SlyD, A Ni(II) Metallochaperone for [NiFe]-Hydrogenase Biosynthesis in Escherichia coli

by

Harini Kaluarachchi

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Department of Chemistry University of Toronto

© Copyright by Harini Kaluarachchi 2011

ii

SlyD, A Ni(II) Metallochaperone for [NiFe]-hydrogenase

biosynthesis in Escherichia coli

Harini Kaluarachchi

Doctor of Philosophy

Department of Chemistry

University of Toronto

2011

Abstract

SlyD is a protein involved in [NiFe]-hydrogenase enzyme maturation and, together with HypB

and HypA proteins, contributes to the nickel delivery step. To understand the molecular details

of this in vivo function, the nickel-binding activity of SlyD was investigated in vitro. SlyD is a

monomeric protein that can chelate up to 7 nickel ions with an affinity in the sub-nanomolar

range. By truncation and mutagenesis studies we show that the unique C-terminal metal-binding

domain of this protein is required for Ni(II) binding and that the protein coordinates this metal

non-cooperatively. This activity of SlyD supports the proposed in vivo role of SlyD in nickel

homeostasis.

In addition to nickel, SlyD can bind a variety of other types of transition metals. Therefore it

was feasible that the protein contributes to homeostasis of metals other than nickel. To test this

hypothesis, the metal selectivity of the protein was examined. The preference of SlyD for the

metals examined could be ordered as follows, Mn(II), Fe(II) < Co(II) < Ni(II) ~ Zn(II) << Cu(I)

indicating that the affinity of SlyD for the different metals follows the Irving-Williams series of

iii

metal-complex stabilities. Although the protein is unable to overcome the large thermodynamic

preference in vitro for Cu(I) and exclude Zn(II) chelation, in vivo studies suggest a Ni(II)-

specific function for the protein.

To understand the function of SlyD as a metallochaperone, its interaction with HypB was

investigated. This investigation revealed that SlyD plays a role in Ni(II) storage in E. coli and

can function as a Ni(II)-donor to HypB. This study also revealed that SlyD can modulate the

metal-binding as well as the GTPase activities of HypB. Based on the experimental data, a role

for the HypB-SlyD complex in [NiFe]-hydrogenase biosynthesis is presented.

iv

Acknowledgments

First and foremost I would like to express my gratitude to my supervisor Dr. Deborah Zamble,

without your encouragement and guidance this would not have been possible. Your knowledge,

expertise, enthusiasm and patience for teaching influenced me tremendously. I would like to

thank Dr. Martin Stillman and Duncan Sutherland for their help with establishing mass

spectrometry methods. I am also thankful to Dr. Ingrid Pickering and Dr. Graham George for

their invaluable help at the synchrotron and taking the time to teach me the basics of XAS. Dr.

Alex Young, you have been a great mentor and I thank you for teaching me most of the things I

know about mass spectrometry. I would also like to thank the members of my committee Dr.

Robert Morris, Dr. Voula Kanelis, Dr. Rebecca Jockusch and Dr. Mark Nitz for the assistance

they provided at all levels of the research project.

Thank you to present members of the Zamble lab and the past members that I crossed paths with

for their help, advice, camaraderie and entertainment. I enjoyed the years I spent in the lab.

I was fortunate enough to have worked with the following undergraduates that helped me in the

course of this work, Sandra Krecisz, Judith F. Siebel and Sonia Sugumar, you are all exceptional

and thank you for your help.

I wish to thank Kim Chan Chung for being an amazing friend in and out of the lab, these past

five years would not have been that memorable without you.

Thank you to my sister Supipi, for not only helping me with bench work but for her love and

support throughout my life. I would also like to thank Simon for his help and encouragement in

the last few years.

A special thank you to Andrew Beharry for always believing in me. I am truly grateful for your

unwavering support and friendship. Thank you for being there to share all the success and

failures in all aspects of my life.

Lastly, thank you to my parents for loving me, encouraging me and supporting me

unconditionally. You have been my inspiration and I dedicate this Thesis to you.

v

Table of Contents

Abstract………………………………………………………………………………….......… .....ii

Acknowledgements…………………………………………………………………………… ....iv

Table of contents…………………………………………………………………………..….… ….v

List of Tables………………………………………………………………………………….. ...vii

List of Figures…………………………………………………………………………….…… ..viii

Abbreviations.…………………………………………………………………………….…… …..xi

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

1.1 Biological relevance of Ni(II)…………………………..……………………........ ….1

1.2 The [NiFe]-hydrogenases……………………………………………………......... ….2

1.3 HypB………………………………………………………………………………. ….5

1.4 SlyD………………………………………………………………………….......... .....8

1.5 Purpose of study……………………………………………………………….….. ...11

Chapter 2. The Ni(II)-binding properties of the metallochaperone SlyD……………….. ...17

2.1 Introduction………………………………..…………………………………........ ...17

2.2 Materials and Methods………………………………………………………......... ...19

2.3 Results……………………………………………………………………………. ...26

2.4 Discussion and Conclusions…..…………………………………………….......... ...43

Chapter 3. The in vitro and in vivo metal specificity of the Ni(II) metallochaperone

SlyD………………………………………………………………………………. ...55

3.1 Introduction………………………………..…………………………………........ ...55

3.2 Materials and Methods………………………………………………………......... ...56

vi

3.3 Results……………………………………………………………………………. ...63

3.4 Discussion………………….…..…………………………………………….......... ...92

Chapter 4. Significance of the HypB-SlyD complex in [NiFe]-hydrogenase

biosynthesis……………………………………………………………………… .102

4.1 Introduction………………………………..…………………………………........ .102

4.2 Materials and Methods………………………………………………………......... .104

4.3 Results……………………………………………………………………………. .109

4.4 Discussion……………………..…………………………………………….......... .125

Chapter 5. Summary and Perspective………………………………………………..…….. .135

5.1 Ni(II)-binding properties of SlyD…………………………………………………. .135

5.2 Metal specificity of SlyD……………………………………………………......... .136

5.3 Significance of the HypB-SlyD complex in [NiFe]-hydrogenase biosynthesis....... .138

5.4 Significance…..…………………………..………………………………….......... .138

5.5 Future directions…………...…..…………………………………………….......... .140

vii

List of Tables

Table 2-1 PCR primers used for mutagenesis………….…………………….............................. ...20

Table 2-2 Masses of SlyD protein…………………...………………………………………… ...21

Table 2-3 Best Ni(II) EXAFS curve-fiiting results for SlyD……….…………...…………..….. ...24

Table 2-4 Stoichiometry of Ni(II) binding to SlyD…………………………………………….. ...28

Table 3-1 Summary of metal concentrations used for metal toxicity studies…………………... ...61

Table 3-2 Primers used for qRT-PCR…………………………………………………………... ...63

Table 3-3 Metal stoichiometry of WT SlyD……………………………………………………. ...82

Table 3-4 KD of SlyD-Metal complexes………………………………………………………... …...93

Table 4-1 Total hydrogenase activity…………………………………………………………... .110

Table 4-2 Cytoplasmic 63

Ni(II) levels….………………………………………………………… .11110

Table 4-3 GTP hydrolysis rates of HypB under saturating conditions............................................ 1.123

viii

List of Figures

Figure 1-1 Biosynthtic pathways of [NiFe] hydrogenase…………………….............................. .....3

Figure 1-2 HypB……………………………………...………………………………………… ….6

Figure 1-3 SlyD……………………………………………………...…………...…………..….. ...10

Figure 2-1 SlyD domain architecture and amino acid sequence……………………………..….. ...18

Figure 2-2 Electronic absorption spectra of wild-type SlyD and mutants……………………..... ...27

Figure 2-3 SlyD binds multiple nickel ions……………………………………………………... ...27

Figure 2-4 Ni(II)-induced 2° structural changes in SlyD………………………………………... ...29

Figure 2-5 Quaternary structure of SlyD………………………………………………………… .......30

Figure 2-6 Analysis of Ni(II) binding to SlyD via ESI-MS……………………………………... ...31

Figure 2-7 Stoichiometry of nickel binding to SlyD……………………………………..……….. .......33

Figure 2-8 SlyD metal sites can compete with EGTA……………………………………........... .....34

Figure 2-9 Ni K near-edge spectra……………………………………………………………… .....35

Figure 2-10 k3-weighted Ni(II) EXAFS data and the Fourier transform of the data……………...….. .....37

Figure 2-11 Comparison of WT SlyD and SlyD1-146 CD spectra…………………………………….. .....38

Figure 2-12 Competition for Ni(II) binding by WT and Triple mutant SlyD……………...………… .....41

Figure 2-13 Overlay of WT EXAFS data with SlyD mutants………………………………...……… .....42

Figure 2-14 Model for Ni(II) binding to SlyD…………………………………………………..…… .....45

Figure 3-1 ESI-MS titration of SlyD with Zn(II)…………………………………………………… ….65

Figure 3-2 Re-evaluating Ni(II) binding to SlyD by ESI-MS………………………………………. ….66

ix

Figure 3-3 Zn(II) stoichiometry & affinity of SlyD………………………………………………… ….67

Figure 3-4 Competition of SlyD with EGTA for Zn(II) binding…………………………………… ….68

Figure 3-5 Determining the Zn(II) dissociation constant (KD) of SlyD using PAR……………...…. ….69

Figure 3-6 Comparison of Ni(II)- vs. Zn(II)-induced secondary structure change in SlyD………… ….71

Figure 3-7 Zn(II) dependent oligomeric state of SlyD……………………………………………… ….71

Figure 3-8 Selectivity of SlyD for Ni(II) vs. Zn(II)………………………………………………… ….72

Figure 3-9 Visualizing Cu(I) binding to SlyD………………………………………………………. ….74

Figure 3-10 ESI-MS titration of SlyD with Cu(I) in the presence of 100 M DTT…………………. ….75

Figure 3-11 Stoichiometry of SlyD for Cu(I)………………………………………………………… ….76

Figure 3-12 Determining the Cu(I) KD for SlyD……………………………………………………... ….77

Figure 3-13 Co(II) binding to SlyD…………………………………………………………………... ….79

Figure 3-14 Resulting d-d transition upon Co(II) binding to SlyD……………………..……………. ….80

Figure 3-15 ESI-MS titration of WT SlyD with Co(II) in the presence of 100 M DTT…………… ….81

Figure 3-16 Affinity of Slyd to Co(II) via Fura-2 competition………………………………………. ….83

Figure 3-17 Zn(II) can replace Co(II) bound to SlyD………………………………………………... ….84

Figure 3-18 Selectivity of SlyD for Ni(II) vs. Fe(II)…………………………………………………. ….85

Figure 3-19 Sensitivity of E. coli to transition metals………………………………………………... ….87

Figure 3-20 Expression profiles of zinc, copper and nickel transporter in aerobic growth………….. ….89

Figure 3-21 Expression of Ni(II) transporters in response to Ni(II)…………………………………. ….89

Figure 3-22 Expression of Ni(II) transporters in the presence of excess Zn(II)……………………... ….90

x

Figure 3-23 Expression profiles of zinc and copper transporters in anaerobic growth………………. ….91

Figure 4-1 Architecture of E. coli SlyD and HypB…………………………………………………. ...103

Figure 4-2 Metal transfer detection via ESI-MS……………………………………………………. 113

Figure 4-3 Ni(II) transfer from SlyD to HypB detected via ESI-MS……………………………….. ...114

Figure 4-4 Analysis of holo HypB by ESI-MS……………………………………………………... ...114

Figure 4-5 SlyD modulates the high-affinity metal site of HypB………………………………… ...116

Figure 4-6 Monitoring metal release from HypB to EGTA in the presence of SlyD variants via

ESI-MS………………………………………………………………………………….. ...118

Figure 4-7 Metal release from HypB using different acceptors…………………………………….. ...119

Figure 4-8 GxP binding to HypB via ESI-MS……………………………………………………… ...120

Figure 4-9 SlyD together with GDP enhances the nickel release rate from HypB…………………. ...121

Figure 4-10 Metal release from HypB to EDTA monitored by electronic absorption

spectroscopy…………………………………………………………………………… ...122

Figure 4-11 SlyD removes Zn(II) from the G-domain of HypB……………………………………... ...123

Figure 4-12 SlyD1-146 cannot competitively remove Zn(II) bound to the G-domain of HypB…….. ...125

Figure 4-13 Proposed model for Ni(II) insertion to the hydrogenase precursor under limited and

abundant exogenous nickel availability.………………………………………….….. .131

xi

Abbreviations

Bca Bicinchoninate

Bcs Bathocuproine sulfonate

CD Circular dichroism

DMG Dimethylgyloxime

DTT Dithiothreitol

EDTA Ethylenediaminetetraacetic acid

EGTA Ethyleneglycoltetraacetic acid

ESI-MS Electrospray ionization mass spectrometry

EXAFS Extended x-ray absorption fine structure

FKBP FK-506 binding proteins

FT Fourier transform

G-domain The GTPase domain of HypB corresponding to residues 77-290

GDP Guanosine diphosphate

GMP-PNP Guanosine 5′-[β,γ-imido]triphosphate

GTP Guanosine triphosphate

HEPES 4-(2-hydroxylethyl)-1 piperazineethanesulfonic acid

HPLC High-performance liquid chromatography

HSID Hot source-induced desolvation

ICP-AES Inductively coupled plasma-atomic emission spectroscopy

IF domain Insertion in the flap domain

IPTG Isopropyl-β-D-1-galactopyranoside

xii

LMCT Ligand to metal charge transfer

MBD Metal-binding domain

NEM N-ethyl maleimide

PAR 4-(2-pyridylazo)resorcinol

PCR Polymerase chain reaction

PPIase Peptidyl prolyl isomerase

qRT-PCR Quantitative real-time polymerase chain reaction

TCEP Tris(2-carboxyethyl)phosphine

WT Wildtype

XAS X-ray absorption spectroscopy

1

Chapter 1 Introduction

1

1.1 Biological relevance of Ni(II)

Nickel is not typically considered to be a biologically important element, however it is used by

fungi, algae, archaebacteria, eubacteria and higher plants (1-4). These organisms exploit nickel

reactivity in the active sites of various enzymes that are important for reactions such as carbon

dioxide and carbon monoxide metabolism, hydrogen uptake and production, and hydrolysis of

urea (5, 6). Many microorganisms employ nickel-mediated processes to enable colonization in

environments that are often regarded as inhospitable. For example, the urease enzyme facilitates

Helicobacter pylori colonization in the human acidic gastric mucosa by producing ammonia to

help neutralize the immediate microenvironment of the microbe (7, 8). Another example is

carbon monoxide dehydrogenase, which oxidizes CO to CO2 and enables the host organism to

grow on this simple and toxic compound (9). Although nickel is absolutely required in these

enzyme systems as a central component of their catalytic mechanisms, it can only be found in

trace amounts in the environment (4, 10). Furthermore, when nickel is available to the organism,

the accumulation of unprotected metal ions must be minimized to prevent potential cellular

damage. Therefore efficient uptake systems, nickel-sensors and regulators, as well as specific

and high affinity nickel-trafficking proteins are necessary to safely meet the enzymatic demand

for nickel (5).

The active sites of nickel-containing enzymes often require a cooperative assembly system of

dedicated accessory proteins to biosynthesize non-protein components and to deliver the metal

cofactors (11). Those proteins that are responsible for shuttling the nickel are termed nickel

chaperones and they are thought to transfer the nickel to the enzyme precursors through protein-

protein interactions. Some of these nickel chaperones may also take on additional roles and

participate in the overall nickel homeostasis of the cell by serving as nickel storage proteins.

How these nickel chaperones bind the nickel ions and how they interact with other accessory

1Sections of this chapter were taken from the following review article: Kaluarachchi, H., Chan Chung, K.

C., Zamble D. B., (2010). Microbial nickel proteins, Nat. Prod. Rep., 27, 681-694. HK and KCC are co-

first authors of this review article.

2

proteins to transport and insert the nickel to the target enzymes are questions that are just

beginning to be answered.

1.2 The [NiFe]-hydrogenases

The [NiFe]-hydrogenase enzymes, expressed in a wide range of bacteria and archaea, catalyze

the formation of hydrogen gas from protons and electrons as well as the reverse reaction (12, 13).

This class of hydrogenase can abate excess reducing equivalents in the cell by using protons as

electron acceptors or they can oxidize hydrogen and couple the resulting electrons to various

electron carriers for energy generation. It is therefore not surprising that this enzyme is found in

a variety of organisms that are able to survive and proliferate in hydrogen-rich environments.

One notable example is in the gastrointestinal tract of humans where colonic fermentation results

in the production of hydrogen gas that is subsequently used as an energy source by [NiFe]-

hydrogenase-containing pathogens (14, 15).

[NiFe] hydrogenase is a heterodimeric enzyme composed of a small subunit, containing iron-

sulfur clusters that relay electrons between the active site and redox partners, as well as a large

subunit (LS) with the embedded binuclear nickel- and iron-containing catalytic center (11). The

binuclear metallocenter is one of the most interesting features of the enzyme with the iron atom

coordinated to one CO and two CN ligands and bridged via two cysteine residues to the nickel

ion that has two additional cysteine ligands (Figure 1-1) (16, 17). Various accessory proteins are

required to assemble the ligands and chaperone the metal cofactors of this intricate metallocenter

onto two CXXC motifs of the LS (5, 11, 18). These accessory factors include proteins that are

encoded by the hyp (hydrogenase pleiotropic) genes hypA-hypF (19) or homologs of the Hyp

proteins that are involved in the maturation of specific hydrogenase isoforms, as well as

additional auxiliary proteins (5).

The current model of [NiFe]-hydrogenase biosynthesis (Figure 1-1) proceeds in two steps

consisting of delivery of the iron complex with the coordinated inorganic ligands followed by

nickel insertion. This order of addition is supported by the purification of hydrogenase from

Ralstonia eutropha with an incomplete metallocenter containing iron but lacking the nickel (20,

21). The formation of the unusual CN ligands is catalyzed by two dedicated maturation proteins,

HypE and HypF (22). HypF is responsible for catalyzing activation of carbamoyl phosphate (23)

3

and through an interaction with HypE a carboxamide group from the substrate is transferred to

the C-terminal cysteine of HypE in a reaction driven by ATP hydrolysis (22, 24). Hydrolysis of

carboxamide and a second ATP subsequently produces a thiocyanate adduct on HypE which is

then transferred to a reduced HypC-HypD complex (25, 26). HypD contains a [4Fe-4S] cluster

and it has been postulated that this could be the source of the iron atom of the hydrogenase

metallocenter and/or it could provide reducing equivalents for the attachment of the ligands to

iron (26, 27). In addition, recent evidence from the R. eutropha H16 system suggests that the

iron center may be loaded with its diatomic ligands prior to transfer to the hydrogenase LS (28).

The insertion of the iron is dependent on a conserved N-terminal cysteine residue on HypC

which is critical for association with either HypD or the LS in separate steps of the maturation

pathway (29).

Figure 1-1. Biosynthetic pathways of [NiFe]-hydrogenase. Working model for the assembly of the

[NiFe] center on the large subunit (LS) of the [NiFe]-hydrogenase, based primarily on our understanding

of the E. coli factors. HycE is the LS of hydrogenase 3. HypC and HypA can be replaced by the

homologous HybG and HybF, respectively, and SlyD is not found in all organisms. HycI is the

hydrogenase 3 isoform-specific protease that senses Ni(II) insertion and cleaves the C-terminus of the LS.

Assembly and association with the small subunit are not shown. The structures shown are the LS and

[NiFe] active site of the ready-oxidized form of the [NiFe]-hydrogenase from Desulfovibrio

fructosovorans (pdb 1YRQ). The presence of the bridging ligand and its identity depend on the state of

the enzyme. The colour scheme is as follows nickel, green; iron, orange; sulfur, yellow; oxygen, red;

nitrogen, blue. Structure was generated with MacPymol.

In addition to a role in iron delivery, it has been suggested that HypC or the homolog HybG act

as a chaperone to keep the LS from misfolding and to maintain a conformational state that

4

enables the nickel ion to access the active site (30). An investigation of the interaction between

HypC and HycE (the LS of Escherichia coli hydrogenase 3) revealed a possible mechanism of

conformational rearrangement of HycE during iron and nickel insertion where HypC masks the

first cysteine residue in the first CXXC motif in HycE through an interaction that cannot be

inhibited by reducing agents (31). This step is then followed by coordination of the second

cysteine in the motif to the iron (31).

The second step in the maturation pathway is insertion of the nickel ion to the active site and is

carried out by the dedicated nickel chaperones HypA, HypB and SlyD. HypA is a metal-binding

protein that is implicated in the nickel insertion step of hydrogenase maturation because lesions

in the hypA gene result in strains that lack hydrogenase activity and the phenotype can be

partially rescued by nickel supplementation to the growth media (32, 33). Interestingly, E. coli

expresses HypA as well as a homolog, HybF, each of which has a certain degree of specificity

for hydrogenase 3 or the hydrogenase 1 and 2 isoenzymes, respectively (33, 34). E. coli HypA

and HybF as well as HypA proteins from other organisms share conserved histidine and

glutamate residues at positions 2 and 3 as well as four conserved cysteine residues that are

localized within a pair of CXXC motifs (34-36). These latter motifs are ideal coordination sites

for structural zinc ions and are commonly found in domains that facilitate interactions with other

biomolecules (37). HypA binds stoichiometric amounts of zinc with nanomolar affinity (34, 38)

and two recently-solved HypA structures verified the tetrahedral coordination of the zinc ion to a

tetrathiolate site (36, 38-40). A second metal-binding site in HypA includes the conserved N-

terminal residues (Met1,His2,Glu3) and binds nickel with micromolar affinity in an octahedral

geometry (34, 36, 38, 41). Given the assignment of separate HypA homologs for the

biosynthesis of the different hydrogenase isoforms in E. coli and also considering its structural

zinc domain, it is feasible that HypA acts as a scaffold protein for the metallocenter assembly

process on the hydrogenase LS (34, 42). Along these lines, it has been noted that a variable

region in the zinc domain of HypA homologs could interact with distinct hydrogenase LSs given

the variability in the hydrogenase enzymes themselves (40). It is possible that HypA is the

immediate source of nickel for the hydrogenase enzyme, and that nickel binding or release

modulates how HypA interacts with the appropriate LS. However, it is unlikely that HypA

functions on its own. HypB (discussed below) is another metal-binding protein required for

nickel incorporation during maturation of the [NiFe]-hydrogenase and complex formation

between HypA and HypB has been demonstrated in vitro (38, 41).

5

Ni(II) insertion by the nickel insertion complex and subsequent coordination of the metal ion by

both cysteines induces cleavage of a C-terminal peptide of HycE by the endopeptidase HycI after

HypC dissociates from the large subunit (43). The exact mechanism of this endoproteolytic

cleavage event is still unknown and the reaction is not inhibited by conventional inhibitors of

serine or metalloproteases (44). What is evident however, is a requirement for nickel loading

prior to C-terminal processing (45). Recent structures of E. coli HycI (specific for hydrogenase

3) are supportive of a model whereby recognition of nickel-bound hydrogenase triggers

conformational changes in the endoprotease resulting in cleavage of the hydrogenase C-terminus

(46, 47). The cleavage site of the [NiFe]-hydrogenase LS occurs at the C-terminal end of a

histidine or arginine residue found three amino acids downstream from the last cysteine in the

DPCXXCXXH(R) motif (44). Once cleavage is completed, a conformational rearrangement

brings the C-terminus inwards and closes the bridge between the two metals by allowing

coordination of the final cysteine at the C-terminus (43, 48).

The assembly of the [NiFe] active site in hydrogenase represents a highly complex, yet

incredibly well organized cooperative mechanism. Thus far, and possibly due to the alluring

nature of the irregular cyanide and carbon monoxide ligands, significant efforts have been

invested in delineating the steps of the biosynthesis pathway revolving around the iron complex.

However, information on the players of the nickel insertion step is beginning to emerge.

1.3 HypB

Hydrogenase maturation and activity is compromised in strains with deletions of hypB and as

with ∆hypA strains, partial enzyme reconstitution is achieved when the growth media is

supplemented with extra nickel (19, 32, 49, 50). HypB has an essential GTPase activity that is

required for hydrogenase maturation and nickel insertion in vivo (49, 51, 52). This activity is

intrinsically low, with a kcat of 0.2 min-1

and a KM of 4-7 μM for the E. coli and Bradyrhizobium

japonicum proteins (53, 54). It is likely that the activity is stimulated in the context of the

complete biosynthetic pathway, but the role of GTP hydrolysis is not yet known. Nucleotide

binding to HypB occurs at the C-terminal domain of the protein (G-domain) and a structure of

HypB from Methanocaldococcus jannaschii revealed that the GTP-binding site bridges a protein

homodimer (Figure 1-2A) (55). Within this G-domain, conserved cysteine and histidine residues

6

Figure 1-2. HypB. (A) Structure of HypB from M. jannaschii (pdb 2HF8). Two GTP molecules are

bound at the interface of the protein dimer. Two zinc ions (silver) are also bound at the interface of the

protein dimer to conserved Cys95 and Cys127 from one monomer and Cys95, Cys127 and His96 from

the other, as well as several solvent molecules. The corresponding E. coli numbering is Cys166, His167

and Cys198. This homolog does not have the N-terminal sequence found in the E. coli protein. Zinc ions

are also bound on the surface to His100 and His104 as well as residues from a molecule of the adjacent

crystal contact (not shown). The color scheme is magnesium, lime; zinc, silver; nickel, green; oxygen,

red; sulfur, yellow; phosphorus, orange; carbons of the GTPS, cyan; and nitrogen, blue. The pictures

were generated by using MacPymol. (B) Truncated sequence alignment of selected HypB proteins

performed with T-Coffee (56). The CXXCGC motifs and conserved cysteines and histidine from the G-

domain are highlighted in yellow and purple, respectively.

A

B

B

7

form a metal-binding site (see Figure 1-2B for alignment) that in the E. coli protein can bind zinc

with a KD of 1 µM and nickel with an order of magnitude weaker affinity (57). Mutations of

C166 or H167 in E. coli HypB disrupts metal-binding to this low-affinity metal site in vitro (57)

and hydrogenase activity was abrogated in strains expressing these HypB mutants (58). The

structure of M. jannaschii HypB also has metal bound to the corresponding residues as well as

the cysteine corresponding to C198, with an asymmetric dinuclear zinc site at the homodimer

interface and a single cysteine from one of the monomers bridging the two zinc metals (Figure 1-

2A) (55).

