzyme T. J Biol Chem 280: 35953-35960.

Desiderio, M.A., Grassilli, E., Bellesia, E., Salomoni, P., and Franceschi, C. 1995. Involvement of ornithine decarboxylase and polyamines in glucocorticoid-induced apoptosis of rat thymocytes. Cell Growth Differ 6: 505-513.

Di Luccia, A., Picariello, G., Iacomino, G., Formisano, A., Paduano, L., and D'Agostino, L. 2009. The in vitro nuclear aggregates of polyamines. FEBS J 276: 2324-2335.

Doma, M.K., and Parker, R. 2006. Endonucleolytic cleavage of eukaryotic mRNAs with stalls in translation elongation. Nature 440: 561-564.

Driscoll, J.A., Brody, S.L., and Kollef, M.H. 2007. The epidemiology, pathogenesis and treatment of Pseudomonas aeruginosa infections. Drugs 67: 351-368.

Emsley, P., and Cowtan, K. 2004. Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr 60: 2126-2132.

Engel, J.N. 2003. Molecular pathogenesis of acute Pseudomonas aeruginosa infections. . Kluwer Academic/Plenum Press, pp. 201-230.

Enninga, J., and Rosenshine, I. 2009. Imaging the assembly, structure and activity of type III secretion systems. Cell Microbiol 11: 1462-1470.

Erwin, B.G., Ewton, D.Z., Florini, J.R., and Pegg, A.E. 1983. Polyamine depletion inhibits the differentiation of L6 myoblast cells. Biochem Biophys Res Commun 114: 944-949.

Eulalio, A., Behm-Ansmant, I., and Izaurralde, E. 2007. P bodies: at the crossroads of post-transcriptional pathways. Nat Rev Mol Cell Biol 8: 9-22.

Finck-Barbancon, V., Goranson, J., Zhu, L., Sawa, T., Wiener-Kronish, J.P., Fleiszig, S.M., Wu, C., Mende-Mueller, L., and Frank, D.W. 1997. ExoU expression by Pseudomonas aeruginosa correlates with acute cytotoxicity and epithelial injury. Mol Microbiol 25: 547-557.

Finck-Barbancon, V., Yahr, T.L., and Frank, D.W. 1998. Identification and characterization of SpcU, a chaperone required for efficient secretion of the ExoU cytotoxin. J Bacteriol 180: 6224-6231.

Frank, D.W. 1997. The exoenzyme S regulon of Pseudomonas aeruginosa. Mol Microbiol 26: 621-629.

137

Fraylick, J.E., La Rocque, J.R., Vincent, T.S., and Olson, J.C. 2001. Independent and coordinate effects of ADP-ribosyltransferase and GTPase-activating activities of exoenzyme S on HT-29 epithelial cell function. Infect Immun 69: 5318-5328.

Fromont-Racine, M., Rain, J.C., and Legrain, P. 1997. Toward a functional analysis of the yeast genome through exhaustive two-hybrid screens. Nat Genet 16: 277-282.

Furuchi, T., Kashiwagi, K., Kobayashi, H., and Igarashi, K. 1991. Characteristics of the gene for a spermidine and putrescine transport system that maps at 15 min on the Escherichia coli chromosome. J Biol Chem 266: 20928-20933.

Gales, A.C., Jones, R.N., Turnidge, J., Rennie, R., and Ramphal, R. 2001. Characterization of Pseudomonas aeruginosa isolates: occurrence rates, antimicrobial susceptibility patterns, and molecular typing in the global SENTRY Antimicrobial Surveillance Program, 1997-1999. Clin Infect Dis 32 Suppl 2: S146-155.

Garrity-Ryan, L., Kazmierczak, B., Kowal, R., Comolli, J., Hauser, A., and Engel, J.N. 2000. The arginine finger domain of ExoT contributes to actin cytoskeleton disruption and inhibition of internalization of Pseudomonas aeruginosa by epithelial cells and macrophages. Infect Immun 68: 7100-7113.

Gatfield, D., and Izaurralde, E. 2004. Nonsense-mediated messenger RNA decay is initiated by endonucleolytic cleavage in Drosophila. Nature 429: 575-578.

Gilmour, S.K. 2007. Polyamines and nonmelanoma skin cancer. Toxicol Appl Pharmacol 224: 249-256.

Giorgi, C., and Moore, M.J. 2007. The nuclear nurture and cytoplasmic nature of localized mRNPs. Semin Cell Dev Biol 18: 186-193.

Gouet, P., Courcelle, E., Stuart, D.I., and Metoz, F. 1999. ESPript: analysis of multiple sequence alignments in PostScript. Bioinformatics 15: 305-308.

Ha, H.C., Woster, P.M., Yager, J.D., and Casero, R.A., Jr. 1997. The role of polyamine catabolism in polyamine analogue-induced programmed cell death. Proc Natl Acad Sci U S A 94: 11557-11562.

Hauser, A.R. 2009. The type III secretion system of Pseudomonas aeruginosa: infection by injection. Nat Rev Microbiol 7: 654-665.

Hayward, S., and Berendsen, H.J. 1998. Systematic analysis of domain motions in proteins from conformational change: new results on citrate synthase and T4 lysozyme. Proteins 30: 144-154.

He, F., and Jacobson, A. 1995. Identification of a novel component of the nonsense-mediated mRNA decay pathway by use of an interacting protein screen. Genes Dev 9: 437-454.

Heby, O., and Emanuelsson, H. 1981. Role of the polyamines in germ cell differentiation and in early embryonic development. Med Biol 59: 417-422.

Heby, O., Sarna, G.P., Marton, L.J., Omine, M., Perry, S., and Russell, D.H. 1973. Polyamine content of AKR leukemic cells in relation to the cell cycle. Cancer Res 33: 2959-2964.

Hermann, H., Fabrizio, P., Raker, V.A., Foulaki, K., Hornig, H., Brahms, H., and Luhrmann, R. 1995. snRNP Sm proteins share two evolutionarily conserved sequence motifs which are involved in Sm protein-protein interactions. EMBO J 14: 2076-2088.

138

Higashi, K., Ishigure, H., Demizu, R., Uemura, T., Nishino, K., Yamaguchi, A., Kashiwagi, K., and Igarashi, K. 2008. Identification of a spermidine excretion protein complex (MdtJI) in Escherichia coli. J Bacteriol 190: 872-878.

Higgins, C.F. 1992. ABC transporters: from microorganisms to man. Annu Rev Cell Biol 8: 67-113.

Holbrook, J.A., Neu-Yilik, G., Hentze, M.W., and Kulozik, A.E. 2004. Nonsense-mediated decay approaches the clinic. Nat Genet 36: 801-808.

Holm, L., Kaariainen, S., Rosenstrom, P., and Schenkel, A. 2008. Searching protein structure databases with DaliLite v.3. Bioinformatics 24: 2780-2781.

Igarashi, K., and Kashiwagi, K. 1999. Polyamine transport in bacteria and yeast. Biochem J 344 Pt 3: 633-642.

Ink, B.S., and Pickup, D.J. 1990. Vaccinia virus directs the synthesis of early mRNAs containing 5' poly(A) sequences. Proc Natl Acad Sci U S A 87: 1536-1540.

Isken, O., and Maquat, L.E. 2007. Quality control of eukaryotic mRNA: safeguarding cells from abnormal mRNA function. Genes Dev 21: 1833-1856.

Jacobson, A., and Peltz, S.W. 1996. Interrelationships of the pathways of mRNA decay and translation in eukaryotic cells. Annu Rev Biochem 65: 693-739.

Jakymiw, A., Lian, S., Eystathioy, T., Li, S., Satoh, M., Hamel, J.C., Fritzler, M.J., and Chan, E.K. 2005. Disruption of GW bodies impairs mammalian RNA interference. Nat Cell Biol 7: 1267-1274.

Kabra, P.M., Lee, H.K., Lubich, W.P., and Marton, L.J. 1986. Solid-phase extraction and determination of dansyl derivatives of unconjugated and acetylated polyamines by reversed-phase liquid chromatography: improved separation systems for polyamines in cerebrospinal fluid, urine and tissue. J Chromatogr 380: 19-32.