Some HypB homologs have a CXXCGC motif at the N-terminus (Figure 1-2B) which

contributes to a second metal-binding site. In E. coli HypB, the cysteines in this sequence bind

nickel with picomolar affinity (57) and mutations of these cysteines also abrogated hydrogenase

production in vivo (58). A combination of X-ray absorption spectroscopy (XAS) analysis,

mutagenesis, and chemical modification were used to demonstrate that the initial methionine is

removed and nickel is bound in a square planar geometry with coordination to the amino

terminal nitrogen as well as the three cysteines (57, 58). The location of this tight nickel site at

the beginning of the primary sequence of HypB makes it tempting to speculate that it competes

for nickel in the E. coli cytoplasm, ensuring that sufficient nickel is directed towards the

hydrogenase enzymes, and that it acts as a flexible nickel delivery arm, perhaps in complex with

HypA. However, this site is not conserved so it must have a function that meets the specific

requirements of only certain microorganisms. Furthermore, if the nickel in this high-affinity site

is passed on to other proteins, there must be a change in the site that activates release of the

metal.

Now that each of the nickel- and zinc-binding properties of HypB has been characterized

individually, the question arises as to how they relate to each other within the function of HypB.

XAS analysis of E. coli HypB suggested that the two sites subtly influence each other, indicating

that the two metal sites are not independent (58) Furthermore, one of the ligands of the low-

affinity metal site is part of the GTPase switch II motif, which contributes to the protein

conformational response to GTP hydrolysis. It has been suggested that the reorganization of the

low-affinity metal-binding site of E. coli HypB upon nickel-binding to the high-affinity site

could lead to conformational changes of the protein that provide a link between the GTPase

activity and the metal-binding sites (58). These observations suggest a role for the G-domain

8

metal-binding site as a structural component that recognizes nickel binding or release from the

N-terminal domain and subsequently activates GTP hydrolysis. Supporting this hypothesis, a

recent analysis of HypB has shown that metal binding to the G-domain has an inhibitory affect

on the GTPase activity of the protein (59). This link between metal-binding and enzyme activity

is essential because mutating the ligands of the G-domain metal site of HypB abrogates all

hydrogenase production in E. coli (60). Therefore, it is likely that the low-affinity site binds

metal in vivo and contributes directly to nickel insertion. It has also been proposed that the

GTPase activity may be required for associations between the nickel-insertion factors or for

nickel-insertion protein complex dissociation following nickel ion delivery to the hydrogenase

(5, 42).

Some HypB homologs also contain a histidine-rich region near the N-terminus capable of

binding multiple nickel ions (35, 50, 54). This polyhistidine stretch is important in nickel

storage, ensuring nickel sequestration for subsequent hydrogenase maturation during nickel

deprivation (61). In the cases of organisms with a HypB that lacks the histidine-rich stretch such

as E. coli and H. pylori it was believed that sufficient levels of nickel were made available by

efficient nickel-transport systems (13, 62). However, the role of nickel-storage factor in E. coli

has since been assigned to SlyD.

1.4 SlyD

The lack of a histidine-rich region in E. coli HypB prompted a search for possible nickel storage

proteins that could contribute to the hydrogenase biosynthetic pathway in this organism.

Experiments using pull-down assays of genomically-tagged HypB resulted in the identification

of SlyD as a partner protein (63), and this interaction has also recently been observed between

the H. pylori proteins (64). A role for SlyD during hydrogenase maturation was established with

∆slyD strains of E. coli, which exhibit deficient hydrogenase activity and overall less nickel

uptake (63).

SlyD was originally discovered as a protein that is required to render E. coli sensitive to lysis

mediated by the phage ΦΧ174 (65). The protein stabilizes the viral lysis protein E and bacteria

carrying mutations in the slyD gene are resistant to lysis, hence its acronym sensitive to lysis D

(65). SlyD is a multi-domain protein with a well folded N-terminal region corresponding to a

9

FKBP domain (FK-506 binding protein) with PPIase activity (peptidyl-prolyl isomerase) and a

chaperone domain that assists in protein folding by binding to unfolded/aggregated proteins

(Figure 1-3A) (66, 67). PPIases are ubiquitous enzymes that catalyzes one of the slowest steps

during protein folding, the trans-to-cis isomerization of Xaa-Pro bonds. Typically these

isomerases are divided into three families which include the cyclophilins, the paruvalins, and

FKBPs, and most prolyl-isomerases that assist in cellular protein folding falls into this last class

(68). The general organization of secondary structure elements classifies SlyD as a typical

FKBP protein although it has low sequence identity to the human FKBP12 (hFKBP12) (67).

The chaperone-like activity of SlyD resides in the characteristic insertion within the PPIase

domain termed the “flap” region in hFKBP12, thus this extended chaperone domain is called the

insertion-in-flap (IF) in SlyD (67). This chaperone module binds to unfolded or aggregated

polypeptides and overrides the inherently high sequence specificity of the prolyl isomerase site

thereby enabling the SlyD to function as a general protein-folding protein (68, 69). Supporting

the chaperone activity of SlyD, recent analysis of SlyD complexes in vivo indicate that this

protein interacts directly with the large subunit of the hydrogenase-3 isozyme, HycE (70). This

interaction occurs even before the initial metallocentre assembly step, the iron insertion. SlyD-

HycE interaction is mediated via the IF domain of SlyD, thus it is assumed to function as a

chaperone by keeping the precursor enzyme in a partially unfolded state to allow metal insertion.

While the chaperone function of SlyD is useful for enzyme maturation, the isomerase activity is

not required for hydrogenase maturation in E. coli because mutations that eliminate the PPIase

activity in vitro do not reduce hydrogenase production to the level of the ∆slyD strain in vivo

(71).

Unlike the conserved N-terminal domains of SlyD, the C-terminal region is variable in SlyD

homologues (72) (Figure 1-3B). In E. coli SlyD the final 50 residues include 28 potential metal-

binding amino acids (6 Cys, 15 His, 2 Glu, 5 Asp) and is therefore called the metal-binding

domain (MBD). This domain is unstructured according to NMR data (66, 67), and has the ability

to bind several nickel ions (73), so the protein is expected to function as a nickel storage protein.

The importance of this unique MBD is further supported because truncation of the metal-binding

domain abrogates the activity of SlyD in hydrogenase production (74). Interestingly, metal

binding at the C-terminus can regulate the PPIase activity of the protein (73) therefore providing

what seems to be a switch-type mechanism to turn off one of its activities upon metal binding.

10

Figure 1-3. SlyD. (A) The solution structure of SlyD from E. coli (67), colour coded according to the

secondary structure elements. The domains are labeled and only part of the unstructured metal-binding

domain (MB domain) corresponding to residues 146-165 is shown. (B) Sequence alignment of SlyD

proteins performed with T-Coffee (56). The His and Cys residues are in green and red respectively. The

shaded sequence corresponds to the variable metal-binding domains of SlyD homologs.

SlyD and HypB form a heterodimer through the proline-containing linker region of HypB

between the high-affinity N-terminal metal-binding site and the G-domain, and a loop in the IF

domain of SlyD (74). Complex formation between the two proteins is important for the

hydrogenase-related function of SlyD in vivo, because disruption of this interaction through

mutagenesis of hypB results in decreased but not complete loss of hydrogenase activity, to the

same extent as that observed in the ∆slyD strain (74). In addition, complex formation with SlyD

accelerates nickel release from the high-affinity N-terminal metal-binding site of HypB (74), an

10 20 30 40 50 60 70 80

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| ....|....| ....|....|

E. coli MKVAK----D LVVSLAYQVR TEDG-VLVDE SPVSAPLDYL HGHGSLISGL ETALEGHEVG DKFDVAVGAN DAYGQYDENL 75

V. cholera MKIEK----N TVASLAYQLT IEDG-VVVDQ STVDAPLDYL HGHNNLITGL ERELEGKVAG DKFTVTIAPE DAYGEHNEDL 75

H. pylori MQNHDLESIK QAALIEYEVR EQGSSIVLDS NISKEPLEFI IGTNQIIAGL EKAVLKAQIG EWEEVVIAPE EAYGVYESSY 80

H. influenzae MKVEK----N VVVSISYQVR TQDG-VLVDE APANQPLEYL QGHNNLVIGL EKALEGKEVG DKFEVRVQPE EGYGAYSENM 75

T. thermophilus MKVGQ----D KVVTIRYTLQ VEG-E-VLDQ G----ELSYL HGHRNLIPGL EEALEGREEG EAFQAHVPAE KAYGPHDPEG 70

*: . . .. : * : :. ::*. *.:: * .:: ** * : * : . : .: ..** :. .

90 100 110 120 130 140 150 160

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| ....|....| ....|....|

E. coli VQRVPKDVFM GVDELQVGMR FLAETD-QGP VPVEITAVED DHVVVDGNHM LAGQNLKFNV EVVAIREATE EELAHGHVHG 154

V. cholera VQRVPADVFQ GVDELEVGMR FLADTD-QGP IPVEITEVDG DEVVVDGNHM LAGQSLTFTV EVVAVRAATE DEIAHGHIHQ 154

H. pylori LQEVPRDQFE GI-ELEKGMS VFGQTEDNQT IQAIIKDFSA THVMVDYNHP LAGKTLAFRF KVLGFREVSE EEILASHHGG 159

H. influenzae VQRVPKDVFQ GVDELEVGMR FLADTD-IGP VPVVITEIDG DEVVVDGNHM LAGQELHFTV EVVAAREATL EEIAHGHVHG 154

T. thermophilus VQVVPLSAFP EDAEVVPGAQ FYAQDMEGNP MPLTVVAVEG EEVTVDFNHP LAGKDLDFQV EVVKVREATP EELLHGHA-- 148

:* ** . * *: * . .: . : : .. .* ** ** ***: * * . :*: * .: :*: .*

170 180 190 200 210

....|....| ....|....| ....|....| ....|....| ....|....| .

E. coli A----HDHHH DHDHD-GC-C GG--HGHDH- GHEHGGEGCC GGKGNGGCGC H 196

V. cholera AGGCGHDHDH DHDHEGGC-C GGEGHGHDHH GHGKKEGGCC G---GGGCGS H 201

H. pylori G--------- ----T-GC-C GG--HGG-H- GGKKG-GG-C G----CSCSH G 185

H. influenzae A----HSHDD DEEGH-GCGC GG--HHHEH- NHEHNHGSC- ------GCGG H 190

T. thermophilus ---------- ---------- ---------- ---------- ---------- H 149

(B)

(A)

PPIase

domain

IF

domain

MB

domain

11

activity of SlyD that requires its metal-binding domain. These observations support an active

role for SlyD in hydrogenase maturation beyond that of a nickel source, where SlyD modulates

the affinity of the nickel site on HypB.

1.5 Purpose of study

It is clear that SlyD plays an important role in [NiFe]-hydrogenase biosynthsis in E. coli and can

also contribute to Ni(II) homeostasis in this bacterium. While in vivo a Ni(II)-associated

function of SlyD has been established, how this protein binds Ni(II) was not well understood.

Therefore, using several spectroscopic techniques, electrospray-ionization mass spectrometry

(ESI-MS) and molecular biology, an in-depth investigation was conducted to understand the in

vitro nickel-binding activity of SlyD as described in Chapter 2. In addition to Ni(II), SlyD was

also known to bind to other transition metals in vitro (75), bringing forth the issue of metal-

selectivity of SlyD. In an effort to understand whether SlyD could selectively bind Ni(II), the

metal-binding ability of SlyD to several transition metals was investigated in vitro and its role in

metal homeostasis in vivo was examined. These studies are presented in Chapter 3. Finally,

SlyD has the ability to bind several nickel ions and is hypothesized to function as a reservoir of

Ni(II) that is used for [NiFe]-hydrogenase assembly. Chapter 4 is dedicated to examining this

hypothesis and understanding how nickel may be mobilized to the active site of the precursor

enzyme. While nickel mobilization to the hydrogenase enzyme is considered to be carried out by

HypA, HypB and SlyD, how the coordinated action of these proteins leads to successful

biogenesis of the enzyme is not well understood. Therefore, the consequences of the interaction

between HypB and SlyD nickel metallochaperones were examined and the importance of this

HypB-SlyD interaction is presented in Chapter 4.

12

References

1. Fontecilla-Camps, J. C. (1998) Biological Nickel, Struct. Bonding 91, 1-30.

2. Gerendás, J., Polacco, J. C., Freyermuth, S. K., and Sattelmacher, B. (1999) Significance

of nickel for plant growth and metabolism, J. Plant Nutr. Soil Sci. 162, 241-256.

3. Mulrooney, S. B., and Hausinger, R. P. (2003) Nickel uptake and utilization by

microorganisms, FEMS Microbiol. Rev. 27, 239-261.

4. Muyssen, B. T. A., Brix, K. V., DeForest, D. K., and Janssen, C. R. (2004) Nickel

essentiality and homeostasis in aquatic organisms, Environ. Rev. 12, 113-131.

5. Li, Y., and Zamble, D. B. (2009) Nickel homeostasis and nickel regulation: an overview,

Chem. Rev. 109, 4617-4643.

6. Ragsdale, S. W. (2009) Nickel-based Enzyme Systems, J. Biol. Chem. 284, 18571-18575.

7. Bauerfeind, P., Garner, R., Dunn, B. E., and Mobley, H. L. (1997) Synthesis and activity

of Helicobacter pylori urease and catalase at low pH, Gut 40, 25-30.

8. Maier, R. J., Benoit, S. L., and Seshadri, S. (2007) Nickel-binding and accessory proteins

facilitating Ni-enzyme maturation in Helicobacter pylori, BioMetals 20, 655-664.

9. Ragsdale, S. W. (2007) Nickel and the carbon cycle, J. Inorg. Biochem. 101, 1657-1666.

10. Nieminen, T. M., Ukonmaanaho, L., Rausch, N., and Shotyk, W. (2007) Biogeochemistry

of nickel and its release into the environment, in Metal Ions in Life Sciences (Sigel, A.,

Sigel, H., and Sigel, R. K. O., Eds.), John Wiley and Sons.

11. Kuchar, J., and Hausinger, R. P. (2004) Biosynthesis of metal sites, Chem. Rev. 104, 509-

525.

12. Vignais, P. M., and Billoud, B. (2007) Occurrence, classification, and biological function

of hydrogenases: an overview, Chem. Rev. 107, 4206-4272.

13. Casalot, L., and Rousset, M. (2001) Maturation of the [NiFe] hydrogenases, Trends

Microbiol. 9, 228-237.

14. Olson, J. W., and Maier, R. J. (2002) Molecular hydrogen as an energy source for

Helicobacter pylori, Science 298, 1788-1790.

15. Maier, R. J. (2005) Use of molecular hydrogen as an energy source substrate by human

pathogenic bacteria, Biochem. Soc. Trans. 33, 83-85.

16. Fontecilla-Camps, J. C., Volbeda, A., Cavazza, C., and Nicolet, Y. (2007)

Structure/function relationships of [NiFe]- and [FeFe]-hydrogenases, Chem. Rev. 107,

4273-4303.

17. Volbeda, A., and Fontecilla-Camps, J. C. (2003) The active site and catalytic mechanism

of NiFe hydrogenases, Dalton Trans. 4030-4038.

18. Böck, A., King, P. W., Blokesch, M., and Posewitz, M. C. (2006) Maturation of

hydrogenases, Adv. Microb. Physiol. 51, 1-71.

19. Jacobi, A., Rossmann, R., and Bock, A. (1992) The hyp operon gene products are

required for the maturation of catalytically active hydrogenase isoenzymes in Escherichia

coli, Arch. Microbiol. 158, 444-451.

20. Winter, G., Buhrke, T., Lenz, O., Jones, A. K., Forgber, M., and Friedrich, B. (2005) A

model system for [NiFe] hydrogenase maturation studies: Purification of an active site-

containing hydrogenase large subunit without small subunit, FEBS Lett. 579, 4292-4296.

21. Löscher, S., Zebger, I., Andersen, L. K., Hildebrandt, P., Meyer-Klaucke, W., and

Haumann, M. (2005) The structure of the Ni-Fe site in the isolated HoxC subunit of the

hydrogen-sensing hydrogenase from Ralstonia eutropha, FEBS Lett. 579, 4287-4291.

13

22. Blokesch, M., Paschos, A., Bauer, A., Reissmann, S., Drapal, N., and Bock, A. (2004)

Analysis of the transcarbamoylation-dehydration reaction catalyzed by the hydrogenase

maturation proteins HypF and HypE, Eur. J. Biochem. 271, 3428-3436.

23. Paschos, A., Bauer, A., Zimmermann, A., Zehelein, E., and Bock, A. (2002) HypF, a

carbamoyl phosphate-converting enzyme involved in [NiFe] hydrogenase maturation, J.

Biol. Chem. 277, 49945-49951.

24. Rangarajan, E. S., Asinas, A., Proteau, A., Munger, C., Baardsnes, J., Iannuzzi, P., Matte,

A., and Cygler, M. (2008) Structure of [NiFe] hydrogenase maturation protein HypE

from Escherichia coli and its interaction with HypF, J. Bacteriol. 190, 1447-1458.

25. Reissmann, S., Hochleitner, E., Wang, H., Paschos, A., Lottspeich, F., Glass, R. S., and

Bock, A. (2003) Taming of a poison: biosynthesis of the NiFe-hydrogenase cyanide

ligands, Science 299, 1067-1070.

26. Blokesch, M., Albracht, S. P., Matzanke, B. F., Drapal, N. M., Jacobi, A., and Bock, A.

(2004) The complex between hydrogenase-maturation proteins HypC and HypD is an

intermediate in the supply of cyanide to the active site iron of [NiFe]-hydrogenases, J.

Mol. Biol. 344, 155-167.

27. Blokesch, M., and Bock, A. (2006) Properties of the [NiFe]-hydrogenase maturation

protein HypD, FEBS Lett. 580, 4065-4068.

28. Ludwig, M., Schubert, T., Zebger, I., Wisitruangsakul, N., Saggu, M., Strack, A., Lenz,

O., Hildebrandt, P., and Friedrich, B. (2009) Concerted action of two novel auxiliary

proteins in assembly of the active site in a membrane-bound [NiFe] hydrogenase, J. Biol.

Chem. 284, 2159-2168.

29. Blokesch, M., and Bock, A. (2002) Maturation of [NiFe]-hydrogenases in Escherichia

coli: the HypC cycle, J. Mol. Biol. 324, 287-296.

30. Drapal, N., and Bock, A. (1998) Interaction of the hydrogenase accessory protein HypC

with HycE, the large subunit of Escherichia coli hydrogenase 3 during enzyme

maturation, Biochemistry 37, 2941-2948.

31. Magalon, A., and Bock, A. (2000) Analysis of the HypC-hycE complex, a key

intermediate in the assembly of the metal center of the Escherichia coli hydrogenase 3, J.

Biol. Chem. 275, 21114-21120.

32. Olson, J. W., Mehta, N. S., and Maier, R. J. (2001) Requirement of nickel metabolism

proteins HypA and HypB for full activity of both hydrogenase and urease in Helicobacter

pylori, Mol. Microbiol. 39, 176-182.

33. Hube, M., Blokesch, M., and Bock, A. (2002) Network of hydrogenase maturation in

Escherichia coli: role of accessory proteins HypA and HybF, J. Bacteriol. 184, 3879-

3885.

34. Blokesch, M., Rohrmoser, M., Rode, S., and Bock, A. (2004) HybF, a zinc-containing

protein involved in NiFe hydrogenase maturation, J. Bacteriol. 186, 2603-2611.

35. Fu, C., and Maier, R. J. (1994) Nucleotide sequences of two hydrogenase-related genes

(hypA and hypB) from Bradyrhizobium japonicum, one of which (hypB) encodes an

extremely histidine-rich region and guanine nucleotide-binding domains, Biochim.

Biophys. Acta. 1184, 135-138.

36. Xia, W., Li, H., Sze, K. H., and Sun, H. (2009) Structure of a nickel chaperone, HypA,

from Helicobacter pylori reveals two distinct metal binding sites, J. Am. Chem. Soc. 131,

10031-10040.

37. Gamsjaeger, R., Liew, C. K., Loughlin, F. E., Crossley, M., and Mackay, J. P. (2007)

Sticky fingers: zinc-fingers as protein-recognition motifs, Trends Biochem. Sci. 32, 63-

70.

14

38. Atanassova, A., and Zamble, D. B. (2005) Escherichia coli HypA is a zinc metalloprotein

with a weak affinity for nickel, J. Bacteriol. 187, 4689-4697.

39. Kennedy, D. C., Herbst, R. W., Iwig, J. S., Chivers, P. T., and Maroney, M. J. (2007) A

dynamic Zn site in Helicobacter pylori HypA: a potential mechanism for metal-specific

protein activity, J. Am. Chem. Soc. 129, 16-17.

40. Watanabe, S., Arai, T., Matsumi, R., Atomi, H., Imanaka, T., and Miki, K. (2009) Crystal

Structure of HypA, a Nickel-Binding Metallochaperone for [NiFe] Hydrogenase

Maturation, J. Mol. Biol. 394.

41. Mehta, N., Olson, J. W., and Maier, R. J. (2003) Characterization of Helicobacter pylori

nickel metabolism accessory proteins needed for maturation of both urease and

hydrogenase, J. Bacteriol. 185, 726-734.

42. Leach, M. R., and Zamble, D. B. (2007) Metallocenter assembly of the hydrogenase

enzymes, Curr. Opin. Chem. Biol. 11, 159-165.

43. Magalon, A., and Bock, A. (2000) Dissection of the maturation reactions of the [NiFe]

hydrogenase 3 from Escherichia coli taking place after nickel incorporation, FEBS Lett.

473, 254-258.

44. Rossmann, R., Maier, T., Lottspeich, F., and Bock, A. (1995) Characterisation of a

protease from Escherichia coli involved in hydrogenase maturation, Eur. J. Biochem.

227, 545-550.

45. Rossmann, R., Sauter, M., Lottspeich, F., and Bock, A. (1994) Maturation of the large

subunit (HYCE) of Escherichia coli hydrogenase 3 requires nickel incorporation

followed by C-terminal processing at Arg537, Eur. J. Biochem. 220, 377-384.

46. Yang, F., Hu, W., Xu, H., Li, C., Xia, B., and Jin, C. (2007) Solution structure and

backbone dynamics of an endopeptidase HycI from Escherichia coli: implications for

mechanism of the [NiFe] hydrogenase maturation, J. Biol. Chem. 282, 3856-3863.

47. Kumarevel, T., Tanaka, T., Bessho, Y., Shinkai, A., and Yokoyama, S. (2009) Crystal

structure of hydrogenase maturating endopeptidase HycI from Escherichia coli, Biochem.

Biophys. Res. Comm. 389, 310-314.

48. Theodoratou, E., Huber, R., and Bock, A. (2005) [NiFe]-Hydrogenase maturation

endopeptidase: structure and function, Biochem. Soc. Trans. 33, 108-111.

49. Maier, T., Lottspeich, F., and Bock, A. (1995) GTP hydrolysis by HypB is essential for

nickel insertion into hydrogenases of Escherichia coli, Eur. J. Biochem. 230, 133-138.

50. Rey, L., Imperial, J., Palacios, J. M., and Ruiz-Argueso, T. (1994) Purification of

Rhizobium leguminosarum HypB, a nickel-binding protein required for hydrogenase

synthesis, J. Bacteriol. 176, 6066-6073.

51. Olson, J. W., and Maier, R. J. (2000) Dual roles of Bradyrhizobium japonicum nickelin

protein in nickel storage and GTP-dependent Ni mobilization, J. Bacteriol. 182, 1702-

1705.

52. Mehta, N., Benoit, S., and Maier, R. J. (2003) Roles of conserved nucleotide-binding

domains in accessory proteins, HypB and UreG, in the maturation of nickel-enzymes

required for efficient Helicobacter pylori colonization, Microb. Pathog. 35, 229-234.

53. Maier, T., Jacobi, A., Sauter, M., and Bock, A. (1993) The product of the hypB gene,

which is required for nickel incorporation into hydrogenases, is a novel guanine

nucleotide-binding protein, J. Bacteriol. 175, 630-635.

54. Fu, C., Olson, J. W., and Maier, R. J. (1995) HypB protein of Bradyrhizobium japonicum

is a metal-binding GTPase capable of binding 18 divalent nickel ions per dimer, Proc.

Nat. Acad. Sci. U. S. A. 92, 2333-2337.

15

55. Gasper, R., Scrima, A., and Wittinghofer, A. (2006) Structural insights into HypB, a

GTP-binding protein that regulates metal binding, J. Biol. Chem. 281, 27492-27502.

56. Notredame, C., Higgins, D., and Heringa, J. (2000) T-Coffee: A novel method for

multiple sequence alignments, J. Mol. Biol. 302, 205-217.

57. Leach, M. R., Sandal, S., Sun, H., and Zamble, D. B. (2005) Metal binding activity of the

Escherichia coli hydrogenase maturation factor HypB, Biochemistry 44, 12229-12238.

58. Chan Chung, K. C., Cao, L., Dias, A. V., Pickering, I. J., George, G. N., and Zamble, D.

B. (2008) A high-affinity metal-binding peptide from Escherichia coli HypB, J. Am.

Chem. Soc. 130, 14056-14057.

59. Cai, F., Ngu, T. T., Kaluarachchi, H., and Zamble, D. B. (2011) Relationship between the

GTPase, metal-binding, and dimerization activities of E. coli HypB, J. Biol. Inorg. Chem.

In Press.

60. Dias, A. V., Mulvihill, C. M., Leach, M. R., Pickering, I. J., George, G. N., and Zamble,

D. B. (2008) Structural and biological analysis of the metal sites of Escherichia coli

hydrogenase accessory protein HypB, Biochemistry 47, 11981-11991.