Kabsch, W., and Sander, C. 1983. Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features. Biopolymers 22: 2577-2637.

Kambach, C., Walke, S., Young, R., Avis, J.M., de la Fortelle, E., Raker, V.A., Luhrmann, R., Li, J., and Nagai, K. 1999. Crystal structures of two Sm protein complexes and their implications for the assembly of the spliceosomal snRNPs. Cell 96: 375-387.

Kashiwagi, K., Hosokawa, N., Furuchi, T., Kobayashi, H., Sasakawa, C., Yoshikawa, M., and Igarashi, K. 1990. Isolation of polyamine transport-deficient mutants of Escherichia coli and cloning of the genes for polyamine transport proteins. J Biol Chem 265: 20893-20897.

Kashiwagi, K., Kuraishi, A., Tomitori, H., Igarashi, A., Nishimura, K., Shirahata, A., and Igarashi, K. 2000. Identification of the putrescine recognition site on polyamine transport protein PotE. J Biol Chem 275: 36007-36012.

Kashiwagi, K., Miyamoto, S., Nukui, E., Kobayashi, H., and Igarashi, K. 1993. Functions of potA and potD proteins in spermidine-preferential uptake system in Escherichia coli. J Biol Chem 268: 19358-19363.

Kashiwagi, K., Pistocchi, R., Shibuya, S., Sugiyama, S., Morikawa, K., and Igarashi, K. 1996. Spermidine-preferential uptake system in Escherichia coli. Identification of amino acids involved in polyamine binding in PotD protein. J Biol Chem 271: 12205-12208.

139

Khan, A.U., Di Mascio, P., Medeiros, M.H., and Wilson, T. 1992. Spermine and spermidine protection of plasmid DNA against single-strand breaks induced by singlet oxygen. Proc Natl Acad Sci U S A 89: 11428-11430.

Khanna, R., and Kiledjian, M. 2004. Poly(A)-binding-protein-mediated regulation of hDcp2 decapping in vitro. EMBO J 23: 1968-1976.

Kim, S.H., Quigley, G., Suddath, F.L., and Rich, A. 1971. High-resolution x-ray diffraction patterns of crystalline transfer RNA that show helical regions. Proc Natl Acad Sci U S A 68: 841-845.

Kramer, D.L., Vujcic, S., Diegelman, P., Alderfer, J., Miller, J.T., Black, J.D., Bergeron, R.J., and Porter, C.W. 1999. Polyamine analogue induction of the p53-p21WAF1/CIP1-Rb pathway and G1 arrest in human melanoma cells. Cancer Res 59: 1278-1286.

Kufel, J., Bousquet-Antonelli, C., Beggs, J.D., and Tollervey, D. 2004. Nuclear pre-mRNA decapping and 5' degradation in yeast require the Lsm2-8p complex. Mol Cell Biol 24: 9646-9657.

Kusunoki, S., and Yasumasu, I. 1978. Inhibitory effect of alpha-hydrazinoornithine on egg cleavage in sea urchin eggs. Dev Biol 67: 336-345.

Kwon, D.H., and Lu, C.D. 2006. Polyamines induce resistance to cationic peptide, aminoglycoside, and quinolone antibiotics in Pseudomonas aeruginosa PAO1. Antimicrob Agents Chemother 50: 1615-1622.

Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685.

Laskowski, R.A., MacArthur, M.W., Moss, D.S., and Thornton, J.M. 1993. PROCHECK: a program to check the stereochemical quality of protein structures. Journal of Applied Crystallography 26: 283-291.

Lejeune, F., Li, X., and Maquat, L.E. 2003. Nonsense-mediated mRNA decay in mammalian cells involves decapping, deadenylating, and exonucleolytic activities. Mol Cell 12: 675-687.

Lejeune, F., and Maquat, L.E. 2005. Mechanistic links between nonsense-mediated mRNA decay and pre-mRNA splicing in mammalian cells. Curr Opin Cell Biol 17: 309-315.

Leveque, J., Foucher, F., Bansard, J.Y., Havouis, R., Grall, J.Y., and Moulinoux, J.P. 2000. Polyamine profiles in tumor, normal tissue of the homologous breast, blood, and urine of breast cancer sufferers. Breast Cancer Res Treat 60: 99-105.

Ling, S.H., Decker, C.J., Walsh, M.A., She, M., Parker, R., and Song, H. 2008. Crystal structure of human Edc3 and its functional implications. Mol Cell Biol 28: 5965-5976.

Liu, H., Rodgers, N.D., Jiao, X., and Kiledjian, M. 2002. The scavenger mRNA decapping enzyme DcpS is a member of the HIT family of pyrophosphatases. EMBO J 21: 4699-4708.

Liu, J., Valencia-Sanchez, M.A., Hannon, G.J., and Parker, R. 2005. MicroRNA-dependent localization of targeted mRNAs to mammalian P-bodies. Nat Cell Biol 7: 719-723.

Lu, C.D., Itoh, Y., Nakada, Y., and Jiang, Y. 2002. Functional analysis and regulation of the divergent spuABCDEFGH-spuI operons for polyamine uptake and utilization in Pseudomonas aeruginosa PAO1. J Bacteriol 184: 3765-3773.

140

Lykke-Andersen, J. 2002. Identification of a human decapping complex associated with hUpf proteins in nonsense-mediated decay. Mol Cell Biol 22: 8114-8121.

Machius, M., Brautigam, C.A., Tomchick, D.R., Ward, P., Otwinowski, Z., Blevins, J.S., Deka, R.K., and Norgard, M.V. 2007. Structural and biochemical basis for polyamine binding to the Tp0655 lipoprotein of Treponema pallidum: putative role for Tp0655 (TpPotD) as a polyamine receptor. J Mol Biol 373: 681-694.

MacRae, M., Kramer, D.L., and Coffino, P. 1998. Developmental effect of polyamine depletion in Caenorhabditis elegans. Biochem J 333 ( Pt 2): 309-315.

Mandel, C.R., Bai, Y., and Tong, L. 2008. Protein factors in pre-mRNA 3'-end processing. Cell Mol Life Sci 65: 1099-1122.

Maquat, L.E. 2004. Nonsense-mediated mRNA decay: splicing, translation and mRNP dynamics. Nat Rev Mol Cell Biol 5: 89-99.

Matsufuji, S., Matsufuji, T., Miyazaki, Y., Murakami, Y., Atkins, J.F., Gesteland, R.F., and Hayashi, S. 1995. Autoregulatory frameshifting in decoding mammalian ornithine decarboxylase antizyme. Cell 80: 51-60.

Meyer, S., Temme, C., and Wahle, E. 2004. Messenger RNA turnover in eukaryotes: pathways and enzymes. Crit Rev Biochem Mol Biol 39: 197-216.

Milovic, V., and Turchanowa, L. 2003. Polyamines and colon cancer. Biochem Soc Trans 31: 381-383.

Mitchell, P., and Tollervey, D. 2000. mRNA stability in eukaryotes. Curr Opin Genet Dev 10: 193-198.

Moteki, S., and Price, D. 2002. Functional coupling of capping and transcription of mRNA. Mol Cell 10: 599-609.

Murshudov, G.N., Vagin, A.A., and Dodson, E.J. 1997. Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr 53: 240-255.

Naidoo, N., Harrop, S.J., Sobti, M., Haynes, P.A., Szymczyna, B.R., Williamson, J.R., Curmi, P.M., and Mabbutt, B.C. 2008. Crystal structure of Lsm3 octamer from Saccharomyces cerevisiae: implications for Lsm ring organisation and recruitment. J Mol Biol 377: 1357-1371.

Nakada, Y., and Itoh, Y. 2003. Identification of the putrescine biosynthetic genes in Pseudomonas aeruginosa and characterization of agmatine deiminase and N-carbamoylputrescine amidohydrolase of the arginine decarboxylase pathway. Microbiology 149: 707-714.

O'Toole, B.A., Huffman, K.W., and Gibson, J.P. 1989. Effects of eflornithine hydrochloride (DFMO) on fetal development in rats and rabbits. Teratology 39: 103-113.