61. Olson, J. W., Fu, C., and Maier, R. J. (1997) The HypB protein from Bradyrhizobium

japonicum can store nickel and is required for the nickel-dependent transcriptional

regulation of hydrogenase, Mol. Microbiol. 24, 119-128.

62. Eitinger, T., and Mandrand-Berthelot, M. A. (2000) Nickel transport systems in

microorganisms, Arch. Microbiol. 173, 1-9.

63. Zhang, J. W., Butland, G., Greenblatt, J. F., Emili, A., and Zamble, D. B. (2005) A role

for SlyD in the Escherichia coli hydrogenase biosynthetic pathway, J. Biol. Chem. 280,

4360-4366.

64. Stingl, K., Schauer, K., Ecobichon, C., Labigne, A., Lenormand, P., Rousselle, J. C.,

Namane, A., and de Reuse, H. (2008) In vivo interactome of Helicobacter pylori urease

revealed by tandem affinity purification, Mol. Cell. Proteomics 7, 2429-2441.

65. Roof, W. D., Horne, S. M., Young, K. D., and Young, R. (1994) SlyD, a host gene

required for X174 lysis, is related to the FK506-binding protein family of peptidyl-

prolyl cis-trans-isomerases, J. Biol. Chem. 269, 2902-2910.

66. Martino, L., He, Y., Hands-Taylor, K. L., Valentine, E. R., Kelly, G., Giancola, C., and

Conte, M. R. (2009) The interaction of the Escherichia coli protein SlyD with nickel ions

illuminates the mechanism of regulation of its peptidyl-prolyl isomerase activity, FEBS J.

276, 4529-4544.

67. Weininger, U., Haupt, C., Schweimer, K., Graubner, W., Kovermann, M., Bruser, T.,

Scholz, C., Schaarschmidt, P., Zoldak, G., Schmid, F. X., and Balbach, J. (2009) NMR

solution structure of SlyD from Escherichia coli: spatial separation of prolyl isomerase

and chaperone function, J. Mol. Biol. 387, 295-305.

68. Jakob, R. P., Zoldak, G., Aumuller, T., and Schmid, F. X. (2009) Chaperone domains

convert prolyl isomerases into generic catalysts of protein folding, Proc. Natl. Acad. Sci.

U. S. A. 106, 20282-20287.

69. Zoldak, G., and Schmid, F. X. Cooperation of the prolyl isomerase and chaperone

activities of the protein folding catalyst SlyD, J. Mol. Biol. 406, 176-194.

70. Chung, K. C. C., and Zamble, D. B. (2011) The Escherichia coli metal-binding

chaperone SlyD interacts with the large subunit of [NiFe]-hydrogenase 3, FEBS Lett. 585,

291-294.

71. Zhang, J. W., Leach, M. R., and Zamble, D. B. (2007) The peptidyl-prolyl isomerase

activity of SlyD is not required for maturation of Escherichia coli hydrogenase, J.

Bacteriol. 189, 7942-7944.

16

72. Erdmann, F., and Fischer, G. (2007) The nickel-regulated peptidyl prolyl cis/trans

isomerase SlyD, in Metal Ions in Life Sciences (Sigel, A., Sigel, H., and Sigel, R. K. O.,

Eds.), pp 501-518, John Wiley and Sons.

73. Hottenrott, S., Schumann, T., Pluckthun, A., Fischer, G., and Rahfeld, J. U. (1997) The

Escherichia coli SlyD is a metal ion-regulated peptidyl-prolyl cis/trans-isomerase, J. Biol.

Chem. 272, 15697-15701.

74. Leach, M. R., Zhang, J. W., and Zamble, D. B. (2007) The role of complex formation

between the Escherichia coli hydrogenase accessory factors HypB and SlyD, J. Biol.

Chem. 282, 16177-16186.

75. Wulfing, C., Lombardero, J., and Pluckthun, A. (1994) An Escherichia coli protein

consisting of a domain homologous to FK506-binding proteins (FKBP) and a new metal

binding motif, J. Biol. Chem.269, 2895-2901.

17

Chapter 2 The Ni(II)-binding properties of the metallochaperone SlyD

2

2.1 Introduction

Nickel plays a significant role in microorganisms, where the partnership between this metal and

proteins is essential for a multitude of diverse biological functions (1-3). Only present at

nanomolar concentrations in the environment (4, 5), many organisms have evolved to utilize this

transition metal ion in a variety of metalloenzymes such as urease, [NiFe]-hydrogenase, a

superoxide dismutase, and acetyl-coenzyme A synthase (1-3). Although essential, the limited

availability and the inherently reactive nature of nickel ions create the necessity for strict control

within the cellular milieu. Hence the biogenesis of many of the intricate nickel-containing

enzyme centres requires sets of proteins that include factors dedicated to the transport and

delivery of the Ni(II) ions to the target proteins (1, 3). These soluble metal-trafficking proteins

are referred to as metallochaperones (6, 7).

The [NiFe]-hydrogenases are crucial for energy metabolism in Escherichia coli where the

enzymes catalyze the oxidation or production of hydrogen gas (8, 9). These bacteria express at

least 3 isozymes of [NiFe]-hydrogenases that all have a nickel- and iron-containing bimetallic

center at the active site (10). Most of the pleiotropic accessory proteins required for the multi-

step biosynthesis of these metallocenters are encoded by the hyp operon (8, 9, 11), and the

responsibilities of these proteins include delivery of both types of metal ions, synthesis of

organic ligands, and protein folding. Although many of the mechanistic details are not yet

known, HypA and HypB are thought to facilitate the Ni(II) insertion step of this pathway (9, 11).

Furthermore, analysis of bacterial multi-protein complexes led to the discovery that HypB forms

a complex with a protein called SlyD, which suggested that SlyD also participates in

2Reproduced with permission from Kaluarachchi, H., Sutherland, D. E. K., Young, A., Pickering, I. J.,

Stillman, M. J., and Zamble, D. B. (2009). The Ni(II)-binding properties of the metallochaperone SlyD, J.

Am. Chem. Soc.,131,18489-18500. Copyright 2009 American Chemical Society. Author contributions:

H.K performed experiments, analyzed data and wrote the paper. D.E.K.S. and A.Y., assisted during ESI-

MS data collection and analysis. I.J.P assisted in XAS data analysis and writing of the XAS sections.

M.J.S. assisted in ESI-MS data analysis and writing of the ESI-MS sections. D.B.Z analyzed data and

wrote the paper. The main text has not been changed. All tables and figures including data presented as

supplementary information have been added to the main text and renumbered.

18

hydrogenase biosynthesis (12). This hypothesis was supported by experiments with ∆slyD

strains of E. coli, which exhibited reduced activity levels of all three [NiFe]-hydrogenases (12).

That this phenotype could be complemented by the addition of excess nickel to the media

indicated that SlyD contributes specifically to the nickel-loading step of the hydrogenase

metallocenter assembly pathway (12).

Originally identified as a protein that sensitizes E. coli to lysis by bacteriophage ΦX174 (13),

SlyD is a member of the FK506-binding protein (FKBP) family of peptidyl-prolyl isomerases

(PPIase) (14, 15). The PPIase domain located at the N-terminus of the protein (Figure 2-1)

catalyzes the intrinsically slow interconversion of peptidyl-prolyl amide bonds, thereby assisting

in protein folding (16-18). In addition to the isomerase function, which is non-essential for

hydrogenase biosynthesis (19), SlyD also possesses general chaperone activity similar to many

other FKBPs (16, 18). This latter property is attributed to an additional domain inserted within a

loop of the PPIase domain termed IF (insertion in the flap) that is observed in a subset of FKBPs

(20, 21). Deletion of several residues in the SlyD IF domain prevents interaction with HypB and

abrogates its function in hydrogenase production in vivo (22).

Figure 2-1. SlyD domain architecture and amino acid sequence. The amino acid sequence is colour

coded according to the domains. The PPIase domain is homologous to the FK506-binding proteins. The

SlyD metal-binding domain is not conserved in all SlyD homologs, although many have sequences of

variable length that are rich in potential metal-binding residues.

The last fifty amino acids of SlyD encompass an unusual domain that is rich in the potential

metal-binding residues histidine, cysteine, aspartate and glutamate. Given the composition of this

domain, it is not surprising that SlyD is capable of coordinating a range of transition metal ions

such as Co(II), Ni(II), and Cu(II), as well as Zn(II) (14, 23). Several HypB homologues are also

capable of binding multiple metal ions due to his-rich sequences, and in these organisms HypB is

thought to contribute to nickel storage in addition to hydrogenase production (24). The lack of

1 196Peptidyl-Prolyl Isomerase Domain Metal-Binding Domain146

IF- Chaperone Domain

76 120MKVAKDLVVS LAYQVRTEDG VLVDESPVSA PLDYLHGHGS LISGLETALE GHEVGDKFDV

AVGANDAYGQ YDENLVQRVP KDVFMGVDEL QVGMRFLAET DQGPVPVEIT AVEDDHVVVD

GNHMLAGQNL KFNVEVVAIR EATEEELAHG HVHGAHDHHH DHDHDGCCGG HGHDHGHEHG

GEGCCGGKGN GGCGCH

19

such a region in E. coli HypB and the ability of SlyD to bind nickel ions led to the proposal that

SlyD is responsible for nickel storage in E. coli and that it serves as a source of nickel for the

production of the hydrogenase enzymes (22, 23). This hypothesis was supported by the

observations that SlyD influences nickel accumulation in this organism and that the C-terminal

domain of SlyD is required for the optimal production of hydrogenase activity (22). SlyD also

influences nickel release from a metal-binding site of HypB, suggesting another, more active task

for SlyD during hydrogenase biosynthesis (22).

The NMR solution structures of both a truncated version of E. coli SlyD (1-165) and the full-

length protein were recently reported (21, 25). The structures reveal that the PPIase and IF

domains of SlyD resemble those of other members of the FKBP family (26), but in both cases

the metal-binding regions were unstructured. To further understand the function of SlyD as a

metallochaperone and its role in nickel homeostasis, a detailed characterization of nickel binding

to SlyD is reported here. Our investigation revealed SlyD to be a monomeric protein containing

multiple metal sites of similar affinities. Nickel binding occurs via a non-cooperative

mechanism and the protein chelates nickel by using a mixture of coordination geometries.

Through the study of a range of mutants we find that the cysteine residues in the C-terminal

domain confer higher affinity as well as increased binding capacity to SlyD. Furthermore,

although this domain is primarily responsible for metal binding to SlyD, the effects of metal

binding are not limited to this domain in the protein.

2.2 Materials and Methods

Materials. Pfu DNA polymerase was purchased from Stratagene. Primers (Table 2-1) were

purchased from Sigma Genosys. All chromatography media were from GE Healthcare.

Kanamycin, tris(2-carboxyethyl)phosphine (TCEP), dithiothreitol (DTT), isopropyl-β-D-1-

galactopyranoside (IPTG) were purchased from BioShop (Toronto, ON). Metal salts were, as a

minimum, 99.9% pure and purchased from Aldrich. The concentrations of the metal stock

solutions were verified by ICP-AES. Several of the metal-binding studies were performed with

atomic absorption standard solutions (AAS grade) where noted. Other reagents were analytical

grade from Sigma. The buffers for all metal assays were treated with Chelex-100 (Bio-Rad) to

20

minimize trace metal contamination. All samples were prepared with Milli-Q water, 18.2 MΩ-

cm resistance (Millipore).

Generation of SlyD mutants. Point mutations of SlyD were generated in the parent SlyD-

pET24b vector (12) by using the QuickChange PCR mutagenesis method and Pfu polymerase

along with forward and reverse primers listed in Table 2-1. The resulting PCR plasmids were

transformed into XL-2 Blue E. coli heat-shock competent cells (Stratagene) and isolated using

Qiagen plasmid mini-prep kit. The pET24b-SlyD mutation constructs were verified by

sequencing in the forward and reverse directions (ACGT, Toronto, ON). Three different mutants

targeting each pair of cysteine residues were created and are as follows: SlyD C167, 168A, SlyD

C184,185A, and SlyD Δ193-196. A fourth mutant containing all of the above-mentioned

mutations resulting in a protein devoid of any cysteines and referred to as the Triple mutant here

onwards, was also prepared.

Table 2-1. PCR primers used for mutagenesis.

Name Sequencea

SlyD C167,168A forward 5‟-CGACCACGACGGTgcCgcCGGCGGTCATGGCC-3‟

SlyD C167,168A reverse 5‟-GGCCATGACCGCCGgcGgcACCGTCGTGGTCG-3‟

SlyD C184,185A forward 5‟-CACGGTGGCGAAGGCgcCGcAGGCGGTAAAGGCAACGG-3‟

SlyD C184,185A reverse 5‟-CCGTTGCCTTTACCGCCtgCgGcGCCTTCGCCACCGTG-3‟

SlyD ∆193-196 forward 5‟-GTAAAGGCAACGGCGGTTaaGGTTGCCACTAATACTCG-3‟

SlyD ∆193-196 reverse 5‟-CGAGTATTAGTGGCAACCttAACCGCCGTTGCCTTTAC-3‟

aThe bases corresponding to the mutated residues are underlined and the mutations are in lower case.

Protein expression and purification. SlyD wild-type (WT) protein and mutants were expressed

according to the method previously described (12), except that the mutants were expressed in a

∆slyD BL21(DE3) strain of E. coli (22). SlyD and SlyD variants were purified by using a nickel-

nitrilotriacetic acid (Ni-NTA, Qiagen) column followed by anion exchange on a MonoQ column

(GE Healthcare). The presence of SlyD in the protein fractions collected at each step was

verified by SDS-PAGE. After purification of the protein via the MonoQ column we noted that

for several fractions containing SlyD (with the exception of the Triple mutant), there was an

unexpected absorption band at 320 nm that varied in intensity. ESI-MS analysis of the protein

fractions that exhibited the 320 nm absorption did not resolve a mass difference (data not

shown), the absorption was not affected by treatment with ethylenediaminetetraacetic acid

21

(EDTA), and bound metal was not detected in metal analysis (27). We also noted that by

collecting only the fractions with A320/A280 < 0.1 (i.e. fractions without the absorption at 320 nm)

we are able to modify all the cysteine residues upon reaction with N-ethyl maleimide (NEM) (see

below for reaction conditions). In contrast, the fractions with the absorption at 320 nm could not

be completely modified by NEM and showed lower nickel binding capacity in nickel titration

experiments compared to the fully reduced protein fractions (data not shown). For these reasons,

the absorption is assigned to an oxidation product of the Cys residues but the identity of this

oxidation product is not yet known. Treatment with several reducing reagents (TCEP, dithionite,

β-meracaptoethanol, DTT) has proven unsuccessful in reducing the Cys residues to the thiolate

form while keeping the protein intact. Thus, only fractions without this absorption were pooled

and further purified via gel-filtration on a Superdex S-75 column equilibrated with 20 mM

HEPES, pH 7.5, 100 mM NaCl, 2 mM TCEP, and used in all the experiments described here.

The purity of the final protein fractions were verified by ESI-MS and the masses obtained are

listed in Table 2-2.

Table 2-2. Masses of SlyD proteins.

Protein Calculated Mass

(Da)

Mass by ESI-MS

(Da)

SlyD WT 20852.8 20852.0

SlyD C167,168A 20788.7 20788.0

SlyD C184,185A 20788.7 20788.0

SlyD ∆193-196 20452.3 20452.0

Triple mutant 20324.1 20324.0

SlyD 1-146 15833.7 15832.0

N-ethyl maleimide (NEM) modification assay. To prepare apo-protein, samples were pre-treated

with 4 mM TCEP and 20 mM EDTA overnight in an anaerobic glove box at 4oC. The protein

was then gel filtered twice through PD-10 columns (GE Healthcare) equilibrated with 20 mM

HEPES, pH 7.5, 100 mM NaCl (buffer A). An aliquot of this gel-filtered sample was reacted

with a freshly prepared NEM solution in the same buffer. The final concentration of NEM in all

reactions was approximately five times the concentration of cysteines in the protein ([NEM] = 5

× # of Cys × [protein]). The reaction samples were incubated overnight at 4 oC and were

desalted prior to analysis by ESI-MS. Only protein samples that could be fully modified at all

cysteines were used in further experiments.

22

Metal-binding experiments. Apo-protein was prepared as described above, and then incubated

with the indicated amount of Ni(II) (AAS grade diluted into buffer A) overnight at 4oC in an

anaerobic glove box. Nickel binding was monitored by using electronic absorption

spectroscopy.

Circular dichroism (CD) spectroscopy. Protein samples were pre-treated with EDTA and TCEP

as described above and buffer exchanged into 10 mM ammonium acetate, pH 7.5, using micro-

centrifugal devices (MWCO of 10,000) (Pall nanosep centrifuge devices) in an anaerobic glove

box. SlyD samples (60 µM) were incubated with known amounts of NiCl2 overnight at 4oC.

Samples were loaded into a 0.1 cm cuvette and capped to minimize exposure to air. The spectra

were collected on a Jasco J-710 spectropolarimeter by scanning in the wavelength range of

205−320 nm at room temperature. The final spectra obtained are averages of 5 scans collected

by using a scan speed of 20 nm/min. The molar ellipticity was calculated by using the equation

[Ө]= ellipticity (mdeg)/[conc. protein (M) x 10 x 0.l (cm)].

Equilibrium dialysis experiments. Microchambers of DIALYZER (Harvard Apparatus Inc.)

were utilized for equilibrium dialysis experiments that were all conducted at 4oC overnight. The

protein chamber contained 250 µL of protein solution (40 µM) in buffer A with 1 mM TCEP and

the ligand chamber was filled with 250 µL of approximately 10-fold excess NiCl2 in the same

buffer. The amount of protein-bound nickel and free nickel were determined by a high-

performance liquid chromatography (HPLC) method described previously (27). For HPLC

analysis, which was conducted in triplicate, > 50 μg of protein was dried by centrifugation under

vacuum, reconstituted with metal-free concentrated HCl (Seastar Chemicals), and hydrolyzed by

incubation overnight at 95 °C. The sample was dried again to remove HCl and reconstituted in

80 μL of MilliQ water. This sample was injected onto an IonPak CS5A column attached to a

metal-free Dionex BioLC HPLC system followed by post-column mixing with 4-(2-

pyridylazo)resorcinol (PAR) and detection at 530 nm. To determine free nickel concentration, an

equivalent volume of sample from the ligand chamber was analyzed as described above.

Analytical gel filtration. Analytical gel filtration experiments were performed on a Superdex 75-

HR column (GE Healthcare) at 4 °C with a flow rate of 0.5 mL/min and an injection volume of

100 µL. A protein concentration of 100 µM containing 2 mM TCEP was used in all gel-

filtration experiments with the column equilibrated with buffer A containing similar amounts of

23

reducing agent. SlyD samples were incubated with the indicated amounts of NiCl2 overnight at 4

°C prior to use and the column buffer for these runs was supplemented with NiCl2 concentrations

similar to that of the protein sample. The apparent molar mass of the protein (Mr) was calculated

from the eluting volume (Ve) using the equation Kav = (Ve-Vo)/(Vt-Vo), where Kav is the partition

coefficient, Vt is the total column volume and Vo is the void volume. The column was calibrated

by injecting 100 µL of the LMW gel-filtration standards kit (GE Healthcare). Calibrations and

the sample runs were conducted in triplicate to ensure reproducibility.

X-ray absorption spectroscopy (XAS) sample preparation. The purified protein solutions were

pre-treated with 2 mM TCEP and 10 mM EDTA overnight and were buffer exchanged into 20

mM HEPES pH 7.5, 100 mM NaCl and 2 mM TCEP to remove the chelator and then

concentrated by using an Amicon Ultra Centrifugal Filter Device (Millipore, MWCO = 3000 Da)

to an approximate volume of 200 μL. All protein samples were placed on ice and 0.8 equivalents

of Ni(II) (AAS grade diluted in buffer A) was added gradually in multiple small aliquots to

suppress protein aggregation. Glycerol was then added to the protein solutions to give final

concentrations of 20% (v/v). The final protein concentrations of the samples were approximately

1 mM and these were subsequently frozen with liquid nitrogen and stored at −80 °C. For XAS

data acquisition the samples were thawed, transferred to 2 mm pathlength (2 × 4 × 20 mm)

Lucite sample cuvettes, re-frozen and were maintained at 10 K using an Oxford Instruments

liquid helium flow cryostat. Analysis of holo-protein samples subjected to several freeze/thaw

cycles by using electronic absorption spectroscopy indicated no changes in the absorption

profiles indicating that this process utilized during XAS sample handling did not lead to the loss

of any metal-protein interactions.

XAS data collection. XAS data were collected with the SPEAR3 storage ring containing 100 to

200 mA at the Stanford Synchrotron Radiation Lightsource (SSRL Beam-line 7-3 for mutant

proteins, beamline 9-3 for wild-type SlyD). Experiments employed a Si(220) double crystal

monochromator with an upstream Rh-coated collimating mirror which also provided harmonic

rejection; beamline 9-3 additionally has a downstream focusing mirror. Incident and transmitted

intensity was measured by using an ion chamber filled with nitrogen gas, and the spectrum of the

sample was measured in fluorescence using a Canberra 30-element detector. For each sample 4-8

scans were accumulated, and the energy was calibrated by reference to the absorption of a nickel

24

metal foil measured simultaneously with each scan, assuming a lowest-energy inflection point of

8331.6 eV for nickel foil.

Table 2-3. Best Ni(II) EXAFS curve-fitting results for SlyD loaded with 0.8 equivalents of metal.a

Fit A-Bs N R (Å) σ2 (Å

2) ΔE (eV) F-factor

1 Ni-S

Ni-N

2

3

2.18(6)

2.11(5)

0.005(6)

0.006(21)

-13(3) 44.4

2 Ni-S

Ni-N

2

4

2.19(3)

2.13(3)

0.004(1)

0.009(15)

-10(5) 44.9

3 Ni-S

Ni-N

1

3

2.20(2)

2.08(2)

0.004(3)

0.004(3)

-9(3) 44.5

4 Ni-S

Ni-N

1

4

2.20(2)

2.10(3)

0.002(3)

0.007(6)

-8(4) 45.4

5 Ni-S

Ni-N

Ni-Imi

1

2

1

2.21(3)

2.08(7)

2.09(8)

0.005(6)

0.003(13)

0.003(36)

-7(3) 44.4

6 Ni-S

Ni-N

Ni-Imi

1

1

2

2.22(7)

2.12(6)

2.08(1)

0.005(8)

0.005(36)

0.002(3)

-3(1) 29.6

7 Ni-S

Ni-N

Ni-Imi

1

3

1

2.20(7)

2.11(6)

2.08(6)

0.003(4)

0.008(21)

0.002(12)

-6(3) 44.8

8b Ni-S

Ni-N

Ni-Imi

1

2

2

2.20(9)

2.16(7)

2.08(1)

0.004(4)

0.005(22)

0.002(2)

-3(1) 28.5

9 Ni-S

Ni-N

Ni-Imi

1

1

3

2.20(4)

2.14(9)

2.09(1)

0.006(6)

0.003(8)

0.003(2)

-2(1) 28.8

10 Ni-S

Ni-N

Ni-Imi

2

1

2

2.21(3)

2.14(11)

2.07(2)

0.011(6)

0.003(9)

0.001(2)

-4(2) 35.2

11 Ni-S

Ni-N

Ni-Imi

1

3

2

2.21(2)

2.14(3)

2.09(1)

0.003(3)

0.018(9)

0.002(1)

-4(1) 30.3

12 Ni-S

Ni-N

Ni-Imi

2

2

2

2.20(6)

2.18(8)

2.09(1)

0.006(3)

0.006(18)

0.002(2)

-4(2) 31.8

aA-Bs denotes absorber and backscatterer interaction; N denotes coordination number; R is given in Å and

represents interatomic distances; σ2 given in Å

2, are the Debye-Waller factors (mean-square deviations in

interatomic distance); the threshold energy shifts, ∆E0 are given in eV. The values in parentheses are three

times the estimated standard deviations of the final digit listed obtained from the diagonal elements of the

covariance matrix. The F-factor or fit-error function is defined as √∑k6(χ(k)calcd –χ (k)exptl)

2/∑k

6χ(k)exptl

2; The

summation is over all data points included in the refinement. bThe 1S2N2His fit is shown in Figure 7.

25

XAS data analysis. XAS data reduction and analysis were performed using the EXAFSPAK suite

programs (http://ssrl.slac.stanford.edu/exafspak.html), employing a Gaussian pre-edge function

and a weighted polynomial spline with normalization correction to extract the EXAFS

oscillations, (k). The EXAFS was quantitatively analyzed by curve-fitting (k) directly as

described (28), using ab initio theoretical phase and amplitude functions calculated using FEFF,

version 8.4 (29). Phase-shifted fourier transforms were calculated by using theoretical phase

functions from the largest EXAFS component. Proximal atoms that could be fit as nitrogen or

oxygen were fit as nitrogen. To fit EXAFS arising from histidine ligands coordinated to Ni(II),

FEFF 8.25 was used to generate theoretical phase and amplitude parameters for full multiple

scattering pathways containing second and third coordination sphere C and N atoms, using a

model abstracted from the EXAFSPAK program mol-opt. All possible scattering pathways were

included during fitting analysis, and no tilt or wag was introduced into the histidine orientation.

The Debye-Waller factor of all of the multiple scattering pathways was varied at a fixed

difference from that of the ligating nitrogen. For clarity, only the first shell Ni-Nimidazole

scattering pathway is listed (Table 2-3).

Sample preparation and instrument conditions for ESI-MS.