Ottmann, C., Yasmin, L., Weyand, M., Veesenmeyer, J.L., Diaz, M.H., Palmer, R.H., Francis, M.S., Hauser, A.R., Wittinghofer, A., and Hallberg, B. 2007. Phosphorylation-independent interaction between 14-3-3 and exoenzyme S: from structure to pathogenesis. EMBO J 26: 902-913.

Ouameur, A.A., and Tajmir-Riahi, H.A. 2004. Structural analysis of DNA interactions with biogenic polyamines and cobalt(III)hexamine studied by Fourier transform infrared and capillary electrophoresis. J Biol Chem 279: 42041-42054.

Packham, G., and Cleveland, J.L. 1994. Ornithine decarboxylase is a mediator of c-Myc-induced apoptosis. Mol Cell Biol 14: 5741-5747.

141

Parchment, R.E., and Pierce, G.B. 1989. Polyamine oxidation, programmed cell death, and regulation of melanoma in the murine embryonic limb. Cancer Res 49: 6680-6686.

Parker, R., and Sheth, U. 2007. P bodies and the control of mRNA translation and degradation. Mol Cell 25: 635-646.

Parker, R., and Song, H. 2004. The enzymes and control of eukaryotic mRNA turnover. Nat Struct Mol Biol 11: 121-127.

Patel, D.D., and Pickup, D.J. 1987. Messenger RNAs of a strongly-expressed late gene of cowpox virus contain 5'-terminal poly(A) sequences. EMBO J 6: 3787-3794.

Pederson, K.J., and Barbieri, J.T. 1998. Intracellular expression of the ADP-ribosyltransferase domain of Pseudomonas exoenzyme S is cytotoxic to eukaryotic cells. Mol Microbiol 30: 751-759.

Pendeville, H., Carpino, N., Marine, J.C., Takahashi, Y., Muller, M., Martial, J.A., and Cleveland, J.L. 2001. The ornithine decarboxylase gene is essential for cell survival during early murine development. Mol Cell Biol 21: 6549-6558.

Phillips, R.M., Six, D.A., Dennis, E.A., and Ghosh, P. 2003. In vivo phospholipase activity of the Pseudomonas aeruginosa cytotoxin ExoU and protection of mammalian cells with phospholipase A2 inhibitors. J Biol Chem 278: 41326-41332.

Pilkington, G.R., and Parker, R. 2008. Pat1 contains distinct functional domains that promote P-body assembly and activation of decapping. Mol Cell Biol 28: 1298-1312.

Pingoud, A., Urbanke, C., Alves, J., Ehbrecht, H.J., Zabeau, M., and Gualerzi, C. 1984. Effect of polyamines and basic proteins on cleavage of DNA by restriction endonucleases. Biochemistry 23: 5697-5703.

Pollard, V.W., and Malim, M.H. 1998. The HIV-1 Rev protein. Annu Rev Microbiol 52: 491-532.

Poole, K. 2004. Efflux-mediated multiresistance in Gram-negative bacteria. Clin Microbiol Infect 10: 12-26.

Quigley, G.J., Teeter, M.M., and Rich, A. 1978. Structural analysis of spermine and magnesium ion binding to yeast phenylalanine transfer RNA. Proc Natl Acad Sci U S A 75: 64-68.

Rabin, S.D., Veesenmeyer, J.L., Bieging, K.T., and Hauser, A.R. 2006. A C-terminal domain targets the Pseudomonas aeruginosa cytotoxin ExoU to the plasma membrane of host cells. Infect Immun 74: 2552-2561.

Reddy, P.R., and Rukmini, V. 1981. Alpha-difluoromethylornithine as a postcoitally effective antifertility agent in female rats. Contraception 24: 215-221.

Rich, A., and Zhang, S. 2003. Timeline: Z-DNA: the long road to biological function. Nat Rev Genet 4: 566-572.

Rocha, C.L., Coburn, J., Rucks, E.A., and Olson, J.C. 2003. Characterization of Pseudomonas aeruginosa exoenzyme S as a bifunctional enzyme in J774A.1 macrophages. Infect Immun 71: 5296-5305.

Ruhl, M., Himmelspach, M., Bahr, G.M., Hammerschmid, F., Jaksche, H., Wolff, B., Aschauer, H., Farrington, G.K., Probst, H., Bevec, D., et al. 1993. Eukaryotic initiation factor 5A is a cellular target of the human immunodeficiency virus

142

type 1 Rev activation domain mediating trans-activation. J Cell Biol 123: 1309-1320.

Sadikot, R.T., Blackwell, T.S., Christman, J.W., and Prince, A.S. 2005. Pathogen-host interactions in Pseudomonas aeruginosa pneumonia. Am J Respir Crit Care Med 171: 1209-1223.

Salgado-Garrido, J., Bragado-Nilsson, E., Kandels-Lewis, S., and Seraphin, B. 1999. Sm and Sm-like proteins assemble in two related complexes of deep evolutionary origin. EMBO J 18: 3451-3462.

Sambrook, J., Fritsch, E.F., Maniatis, T. 1987. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Sato, H., Frank, D.W., Hillard, C.J., Feix, J.B., Pankhaniya, R.R., Moriyama, K., Finck-Barbancon, V., Buchaklian, A., Lei, M., Long, R.M., et al. 2003. The mechanism of action of the Pseudomonas aeruginosa-encoded type III cytotoxin, ExoU. EMBO J 22: 2959-2969.

Sayner, S.L., Frank, D.W., King, J., Chen, H., VandeWaa, J., and Stevens, T. 2004. Paradoxical cAMP-induced lung endothelial hyperpermeability revealed by Pseudomonas aeruginosa ExoY. Circ Res 95: 196-203.

Scheller, N., Resa-Infante, P., de la Luna, S., Galao, R.P., Albrecht, M., Kaestner, L., Lipp, P., Lengauer, T., Meyerhans, A., and Diez, J. 2007. Identification of PatL1, a human homolog to yeast P body component Pat1. Biochim Biophys Acta 1773: 1786-1792.

Schipper, R.G., Romijn, J.C., Cuijpers, V.M., and Verhofstad, A.A. 2003. Polyamines and prostatic cancer. Biochem Soc Trans 31: 375-380.

Schumacher, M.A., Pearson, R.F., Moller, T., Valentin-Hansen, P., and Brennan, R.G. 2002. Structures of the pleiotropic translational regulator Hfq and an Hfq-RNA complex: a bacterial Sm-like protein. EMBO J 21: 3546-3556.

Schwartz, D., Decker, C.J., and Parker, R. 2003. The enhancer of decapping proteins, Edc1p and Edc2p, bind RNA and stimulate the activity of the decapping enzyme. RNA 9: 239-251.

Schwartz, D.C., and Parker, R. 1999. Mutations in translation initiation factors lead to increased rates of deadenylation and decapping of mRNAs in Saccharomyces cerevisiae. Mol Cell Biol 19: 5247-5256.

Schwartz, D.C., and Parker, R. 2000. mRNA decapping in yeast requires dissociation of the cap binding protein, eukaryotic translation initiation factor 4E. Mol Cell Biol 20: 7933-7942.

Schweizer, H.P. 2003. Efflux as a mechanism of resistance to antimicrobials in Pseudomonas aeruginosa and related bacteria: unanswered questions. Genet Mol Res 2: 48-62.

Schwer, B., Visca, P., Vos, J.C., and Stunnenberg, H.G. 1987. Discontinuous transcription or RNA processing of vaccinia virus late messengers results in a 5' poly(A) leader. Cell 50: 163-169.

Seiler, N. 2004. Catabolism of polyamines. Amino Acids 26: 217-233. Seraphin, B. 1995. Sm and Sm-like proteins belong to a large family: identification of

proteins of the U6 as well as the U1, U2, U4 and U5 snRNPs. EMBO J 14: 2089-2098.

143

Shafikhani, S.H., and Engel, J. 2006. Pseudomonas aeruginosa type III-secreted toxin ExoT inhibits host-cell division by targeting cytokinesis at multiple steps. Proc Natl Acad Sci U S A 103: 15605-15610.

Sheth, U., and Parker, R. 2003. Decapping and decay of messenger RNA occur in cytoplasmic processing bodies. Science 300: 805-808.