(i) Direct Ni(II) titration of SlyD:

Protein samples were pre-treated with 10 mM EDTA and 2 mM TCEP for 24 hrs at 4oC in an

anaerobic glove box. The samples were then gel-filtered via PD-10 columns using 10 mM

ammonium acetate, pH 7.5. To minimize the sodium ion content in the sample the gel-filtered

samples were subjected to further buffer exchanges into the same buffer via nanosep centrifugal

devices MWCO 10000 (PALL Lifesciences). The metal solutions and the protein samples

prepared under anaerobic conditions were septum capped to minimize oxidation of the protein

prior to ESI-MS measurements. Direct Ni(II) titration results were obtained by injecting the

protein sample with a known amount of Ni(II) acetate solution and allowing the sample to

equilibrate for 3-4 min prior to acquiring the mass spectra. The mass spectra were acquired on a

AB/Sciex QStarXL mass spectrometer equipped with an ion spray source in the positive ion

mode. Ions were scanned from m/z 1200-3500 with accumulation of 1 s per spectrum with no

interscan time delay and averaged for 35 cycles under MCA mode. The instrument parameters

are as follows: ion source gas 50.0 psi; curtain Gas 45.0 psi; ion spray voltage 5500.0 V;

26

declustering potential 150.0 V; focusing potential 210 V; collision gas 3.0; MCP (detector)

2200.0 V. The nickel titration of SlyD1-146 was conducted in a similar manner under aerobic

conditions, and the ions were scanned in the m/z range of 650-2400 with an ion spray voltage set

to 5300 V and a declustering potential of 50.0 V. The spectra were deconvoluted using the

Bayesian protein reconstruction program over a mass range of 20000 – 22000 Da and 15000 –

17000 Da for SlyD WT and SlyD1-146 respectively. A step mass of 1 Da, signal/noise ratio of 10

and the minimum intensity detected set to 1 % was used during reconstruction of the data. The

abundance of apo and holo SlyD species were calculated from the signal intensities of the

reconstructed spectra assuming that all different metalloforms are ionized to the same extent.

Only the peaks corresponding to salt free SlyD were used for calculating the metallation states of

SlyD.

(ii) Titration of nickel-bound SlyD with competitors (EGTA and glycine)

Fractions of the apo-protein in 10 mM ammonium acetate, pH 7.5, were incubated with excess

nickel-acetate overnight at 4oC under anaerobic conditions. The protein sample was then gel-

filtered via PD10 column to remove excess metal. The collected holo-protein fractions were then

incubated with known amounts of the competitors, each prepared in chelexed MilliQ water and

adjusted to a pH of 7.5 by using ammonium hydroxide (metal-free grade). Samples were

incubated overnight at 4oC under an anaerobic environment prior to analysis via ESI-MS. The

mass spectra were collected under identical instrument parameters as mentioned above except

the ions were scanned in the m/z range of 1400-3500.

2.3 Results

Characterization of Ni(II) binding to SlyD. The electronic absorption spectrum of SlyD

incubated with Ni(II) revealed an absorption feature at 315 nm accompanied by an increase in

the absorption at 280 nm (Figure 2-2). These bands are similar to the sulfur-to-metal charge

transfer bands observed in the absorption spectra of small molecule nickel-thiolate complexes

(30, 31) as well as other proteins that bind Ni(II) via Cys residues (32-35). Titration of wild-type

SlyD with nickel produced an increase in the absorption at 315 nm that saturated upon the

addition of 4-5 equivalents of metal, with the protein remaining soluble in solution even after the

addition of a large excess of nickel ions (Figure 2-3). It was unclear whether nickel was binding

27

quantitatively to SlyD, so equilibrium dialysis experiments were conducted to obtain the

stoichiometry of nickel binding to the protein. These experiments demonstrated that the protein

coordinates 4.2 ± 0.2 Ni(II) ions/monomer (Table 2-4).

Figure 2-2. Electronic absorption spectra of wild-type SlyD and mutants. Upon incubation of SlyD (25

µM) with 7 molar equivalents of nickel an absorption band is observed at 315 nm due to LMCT from

sulfur to the Ni(II) centre. The spectra of the mutants are different from that of WT suggesting alterations

in the Ni(II) coordination environment as cysteines are replaced with alanines. Apo-WT (black solid

line), holo-WT (blue dashed line), holo-SlyD C167,168A mutant (red dashed line), holo-SlyD C184,185A

mutant (green dashed line), and holo-SlyD Δ193-196 (purple dashed line).

Figure 2-3. SlyD binds multiple nickel ions. Ni(II) titration curve obtained by incubating SlyD (25 µM)

in 20 mM HEPES, pH 7.5, 100 mM NaCl with increasing concentrations of Ni(II). Metal binding to the

protein was noted by the increasing intensity of the LMCT that appears at 315 nm (Figure 2-2). The

normalized change in absorbance is obtained by subtracting the holo-SlyD spectrum from the apo-SlyD

spectrum and dividing the resulting number by the concentration of the protein used for the titration.

0

0.01

0.02

0.03

0.04

0.05

240 280 320 360 400 440 480

No

rma

lize

d A

bso

rba

nce

(

M-1

)

Wavelength (nm)

0

0.003

0.006

0.009

0.012

0.015

0 4 8 12 16 20 24 28 32

No

rma

lize

d Δ

Ab

s 315

(µM

-1)

Equivalents of Ni(II)

28

Table 2-4. Stoichiometry of Ni(II) binding to SlyD.

Average Ni(II)

Stoichiometrya SD

b

Wild-type SlyD 4.2 0.2

SlyD C167,168A 2.6 0.2

SlyD C184,185A 2.8 0.4

SlyD Δ193-196 2.8 0.3

Triple mutant 1.9 0.4 aThe averages reported are from three independent trials of equilibrium dialysis.

bStandard deviation.

To determine whether nickel binding induces changes in the secondary structure of SlyD, CD

spectroscopy was used. Two minima are observed in the spectrum of the apo-protein at 215 nm

and 227 nm (Figure 2-4). The 215 nm absorption band signifies the existence of β-sheet

structure in the protein, which is expected because of the predominance of β-sheets in both the

PPIase and IF domains (21, 25), whereas the 227 nm band indicates a small amount of β-turn

type structure (23, 36). As Ni(II) is titrated into the protein solution the molar ellipticity

gradually decreases, in agreement with the trend that has been previously observed (23). The

changes in the CD signal, which reach a plateau upon addition of four molar equivalents of

Ni(II), are consistent with an increasing number of β-turns as SlyD chelates Ni(II) ions (23, 36).

In addition, a near UV-CD band emerges at 270 nm as SlyD is incubated with nickel, with

saturation reached at 2 equivalents of nickel. At this longer wavelength the likely chromophores

are aromatic residues that are extremely sensitive to even minor structural perturbations due to

events such as ligand binding, domain rearrangements, and protein-protein interactions (37).

Metal-protein complexes can also generate near UV-CD signals as a result of charge transfer

between the metal ion and aromatic residues in the proteins, as observed in several cases (38,

39). Given that the aromatic residues in SlyD (tyrosine and phenylalanines) are exclusively

located in the N-terminal PPIase and IF domains, these CD results suggest that the effects of

nickel binding to SlyD are not limited to the metal-binding (MB) domain.

To investigate the quaternary structure of SlyD and the effects of Ni(II) on its oligomeric state,

analytical gel filtration experiments were conducted (Figure 2-5). Based on the partition

coefficient an apparent molar mass (Mr,app) of 25 kDa was calculated for apo-SlyD. This mass is

slightly higher than the theoretical molecular weight of 21 kDa, a mass that was confirmed by

using ESI-MS (Table 2-2). However, elution profiles corresponding to higher than the expected

29

molecular masses have been observed for other histidine-rich proteins analyzed by size exclusion

chromatography (40, 41). Hence, it can be concluded that SlyD exists as a monomeric protein at

the concentrations (100 µM) used for analytical gel-filtration. Incubation of SlyD with 1 molar

and 10 molar equivalents of Ni(II) prior to gel filtration did not have a pronounced effect on the

oligomeric state of SlyD (Figure 2-5), resulting in only slightly higher Mr,app of 27 kDa and 28

kDa, respectively.

Figure 2-4. Ni(II)-induced 2° structural changes in SlyD. The intensity of the mean residue ellipticity in

the far-UV decreases compared to that in the spectrum of the apo-protein (blue diamonds) as the protein is

incubated with 1 equivalent of Ni(II) (green diamonds), 2 equivalents (purple diamonds), 4 equivalents

(black diamonds), and 8 equivalents (red diamonds), suggesting the formation of β-turn type structure.

-24

-20

-16

-12

-8

-4

0

4

205 225 245 265 285 305

ӨM

RE

x 1

02

(de

g c

m2

dm

ol-1

)

Wavelength (nm)

30

Figure 2-5. Quaternary structure of SlyD. Analytical gel-filtration of 100 uM SlyD without

Ni(II) (purple), with 1 molar equivalent of Ni(II) (green) and 8 equivalents of Ni(II) (blue). The

retention volumes correspond to a monomeric protein and the presence of metal does not affect

the quaternary structure of SlyD. All experiments were performed at 4oC using a 20 mM HEPES,

100 mM NaCl and 1 mM TCEP, pH 7.5, buffer containing the same concentrations of nickel as

the protein sample.

Analysis of Ni(II) binding to SlyD by ESI-MS. To characterize the nickel-binding activity of

SlyD in more detail, the protein was analyzed by ESI-MS as it was titrated with increasing

amounts of nickel. This is the first report of utilizing ESI-MS to directly monitor the formation

of nickel-protein complexes. The observed charge states and the corresponding deconvoluted

spectra for a representative Ni(II) titration as well as a summary of the different species present

at each titration point are shown in Figure 2-6. The data reveal the presence of several

metalloforms that differ in nickel stoichiometries at each titration point, indicating that SlyD has

multiple metal-binding sites with similar affinities for Ni(II). Furthermore, as in the case with

the cysteine-rich metallothioneins (42), the progressive detection of all of the possible metallated

species signifies that Ni(II) binding to SlyD occurs via a non-cooperative mechanism.

0

10

20

30

40

50

13 14 15 16 17 18 19

Ab

so

rba

nce

(A

U)

Retention Volume (mL)

31

Figure 2-6. Analysis of Ni(II) binding to SlyD via ESI-MS. (A) Mass spectra recorded as increasing

amounts of Ni(II) (0.0, 0.6 1.1, 2.2, 3.5 and 8 molar equivalents, top to bottom) were added to apo-SlyD

(34 µM). The associated deconvoluted spectra are shown on the right. (B) Graphical representation of

the Ni(II) titration in which several metalloforms are observed at each titration point, indicating a non-

cooperative metal-binding mechanism.

Mass reconstruction of +TOF MS: 0 MCA scans from SlyDJan2109_1.wiff Max. 2.5e4 cps.

2.09e4 2.10e4 2.11e4 2.12e4 2.13e4 2.14e4 2.15e4Mass, amu

0.0

1000.0

2000.0

3000.0

4000.0

5000.0

6000.0

7000.0

8000.0

9000.0

1.0e4

1.1e4

1.2e4

1.3e4

1.4e4

1.5e4

1.6e4

1.7e4

1.8e4

1.9e4

2.0e4

2.1e4

2.2e4

2.3e4

2.4e4

2.5e42.5e4

Intensi

ty, cps

20853.00

Mass reconstruction of +TOF MS: 25 MCA scans from Sample 2 (SlyD_jan3009) of SlyD_jan3009_04.wiff Max. 6677.4 cps.

2.09e4 2.10e4 2.11e4 2.12e4 2.13e4 2.14e4 2.15e4Mass, amu

0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

3200

3400

3600

3800

4000

4200

4400

4600

4800

5000

5200

5400

5600

5800

6000

6200

6400

6600

Intensi

ty, cps

20910.00

Mass reconstruction of +TOF MS: 0 MCA scans from SlyD_jan3009_07.wiff Max. 3513.1 cps.

2.09e4 2.10e4 2.11e4 2.12e4 2.13e4 2.14e4 2.15e4Mass, amu

0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

3200

3400

3513

Intensi

ty, cps

20910.00

Mass reconstruction of +TOF MS: 0 MCA scans from SlyD_jan3009_08.wiff Max. 3932.2 cps.

2.09e4 2.10e4 2.11e4 2.12e4 2.13e4 2.14e4 2.15e4Mass, amu

0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

3200

3400

3600

3800

3932

Intensi

ty, cps

21023.00

Mass reconstruction of +TOF MS: 0 MCA scans from SlyD_jan3009_09.wiff Max. 2983.8 cps.

2.09e4 2.10e4 2.11e4 2.12e4 2.13e4 2.14e4 2.15e4Mass, amu

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

2000

2100

2200

2300

2400

2500

2600

2700

2800

2900

2984

Intensi

ty, cps

21081.00

Mass reconstruction of +TOF MS: 25 MCA scans from Sample 2 (SlyD_jan3009) of SlyD_jan3009_10.wiff Max. 1911.8 cps.

2.09e4 2.10e4 2.11e4 2.12e4 2.13e4 2.14e4 2.15e4Mass, amu

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

Inten

sity, cp

s

21195.00

Mass (Da)20800 21100 21300 21500

0 1 2 3 4 5 6 7 8 9 +TOF MS: 0 MCA scans from SlyDJan2109_1.wiffa=3.56209262161544440e-004, t0=5.36370156367083840e+001

Max. 1027.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0

50

100

150

200

250

300

350

400

450

500

550

600

650

700

750

800

850

900

950

1000

1027

Intensit

y, coun

ts

+TOF MS: 25 MCA scans from Sample 2 (SlyD_jan3009) of SlyD_jan3009_04.wiffa=3.56201640895968570e-004, t0=5.33165838415952750e+001

Max. 491.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0

20

40

60

80

100

120

140

160

180

200

220

240

260

280

300

320

340

360

380

400

420

440

460

480

491

Intensit

y, coun

ts

+TOF MS: 0 MCA scans from SlyD_jan3009_07.wiffa=3.56201640895968570e-004, t0=5.33165838415952750e+001

Max. 291.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

170

180

190

200

210

220

230

240

250

260

270

280

290

Intensit

y, coun

ts

+TOF MS: 0 MCA scans from SlyD_jan3009_08.wiffa=3.56201640895968570e-004, t0=5.33165838415952750e+001

Max. 395.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0

20

40

60

80

100

120

140

160

180

200

220

240

260

280

300

320

340

360

380

395

Intensit

y, coun

ts

+TOF MS: 0 MCA scans from SlyD_jan3009_09.wiffa=3.56201640895968570e-004, t0=5.33165838415952750e+001

Max. 330.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

170

180

190

200

210

220

230

240

250

260

270

280

290

300

310

320

330

Intensit

y, coun

ts

+TOF MS: 25 MCA scans from Sample 2 (SlyD_jan3009) of SlyD_jan3009_10.wiffa=3.56201640895968570e-004, t0=5.33165838415952750e+001

Max. 250.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

170

180

190

200

210

220

230

240

250

Intensit

y, coun

ts

No

rmal

ized

Inte

nsi

ty

1200 1600 1800 2000 2200 2400

m/z (Da)

+16 +14 +12 +10 +9 +8 +7(A)

0.0

0.4

0.6

1.1

2.2

3.5

8.0

0

0.2

0.4

0.6

0.8

1

0

1

2

3

4

5

6

7

8

9

(B)

32

It has been established that the appearance of specific charge states is associated with the

conformational status of a protein (43). For metalloproteins, a change in the maximum and/or

distribution of charge states can be associated with alterations in the size or conformation of the

protein due to chemical changes such as metallation (44, 45). The m/z spectrum of apo-SlyD

provides evidence for the presence of two different conformations of SlyD in solution (Figure 2-

6). The fact that the relative peak intensities are higher for the smaller charge states suggests that

the majority of the protein adopts a more folded conformation. The small population of lower

m/z species observed in the spectrum of the apo-protein decreases in intensity upon Ni(II)

addition, which suggests that this population of SlyD adopts a more closed conformation upon

Ni(II) binding. Thus, the MS data provide evidence for a Ni(II)-induced folding event, which is

consistent with the Ni(II)-dependent structural changes observed via CD spectroscopy.

Upon analysis of the nickel titrations by MS a definitive end point could not be obtained before

the concentration of salt interfered with the quality of the spectral data. Thus, a different

approach was used to measure the stoichiometry of Ni(II) binding to SlyD. The protein was

incubated with a 10-fold excess of metal overnight and then gel-filtered to remove any loosely

bound Ni(II). The mass spectrum indicated the presence of a mixed population corresponding to

a range of two to seven Ni(II) ions coordinated to the protein, with the dominant peaks arising

from the four- and five-coordinated species (Figure 2-7). It is apparent from the MS data that

holo-SlyD is a mixture of several metalloforms that exist concurrently, which likely leads to the

average stoichiometry of 4.2 ± 0.2 obtained via equilibrium dialysis. This hypothesis is

corroborated by the weighted sum of the metal-bound peaks observed in the MS that yields an

average of 4.6 ± 0.1 nickel ions per SlyD.

To estimate the relative affinities of the multiple metal sites observed in the ESI-MS

experiments, holo-SlyD was prepared by incubating apo-protein with excess nickel followed by

gel filtration and then titration with increasing amounts of glycine prior to MS analysis. The

number of metals bound to SlyD remained unaffected even in the presence of 128 molar

equivalents of the competitor and the mass spectrum was almost identical to that of the protein in

the absence of competitor (data not shown). Therefore, it can be deduced that the affinities of the

detected metal sites are tighter than that of a Ni-(glycine)2 complex (Ka=105.74

) (46). A similar

experiment with EGTA, which is a competitor that forms a 1:1 complex with Ni(II) and has a

33

higher affinity for the metal (Ka = 1013.5

) (46), was also conducted. The SlyD metal-binding sites

were able to compete with EGTA (Figure 2-8), although to a lesser extent than with glycine

because an excess of EGTA could pull off almost all of the nickel from SlyD. Furthermore, the

mass spectra recorded for the EGTA competition experiment provide evidence for the formation

of a small population of ternary complex (approximately 1/6 of the intensity of the Ni-SlyD

peaks, data not shown). The presence of the protein-Ni(II)-EGTA complex implies that metal

release from SlyD does not solely rely on diffusion but that the chelator actively strips the metal

from the protein via formation of a transient ternary complex. This observation also suggests

that at least one of the metal sites is likely to be at the surface of the protein where it is accessible

to the chelator.

Figure 2-7. Stoichiometry of nickel binding to SlyD. Prior to analysis by ESI-MS, wild-type SlyD was

incubated with a 10-fold molar excess of nickel overnight at 4 oC under anaerobic conditions (Left). The

deconvoluted spectrum reveals 4-5 Ni(II) ions bound to SlyD as the predominant species present after gel

filtration (right).

Mass (Da)m/z (Da)

Rela

tive

In

ten

sity

(%

)

Rela

tive

In

ten

sity

(%

)

+TOF MS: 0 MCA scans from SlyD10feb09_gel fil DP150.wiffa=3.56218046144502220e-004, t0=5.35462600039142060e+001

Max. 218.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

55%

60%

65%

70%

75%

80%

85%

90%

95%

100%

Re

l. I

nt.

(%

)

2343.3

2635.6

2650.0

2358.2

2342.2

2633.6

2663.0

2370.1

2114.62265.5

Mass reconstruction of +TOF MS: 0 MCA scans from SlyD10feb09_gel fil DP150.wiff Max. 2451.0 cps.

2.090e4 2.095e4 2.100e4 2.105e4 2.110e4 2.115e4 2.120e4 2.125e4 2.130e4 2.135e4 2.140e4 2.145e4 2.150e4Mass, amu

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

55%

60%

65%

70%

75%

80%

85%

90%

95%

100%

Re

l. I

nt.

(%

)

21140.0

2

3

45

6

7

21000 21200 21400

+9

+8

+10

1600 2400 3200

34

Figure 2-8. SlyD metal sites can compete with EGTA. Titration of holo-SlyD (44.7 µM) with increasing

amounts of EGTA in 10 mM ammonium acetate, pH 7.5. The samples were allowed to equilibrate for 5

hrs at room temperature under anaerobic conditions. The relative intensities are representative of the

abundance of each species at a given titration point. Several metalloforms are observed at each titration

point implicating very similar affinities of the multiple metal sites present. Note that the x-axis ([EGTA]

values) is not linear.

X-ray absorption near edge spectroscopy. To obtain structural information about the metal

coordination environment in SlyD, WT protein loaded with 0.8 equivalents of Ni(II) was

analyzed by X-ray absorption spectroscopy (XAS). The absorption edge around 8341 eV

correlates well with the K-edge energy of previously analyzed Ni(II) complexes (Figure 2-9),

and the general shape of the curve for the Ni(II) edge reflects a mixture of sites with either a

square pyramidal or octahedral environment as the major component (47-50). The near-edge of

SlyD exhibits two transitions of weak intensity, at 8332 eV and 8335 eV, as well as a shoulder at

ca. 8342 eV. The small pre-edge feature at 8332 eV can be assigned to a 1s → 3d transition, and

the peak area of this particular transition usually provides an indication of the geometry of the

metal site. The 1s → 3d transition is formally dipole-forbidden but weakly quadrupole-allowed

00.02

0.050.1

0.20.4

0.60.8

15

0

0.1

0.2

0.3

0.4

0.5

0.6

01

23

4

5

6

7

Rela

tive Inte

nsi

ty

35

and thus the expected intensity for this transition is minimal for rigorously centrosymmetric

environments (i.e. square planar or Oh). However, for non-centrosymmetic complexes (5-

coordinate and Td) the transition gains in intensity from dipole-allowed character due to

increased p-d mixing (47, 51). The weak intensity in the SlyD spectrum suggests the presence of

centrocymmetric complexes. The six and four-coordinate environments can usually be

distinguished by a well-resolved peak at 8335 eV associated with the 1s → 4pz transition (with

shakedown contributions) of the latter complexes, and the weak peak at 8335 eV in the SlyD

spectrum suggests a predominance of octahedral complexes with a minor amount of square-

planar or distorted square-planar sites. The 1s → 4pz transition is also observed in spectra of five

coordinate square pyramidal complexes in the form of a shoulder-like feature at 8342 eV prior to

the main edge peak (47), as observed for SlyD. Furthermore, the height of the edge maximum is

indicative of an environment with a significant number of O/N ligands (52). Thus the XAS data

clearly suggest the presence of multiple nickel sites with different geometries and coordination

numbers. These data support the model in which nickel can bind to SlyD in several alternative

sites with similar affinity.

Figure 2-9. Ni K near-edge spectra. (A) The overlaid spectra have been normalized to the edge jump.

The samples include WT SlyD loaded with 0.8 equivalents of Ni(II) (solid black line), as well as the

mutants C167,168A (dashed red line), C184,185A (dashed green line), ∆193-196 (dashed purple line),

and Triple mutant (dashed blue line) loaded with the same amount of metal. (B) The overlaid spectra have

been normalized to the edge jump and are as follows: WT SlyD loaded with 0.8 equivalents of Ni(II)

(solid black line), square planar Ni(Mnt)2 complex (green dashed line), tetrahedral Ni(S)4 complex (red

dashed line) and octahedral Ni(en)3 complex (purple dashed line).

Energy (eV)

No

rmalized F

luo

rescen

ce

8330 8340 8350 8360

0.2

0.6

1.0

1.4

8330 8338

0.3

0

8330 8340 8350 8360

No

rmalized F

luo

rescen

ce

0.2

0.6

1.0

1.4

Energy (eV)

(A) (B)

36

Extended X-ray absorption fine structure (EXAFS). Given that the near-edge data as well as

the ESI-MS results indicate that even sub-stoichiometric amounts of nickel bound to SlyD are

distributed in several sites, obtaining an exact model of the nickel coordination environment

from the EXAFS data of Ni(II)-WT SlyD may not be possible. Nevertheless the data were fit in

an effort to provide qualitative information about the composition of the ligand sets. In order to

minimize the number of variables, the mean coordination number of each type of ligand was

systematically varied as integer values in each fit, although it is possible that the average number

of a certain type of ligand around the different nickel sites may not be a whole number.

Moreover, because the electronic absorption spectrum of wild-type SlyD indicates the presence

of at least one nickel-thiolate interaction, all fits included at least one cysteine ligand. Initial

attempts to fit the data with only a single shells of scatterers resulted in poor fits (selected fits are

listed in Table 2-3) and clearly did not model the intensity at 3-4 Å in the Fourier Transform

(FT) of the EXAFS data (Figure 2-10 and data not shown), which is due to atoms outside the

primary coordination sphere of the Ni(II). This set of peaks is suggestive of one or more

imidazole moieties chelating the Ni(II) ion (51, 53), and the distinct shoulder or “camel hump”

feature observed in the EXAFS also supports histidine ligation of metal (Figure 2-10) (54, 55). It

is also possible that some of the signal in the 4 Å region is caused by a Ni(II)-Ni(II) interaction,

which is feasible given that the ESI-MS demonstrates that the nickel can simultaneously fill

several metal sites on the protein, but this possibility was not included in any fits.

The best fit of the wild-type SlyD EXAFS data was achieved with a 5-ligand set including a

single cysteine at 2.20 Å, two imidazoles with Ni-N distances of 2.08 Å, as well as two

additional nitrogen or oxygen ligands at 2.16 Å (Table 2-3 and Figure 2-10). The goodness of

fit, as indicated by the F-factor, did not change significantly upon fitting to three imidazoles, but

visual inspection of the FT suggested that the histidine contribution was too high (data not

shown). Reducing the number of histidines to one resulted in an increase in the F-factor, as did

increasing the number of cysteines to two. Reasonable fits were also achieved with four and six

ligand sets, but in both cases the best fits also included two imidazoles. The Ni-S distance of ≈

2.2 Å that was observed in all of the best fits is in agreement with other 5-coordinate nickel

compounds (48), but it is also reasonable for four-coordinate complexes and has been observed

in several nickel proteins with octahedral ligation (48, 52). Similarly, the Ni-N distances of the

imidazole ligands (2.07-2.08 Å) are consistent with those of other nickel proteins (56, 57). The

37

length of the additional bonds (> 2.1 Å, either nitrogen or oxygen coordination) is unusually

long, and some correlation between parameters increases the uncertainty of this distance in the

fits. However, it is not atypical of nickel(II) complexes with cyclic ligands so it may indicate

some constraints in the SlyD ligands (58-60), or be a reflection of fitting a mixture of

coordination environments.

Figure 2-10. k3-weighted Ni(II) EXAFS data (left) and the Fourier transform of the data (right). Fourier

transforms were phase corrected for the first shell interaction. The raw data for wild-type SlyD loaded

with 0.8 equivalents of Ni(II) is in black and the best fit model (fit 8, 1S2N2His Table 2-3) is in green.