Sheth, U., and Parker, R. 2006. Targeting of aberrant mRNAs to cytoplasmic processing bodies. Cell 125: 1095-1109.

Silva, A.L., Ribeiro, P., Inacio, A., Liebhaber, S.A., and Romao, L. 2008. Proximity of the poly(A)-binding protein to a premature termination codon inhibits mammalian nonsense-mediated mRNA decay. RNA 14: 563-576.

Sitkiewicz, I., Stockbauer, K.E., and Musser, J.M. 2007. Secreted bacterial phospholipase A2 enzymes: better living through phospholipolysis. Trends Microbiol 15: 63-69.

Sonenberg, N., and Hinnebusch, A.G. 2009. Regulation of translation initiation in eukaryotes: mechanisms and biological targets. Cell 136: 731-745.

Spotheim-Maurizot, M., Ruiz, S., Sabattier, R., and Charlier, M. 1995. Radioprotection of DNA by polyamines. Int J Radiat Biol 68: 571-577.

Steglich, C., and Scheffler, I.E. 1982. An ornithine decarboxylase-deficient mutant of Chinese hamster ovary cells. J Biol Chem 257: 4603-4609.

Steitz, T.A. 2008. A structural understanding of the dynamic ribosome machine. Nat Rev Mol Cell Biol 9: 242-253.

Sugiyama, S., Matsuo, Y., Maenaka, K., Vassylyev, D.G., Matsushima, M., Kashiwagi, K., Igarashi, K., and Morikawa, K. 1996a. The 1.8-A X-ray structure of the Escherichia coli PotD protein complexed with spermidine and the mechanism of polyamine binding. Protein Sci 5: 1984-1990.

Sugiyama, S., Vassylyev, D.G., Matsushima, M., Kashiwagi, K., Igarashi, K., and Morikawa, K. 1996b. Crystal structure of PotD, the primary receptor of the polyamine transport system in Escherichia coli. J Biol Chem 271: 9519-9525.

Sun, J., and Barbieri, J.T. 2003. Pseudomonas aeruginosa ExoT ADP-ribosylates CT10 regulator of kinase (Crk) proteins. J Biol Chem 278: 32794-32800.

Tabor, C.W., Tabor, H., and Xie, Q.W. 1986. Spermidine synthase of Escherichia coli: localization of the speE gene. Proc Natl Acad Sci U S A 83: 6040-6044.

Tabor, H. 1961. The stabilization of Bacillus subtilis transforming principle by spermine. Biochem Biophys Res Commun 4: 228-231.

Talbot, G.H., Bradley, J., Edwards, J.E., Jr., Gilbert, D., Scheld, M., and Bartlett, J.G. 2006. Bad bugs need drugs: an update on the development pipeline from the Antimicrobial Availability Task Force of the Infectious Diseases Society of America. Clin Infect Dis 42: 657-668.

Tam, R., and Saier, M.H., Jr. 1993. Structural, functional, and evolutionary relationships among extracellular solute-binding receptors of bacteria. Microbiol Rev 57: 320-346.

Tange, T.O., Nott, A., and Moore, M.J. 2004. The ever-increasing complexities of the exon junction complex. Curr Opin Cell Biol 16: 279-284.

Tarun, S.Z., Jr., and Sachs, A.B. 1996. Association of the yeast poly(A) tail binding protein with translation initiation factor eIF-4G. EMBO J 15: 7168-7177.

144

Teixeira, D., Sheth, U., Valencia-Sanchez, M.A., Brengues, M., and Parker, R. 2005. Processing bodies require RNA for assembly and contain nontranslating mRNAs. RNA 11: 371-382.

Tharun, S., He, W., Mayes, A.E., Lennertz, P., Beggs, J.D., and Parker, R. 2000. Yeast Sm-like proteins function in mRNA decapping and decay. Nature 404: 515-518.

Tharun, S., Muhlrad, D., Chowdhury, A., and Parker, R. 2005. Mutations in the Saccharomyces cerevisiae LSM1 gene that affect mRNA decapping and 3' end protection. Genetics 170: 33-46.

Thomas, T., and Thomas, T.J. 1994. Regulation of cyclin B1 by estradiol and polyamines in MCF-7 breast cancer cells. Cancer Res 54: 1077-1084.

Thomas, T., and Thomas, T.J. 2001. Polyamines in cell growth and cell death: molecular mechanisms and therapeutic applications. Cell Mol Life Sci 58: 244-258.

Thompson, J.D., Higgins, D.G., and Gibson, T.J. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673-4680.

Toro, I., Thore, S., Mayer, C., Basquin, J., Seraphin, B., and Suck, D. 2001. RNA binding in an Sm core domain: X-ray structure and functional analysis of an archaeal Sm protein complex. EMBO J 20: 2293-2303.

van Hoof, A., Frischmeyer, P.A., Dietz, H.C., and Parker, R. 2002. Exosome-mediated recognition and degradation of mRNAs lacking a termination codon. Science 295: 2262-2264.

Vassylyev, D.G., Tomitori, H., Kashiwagi, K., Morikawa, K., and Igarashi, K. 1998. Crystal structure and mutational analysis of the Escherichia coli putrescine receptor. Structural basis for substrate specificity. J Biol Chem 273: 17604-17609.

Visca, P., Leoni, L., Wilson, M.J., and Lamont, I.L. 2002. Iron transport and regulation, cell signalling and genomics: lessons from Escherichia coli and Pseudomonas. Mol Microbiol 45: 1177-1190.

Wahl, M.C., Will, C.L., and Luhrmann, R. 2009. The spliceosome: design principles of a dynamic RNP machine. Cell 136: 701-718.

Wallace, H.M. 2003. Polyamines and their role in human disease--an introduction. Biochem Soc Trans 31: 354-355.

Walsh, T.R., Toleman, M.A., Poirel, L., and Nordmann, P. 2005. Metallo-beta-lactamases: the quiet before the storm? Clin Microbiol Rev 18: 306-325.

Wang, J.Y. 2007. Polyamines and mRNA stability in regulation of intestinal mucosal growth. Amino Acids 33: 241-252.

Watson, N., Dunyak, D.S., Rosey, E.L., Slonczewski, J.L., and Olson, E.R. 1992. Identification of elements involved in transcriptional regulation of the Escherichia coli cad operon by external pH. J Bacteriol 174: 530-540.

Welch, E.M., Barton, E.R., Zhuo, J., Tomizawa, Y., Friesen, W.J., Trifillis, P., Paushkin, S., Patel, M., Trotta, C.R., Hwang, S., et al. 2007. PTC124 targets genetic disorders caused by nonsense mutations. Nature 447: 87-91.

Wen, J., and Brogna, S. 2008. Nonsense-mediated mRNA decay. Biochem Soc Trans 36: 514-516.

145

Wilson, R.W., and Bloomfield, V.A. 1979. Counterion-induced condesation of deoxyribonucleic acid. a light-scattering study. Biochemistry 18: 2192-2196.

Yahr, T.L., Vallis, A.J., Hancock, M.K., Barbieri, J.T., and Frank, D.W. 1998. ExoY, an adenylate cyclase secreted by the Pseudomonas aeruginosa type III system. Proc Natl Acad Sci U S A 95: 13899-13904.

Yahr, T.L., and Wolfgang, M.C. 2006. Transcriptional regulation of the Pseudomonas aeruginosa type III secretion system. Mol Microbiol 62: 631-640.

Zaric, B., Chami, M., Remigy, H., Engel, A., Ballmer-Hofer, K., Winkler, F.K., and Kambach, C. 2005. Reconstitution of two recombinant LSm protein complexes reveals aspects of their architecture, assembly, and function. J Biol Chem 280: 16066-16075.

Zhang, Y., Deng, Q., and Barbieri, J.T. 2007. Intracellular localization of type III-delivered Pseudomonas ExoS with endosome vesicles. J Biol Chem 282: 13022-13032.

Zhou, L., Wang, J., and Zhang, L.H. 2007. Modulation of bacterial Type III secretion system by a spermidine transporter dependent signaling pathway. PLoS One 2: e1291.

Zuk, D., and Jacobson, A. 1998. A single amino acid substitution in yeast eIF-5A results in mRNA stabilization. EMBO J 17: 2914-2925.