Metal chelation to the PPIase and IF domains. The CD spectra of the nickel titration of WT

SlyD indicated a metal-dependent structural perturbation in the N-terminal domains of the

protein. These domains contain several potential metal-binding residues (Figure 2-1), so it is

possible that the structural changes are independent of the metal-binding domain (25). To test

this possibility, a truncated variant of SlyD lacking the MB domain, SlyD1-146, was analyzed by

CD spectroscopy. The far-UV CD spectrum of SlyD1-146 exhibits significantly more intense

negative molar ellipticity, with a minimum at 208 nm and another transition at 222 nm (Figure 2-

11). These features may indicate a relatively higher degree of secondary structure, mainly alpha

helical, in comparison to the full-length protein. However, they could also be due to a decrease in

structural elements such as β-turns and a partial opening up of the structure (36, 61). Regardless,

it is clear that the metal-binding domain does have a significant effect on the overall structure of

the protein even in the absence of metal. Incubation of SlyD1-146 with increasing amounts of

χ(k

) x k

3

k (Å-1) R+∆ (Å)

Tran

sfor

m M

agni

tude

6

4

2

0

-2

-4

1.4

1.2

1.0

0.6

0.4

0.2

0.8

0 1 2 3 4 52 4 6 8 10 12

38

Ni(II) (up to 5 equivalents, Figure 2-11 and data not shown) did not change the CD spectrum at

any wavelength measured, indicating that the metal-binding C-terminal domain is necessary to

observe the structural changes in the N-terminal domains and suggesting that there is no direct

nickel binding to the isolated N-terminal domain.

Figure 2-11. Comparison of WT SlyD and SlyD1-146 CD spectra. The spectra are an average of 4 scans

collected on SlyD (60 µM in 10 mM ammonium acetate) incubated with and without Ni(II) overnight at

4oC. The SlyD1-146 variant that contains only the PPIase and the IF domains exhibits higher negative

ellipticity and the protein structure is unperturbed by addition of nickel. Apo-WT (black x), WT with 2

equivalents of Ni(II) (green diamonds), Apo-SlyD1-146 (red x), SlyD1-146 with 2 equivalents of Ni(II)

(purple diamonds).

To test directly whether the N-terminal domains of SlyD have an independent metal-binding

capacity, a Ni(II) titration of SlyD1-146 was conducted in a similar manner as for the WT protein

and analyzed by using ESI-MS. As observed for the full-length protein, the MS spectrum of

apo-SlyD1-146 revealed that there are two populations of conformers differing in overall charge

(data not shown). However, unlike the WT protein, the apo-SlyD1-146 spectrum was dominated

by lower m/z species suggesting a more open conformation of the protein (data not shown), in

agreement with the CD spectroscopy and confirming that the MB domain influences the overall

structure of the protein. Furthermore, unlike the full-length protein, the population distribution

did not change upon addition of nickel. Some nickel binding was observed in the MS but a

Ni(II)-bound protein peak was only detected upon addition of at least 2 molar equivalents of

Ni(II), and following incubation of the protein with 4 equivalents of Ni(II) at least ~30% of

-90

-70

-50

-30

-10

10

195 215 235 255 275 295

ӨM

RE

x 10

2(d

eg

cm

2d

mo

l-1)

Wavelength (nm)

39

SlyD1-146 still remained in the apo form (data not shown). These results suggested that the Ni(II)-

protein complexes detected via MS were a result of non-specific interactions between the metal

and the protein. To test this hypothesis, the protein sample was incubated with 8 molar

equivalents of Ni(II) overnight, then analyzed by MS following gel filtration. Only a peak

corresponding to apo-SlyD1-146 was detected in the gel-filtered samples, confirming that the

metal-protein signals observed during the direct Ni(II) titration experiment arise from non-

specific, weak interactions with the N-terminal domains of SlyD.

Characterization of SlyD mutants. To gain more insight into the involvement of the cysteine

residues in metal binding, as suggested by the spectroscopic experiments, site-directed

mutagenesis was employed. The six cysteines in SlyD are distributed throughout the metal-

binding domain in pairs (Figure 2-1), so three mutants were generated in which each pair of Cys

residues was either mutated to alanine (SlyD C167,168A and SlyD C184,185A) or removed

(C193 and C195 are absent in the SlyD ∆193-196 truncation) (Table 2-1). Electronic absorption

spectroscopy revealed that all three mutants bind Ni(II) (Figure 2-2) and titration experiments

produced binding curves similar to that of the WT SlyD (data not shown). However, the

electronic absorption signals of the mutants saturate with less nickel than wild-type SlyD, and the

spectra from the mutants are altered. Less intense, slightly red-shifted LMCT bands are

observed for the C167,168A and C184,185A mutants, whereas an increase in the intensity of the

LMCT band is observed for the Δ193-196 protein (Figure 2-2). This preliminary analysis of the

mutants implies changes in ligand environment and/or coordination geometry of the Ni(II) sites.

The CD spectra of the apo- and holo-mutant proteins were similar to those of wild-type SlyD,

indicating that mutations did not perturb the overall secondary structure of the protein (data not

shown). Equilibrium dialysis experiments demonstrated that the nickel-binding capacity of all

the mutants was reduced (Table 2-4), supporting a model in which all six Cys residues are

involved in metal binding.

To determine if cysteines are necessary for any metal binding to SlyD, a fourth mutant was

generated that lacked all six Cys residues by introducing all three of the sets of mutations

mentioned above into slyD (referred to as the Triple mutant). As expected, no change was

observed in the electronic absorption spectrum of the Triple mutant incubated with increasing

amounts of Ni(II) (data not shown). However, the Triple mutant protein binds 2 Ni(II) per

40

monomer in equilibrium dialysis experiments (Table 2-4). Given that the electronic absorption

spectroscopy was performed with similar protein concentrations as the equilibrium dialysis

experiments, the lack of a change in the absorption spectrum clearly denotes that the Ni(II) ions

bind to donor atoms for which the LMCT is shifted well into the blue region of the spectrum.

That the Triple mutant was chelating Ni(II) as a monomeric species was verified by analytical

gel-filtration experiments (data not shown). MS analysis of the apo-Triple mutant indicated the

presence of two different conformers with similar populations of m/z species as observed for the

wild-type SlyD (data not shown). A Ni(II) titration revealed that the Triple mutant also adapts a

more folded conformation when bound to nickel and that metal-binding occurs via a non-

cooperative manner comparable to that of wild-type SlyD. However, in agreement with the

equilibrium dialysis data the MS indicated reduced nickel-binding capacity for the Triple mutant.

To gain a better understanding of the importance of the thiolate ligands for metal binding, Ni(II)

titrations of WT SlyD were performed in the presence of equimolar Triple mutant, followed by

ESI-MS analysis. Upon addition of 18 M Ni(II) (≈ 36% of the total protein concentration),

peaks corresponding to 1 and 2 Ni(II) ions bound to WT appeared whereas only a single ion

bound was apparent for the Triple mutant (Figure 2-12), and less of WT was apo-protein in this

competitive environment. In addition, the peak for apo-protein disappear at lower Ni(II)

concentrations for the WT protein than the mutant. No clear peaks could be distinguished in the

mass spectrum past the 246 M Ni(II) titration point for the WT SlyD due to interferences

caused by the increased salt concentration. Nevertheless, wild-type SlyD clearly binds more

nickel than the mutant, indicating that thiol moieties impart better metal sequestering ability to

SlyD but that the protein lacking cysteines is still competent at nickel binding.

XAS of SlyD mutants. An overlay of the near-edge X-ray absorption spectra generated from

WT SlyD and the mutant proteins provides clear evidence that the mutations alter the SlyD

Ni(II)-binding sites (Figure 2-9). Although the overall shapes of the edge spectra are similar to

that of WT, all the mutants exhibit an edge peak that is shifted to a higher energy. Furthermore,

an increase in the intensity of the normalized fluorescence relative to the WT is also evident.

These changes correlate well with the trend noted for Ni(II) complexes as their S-donor ligands

are substituted with N,O-donor ligands (47). Very much like WT SlyD, the SlyD C167,168A

and the SlyD Δ193-196 mutants‟ edge spectra exhibit features for a mixture of octahedral (8332

41

eV), square pyramidal geometries (shoulder at 8342 eV) with some minor contributions from 4-

coordinate planar Ni(II) sites (8335 eV). The SlyD C184,185A mutant on the other hand appears

to contain Ni(II) sites only in centrosymmetric geometries: octahedral and square planar.

However, only a small fraction of this mutant forms the square planar nickel sites since the

intensity of the 1s → 4pz transition is very small. The spectrum of the Triple mutant produced

the highest normalized intensity with a Ni(II) coordination environment that is expected to solely

contain oxygen and nitrogen donor ligands. The near-edge spectrum for this mutant contains a

low intensity 1s → 3d transition at 8332 eV as well as a shoulder at 8342 eV corresponding to 1s

→ 4pz transitions, which are suggestive of a mixture of metal sites with six-coordinate and

square pyramidal geometry, and lacks the peak at 8335 eV assigned to square planar complexes.

Therefore, the XAS data suggest the necessity of thiolate ligands to form low coordination

number, square planar Ni(II) centers in SlyD. The FT of the EXAFS of the SlyD mutants loaded

with nickel are similar to that of wild-type protein (Figure 2-13), but the FT magnitude in the 3-4

Å region is larger for the Triple mutant compared to that of WT SlyD, suggesting that upon

deletion of all Cys ligands the protein compensates with a larger number of histidine ligands in

the coordination sphere.

Figure 2-12. Competition for Ni(II) binding by WT and Triple mutant SlyD. An equimolar solution (25

µM each) of WT SlyD (solid bars) and Triple mutant (striped bars) was titrated with nickel in 10 mM

ammonium acetate, pH 7.5. The relative intensities of different metalloforms at each titration point

correlates to the relative abundance and is represented as follows: Apo (blue); 1, 2, 3, 4, 5, 6 and 7 Ni(II)

ions bound: green, red, black, purple, yellow, grey and light blue bars, respectively.

0

0.2

0.4

0.6

0.8

1

Rela

tive I

nte

nsi

ty

Ni(II) µM

0 18 37 78 111 169 246

42

Figure 2-13. Overlay of WT EXAFS data with SlyD mutants. The WT data are represented as black

solid lines in each spectrum. The k3-weighted Ni(II) EXAFS data are shown in the left and the respective

Fourier transform of the data is on the right. From top to bottom: SlyD C167168A, SlyD C184,185A,

SlyD Δ193-196, and Triple mutant

χ(k

) x

k3

Tra

ns

form

Ma

gn

itu

de

k (Ǻ-1) R+∆ (Ǻ)

χ(k

) x k

3

k (Å-1) R+∆ (Å)

Tra

nsfo

rm M

ag

nitu

de

43

2.4 Discussion and Conclusions

SlyD is an accessory protein required for achieving optimal levels of [NiFe]-hydrogenase

activity in Escherichia coli (12, 22). Together with auxiliary proteins HypA and HypB, it is

thought to act as a nickel chaperone that facilitates the Ni(II) insertion step in the maturation

pathway of this enzyme (11, 12, 22), and it may also serve as a nickel storage protein in this

organism (12). Of the three functional domains in SlyD, the PPIase domain, the molecular

chaperone domain, and the C-terminal metal-binding domain, the latter two are essential for the

role of SlyD in the enzyme biosynthetic pathway (19, 22). Although the molecular chaperone

activity of SlyD has been investigated previously (18, 62), a detailed characterization of the

nickel-binding properties has not been reported. Therefore, to obtain a better understanding of

SlyD as a Ni(II)-binding protein and its function as a metallochaperone, nickel coordination to

the protein was examined in detail by using a range of spectroscopic and spectrometric methods.

Consistent with the results from previous studies, we find that SlyD binds multiple metal ions but

with a higher average nickel stoichiometry than previously noted (14, 23). The discrepancy in

the data is likely due to the oxidation of the cysteine residues under atmospheric conditions, a

hypothesis that is supported by the lower nickel-binding stoichiometries observed with the

mutants in which cysteines were replaced with alanines (Table 2-4). Under physiological

conditions SlyD should retain its full binding capacity given that the protein is located in the

reducing cytosolic compartment and functions under anaerobic growth conditions, at least for its

role in hydrogenase biosynthesis (8). All of the experimental evidence support the importance of

the thiolate ligands for nickel binding. Not only do they impart higher nickel-binding capacity to

SlyD, the Cys ligands enable the formation of metal sites with increased affinity for Ni(II).

Although the amount of free nickel in the E. coli cytoplasm has not been measured, it is likely

that it is a very competitive environment with the nickel ion distribution tightly controlled (63-

65), so the cysteine ligands would allow SlyD to sequester the potentially toxic nickel ions away

from any weaker adventitious sites.

Ni(II) ions fill the binding sites on SlyD in a progressive and overlapping fashion without

altering the oligomeric state of the protein, indicating that metal binding is not cooperative.

Although we anticipated preferential nickel binding to at least one higher-affinity site, the

detailed electrospray mass spectrometry experiments show that SlyD exhibits multiple metal

44

sites of similar affinity, and the X-ray absorption data reveal that even sub-stoichiometric

amounts of nickel produce a mixture of nickel coordination environments. Up to seven nickel

ions remain bound to full-length SlyD following gel filtration chromatography and competition

with glycine, indicating that all of the seven sites have at least low micromolar affinity for Ni(II)

(KD < 1.8 μM). The observation of non-specific protein-metal complexes at high metal

concentrations is an inherent issue of the ionization process utilized during positive mode metal-

protein complex detection. However, the fact that the WT SlyD is able to chelate seven nickel

ions even in the presence of glycine or the Triple mutant clearly indicates that all seven nickel

ions are specifically bound by the protein. Furthermore, this is indicative that all seven metal

sites may be utilized by the protein to bind nickel ions competitively in the bacterial cytosol.

Although the MS data do not allow us to distinguish between nickel binding initially to one site

and then redistribution to other sites versus concurrent filling of multiple sites, it does reveal the

extreme plasticity of SlyD for metal binding and its ability to load a substantial amount of nickel

ions. A similar metal coordination pattern has been observed for human metallothionein (MT-1),

which exists in a dynamic mixture of metalloforms that is dependent on metal concentration in

solution (66-68).

That we were unable to achieve complete saturation via equilibrium dialysis at the maximal

stoichiometry of 7 nickel ions bound that was observed in the MS suggests that not all of the

metal sites on SlyD are equally accessible. A possible explanation for this result can be derived

from a model in which the metal-binding capacity of SlyD is governed by the sequence of Ni(II)

binding to the seven metal sites (Figure 2-14). Given that all sites seem to be of similar affinity,

binding to any of the sites would be equally probable initially. However, Ni(II) binding to a

particular site (e.g. sites 6 & 7 instead of sites 1 & 2) may subsequently hinder the binding of

Ni(II) to other sites (i.e. 1-4) due to structural constraints imposed upon metal chelation, whereas

binding to a different sets of site (i.e. sites 1 & 2) allows the protein to chelate the maximum

number of nickel ions. The changes in conformation brought about by metallation along „dead-

end‟ pathways would be observed by very slow binding rates where complete metallation might

not be realised within the hours of a typical experiment (68). Kinetic experiments will be

required to test this hypothesis.

45

Figure 2-14. Model for Ni(II) binding to SlyD. The order of nickel binding to individual sites

determines the maximum stoichiometry of the protein. For example, binding to sites 6 and 7

hinders the formation of sites 1-4 or inhibits access to these sites whereas binding to sites 1 and 2

allows the protein to coordinate the metal to its full capacity, thereby leading to an equilibrium of

several metalloforms.

Competition experiments with EGTA provide additional evidence for the similar affinity of the

metal sites because the selective depletion of individual metalloforms (resulting in populations

with only 4 Ni(II) ions bound to SlyD, for example) is not observed at any concentration of the

competitor. These experiments also support the conclusion that the protein is able to chelate

nickel in a competing cellular environment, which would be essential for its function as a source

of nickel for hydrogenase biosynthesis even under nickel-limiting conditions. The equilibrium

dissociation constant of nickel binding to SlyD has an upper limit of low micromolar, placing it

in a range that is similar to or tighter than that of nickel-binding proteins from other organisms

such as Hpn, Hpn-like or HspA from Helicobacter pylori (40, 41, 69), UreE (55, 70), and

histidine-rich HypB homologs (71, 72). This latter comparison, coupled with the fact that SlyD

forms a complex with E. coli HypB, supports the hypothesis that one of the purposes of SlyD is

to fill the role occupied by histidine-rich versions of HypB in other systems.

The CD spectroscopy, together with charge state distributions detected during the mass

spectrometry experiments, are indicative of a protein capable of metal binding with small but

distinctive changes in protein structure. The CD suggests that the changes induced by nickel

binding are dominated by the formation of β-turn type structures that are localized to the C-

terminal domain of the protein. A similar metal-induced structure formation has been previously

observed for other proteins (67, 73, 74). An analysis of a collection of proteins indicates that β-

1

2 3

4

56

7

Ni(II) Ni(II)

46

turns predominantly consist of hydrophilic amino acid residues and are concentrated at the

protein surface (36). Thus it is likely that the metal-binding domain is located at the surface of

the protein, an optimal position for facile nickel binding, and a position supported by the

unstructured nature of this region in apo-SlyD (21, 25). Furthermore, in such a location the

nickel ions could be tightly bound but not completely buried, allowing for direct transfer of

nickel ions between SlyD and other proteins via ligand exchange reactions that would not entail a

substantial conformational rearrangement in SlyD.

Considering the properties of SlyD, including a chaperone domain that enables protein-protein

complex formation and an accessible metal-binding domain at the surface, it is evident that SlyD

could participate in hydrogenase biosynthesis in one or more capacities (11, 12, 22). The fact

that it can bind many nickel ions suggests that SlyD functions as a store of nickel, which could

be made available specifically to the hydrogenase pathway through complex formation with

other accessory proteins such as HypB. HypB has an N-terminal metal-binding site with sub-

picomolar affinity for nickel (34), so from a thermodynamic perspective it is feasible that it could

abstract nickel from SlyD via a ternary complex such as that observed in the competition

experiments with EGTA. However, HypB is not the final destination of the nickel ion in this

pathway and in vitro experiments demonstrate that the MB domain of SlyD increases the release

of HypB-bound nickel to a third chelator (22), so an alternative possibility is that SlyD helps to

make this nickel available. The route of the nickel ion on its way to the hydrogenase precursor

protein and whether other components of the nickel insertion complex participate in nickel

delivery, are issues that remain to be defined by future experiments.

Unlike the full-length protein, any adventitious nickel ions bound to SlyD1-146 were completely

stripped away by gel filtration chromatography. The conclusion that the isolated PPIase domain

does not chelate Ni(II) is supported by the CD experiments demonstrating that its structure is

unaffected by the presence of Ni(II) in solution. Thus the MB domain is required for tight nickel

binding to SlyD. Although the metals bind to the C-terminus of SlyD, the N-terminal domains

are also perturbed by this activity, as is evident from the appearance of the near-UV CD band

upon nickel titration of the full-length protein. This perturbation could either result from metal

binding to one or more sites that bridge the MB and the N-terminal domains, or from nickel

binding to the MB domain producing an allosteric effect that propagates into the N-terminus.

47

Experiments with the isolated MB domain designed to resolve this issue have been precluded to

date due to excessive degradation. These conformational changes observed upon nickel binding

may cause the nickel-induced inhibition of the PPIase activity that has previously been reported

(23), an effect not observed in the absence of the metal-binding domain. This connection is in

agreement with a recent NMR study that revealed that the addition of nickel perturbs residues

around the catalytic pocket of the PPIase domain (25).

The fact that nickel binding to SlyD functions as a reversible switch that regulates the PPIase

provides a link between the two activities of the protein. However, as with many PPIase

enzymes (17), it is not known whether this activity of SlyD has a cellular function, so the

physiological relevance of this connection is unclear. The PPIase activity would only be blocked

under anaerobic growth conditions because nickel import by E. coli is activated in the absence of

oxygen (75), the environmental conditions that also necessitate hydrogenase activity (8). The

PPIase activity of SlyD is not critical for hydrogenase biosynthesis (19), but it has been proposed

that it may be detrimental (25), and nickel binding could be a mechanism of preventing

undesired isomerizations of prolines in the other hydrogenase accessory proteins or the enzyme

itself. A similar metal-dependent switch mechanism has been observed for another bacterial

heat-shock chaperone protein Hsp33 (76). Hsp33 normally exists in a deactivated form where

Cys residues in the protein are coordinated to a zinc ion. When the bacteria undergo oxidative

stress the zinc ion is displaced and the protein exhibits folding chaperone activity in its apo-form

by forming an oxidized dimer (76, 77). Whether SlyD is modulated by such a metal-dependent

switch, in which it functions as a general protein folding enzyme under aerobic conditions and

switches to the more specific task of a Ni(II) chaperone for hydrogenase biosynthesis under

anaerobic expression, has yet to be determined.

It is not uncommon for a chaperone protein to perform cellular functions in addition to its protein

folding activity and these extra functions are usually integrated into the chaperone in the form of

an extra domain. For instance, the heat shock protein A (HspA) in H. pylori contains a metal-

binding domain at the C-terminus that is beneficial for Ni(II) resistance in this organism (40).

Calreticulin is a chaperone involved in glycoprotein folding that also participates in the

regulation of intracellular Ca(II) homeostasis (78). In the case of the aforementioned Hsp33,

incorporation of an oxygen-sensitive metal-binding domain at the C-terminus allows this protein

48

to function as a chaperone in response to oxidative stress (77). Augmentation of a constitutively-

expressed chaperone such as SlyD (15) with a metal-binding domain would be advantageous

because a readily available sink for Ni(II) would be present when the bacteria switch to

anaerobic metabolism and start importing nickel. In this manner, the cell would be protected

from the toxic properties of this element while ensuring that the nickel is directed to where it is

needed in the hydrogenase biosynthetic pathway.

SlyD homologs have been annotated in a large number of prokaryotic genomes (79), and

although the C-terminal domains are not well-conserved, many of the homologs have significant

extensions rich in metal-binding residues. Thus SlyD may be a key player in prokaryotic nickel

homeostasis. For example, both biochemical and genetic experiments have implicated SlyD in

the biosynthetic pathway for urease in the human pathogen H. pylori (80, 81). In view of the role

of SlyD in nickel homeostasis in E. coli, the non-cooperative, adaptable metal-binding nature of

SlyD is analogous to metallothioneins that are proposed to be involved in the homeostasis of zinc

levels in the human brain (82, 83). Although SlyD has been characterized in the context of a

nickel-binding protein there are several questions that remain to be resolved. With its high

content of potential metal-binding residues can the protein function as a general metal detoxifier

similar to metallothioneins? What is the metal-specificity of SlyD and how does this specificity

correlate with optimal biogenesis of the hydrogenase enzyme? Is it possible for SlyD to function

as a metallochaperone for other enzymes? Answers to these issues are presented in Chapter 3.

49

References

1. Mulrooney, S. B., and Hausinger, R. P. (2003) Nickel uptake and utilization by

microorganisms, FEMS Microbiol. Rev. 27, 239-261.

2. Ragsdale, S. W. (2009) Nickel-based enzyme systems, J. Biol. Chem. 284, 18571-18575.

3. Li, Y., and Zamble, D. B. (2009) Nickel Homeostasis and Nickel Regulation: An

Overview, Chem Rev.

4. Barceloux, D. G. (1999) Nickel, Clin. Toxicol. 37, 239-258.

5. Muyssen, B. T. A., Brix, K. V., DeForest, D. K., and Janssen, C. R. (2004) Nickel

essentiality and homeostasis in aquatic organisms, Environ. Rev. 12, 113-131.

6. O'Halloran, T. V., and Culotta, V. C. (2000) Metallochaperones, an intracellular shuttle

service for metal ions, J. Biol. Chem. 275, 25057-25060.

7. Rosenzweig, A. C. (2002) Metallochaperones: Bind and deliver, Chem.Biol. 9, 673-677.

8. Böck, A., King, P. W., Blokesch, M., and Posewitz, M. C. (2006) Maturation of

hydrogenases, Advances in Microbial Physiology 51 (Suppl.), 1-71.

9. Forzi, L., and Sawers, R. G. (2007) Maturation of [NiFe]-hydrogenases in Escherichia

coli, Biometals 20, 565-578.

10. Vignais, P. M., and Billoud, B. (2007) Occurrence, classification, and biological function

of hydrogenases: an overview, Chem. Rev. 107, 4206-4272.

11. Leach, M. R., and Zamble, D. B. (2007) Metallocenter assembly of the hydrogenase

enzymes, Curr. Opin. Chem. Biol. 11, 159-165.

12. Zhang, J. W., Butland, G., Greenblatt, J. F., Emili, A., and Zamble, D. B. (2005) A role

for SlyD in the Escherichia coli hydrogenase biosynthetic pathway, J. Biol. Chem. 280,

4360-4366.

13. Roof, W. D., and Young, R. (1993) Phi X174 E complements lambda S and R

dysfunction for host cell lysis, J. Bacteriol. 175, 3909-3912.

14. Wulfing, C., Lombardero, J., and Pluckthun, A. (1994) An Escherichia coli protein

consisting of a domain homologous to FK506-binding proteins (Fkbp) and a new metal-

binding motif, J. Biol. Chem. 269, 2895-2901.

15. Roof, W., Horne, S., Young, K., and Young, R. (1994) slyD, a host gene required for phi

X174 lysis, is related to the FK506- binding protein family of peptidyl-prolyl cis-trans-

isomerases, J. Biol. Chem. 269, 2902-2910.

16. Kang, C. B., Hong, Y., Dhe-Paganon, S., and Yoon, H. S. (2008) FKBP family proteins:

immunophilins with versatile biological functions, Neurosignals 16, 318-325.

17. Gothel, S. F., and Marahiel, M. A. (1999) Peptidyl-prolyl cis-trans isomerases, a

superfamily of ubiquitous folding catalysts, Cell. Mol. Life Sci. 55, 423-436.

18. Scholz, C., Eckert, B., Hagn, F., Schaarschmidt, P., Balbach, J., and Schmid, F. X. (2006)

SlyD proteins from different species exhibit high prolyl isomerase and chaperone

activities, Biochemistry 45, 20-33.

19. Zhang, J. W., Leach, M. R., and Zamble, D. B. (2007) The peptidyl-prolyl isomerase

activity of SlyD is not required for maturation of Escherichia coli hydrogenase, J.

Bacteriol. 189, 7942-7944.

20. Furutani, M., Iida, T., Yamano, S., Kamino, K., and Maruyama, T. (1998) Biochemical

and genetic characterization of an FK506-sensitive peptidyl prolyl cis-trans isomerase

from a thermophilic archaeon, Methanococcus thermolithotrophicus, J. Bacteriol. 180,

388-394.

50

21. Weininger, U., Haupt, C., Schweimer, K., Graubner, W., Kovermann, M., Bruser, T.,

Scholz, C., Schaarschmidt, P., Zoldak, G., Schmid, F. X., and Balbach, J. (2009) NMR

solution structure of SlyD from Escherichia coli: spatial separation of prolyl isomerase

and chaperone function, J. Mol. Biol. 387, 295-305.

22. Leach, M. R., Zhang, J. W., and Zamble, D. B. (2007) The role of complex formation

between the Escherichia coli hydrogenase accessory factors HypB and SlyD, J. Biol.

Chem. 282, 16177-16186.

23. Hottenrott, S., Schumann, T., Pluckthun, A., Fischer, G., and Rahfeld, J. U. (1997) The

Escherichia coli SlyD is a metal ion-regulated peptidyl-prolyl cis/trans-isomerase, J. Biol.

Chem. 272, 15697-15701.

24. Kuchar, J., and Hausinger, R. P. (2004) Biosynthesis of Metal Sites, Chem. Rev. 104,

509-526.

25. Martino, L., He, Y., Hands-Taylor, K. L. D., Valentine, E. R., Kelly, G., Giancoloa, C.,

and Conte, M. R. (2009) The interaction of the Escherichia coli protein SlyD with nickel

ions illuminates the mechanism of regulation of its peptidyl-prolyl isomerase activity,

FEBS J. 276, 4529-4544.

26. Suzuki, R., Nagata, K., Yumoto, F., Kawakami, M., Nemoto, N., Furutani, M., Adachi,

K., Maruyama, T., and Tanokura, M. (2003) Three-dimensional solution structure of an

archaeal FKBP with a dual function of peptidyl prolyl cis-trans isomerase and chaperone-

like activities, J. Mol. Biol. 328, 1149-1160.

27. Atanassova, A., Lam, R., and Zamble, D. B. (2004) A high-performance liquid

chromatography method for determining transition metal content in proteins, Anal.

Biochem. 335, 103-111.

28. George, G. N., Garrett, R. M., Prince, R. C., and Rajagopalan, K. V. (1996) The

molybdenum site of sulfite oxidase: a comparison of wild-type and the cystein 207 to

serine mutant using x-ray absorption spectroscopy, J. Am. Chem. Soc. 118, 8588-8592.

29. Rehr, J. J., Mustre de Leon, J., Zabinsky, S. I., and Albers, R. C. (1991) Theoretical X-

ray absorption fine structure standards, J. Am. Chem. Soc. 113, 5135-5140.

30. Ragunathan, K. G., and Bharadwaj, P. K. (1992) Nickel(Ii) Complexes with Tripodal

Ligands - Synthesis, X-Ray Structural and Spectroscopic Studies, J. Chem. Soc. Dalton

Trans., 2417-2422.

31. Rosenfield, S. G., Berends, H. P., Gelmini, L., Stephan, D. W., and Mascharak, P. K.

(1987) New Octahedral Thiolato Complexes of Divalent Nickel - Syntheses, Structures,

and Properties of (Et4n)[Ni(Sc5h4n)3] and (Ph4p)[Ni(Sc4h3n2)3].Ch3cn, Inorg. Chem.

26, 2792-2797.

32. Chivers, P. T., and Sauer, R. T. (2002) NikR repressor: High-affinity nickel binding to

the C-terminal domain regulates binding to operator DNA, Chem. Biol. 9, 1141-1148.

33. Iwig, J. S., Leitch, S., Herbst, R. W., Maroney, M. J., and Chivers, P. T. (2008) Ni(II) and

Co(II) sensing by Escherichia coli RcnR, J. Am. Chem. Soc. 130, 7592-7606.

34. Leach, M. R., Sandal, S., Sun, H. W., and Zamble, D. B. (2005) Metal binding activity of

the Escherichia coli hydrogenase maturation factor HypB, Biochemistry 44, 12229-

12238.

35. Wang, S. C., Dias, A. V., Bloom, S. L., and Zamble, D. B. (2004) Selectivity of metal

binding and metal-induced stability of Escherichia coli NikR, Biochemistry 43, 10018-

10028.

36. Perczel, A., and Hollosi, M. (1996) Turns, in Circular Dichroism and the Conformational

Analysis of Biomolecules (Fasman, G. D., Ed.), pp 285-380, Plenum Press, New York.

51

37. Woody, R. W., and Dunker, A. K. (1996) Aromatic and Cystine Side-Chain Circular

Dichroism in Proteins, in Circular Dichroism and the Conformational Analysis of

Biomolecules (Fasman, G. D., Ed.), pp 109-158, Plenum Press, New York.

38. Beltramini, M., Bubacco, L., Salvato, B., Casella, L., Gullotti, M., and Garofani, S.

(1992) The aromatic circular dichroism spectrum as a probe for conformational changes

in the active site environment of hemocyanins, Biochim. Biophys. Acta. 1120, 24-32.

39. Vasak, M., Kagi, J. H., Holmquist, B., and Vallee, B. L. (1981) Spectral studies of cobalt

(II)- and Nickel (II)-metallothionein, Biochemistry 20, 6659-6664.

40. Cun, S., Li, H., Ge, R., Lin, M. C., and Sun, H. (2008) A histidine-rich and cysteine-rich

metal-binding domain at the C terminus of heat shock protein A from Helicobacter

pylori: implication for nickel homeostasis and bismuth susceptibility, J. Biol. Chem. 283,

15142-15151.

41. Ge, R., Watt, R. M., Sun, X., Tanner, J. A., He, Q.-Y., Huang, J.-D., and Sun, H. (2006)

Expression and characterization of a histidine-rich protein, Hpn: potential for Ni2+

storage

in Helicobacter pylori, Biochem. J. 393, 285-293.

42. Sutherland, D. E., and Stillman, M. J. (2008) Noncooperative cadmium(II) binding to

human metallothionein 1a, Biochem. Biophys. Res. Comm. 372, 840-844.

43. Kaltashov, I. A., and Abzalimov, R. R. (2008) Do ionic charges in ESI MS provide useful

information on macromolecular structure?, J. Am. Soc. Mass. Spectrom. 19, 1239-1246.

44. Rigby Duncan, K. E., Kirby, C. W., and Stillman, M. J. (2008) Metal exchange in

metallothioneins: a novel structurally significant Cd(5) species in the alpha domain of

human metallothionein 1a, FEBS J. 275, 2227-2239.

45. Pluym, M., Vermeiren, C. L., Mack, J., Heinrichs, D. E., and Stillman, M. J. (2007)

Heme binding properties of Staphylococcus aureus IsdE, Biochemistry 46, 12777-12787.

46. Martell, A. E., and Smith, R. M. (2003) NIST Standard Reference Database 46, version

7.0, National Institute of Standards and Technology, Rockville, MD.

47. Colpas, G. J., Maroney, M. J., Bagyinka, C., Kumar, M., Willis, W. S., Suib, S. L.,

Baidya, N., and Mascharak, P. K. (1991) X-Ray Spectroscopic Studies of Nickel-

Complexes, with Application to the Structure of Nickel Sites in Hydrogenases, Inorg.

Chem. 30, 920-928.

48. Gu, W. W., Jacquamet, L., Patil, D. S., Wang, H. X., Evans, D. J., Smith, M. C., Millar,

M., Koch, S., Eichhorn, D. M., Latimer, M., and Cramer, S. P. (2003) Refinement of the

nickel site structure in Desulfovibrio gigas hydrogenase using range-extended EXAFS

spectroscopy, J. Inorg. Biochem. 93, 41-51.

49. Musgrave, K. B., Laplaza, C. E., Holm, R. H., Hedman, B., and Hodgson, K. O. (2002)

Structural characterization of metallopeptides designed as scaffolds for the stabilization

of nickel(II)-Fe(4)S(4) bridged assemblies by X-ray absorption spectroscopy, J. Am.

Chem. Soc. 124, 3083-3092.

50. Tan, G. O., Ensign, S. A., Ciurli, S., Scott, M. J., Hedman, B., Holm, R. H., Ludden, P.

W., Korszun, Z. R., Stephens, P. J., and Hodgson, K. O. (1992) On the structure of the

nickel/iron/sulfur center of the carbon monoxide dehydrogenase from Rhodospirillum

rubrum: an x-ray absorption spectroscopy study, Proc. Natl. Acad. Sci. U. S. A. 89, 4427-

4431.

51. Davidson, G., Clugston, S. L., Honek, J. F., and Maroney, M. J. (2001) An XAS

investigation of product and inhibitor complexes of Ni-containing GlxI from Escherichia

coli: Mechanistic implications, Biochemistry 40, 4569-4582.

52

52. Haumann, M., Porthun, A., Buhrke, T., Liebisch, P., Meyer-Klaucke, W., Friedrich, B.,

and Dau, H. (2003) Hydrogen-induced structural changes at the nickel site of the

regulatory [NiFe] hydrogenase from Ralstonia eutropha detected by X-ray absorption

spectroscopy, Biochemistry 42, 11004-11015.

53. Ferreira, G. C., Franco, R., Mangravita, A., and George, G. N. (2002) Unraveling the

substrate-metal binding site of ferrochelatase: an X-ray absorption spectroscopic study,

Biochemistry 41, 4809-4818.

54. Diakun, G. P., Fairall, L., and Klug, A. (1986) EXAFS study of the zinc-binding sites in

the protein transcription factor IIIA, Nature 324, 698-699.

55. Stola, M., Musiani, F., Mangani, S., Turano, P., Safarov, N., Zambelli, B., and Ciurli, S.

(2006) The nickel site of Bacillus pasteurii UreE, a urease metallo-chaperone, as revealed

by metal-binding studies and X-ray absorption spectroscopy, Biochemistry 45, 6495-

6509.

56. Al-Mjeni, F., Ju, T., Pochapsky, T. C., and Maroney, M. J. (2002) XAS investigation of

the structure and function of Ni in acireductone dioxygenase, Biochemistry 41, 6761-

6769.

57. Colpas, G. J., Brayman, T. G., McCracken, J., Pressler, M. A., Babcock, G. T., Ming, L.-

J., Colangelo, C. M., Scott, R. A., and Hausinger, R. P. (1998) Spectroscopic

characterization of metal binding by Klebsiella aerogenes UreE urease accessory protein,

J. Biol. Inorg. Chem. 3, 150-160.

58. Chandrasekhar, S., and McAuley, A. (1992) Synthesis of an N2S-cyclodecane

macrocycle and its nickel(II) complexes, J. Chem. Soc. Dalton Trans., 2967-2970.

59. Goodman, D. C., Reibenspies, J. H., Goswami, N., Jurisson, S., and Darensbourg, M. Y.

(1997) A new macrocycle N3S2 ligand and its nickel(II), cobalt(II), rhodium(III)-103,

and rodium(III)-105 complexes, J. Am. Chem. Soc. 119, 4955-4963.

60. Halfen, J. A., and Young, V. G. J. (2003) Efficient preparation of 1,4,8-trimethlcyclam

and its conversion into thioalkyl-pendant pentadentate chelate, Chem. Comm. 2894-2895.

61. Perczel, A., and Fasman, G. D. (1992) Quantitative analysis of cyclic β-turn models,

Protein Sci. 1, 378-395.

62. Knappe, T. A., Eckert, B., Schaarschmidt, P., Scholz, C., and Schmid, F. X. (2007)

Insertion of a chaperone domain converts FKBP12 into a powerful catalyst of protein

folding, J. Mol. Biol. 368, 1458-1468.

63. Bloom, S. B., and Zamble, D. B. (2004) The metal-selective DNA-binding response of

Escherichia coli NikR, Biochemistry 43, 10029-10038.

64. Iwig, J. S., Rowe, J. L., and Chivers, P. T. (2006) Nickel homeostasis in Escherichia coli

- the rcnR-rcnA efflux pathway and its linkage to NikR function, Mol. Microbiol. 62,

252-262.

65. Rowe, J. L., Starnes, G. L., and Chivers, P. T. (2005) Complex transcriptional control

links NikABCDE-dependent nickel transport with hydrogenase expression in Escherichia

coli, J. Bacteriol. 187, 6317-6323.

66. Salgado, M. T., Bacher, K. L., and Stillman, M. J. (2007) Probing structural changes in

the alpha and beta domains of copper- and silver-substituted metallothionein by emission

spectroscopy and electrospray ionization mass spectrometry, J. Biol. Inorg. Chem. 12,

294-312.

67. Rigby Duncan, K. E., and Stillman, M. J. (2006) Metal-dependent protein folding:

metallation of metallothionein, J. Inorg. Biochem. 100, 2101-2107.

53

68. Ngu, T. T., Easton, A., and Stillman, M. J. (2008) Kinetic analysis of arsenic-metalation

of human metallothionein: significance of the two-domain structure, J. Am. Chem. Soc.

130, 17016-17028.

69. Zeng, Y. B., Zhang, D. M., Li, H., and Sun, H. (2008) Binding of Ni(2+) to a histidine-

and glutamine-rich protein, Hpn-like, J. Biol. Inorg. Chem. 13, 1121-1131.

70. Lee, M. H., Pankratz, H. S., Wang, S., Scott, R. A., Finnegan, M. G., Johnson, M. K.,

Ippolito, J. A., Christianson, D. W., and Hausinger, R. P. (1993) Purification and

characterization of Klebsiella aerogenes UreE protein: a nickel-binding protein that

functions in urease metallocenter assembly, Protein Sci 2, 1042-1052.

71. Fu, C., Olson, J. W., and Maier, R. J. (1995) HypB protein of Bradyrhizobium japonicum

is a metal-binding GTPase capable of binding 18 divalent nickel ions per dimer, Proc.

Natl. Acad. Sci. U.S.A. 92, 2333-2337.

72. Rey, L., Imperial, J., Palacios, J. M., and Ruiz-Argueso, T. (1994) Purification of

Rhizobium leguminosarum HypB, a nickel-binding protein required for hydrogenase

synthesis, J. Bacteriol. 176, 6066-6073.

73. Yang, S., Thackray, A. M., Fitzmaurice, T. J., and Bujdoso, R. (2008) Copper-induced

structural changes in the ovine prion protein are influenced by a polymorphism at codon

112, Biochim. Biophys. Acta. Prot. Proteom. 1784, 683-692.

74. Munoz, A., Dughish, M., Specher, T., Shaw, C. F., and Petering, D. H. (1999) Formation

and reactions of copper metallothionein in vitro and in cells, J. Inorg. Biochem. 74, 302-

302.

75. Wu, L.-F., Mandrand-Berthelot, M.-A., Waugh, R., Edmonds, C. J., Holt, S. E., and

Boxer, D. H. (1989) Nickel deficiency gives rise to the defective hydrogenase phenotype

of hydC and fnr mutants in Escherichia coli, Mol. Microbiol. 3, 1709-1718.

76. Jakob, U., Muse, W., Eser, M., and Bardwell, J. C. A. (1999) Chaperone activity with a

redox switch, Cell 96, 341-352.

77. Won, H. S., Low, L. Y., De Guzman, R., Martinez-Yamout, M., Jakob, U., and Dyson, H.

J. (2004) The zinc-dependent redox switch domain of the chaperone Hsp33 has a novel

fold, J. Mol. Biol. 341, 893-899.

78. Martin, V., Groenendyk, J., Steiner, S. S., Guo, L., Dabrowska, M., Parker, J. M. R.,

Muller-Esterl, W., Opas, M., and Michalak, M. (2006) Identification by mutational

analysis of amino acid residues essential in the chaperone function of calreticulin, J. Biol.

Chem. 281, 2338-2346.

79. Erdmann, F., and Fischer, G. (2007) The nickel-regulated peptidyl prolyl cis/trans

isomerase SlyD, in Metal Ions in Life Sciences (Sigel, A., Sigel, H., and Sigel, R. K. O.,

Eds.), pp 501-518, John Wiley and Sons.

80. Benanti, E. L., and Chivers, P. T. (2009) An intact urease assembly pathway is required

to compete with NikR for nickel ions in Helicobacter pylori, J. Bacteriol. 191, 2405-

2408.

81. Stingl, K., Schauer, K., Ecobichon, C., Labigne, A., Lenormand, P., Rousselle, J.-C.,

Namane, A., and De Reuse, H. (2008) In vivo interactome of Helicobacter pylori urease

revealed by tandem affinity purification, Mol. Cell. Proteom. 7, 2429-2441.

82. Palumaa, P., Eriste, E., Njunkova, O., Pokras, L., Jornvall, H., and Sillard, R. (2002)

Brain-specific metallothionein-3 has higher metal-binding capacity than ubiquitous

metallothioneins and binds metals noncooperatively, Biochemistry 41, 6158-6163.

54

83. Palumaa, P., Tammiste, I., Kruusel, K., Kangur, L., Jornvall, H., and Sillard, R. (2005)

Metal binding of metallothionein-3 versus metallothionein-2: lower affinity and higher

plasticity, Biochim. Biophys. Acta. Prot. Proteom.1747, 205-211.

55

Chapter 3 The in vitro metal selectivity and in vivo metal specificity of the

nickel metallochaperone SlyD

3.1 Introduction

Metal selectivity in biology is a complex and essential process that is achieved through the action

of a number of protein factors. These proteins involved in metal homeostasis ensure that metal-

dependent proteins are paired with the correct metal-ion to produce a biologically active protein.

Proteins involved in metal homeostasis include membrane importers/exporters, metal storage

proteins that are responsible for safe sequestration, metallochaperones dedicated to targeted

delivery of metal ions, as well as metal-sensing transcription factors which together control the

availability of a specific metal in the cytoplasm of the cell. The fidelity of the proteins involved

in metal homeostasis is at the top of a hierarchy of metal selectivity, as their specificity is now

proposed to ultimately influence the metal occupancy of all other proteins that require a metal

ion cofactor for function (1, 2). Therefore, knowledge of proteins involved in metal homeostasis

and their ability to recognize and chelate a cognate metal partner is a pre requisite to

understanding how metal specificity is achieved in the context of an entire cell.

The Escherichia coli protein SlyD is an example of a protein that is involved in Ni(II)

homeostasis. SlyD contributes to nickel accumulation in this organism and is also necessary for

energy metabolism because it participates in the nickel insertion step during [NiFe]-hydrogenase

metallocenter assembly (3, 4). NMR solution structures of E. coli SlyD revealed that the N-

terminal region consists of two well-defined domains, the FKBP (FK-506 binding protein)

domain and an IF (insert in the flap) domain that recognizes and binds to unfolded protein (5, 6).

The FKBP domain shares homology with other members of the FKBP family of peptidyl-prolyl

isomerases (PPIases) and catalyzes the otherwise slow isomerization of proline peptide bonds

during protein folding in vivo (7, 8). In E. coli SlyD the final 50 residues at the C-terminus

correspond to the metal-binding domain (MBD) and includes 28 potential metal-binding amino

acids (6 Cys, 15 His, 2 Glu, 5 Asp). This C-terminal domain is unstructured according to NMR

data and is variable between SlyD homologues (9). A vital role for the metal-binding domain

was established when truncations of this region resulted in compromised hydrogenase production

56

in E. coli (3). In support of the nickel-related in vivo function of SlyD, a recent analysis of the

protein revealed that it can bind up to 7 nickel ions with an affinity in the nanomolar range (10).

This in vitro investigation also revealed that SlyD coordinates nickel through a non-cooperative

mechanism and exists in a mixture of metalloforms at a given metal concentration. The ability of

the protein to sequester a substantial amount of Ni(II) in the unique MBD has been proposed to

contribute to Ni(II) storage in E. coli (11).

Although SlyD has been characterized in detail in the context of its nickel-binding capabilities,

this protein can bind several other types of transition metals such as Zn(II), Cu(II) and Co(II) in

vitro (12). However, this activity is not well characterized. Therefore, to establish whether SlyD

is specific for Ni(II), and if so, to gain an understanding of how this specificity is achieved, we

explored the metal binding activity of SlyD to several biologically relevant first row transition

metals, Mn(II), Fe(II), Co(II), Cu(I), and Zn(II) (13). Our results indicate that while SlyD is

unable to bind Mn(II) or Fe(II) with appreciable affinity it can bind to Co(II), Cu(I) and Zn(II).

We also find that Ni(II) can selectively bind to SlyD over Co(II) but that selectivity for nickel is

not observed in the presence of Cu(I) or Zn(II). Although SlyD did not exhibit selectivity

towards nickel in vitro, in vivo experiments reveal that SlyD can selectively influence the

balance of nickel ions in E. coli under anaerobic conditions.

3.2 Materials & Methods

Materials. The following metal salts, ZnSO4, [Cu(CH3CN)4]PF6, Ni(Acetate)2, CoSO4, FeSO4

and MnCl2 were purchased from Aldrich as a minimum 99.9% pure. Except for copper, all metal

solutions were prepared by dissolving the salts as purchased in Milli-Q water that was stirred in

an anaerobic glovebox (O2 < 1 ppm) for at least 24 hrs to minimize oxygen content.

Concentrations of metal stock solutions were verified by ICP-AES. The Cu(I) salts were re-

crystallized as described previously to remove any trace amounts of Cu(II) and the resulting

powder was dissolved in an Ar(g) saturated 30% (v/v) acetonitrile/water solution (14). The

concentrations of the copper solutions were established by using bathocuproine sulfonate (Bcs)

as described previously (15). Buffers for all metal assays were prepared with Milli-Q water that

was treated with Chelex-100 (Bio-Rad) and they were stirred in an anaerobic glovebox (O2 < 1

ppm) for at least 24 hrs prior to use. All titrations and assays were conducted at least in triplicate

57

to ensure reproducibility. The metal chelators ethyleneglycoltetraacetic acid (EGTA),

bicinchoninate (Bca), Bcs, glycine were purchased from Sigma-Aldrich as a minimum 99% pure

and the pH of the stock solutions were adjusted to match that of the protein buffer.

Protein expression, purification and preparation. SlyD wild-type (WT) and the triple mutant

proteins (triple mutant is devoid of any Cys residues and has the following mutation/deletions:

Cys 167,168,184,185 Ala and Δ193-196) were expressed and purified as mentioned previously

(10). Preparation of apo protein and determining the reduction state of the protein via N-Ethyl

maleimide modifications were carried out as described previously (10).

Metal binding via electronic absorption spectroscopy. Protein samples were buffer exchanged

into 20 mM HEPES pH 7.5, 100 mM NaCl and 90 L aliquots of apo SlyD were incubated with

the appropriate metal concentrations as noted at 4 oC under anaerobic conditions. Samples were

analyzed under aerobic conditions except for Fe(II) and Cu(I) titrations, which were measured in

an anaerobic cuvette to minimize exposure to air. Due to significant background signals from

Fe(II) and Cu(I) in HEPES buffer, direct metal-binding experiments for these two metals were

conducted in 10 mM ammonium acetate, pH 7.5 buffer. Experiments involving Cu(I) and

Bcs/Bca were conducted in 20 mM HEPES buffer pH 7.5, 100 mM NaCl and spectroscopic

analysis was conducted under aerobic conditions.

Equilibrium dialysis. SlyD samples of 40 M were dialyzed against an equal volume of

appropriate metal solutions containing 320 M metal at 4 oC overnight in an anaerobic glovebox

using Microchambers of DIALYZER (Harvard apparatus). At the end of the equilibration period

samples from the protein chamber (metal bound to protein + free metal) and the metal ion

chamber (free metal) were measured in triplicate via a previously described HPLC method (16).

Metal binding via ESI-MS. Protein samples were buffer exchanged using 500 L aliquots of 10

mM ammonium acetate, pH 7.5 in six dilution steps using 10K nanosep centrifugal devices

(PALL Science). Direct metal titration results for zinc were obtained by equilibrating the protein

samples (10 M each) with a known amount of ZnSO4 solutions. Titration of SlyD with Cu(I),

and Co(II) was obtained by incubating the protein samples (10 M) with [Cu(MeCN)4]+ and

CoSO4 respectively in the presence of 100 M DTT. Samples were allowed to equilibrate

overnight at 4 oC in an anaerobic glovebox, followed by transfer to septum capped vials for MS

analysis.

58

ESI-MS instrumental parameters and data acquisition/processing. The mass spectra were

acquired on a AB/Sciex QStarXL mass spectrometer equipped with an ion spray source in the

positive ion mode and a Hot Source-Induced Desolvation (HSID) interface (Ionics Mass

spectrometry Group Inc.). Ions were scanned from m/z 1300-3000 with accumulation of 1 s per

spectrum with no interscan time delay and averaged for a 2-4 min period. The instrument

parameters are as follows: ion source temperature 200 oC, ion source gas 50.0 psi; curtain Gas

50.0 psi; ion spray voltage 5000.0 V; declustering potential 60.0 V; focusing potential 60 V;

collision gas 3.0; MCP (detector) 2200.0 V. The spectra were deconvoluted using the Bayesian

protein reconstruction program over a mass range of 20000 – 22000 Da and a step mass of 1 Da,

signal/noise ratio of 20 and the minimum intensity detected set to 1 % were used during

reconstruction of the data. Addition of DTT to protein samples decreases the intensity of the

signal detected and at metal concentrations at or above 2 equivalents of metal (i.e. 20 M or

higher) the signal is further suppressed due to high salt conditions. Therefore, samples

containing DTT were acquired with similar instrument conditions listed above except the Q1

transmission was enhanced at m/z of 2500 Da. The amount of apo and holo SlyD species were

calculated from the signal intensities of the reconstructed spectra assuming that all different

metalloforms are ionized to the same extent. Only the peaks corresponding to salt free SlyD

were used for calculating the metallation states of SlyD.

Stoichiometry of metal binding via ESI-MS. To obtain the total number of metal-binding sites, a

100 L aliquot of 100 M SlyD in 10 mM ammonium acetate, pH 7.5 was incubated with the

indicated amount of the appropriate metal and was left to equilibrate overnight at 4 oC in an

anaerobic glovebox. The samples were then gel-filtered via PD-10 columns (GE Healthcare)

using 10 mM ammonium acetate, pH 7.5 to remove unbound or loosely-bound metal ions and

eluting fractions were analyzed via ESI-MS.

Competition experiments via ESI-MS. Fractions of the apo-protein in 10 mM ammonium acetate,

pH 7.5, were incubated with 10 equivalents of metal overnight at 4 oC under anaerobic

conditions. The protein samples were then gel-filtered via PD-10 column to remove excess

metal. The collected holo-protein fractions were incubated with known amounts of the

competitors, each prepared in chelexed MilliQ water and adjusted to a pH of 7.5 by using

ammonium hydroxide (metal-free grade). Samples were incubated overnight at 4 oC under an

anaerobic environment prior to analysis via ESI-MS. The metal affinity constants, protonation

59

constants of the chelators, metal bound chelators and aqueous metal ions at 25 oC were obtained

from the NIST (National Institute of Science and Technology) database version 8.0 (17). The

apparent constants at pH 7.5 were then calculated by using the method described by Fahrni et. al.

(18).

PAR competition experiments. A competition reaction was set up as noted in equation (A) by

adding increasing amounts of apo SlyD to separate aliquots of 400 M PAR containing 8 M

Zn(II) in 20 mM HEPES pH 7.5, 100 mM NaCl.

Zn(PAR)2 + SlyD ⇄ Zn-SlyD + 2PAR Kex1 (A)

Samples were equilibrated overnight at 4 oC and subsequently analyzed by electronic absorption

spectroscopy at 500 nm. Titrations of 400 M PAR with Zn(II) under identical conditions were

used to calculate the amount of [Zn-Par2] formed in the competition reactions described in (A).

As the pH of the solution greatly affects the formation constant of Zn-PAR2 complex as well as

the molar absorption coefficient () of this complex, the metal-binding affinity of PAR under

these particular buffer conditions were calibrated using EGTA as described in detail by

Zimmerman et al. (15). Briefly, increasing amounts of EGTA were titrated into 400 M PAR

containing 8 M Zn(II) to achieve a competition as noted in (B).

Zn(PAR)2 + EGTA ⇄ Zn-EGTA + 2PAR Kex2 (B)

An average KD of SlyD for zinc was then obtained from equation (C) (derived from combining

reactions A and B) where the known formation constant for Zn-EGTA corrected for pH (Ka =

109.39

) was used (18).

KD (Zn-SlyD) = KD (Zn-EGTA) x (Kex2/Kex1) (C)

Circular dichroism (CD) spectroscopy. Protein samples were pre-treated with EDTA and TCEP

as described above and buffer exchanged into 10 mM ammonium acetate pH 7.5 using micro-

centrifugal devices (MWCO of 10,000) (Pall nanosep centrifuge devices) in an anaerobic

glovebox. SlyD samples (60 µM) were incubated with known amounts of metal overnight at

4oC. Samples were loaded into a 0.1 cm cuvette and capped to minimize exposure to air. The

spectra were collected on a Jasco J-710 spectropolarimeter by scanning in the wavelength range

60

of 205−320 nm at room temperature. The final spectra obtained are averages of 5 scans collected

by using a scan speed of 20 nm/min.

Analytical gel filtration chromatography. Analytical gel filtration experiments were performed

on an analytical Superdex 75-HR column (GE Healthcare) at 4 °C with a flow rate of 0.5 ml/min

and using an injection loop of 100 µL. Protein sample of 75 µM were prepared and injected into

the column that was equilibrated with 20 mM HEPES, pH 7.5, 100 mM NaCl, 1 mM TCEP.

Samples were incubated with the indicated amount of metal at 4 °C overnight in an anaerobic

glovebox prior to use. Triple mutant samples were prepared under aerobic conditions. The

column was calibrated by injecting 100 µL of the LMW gel filtration standards kit (GE

Healthcare).

Cu(I)-binding stoichiometries and affinities using Bca and Bcs. All experiments were set up and

equilibrated in an anaerobic glovebox using buffers that were degassed with Argon for several

hours. Apo protein samples were titrated into a solution of [Cu(L)2]3-

(L = Bcs or Bca) of

defined molar ration L:Cu(I) ≥ 2.5, where the chelator concentration was kept high enough to

ensure negligible contribution from the 1:1 Cu:L complex. In these mixtures, the ligand and

copper concentrations were held constant while varying the concentration of the protein.

Transfer of Cu(I) from the chelator to protein or vice versa can be detected by the change in

absorbance at 483 nm for Bcs and 562 nm for Bca (15). By systematically varying the

concentration of the chelators and the protein, conditions that favoured either competitive or non-

competitive conditions were achieved. From the resulting spectra the KD and the stoichiometry

of SlyD for Cu(I) were determined.

Co(II) affinity using Fura2. A stock solution of 35 M Fura-2 (Invitrogen) containing 8-12 M

Co(II) was obtained by incubating the metal with the chelator for 30 min at room temperature.

The free dye concentration was determined by using absorption spectroscopy and the reported

extinction coefficient of 28000 M-1

cm-1

at 363 nm (19). To achieve competitive conditions as

noted in reaction (D), increasing amounts of apo SlyD were added to separate aliquots of this

cobalt-Fura2 solution and allowed to equilibrate overnight at 4 oC in an anaerobic glovebox.

[Co(II)-Fura2] + SlyD ⇄ Co-SlyD + Fura2 Kex3 (D)

KD (Co-SlyD) = (Kex3)-1

KD (Co-Fura2) (E)

61

The resulting solutions were analyzed using electronic absorption spectroscopy at 368 nm, which

indicates the amount of chelator bound to cobalt ion. An extinction coefficient of -24000 M-1

cm-1

at 368 nm was measured by titration of Fura2 with the CoSO4 alone under identical

conditions and used for calculating the concentration of Co(II)-Fura-2 present at each titration

point. An average KD of Co-SlyD was then calculated using data from several replicates via

equation (E).

Metal toxicity studies

(A) In LB media. Wild type and ΔslyD strains from the BW25113 KEIO gene deletion

collection were used in this assay (20). The growth rates of wild type and ΔslyD strains, under

different metal concentrations were assessed by measuring the OD600 over 24 h. The following

metal salts were utilized to supplement the growth media; NiCl2, CoSO4, ZnSO4 and CuSO4. All

cultures were maintained at 37 oC in an aerobic environment. The concentrations tested and the

observed effects are summarized in Table 3-1.

Table 3-1. Summary of metal concentrations used for metal toxicity studies.

Growth

Condition Metal tested [Metal(II))] M

LB media

Aerobic

Co(II) 100, 200, 500, 1000, 10000*

Ni(II) 100, 200, 500, 1000, 10000*

Cu(II) 100, 200, 500, 1000*

Zn(II) 100, 200, 500, 1000, 10000*

Minimal Media

Aerobic

Co(II) 10, 100*

Ni(II) 10, 30*, 100*

Cu(II) 10, 100, 250*†

Zn(II) 10, 150, 300, 400, 700*†

*No growth was detected for the wildtype and ΔslyD strains indicating lethal metal concentrations.

† A slight precipitate was detected upon addition of the metal to minimal media.

(B) In minimal media. All solutions used in this assay were treated with chelex-100 (Biorad)

overnight to remove any trace metals. Metal stocks were prepared from chelex-treated Milli-Q

water. The M9 media (1x) (Sigma) was supplemented with 0.2% glucose, 1 mM MgSO4 and

100 M CaCl2 and sterile filtered. Two bacterial strains were used in these toxicity studies: the

above mentioned BW25113 and the MC4100. The growth rate was monitored in M9 media by

measuring the OD600 over 24 h. All cultures were maintained at 37 oC under aerobic conditions.

The concentrations of metals tested and the observed effects are summarized in Table 3-1.

62

Growth conditions for quantitative real-time PCR (qRT-PCR).

(A) Aerobic. All experiments were conducted using the bacterial strain MC4100 and the ΔslyD

strain in the same background (21). Cultures were grown at 37 oC, shaking at 250 rpm

aerobically until the cultures reached exponential growth phase (approx. 16 hrs) followed by

treatment with a known amount of metal for 30 minutes.

(B) Anaerobic. For anaerobic growth experiments MC4100 wildtype, ΔslyD strains were used.

The M9 salts (1x) were chelex treated overnight to remove any trace metal and then the

following nutrients were added back to the media: 30 mM sodium formate, 0.5% glucose (v/v), 1

mM MgCl2, 100 nM CaCl2, 100 nM (NH4)2MoO4, 100 nM NaSeO3. Cultures were grown in a

capped bottle without any headspace for 17 hrs at 37 oC allowing the cultures to reach a mid-log

phase (OD600 =0.5 – 0.6). These anaerobically-grown cultures were then divided into sterile

falcon tubes containing no metal or with 10 M NiCl2 in an anaerobic glovebox. The small

cultures were then capped and allowed to grow for an additional 30 min at 37 oC. For analysis of

Zn(II) and Cu(II) under anaerobic conditions an identical procedure was followed, except the

chelex-treated M9 media was supplemented with only 0.2 % glucose (v/v), 1 mM MgCl2, 100

nM CaCl2.

RNA isolation and qRT-PCR. Total RNA was isolated by using the RNeasy Mini Kit (Qiagen)

according to the manufacturers’ protocol for the mechanical disruption and purification of

bacterial RNA. The QuantiTect Reverse Transcription Kit (Qiagen) was used for synthesis of

cDNA from ~0.5-1 g of RNA and to eliminate contaminating genomic DNA. qPCR reactions

were performed on a LightCycler 480 Real-time PCR block (Roche) using SYBR green

(Finnzymes) and primers specific for each transporter as listed in Table 3-2. holB mRNA

encoding for a DNA gyrase was used as an internal control as the expression level of this protein

was shown to be unaffected in the presence of metal (22).

63

Table 3-2. Primers used for qRT-PCR.

Primer Name Sequences (5’ 3’)

znuA forward ACATGCATCTTTGGCTTTCC

znuA reverse TGCCTCAAAATCCTTCAGGT

zntA forward AGAAAGCCCCTCAATTTGCT

zntA reverse CCGGAGACGTTTTCAGAGAG

copA forward CAAGCCAGAAATCGGTCAGC

copA reverse CAAAGAAATACCAGATTGCCGC

cusF forward ATGAAACCATGAGCGAAGCACA

cusF reverse CGGATCGTGATGGATGGTGAT

nikA forward CAGTGCTCGATAACCGTCAA

nikA reverse GCGCTTTTCAGGGTAATTTG

rcnA forward GAACCAGGGCACTCAAAAAC

rcnA reverse TGCGGTATGCGAAATAGTTG

holB forward GCTTAGGTGGTGCGAAAGTC

holB reverse GTTTCTGCTGGTGGCTCTTC

3.3 Results

The metal-binding domain (MBD) of SlyD is unusual because more than 50% of the residues in

this C-terminal region can potentially coordinate to a metal ion and contains a mixture of

thiolates, imidazoles and carboxylates. This distinctive amino acid composition raises the

question as to whether this protein can selectively bind Ni(II), the only metal ion thus far proven

to be associated with a SlyD function in vivo (21). To address this issue we characterized the in

vitro metal-binding activity of SlyD to several transition metal ions that are known to be

accumulated by E. coli (23).

Zn(II) binding to SlyD.

Zinc is one of the most abundant transition metals avidly acquired by E. coli and the intracellular

concentrations of this metal ion can reach up to ~0.1 mM (23). Given the abundance of zinc in

the cytoplasm and the ligand composition of MBD of SlyD, it is feasible that Zn(II) binds to

SlyD in vivo. We first examined Zn(II) binding to SlyD in vitro using mass spectrometry. Apo

protein samples were titrated with increasing amounts of ZnSO4 followed by ESI-MS analysis.

The observed charge states and the corresponding deconvoluted spectra for a representative

Zn(II) titration are shown in Figure 3-1. The data reveal the presence of several metalloforms of

the protein that differ in the amount of zinc bound at each titration point, indicating that SlyD

64

contains multiple metal-binding sites with similar affinities for zinc. The existence of these

stable, partially-metallated intermediates indicate a non-cooperative metallation mechanism

similar to that observed when SlyD chelates Ni(II) ions (10, 24). Changes in the charge state

distribution observed during ESI-MS analysis can be used to infer conformational changes of a

protein upon ligand binding (25). As previously noted (10), the m/z spectrum of apo-SlyD shows

the presence of two different populations of SlyD in solution (Figure 3-1), with an abundance of

protein carrying a higher charge (i.e. +16 signal is more intense than the signal for +9 species)

indicating that apo SlyD predominantly exists in a more unfolded state under these experimental

conditions. Upon titration with Zn(II) the relative peak intensities for the lower m/z species

decreases with increasing metal ion concentrations suggesting a metal-induced conformational

change in which the addition of Zn(II) shifts the equilibrium of SlyD to the more folded

conformer. A similar but more dramatic nickel-induced folding was noticed for SlyD in our

previous investigation, which was conducted on the same instrument under identical

experimental conditions but prior to installation of the HSID interface. The HSID is an

atmospheric pressure interface that is used for efficient dissolvation of sample, which is

accomplished via energy transfer from the hot gas as the ions travel through multiple flow

regions. The resulting ions are orthogonally introduced to the mass spectrometer and the

interface is expected to lead to greater sensitivity and stability of the signal while operating

similar to a conventional ion source (http://www.ionicsmsv.eu/ionicsmsveu.swf). To allow

comparison of data acquired under the same instrumental settings, nickel coordination to SlyD

was re-evaluated on the mass spectrometer now equipped with the interface. Slight changes in

the m/z spectra were noted; the apo SlyD spectrum is now predominantly populated by lower m/z

species (Figure 3-2) whereas without the interface the apo protein exhibited a higher abundance

of lower charge states (10). This difference in populations could be due to partial unfolding of

the protein in this HSID due to the higher temperatures used. However, as previously reported

the addition of nickel caused a shift in the charge state envelopes indicating metal-induced

folding. (Figure 3-2 and (10)). The resulting reconstructed data acquired on the mass

spectrometer with the HSID interface are comparable to the data obtained for SlyD previously,

revealing that the protein coordinates a variable number of nickel ions at a given metal

concentration, indicative of a non-cooperative metal-binding mechanism.

65

Figure 3-1. ESI-MS titration of SlyD with Zn(II). The addition of increasing amounts of Zn(II) to

SlyD (10 M) leads to concurrent filling of multiple metal sites indicating a non-cooperative metal-

binding mechanism. From top to bottom the equivalents of zinc added relative to protein concentration

are as follows: 0, 0.5, 1.5, 2, 4 and 6.3. The m/z spectra are shown on the left with their respective

reconstructed spectra in the right panels. The numbers heading the dotted-lines indicate the amount of

Zn(II) bound to SlyD (right panels) and the charge state of the protein (left panels).

m/z

Rel

ativ

e in

tens

ity

+16 +14 +12 +10 +9 +8

Mass (Da)

20800 21000 21400

0 1 2 3 4 5 6 7

1600 2000 2400 2800 21200

66

Figure 3-2. Re-evaluating Ni(II) binding to SlyD by ESI-MS. The m/z spectra (left panels) and the

reconstructed spectra (right panels) for a Ni(II) titration analyzed on the mass spectrometer equipped with

the HSID interface. Addition of increasing concentrations of nickel acetate to the apo protein (10 M)

results in simultaneous metal loading to multiple metal sites, which is indicative of a non-cooperative

Ni(II)-binding mechanism. The molar equivalents of nickel added relative to protein concentration from

top to bottom is as follows: 0, 0.6, 2.1, 3.5, 4.2, 6 The numbers leading the dotted-line represent the

amount of nickel bound to the protein (right panels) and the charge state of SlyD (left panels).

m/z

Re

lative

in

ten

sity

+16 +14 +12 +10 +9 +8

Mass (Da)

20800 21000 21400

0 1 2 3 4 5 6 7

1600 2000 2400 2800 21200

67

A definitive endpoint to the Zn(II) titration cannot be achieved because the concentration of salt

interferes with the quality of the MS data. Therefore, the total number of tight binding sites on

SlyD was established by analyzing a protein sample that was treated with an 8-fold excess of

Zn(II) overnight followed by gel-filtration to remove excess or loosely-bound Zn(II). The mass

spectrum indicates the presence of a mixed population of metallated species with a total of five

metal-binding sites with dominant peaks arising from the four and five Zn(II)-coordinated

species (Figure 3-3).

Figure 3-3. Zn(II) stoichiometry & affinity of SlyD. Mass spectra of a SlyD sample (100 M)

incubated with 800 M zinc followed by gel filtration to remove any weak or non-specifically bound

metal indicates a maximum capacity of 5 Zn(II) sites (Top panels). Treatment with 1 mM glycine does

not lead to any significant change in the amount of metal bound to the protein (bottom panels) indicating

that the KD (Zn-SlyD) is tighter than that of glycine for zinc ions.

To estimate the relative affinities of these multiple metal sites, this sample was incubated with an

excess of glycine. Analysis of the holo SlyD samples containing the competitor resulted in a

spectrum that was almost identical to the spectrum without any chelator added. Thus, the MS

data signify that SlyD is capable of binding up to 5 Zn(II) ions with an affinity greater than a Zn-

+TOF MS: 35 MCA scans from Sample 2 (ZnSlyD-Gelfil_27Mar2009) of ZnSlyD-Gelfil_27Mar2009.wiffa=3.56210051967030970e-004, t0=5.32468411187073800e+001

Max. 458.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0

20

40

60

80

100

120

140

160

180

200

220

240

260

280

300

320

340

360

380

400

420

440

458

Inte

ns

ity

, c

ou

nts

Mass reconstruction of +TOF MS: 35 MCA scans from Sample 2 (ZnSlyD-Gelfil_27Mar2009) of ZnSlyD-Gelfil_27Mar2009.wiff Max. 5991.3 cps.

2.090e4 2.095e4 2.100e4 2.105e4 2.110e4 2.115e4 2.120e4 2.125e4 2.130e4 2.135e4 2.140e4 2.145e4 2.150e4Mass, amu

0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

3200

3400

3600

3800

4000

4200

4400

4600

4800

5000

5200

5400

5600

5800

5991

Inte

ns

ity

, c

ps

21106.0000

+TOF MS: 35 MCA scans from Sample 3 (ZnSlyDGelFil_100uMGly_30March2009) of ZnSlyDGelFil_100uMGly_30March2009.wiffa=3.56210051967030970e-004, t0=5.32468411187073800e+001

Max. 209.0 counts.

1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500m/z, amu

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

170

180

190

200

209

Inte

ns

ity

, c

ou

nts

Mass reconstruction of +TOF MS: 35 MCA scans from Sample 3 (ZnSlyDGelFil_100uMGly_30March2009) of ZnSlyDGelFil_100uMGly_... Max. 1818.1 cps.

2.090e4 2.095e4 2.100e4 2.105e4 2.110e4 2.115e4 2.120e4 2.125e4 2.130e4 2.135e4 2.140e4 2.145e4 2.150e4Mass, amu

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

Inte

ns

ity

, c

ps

21106.0000

Zn(II)

3

Zn(II)

4 Zn(II)

5

Zn(II)

3

Zn(II)

4Zn(II)

5

Rel

ativ

e in

tens

ity

1200 1600 2000 2400 2800 3200

m/z (Da) Mass (Da)

20800 21100 21300 21500

68

(glycine)2 complex (Ka of 104.96

) (17). On the other hand, the addition of the stronger Zn(II)

chelator EGTA (apparent KD at pH 7.5 of 4.25 x 10-10

M), to zinc-loaded SlyD results in a

progressive decrease in the amount of metal bound to SlyD and no distinct metalloform (i.e..

only 3 Zn(II) bound or only 2 Zn(II) bound) can be isolated at any given chelator concentrations,

providing evidence for multiple metal sites of similar affinity (Figure 3-4). Furthermore, this

competition indicated that SlyD can compete with EGTA for zinc coordination because the

protein is still bound to Zn(II) upon equilibration with ~80-fold excess of competitor.

Figure 3-4. Competition of SlyD with EGTA for Zn(II) binding. Titration of holo-SlyD (19.5 µM,

obtained via gel filtration of apo SlyD incubated with 8-fold excess metal) with increasing amounts of

EGTA in 10 mM ammonium acetate, pH 7.5. The samples were allowed to equilibrate for 5 hrs at room

temperature under anaerobic conditions. The relative intensities are representative of the abundance of

each species at a given titration point. Several metalloforms are observed at each titration point

implicating very similar affinities of the multiple metal sites present. Note that the x-axis ([EGTA]

values) is not linear.

To estimate the affinity of SlyD for zinc ions quantitatively, a competition assay was performed

using the colorimetric indicator PAR. This chelator forms a 2:1 PAR to Zn(II) complex in

solutions that results in an increased absorbance at 500 nm (26). Titration of increasing amount

00.195

0.5850.975

1.56

0

0.2

0.4

0.6

0

1

2

3

4

5

Rela

tive I

nte

nsity

69

of apo SlyD to a 400 M PAR solution containing 8 M Zn(II) results in a gradual decrease in

the signal at 500 nm (as opposed to a linear decrease) indicative of an effective competition

between the protein and the ligand as described in reaction (A) (Figure 3-5). An identical

titration with EGTA instead of protein was also conducted to calibrate the affinity of the Zn-

PAR2 under the buffer conditions used, since the formation constant of this complex is affected

by the pH (15). The resulting data from these two titrations were fitted to a model assuming a

single binding event to obtain an apparent KD of 1.0 ± 0.4 x 10-10

M for demonstrating that SlyD

binds zinc more tightly than EGTA at pH 7.5 (Figure 3-5).

Figure 3-5. Determining the Zn(II) dissociation constant (KD) of SlyD using PAR. Change in the

amount of Zn(II)-Par2 complex upon addition of increasing amounts of apo SlyD (filled diamonds) or

EGTA (unfilled squares). Change in the Zn(II)-PAR2 complex is proportional to the absorption at 500

nm. Titrations were conducted in 20 mM HEPES pH 7.5, 100 mM NaCl with a [PAR]Total = 400 M and

[Zn(II)]Total = 8 M. SlyD is more effective than EGTA at pH 7.5 at competing with PAR for zinc,

indicating KD(Zn-SlyD) < KD (Zn-EGTA) of 10-9.3

M.

For comparison, the affinity of Ni(II)-SlyD complexes were also evaluated. Exploratory

experiments indicated that the protein could out compete Fura-2 (KD 2.4 x 10-8

M, (27)) for

Ni(II) coordination (data not shown) establishing a quantitative upper-limit to nickel-SlyD

0

0.2

0.4

0.6

0.8

1

0 0.4 0.8 1.2 1.6 2

Fra

ction o

f Z

n(I

I)-P

AR

2C

om

ple

x

SlyD or (EGTA):Zn(II)

70

complexes. Combining these results with the fact that Ni(II) can bind to SlyD concurrently with

Zn(II) (see below), a KD(Ni(II)-SlyD) of 10-10

M can be roughly estimated.

To determine whether zinc binding could induce structural changes in the protein as suggested

by ESI-MS, a Zn(II) titration was analyzed by using CD spectroscopy. Two minima are

observed for the apo protein at 215 nm and 227 nm (Figure 3-6). Typically proteins with β-sheet

type structures give rise to the band at 215 nm, as would be expected for SlyD because both the

PPIase and IF domains are predominantly composed of β-sheets (6), whereas the 227 nm band

suggests the presence of β-turn type structure (28, 29). As SlyD is titrated with increasing

amounts of Zn(II) a gradual decrease in the molar ellipticity is observed in the far-UV region that

reaches a plateau upon addition of 4 mol equiv of Zn(II). Although this trend is similar to that

observed when SlyD chelates nickel, the overall degree of change in the molar ellipticity is

significantly lower for Zn(II). Therefore, the CD spectra suggests that the zinc-induced

secondary structure for SlyD is likely different from that of Ni(II). Additional evidence for the

adaptation of a different conformation by SlyD when binding zinc is the lack of a signal at 270

nm. The chromophores at these longer wavelengths are usually aromatic residues (tyrosine and

phenylalanines in the case of SlyD) that are exclusively located in the PPIase and the IF-domains

of SlyD. These residues can generate a signal in response to even a very minor change in their

environment that may occur as a result of ligand binding (30). This lack of a charge transfer

band at 270 nm region implies that Zn(II) binding gives rise to a tertiary conformation that is

distinct from that of nickel(II)-bound SlyD. However, considering that this 270 nm signal could

also arise from charge transfer between metal and an aromatic residue (31), the lack of such a

transfer processes with a closed shell d10

metal such as zinc could also be a likely explanation for

this difference.

SlyD is a monomeric protein that can bind Ni(II) without any quaternary structure changes (10).

In order to investigate whether SlyD bound Zn(II) is also a monomer, analytical gel-filtration

chromatography was utilized. The addition of up to 4 mol equiv of Zn(II) did not change the

oligomeric state of SlyD (Figure 3-7). However, addition of 6 mol equiv of Zn(II) and higher

resulted in a very broad elution profile with no distinct peaks except for the monomer indicating

that SlyD is susceptible to form oligomers at these higher Zn(II) concentrations. Therefore,

unlike with nickel, higher zinc concentrations can influence the quaternary structure of the

protein such that SlyD is in an equilibrium of monomer and oligomers.

71

Figure 3-6. Comparison of Ni(II)- vs. Zn(II)-induced secondary structure changes in SlyD. A

decrease in the mean residue ellipticity is detected in the far-UV region as compared to the apo protein

(60 M) spectrum (filled circles) when the protein is incubated with metal. The spectra for the metal

loaded SlyD are as follows: 2 equiv. Zn(II) (filled diamonds), 2 equiv. Ni(II) (unfilled diamonds), 4 equiv

Zn(II) (filled triangles), 4 equiv Ni(II) (unfilled triangles) and 8 equiv Zn(II) (black crosses).

Figure 3-7. Zn(II) dependent oligomeric state of SlyD. Analytical gel-filtration chromatography of 75

M apo SlyD (black line), SlyD with 2 molar equiv. of Zn(II) (green line), 4 molar equiv. of Zn(II) (red

line), and 6 mol equiv of Zn(II) (blue line). The elution volume of ~18 mL corresponds to monomeric

SlyD species and addition of excess zinc (6 mol equiv. and higher) results in a broad elution profile

suggestive of zinc-induced oligomer formation.

-24

-19

-14

-9

-4

1

6

205 225 245 265 285 305

ӨM

RE

x 1

04

(de

g c

m2

dm

ol-1)

Wavelength (nm)

0

2

4

6

8

10

12

14

11 13 15 17 19 21

Abso

rba

nce

(A

U)

Elution Volume (mL)

← Monomer

Oligomer →

72

To assess the selectivity of SlyD for Zn(II) or Ni(II), a competition titration was conducted in

which apo SlyD was incubated with equivalent amounts of Ni(II) and Zn(II) overnight. Binding

of zinc to the protein could be monitored indirectly because replacement of nickel by zinc results

in a decrease in the ligand to metal charge transfer (LMCT) band at 315 nm produced by the

nickel-SlyD complex (10). A less intense signal is observed when apo SlyD is titrated with an

equimolar solution of nickel and zinc in comparison to titration of apo SlyD with Ni(II) alone

(Figure 3-8A). Thus the data provide evidence for metal-binding sites on SlyD that

preferentially sequester Zn(II) in the presence of Ni(II). In order to assess whether the order of

metal addition affects the metallation state of SlyD, two competition titrations were carried out.

Figure 3-8. Selectivity of SlyD for Ni(II) vs. Zn(II). (A) Addition of equimolar solution of Ni(II) and

Zn(II) to SlyD (filled diamonds) results in lower intensity LMCT at 315 nm, when compared to a titration

with equivalent amounts of Ni(II) alone (unfilled diamonds). The x-axis indicates the total amounts of

Ni(II) added relative to apo SlyD (15 M). (B) Change in A315 as Ni(II) is added to a SlyD sample (15

M) pre-incubated with 8 mol equiv of Zn(II) (filled diamonds) or addition of Zn(II) into a SlyD sample

(15 M) pre-incubated with 8 mol equiv. of Ni(II) (unfilled diamonds). The absorbance was normalized

by dividing the absorbance at 315 nm by the concentration of the protein used.

Firstly, SlyD was pre-incubated with 8 fold excess Ni(II) for 24 h followed by a Zn(II) titration

and replacement of Ni(II) was detected as a decrease in the intensity of the LMCT band at 315

nm. Secondly, SlyD was pre-treated with 8 mol excess of Zn(II) for 24 h followed by a Ni(II)

titration and replacement of the Zn(II) by nickel could be detected as an increase in the intensity

of LMCT band at 315 nm. A representative data set for each of the competition titrations that

were allowed to equilibrate for 24 h after the addition of the second metal ion is shown in Figure

3-8B. Increasing the incubation period after addition of the second metal to 48 h or 72 h

0

0.003

0.006

0.009

0.012

0.015

0 4 8 12 16 20

Nor

mal

ized

ΔA

bs31

5(

M-1

)

[Metal(II)]/[SlyD]

0

0.003

0.006

0.009

0.012

0.015

0 1 2 3 4 5 6 7 8

Nor

mal

ized

Abs

315

(M

-1)

[Ni(II)]/[SlyD]

(A) (B)

73

generated similar results indicating that 24 h is sufficient to reach a thermodynamic equilibrium

(data not shown). The results are a clear indication of the fact that nickel is able to replace a

fraction of the Zn(II) bound to SlyD and vice versa. Furthermore, the two titration curves do not

intersect at any of the metal concentrations tested, highlighting that the metallation state of SlyD

is dependent on the order of metal addition when considering Ni(II) and Zn(II).

Cu(I) binding to SlyD.

Compared to zinc the intracellular copper quota is lower in E. coli, reaching only about 10 M

(23). As SlyD is a cytoplasmic protein we only characterized Cu(I) (not Cu(II)) binding to the

protein because this is likely the prevalent form of copper present in the reducing cytosolic

environment (32). Cu(I) addition to SlyD is accompanied by a charge transfer band centred

around 254 nm that increases in intensity as a function of the metal content added, reaching

saturation around 4-5 equivalents of metal added (Figure 3-9). There is precedence for binding

of electron rich Cu(I) to the softer thiolate ligands and the 254 nm signal is ascribed to a S

Cu(I) charge transfer band (33, 34). To verify that the 254 nm LMCT is due to the formation of

the Cu(I)-thiolate bond, a similar titration was conducted using the SlyD mutant devoid of all

Cys residues (denoted as the triple mutant here onwards). Addition of Cu(I) to the mutant SlyD

failed to generate a change in the absorption spectrum, providing support to the fact that the

LMCT bands are due to Cu(I) binding to SlyD via thiolate residues (data not shown).

To gain an understanding of the Cu(I)-binding mechanism of SlyD, copper titrations were

analyzed via ESI-MS. Although all possible precautions were taken to ensure anaerobic

conditions, the addition of Cu(I) resulted in the appearance of protein-dimers that increased in

intensity with the addition of increasing amount of Cu(I) (data not shown). The dimer formation

was accompanied by a decrease in the protein mass by 6 Da suggesting oxidation of the protein

because the oxidized form of SlyD also exhibits this decrease in mass from 20852 Da to 20846

Da (data not shown). Therefore, copper titrations of SlyD were re-evaluated in the presence of

100 M DTT. Inclusion of the reducing agent did indeed prevent the formation of protein-

dimers and the accompanying loss of mass indicating that in the presence of Cu(I) the protein is

extremely susceptible to oxidation (Figure 3-10). Addition of DTT to the protein buffer

significantly reduced the ion signal detected resulting in spectra with poor resolution (data not

shown). Therefore, to improve the quality of the spectra, Cu(I) titrations containing DTT in the

74

buffer were acquired with the ion signal enhanced at a m/z of 2500 Da. Representative spectra

for a copper titration acquired under these conditions are shown in Figure 3-10, and demonstrate

that the addition of Cu(I) results in the appearance of several Cu(I)-bound SlyD species. At

metal concentrations below 4 molar equivalents of Cu(I), the predominant species observed

correspond to apo SlyD, SlyD with a single copper ion bound and SlyD with 4 Cu(I) bound.

Assuming that all metal-SlyD species are ionized to the same extent, the metallation pattern

implies that binding of a single Cu(I) ion facilitates the binding of the 2nd

, 3rd

and the 4th

Cu(I)

ion to the protein. Therefore, unlike Ni(II) or Zn(II) coordination, SlyD appears to bind Cu(I)

through a cooperative mechanism. The binding process is not fully cooperative as minor peaks

corresponding to intermediate states can still be observed (i.e. 2 Cu(I) or 3 Cu(I)-bound SlyD).

The addition of 6 molar equivalents of Cu(I) results in the appearance of metalloforms with more

than 4 Cu(I) ions bound indicating additional metal-binding sites on SlyD. However a definite

saturation point could not be obtained through this direct titration method as these concentrations

of salt interfere with the quality of the MS data. It should be noted that significant changes in the

population of charge states are not detected when Cu(I) is added. This is due to the above

mentioned enhancement of the signal at m/z 2500 Da because Ni(II) titrations analyzed under

these instrument settings did not lead to the expected shift in the m/z spectra that is indicative of

metal-induced folding (Figure 3-2 and data not shown).

Figure 3-9. Visualizing Cu(I) binding to SlyD. An anaerobic titration of apo SlyD (25 M) with

increasing Cu(I) results in the appearance of charge a transfer band ~ 254 nm, indicative of Cu(I) binding

to thiolate ligands (inset). A plot of the change in Abs254 (normalized relative to the protein

concentration) versus equivalents of copper indicates saturation around ~3.5-4 equivalents of Cu(I) added.

0

0.004

0.008

0.012

0.016

0.0 2.0 4.0 6.0 8.0

Norm

aliz

ed

Δ

Ab

s2

54

(

M-1

)

[Cu(I)]/[SlyD]

0

0.2

0.4

0.6

0.8

1

1.2

1.4

220 260 300 340 380 420

Ab

so

rba

nce

Wavelength (nm)

75

Figure 3-10. ESI-MS titration of SlyD with Cu(I) in the presence of 100 M DTT. The addition of

increasing amounts of Cu(I) to SlyD (10 M) leads to appearance of several Cu(I)-SlyD species. From

top to bottom the equivalents of Cu(I) added relative to protein concentrations are as follows: 0, 0.3, 0.8,

2.3, 3.9, 6 The m/z spectra are shown on the left with their respective reconstructed spectra on the right.

The numbers heading the dotted lines indicate the amount of Cu(I) bound to SlyD (right panels) and the

charge state of the protein (left panels). The most intense peaks correspond to apo, 1 Cu(I) and 4 Cu(I)-

bound SlyD species (at Cu(I) equivalents < 4) suggestive of a cooperative metal-binding mechanism.

Given that ESI-MS could not be used to determined the Cu(I) stoichiometry, coupled with the

fact that Cu(I) binding to SlyD was analyzed in the presence of DTT, which has a considerable

affinity for Cu(I) (35), the stoichiometry of the metal complex was re-evaluated by using the

small molecule chelator Bca. This copper probe forms an air stable [Cu(Bca)2]3-

complex with a

formation constant of β2 = 1017.2

that gives rise to a readily detectable charge transfer band in the

visible region with a maximum at 562 nm ( = 7900 M-1

cm-1

) (15). Titration of apo wild-type

protein into a solution of 40 M Cu(I) and 100 M Bca resulted in a linear decrease in the signal

m/z

Rel

ativ

e in

tens

ity+16 +14 +12 +10 +9 +8

Mass (Da)20800 21200 21600

0 1 2 3 4 5 6

1600 2000 2400 2800

76

at 562 nm indicating that the conditions used resulted in non-competitive conditions (Figure 3-11

inset). Increasing the Bca concentrations to 300 M while keeping the copper concentration at

40 M resulted in the same linear decrease in absorbance at 562 nm, indicating that the protein

binds Cu(I) with a much higher affinity than Bca. A linear fit of the data indicate that 10 M

SlyD is sufficient to remove the 40 M Cu(I) bound to Bca, thereby confirming that Bca

concentrations ≤ 300 M cannot compete with SlyD and that the protein binds 4 equivalents of

Cu(I) (Figure 3-11).

Figure 3-11. Stoichiometry of SlyD for Cu(I). Titration of apo SlyD into a solution of 100 M Bca

containing 40 M Cu(I) results in linear decrease in absorbance at 562 nm. A linear fit of the data reveals

that at 0.25 [SlyD]/[Cu(I)] all the Cu(I) is removed from Bca by the protein, indicating a stoichiometry of

4 Cu(I)/SlyD monomer.

To determine the KCu(I) of SlyD, competitive conditions were achieved by using the Bcs chelator.

Similar to the Bca, Bcs also coordinates Cu(I) to form an air-stable [Cu(Bcs)2]3-

, albeit with a

higher formation constant (β2 = 1019.8

), which can be detected by a charge transfer band with a

maximum at 483 nm ( = 13000 M-1

cm-1

) (15). Addition of increasing amount of apo SlyD to a

mixture containing 40 M Cu(I) and 100 M Bcs revealed that SlyD competes for copper at this

concentration of the chelator whereas increasing the Bcs concentration to 500 M while keeping

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 0.05 0.1 0.15 0.2 0.25 0.3

Absorb

ance 562 n

m

[SlyD]: [Cu(I)]

0

0.1

0.2

0.3

0.4

0 0.2 0.4 0.6

Absorb

ance 562 n

m

[SlyD]/[Cu(I)]

77

the copper concentration constant suggested that SlyD is unable to compete effectively at this

high chelator concentration (Figure 3-12). Thus titration of protein into a solution with constant

concentrations of Bcs and copper at 200 M and 40 M respectively was used to determine the

affinity of Cu(I) for SlyD (Figure 3-12). The data were fitted assuming a single binding scheme

to obtain an apparent KD of (1.5 ± 0.3) x 10-17

M for Cu(I) binding to SlyD.

Figure 3-12. Determining the Cu(I) KD for SlyD. Change in the absorbance at 483 nm upon the

addition of apo SlyD to solutions with a constant Cu(I) concentration of 40 M and [Bcs]total = 100 M

(squares), 300 M (circles), 500 M (triangles). The non-linear decrease in the absorption with increasing

protein concentration is indicative of competitive conditions for Cu(I) binding.

To evaluate the metal selectivity of SlyD with respect to Cu(I), a second series of experiments

were conducted using the weaker copper probe, Bca. To establish whether the reaction described

in equation (F) can occur, 10 M SlyD was first incubated with 50 M Cu(I) for 1 h, followed

by addition of 200 M Bca. Then to this resulting mixture 80 M of Ni(II) or Zn(II) was added

that was allowed to equilibrate for 24 h .

Cu4ISlyD + M

II + 2 Bca

2- ⇄ M

IISlyD + 4Cu

I(Bca)2

3- (F)

As expected, the addition of Bca to the SlyD sample containing 5 molar excess of Cu(I) leads to

an increase in the absorption at 562 nm (data not shown). The absorbance corresponds to a

concentration of 10 M CuI(Bca)2

3-, because SlyD can only bind 40 M of the added Cu(I) in

0

0.1

0.2

0.3

0.4

0.5

0.6

0 0.4 0.8 1.2 1.6 2

Ab

so

rba

nce

48

3 n

m

[SlyD]/[Cu(I)]

78

the presence of this chelator (discussed above and see figure 3-11). The ability of either Ni(II) or

Zn(II) to replace the bound Cu(I) was then measured indirectly in these competition experiments,

because the displaced Cu(I) will form a complex with the Bca leading to an increased absorption

at 562 nm. Excess Ni(II) or Zn (II) did not change the signal at 562 nm, demonstrating that the

forward reaction in (F) did not occur under these experimental conditions (data not shown).

Control experiments were conducted with Bca and Ni(II) or Zn(II) alone to ensure that addition

of these metals did not alter the spectrum of Bca. To determine if the reverse reaction could be

achieved, a 10 M protein sample was pre-incubated with 8 equivalents of Ni(II) or Zn(II) for 24

h, followed by addition of 5 equivalents of Cu(I) and 200 M Bca and equilibration for an

additional 24 h. The resulting absorbance of these samples at 562 nm corresponded to only 10

M CuI(Bca)2

3- suggesting that the rest of the 40 M Cu(I) was bound by SlyD (data not

shown). Therefore, these results support tight binding of 4 Cu(I) ions even in the presence of

Ni(II) or Zn(II). This preferential binding of Cu(I) corroborates the thermodynamic parameters

measured for the different metals thus far, where KD (Cu(I)-SlyD) < KD (Ni(II)-SlyD), KD (Zn-SlyD).

Co(II) binding to SlyD.

Although cobalt is not widely used by E. coli this transition metal ion can still be found in low

abundance in the cytosol of this bacteria (13). In addition, Co(II) was of particular interest to us

because two other proteins involved in nickel homeostasis, RcnR and RcnA are known to also

bind and respond to cobalt in vivo (36). To observe whether SlyD could coordinate Co(II), a

direct cobalt titration of apo SlyD was analyzed via electronic absorption spectroscopy. The

resulting spectra indicate that addition of Co(II) leads to an increase in the absorption at 280 nm

accompanied by a broad shoulder in the 300-390 nm region that saturates upon addition of 3-4

equivalents of the metal (Figure 3-13). The absorption spectrum of the Co(II)-saturated SlyD

sample reveals the presence of an LMCT band at 293 nm (c = 7800 M-1

cm-1

) and a broader

charge transfer band centred around 360 nm (c = 3418 M-1

cm-1

) (the linear increase in

absorption as a result of increasing Co(II) concentrations was used to determine the molar

absorptivity of the metal-protein complex). These near UV-Vis bands have been noted for

several proteins that chelate Co(II) in vitro, and are attributed to ligand-to-metal charge transfer

from thiolates in Cys residues to the metal centre (36, 37). This assignment of bands is

supported by the lack of such a CT band when Co(II) is added to the triple mutant (data not

79

shown). The intensity of these near UV bands can be used to estimate the number of S-Co(II)

interactions with an expected 900-1200 M-1

cm-1

per S-Co(II). Based on the extinction

coefficient of 7800 M-1

cm-1

calculated for SlyD it is likely all six Cys residue contribute to

cobalt coordination.

Figure 3-13. Co(II) binding to SlyD. Difference spectrum generated by subtracting apo SlyD (25 M)

from a protein sample incubated with 7 equivalents of Co(II). Inset: Plot of molar absorptivity at 320 nm

vs. equivalents of Co(II) indicates saturation around 3-4 equivalents of Co(II) added.

The visible regions of Co(II) complexes typically have bands due to d-d transitions that are

diagnostic of the coordination environment of the metal within the protein (38). As the molar

absorption coefficient for these bands are expected to be ten-fold lower than the LMCTs (38), a

300 M protein sample was titrated with Co(II) under anaerobic conditions. Titration with

increasing amounts of cobalt resulted in an absorption band centred at 650 nm that exhibit

maximum absorbance upon addition of 2.5 equivalents of metal (Figure 3-14). Extinction

coefficients for Co(II) d-d transitions are useful in determining the geometry of the metal site

such that four coordinate tetrahedral sites exhibit d-d bands with > 300 M-1

cm-1

, five-

coordinate ion has 50 M-1

cm-1

< < 300 M-1

cm-1

and an octahedral site exhibit transition bands

with < 50 M-1

cm-1

(39). An extinction coefficient of 235 M-1

cm-1

was obtained for SlyD

(based on the linear region of the titration curve, with up to 2 equivalents of Co(II) added)

suggestive of formation of predominantly five coordinate Co(II) centres in SlyD. Further cobalt

0

0.1

0.2

0.3

0.4

0.5

240 340 440 540 640 740

Ab

so

rba

nce

Wavelength (nm)

0

0.002

0.004

0.006

0.008

0 2 4 6 8 10

Norm

aliz

ed

Δ

Abs

32

0(

M-1

)

Equivalents of Co(II)

80

addition is accompanied by blue shifting of these spectra with subsequent decrease in the A650

signal that plateaus around 4-5 equivalents of Co(II), indicating changes in the average

coordination sphere at higher cobalt concentrations (Figure 3-14).

Figure 3-14. Resulting d-d transitions upon Co(II) binding to SlyD. (A) The addition of Co(II) to a

300 M SlyD sample leads to the appearance of d-d transition bands centred around 650 nm. (B) A plot

of the change in absorbance at 650 nm versus the total concentration of cobalt indicates that maximum

absorbance is achieved upon addition of ~3 equivalents of Co(II). Co(II) concentrations beyond 3

equivalents leads to blue shifting and significant changes in the absorption profile (dashed lines in (A))

that plateaus around 4-5 equivalents of cobalt ions added (B).

To evaluate Co(II) binding in more depth, the binding of this metal to the protein was further

analyzed by ESI-MS. Similar to copper titrations mentioned above, the addition of cobalt ions

led to a concomitant decrease in mass of the protein by 6 Da and the appearance of dimer species

(data not shown), indicating the oxidation of the protein. Therefore, titrations were carried out in

the presence of 100 M DTT. Mass spectrometry data revealed that similar to zinc and nickel

ions, Co(II) binding by the protein also occurred through a non-cooperative metal-binding

mechanism as mixtures of metallated species were detected at all Co(II) concentrations tested

(Figure 3-15). At a given amount of Co(II), less Co(II)-SlyD species were observed for samples

containing DTT than samples without DTT indicating that the reducing agent was competing

with SlyD for cobalt ion binding (data not shown). Therefore, the stoichiometry of the protein

for Co(II) was established through equilibrium dialysis which indicated an average of 4 Co(II)

per monomer, a capacity analogous to that of nickel ions for SlyD (Table 3-3).

0

0.0002

0.0004

0.0006

0.0008

0.0 2.0 4.0 6.0 8.0 10.0N

orm

aliz

ed A

bs

650

(M

-1)

[Co(II)]/SlyD

(B)

0

0.05

0.1

0.15

0.2

0.25

0.3

400 500 600 700 800 900

Absorb

ance

Wavelength (nm)

(A)

81

Figure 3-15. ESI-MS titration of WT SlyD with Co(II) in the presence of 100 M DTT. Addition of

increasing amounts of Co(II) to SlyD (10 M) leads to concurrent filling of multiple metal sites indicating

a non-cooperative metal-binding mechanism. From top to bottom the equivalents of Co(II) added relative

to protein concentration are as follows: 0, 1, 2.5, 3.5, 4.4, 7 The m/z spectra are shown on the left with

their respective reconstructed spectra on the right panel. The numbers heading the dotted-line indicate the

amount of Co(II) bound to SlyD (right panels) and the charge states of the protein (left panels). The

amount of Co(II)-SlyD appears to be less than the total Co(II) added to the solution when DTT is included

in the protein buffer, suggesting the reducing agent competes with SlyD for cobalt coordination.

Mass (Da)

20800 21200 21600

Rel

ativ

e in

tens

ity

+16 +14 +12 +10 +9 +8 0 1 2 3 4 5 6

1600 2000 2400 2800

m/z

82

Table 3-3. Metal stoichiometry of WT SlyD†.

Metal Ni (II) Mn (II) Co(II)

Mn(II) 0.2 ± 0.1

Co(II) 3.9 ± 0.1

Ni(II) 4.2 ± 0.3

Ni(II) & Mn(II) 4.0 ± 0.2 n/d*

Ni(II) & Co(II) 4.4 ± 0.1 n/d*

†Stoichiometries were determined by using equilibrium dialysis in which 40 M apo SlyD samples were dialyzed

against an equal volume of 320 M metal each, over night at 4 oC in an anaerobic glovebox. *n/d, a protein-metal

complex for the particular metal was not detected.

To establish the affinity of SlyD for Co(II), competition experiments with the fluorescence

indicator Fura-2 were performed. Although Fura-2 was originally designed as a molecular

sensor for Ca(II), this chelator has been previously used for determining the affinity of several

proteins for various transition metals including Co(II) (19, 37). The chelator forms a 1:1

complex with Co(II) that can be visualized by the decrease in absorbance at 368 nm.

Competitive conditions required for determining the dissociation constant of SlyD for cobalt ions

were achieved by titrating apo protein into a constant concentration of [Co(II)-Fura2], leading to

an increase in the signal at 368 nm (indicative of a decreasing [Co(II)-Fura2]- in solution) (Figure

3-16). To determine the affinity, data with a fractional saturation between 0.8-0.2 (with respect

to the Fura-2) from several replicates were compiled to obtain an average KD (SlyD-Co(II)) of 4 ± 1 x

10-9

M. It should be noted that we used the KD of 8.64 x 10-9

M reported for the [Co(II)-Fura2]

complex at pH 7.0 (40), as the affinity of this chelator to metal is constant near neutral pH (41).

As both Co(II)- and Ni(II)-bound SlyD give rise to spectra with transitions in the near UV-

Visible region, metal selectivity of the protein was established by using equilibrium dialysis.

Analysis of samples equilibrated with a solution containing 8-fold excesses of Co(II) and Ni(II)

revealed that the protein was exclusively bound to nickel with an average stoichiometry of 4.4

(Table 3-3). As SlyD binds nickel with an average nickel stoichiometry of ~4 when equilibrated

with Ni(II) alone (Table 3-3 and (10)) these data indicate that SlyD selectively coordinates nickel

to its’ full capacity in the presence of Co(II). Specificity of the protein for Co(II) versus Zn(II)

was established by taking advantage of the absence of LMCT bands when Zn(II) chelates to

SlyD. Incubation of SlyD with 7 mol equivalents of Zn(II) for 24 h followed by addition of

increasing amounts of Co(II) that was allowed to equilibrate further up to 72 h failed to generate

83

any CT bands (data not shown). Thus, these results indicate that SlyD binds Zn(II) preferentially

over cobalt as expected given the measured affinity of SlyD for the individual metals. Similarly,

as Zn(II) was titrated into a sample of SlyD pre-incubated with 8-fold excess of Co(II) the

LMCT band at 320 nm decreased in intensity, and disappeared completely at 6 equivalents of

Zn(II) added indicating effective replacement of Co(II) with Zn(II) at metal sites composed of

Cys ligands (Figure 3-17).

Figure 3-16. Affinity of SlyD to Co(II) via Fura-2 competition. Titration of apo SlyD into a solution

of Co(II) (10 M) and Fura-2 (35 M) leads to a decrease in the amount of [Co(II)-Fura2]2-

in solution

that is detected by the absorbance at 368 nm. From competition titrations such as that shown an apparent

KD (Co(II)-SlyD) of 4 ± 1 x 10-9

M was obtained.

0

0.2

0.4

0.6

0.8

1

0 4 8 12 16 20 24

Fra

ction o

f [C

o(I

I)-F

ura

2]

com

ple

x

[SlyD]Total (M)

84

Figure 3-17. Zn(II) can replace Co(II) bound to SlyD. Addition of increasing amounts of Zn(II) to a

SlyD sample (25 M) saturated with Co(II) (200 M) leads to a decrease in the signal at 320 nm,

indicating that Zn(II) can effectively replace all the Co(II) from metal sites composed of thiolate ligands.

Fe(II) binding to SlyD.

Given the high abundance of the E. coli iron quota, ~ 0.1 mM found in the cytosol (13), and the

propensity of this metal to coordinate to Cys residues, Fe(II) binding to SlyD was also

investigated. Electronic absorption spectroscopic analysis of SlyD protein samples mixed with

increasing amounts of Fe(II) did not reveal any changes in the spectrum (data not shown). As

Fe(II) binding to thiolate ligands is expected to generate LMCT bands in the visible region of an

absorption spectrum (42), the lack of such CT bands were indicative of the fact that SlyD did not

bind Fe(II). To establish that Ni(II) could effectively bind to SlyD even in the presence of

Fe(II), a protein sample pre-incubated with 10 molar excess of Fe(II) was used for a Ni(II)

titration. Ni(II) binding to SlyD was unaffected as the binding curve was comparable to that

obtained from incubation of SlyD with Ni(II) alone (Figure 3-18).

0

0.002

0.004

0.006

0.008

0 2 4 6 8 10 12

Norm

aliz

ed

Δ

Abs

320

(M

-1)

Equivalents of Zn(II)

85

Figure 3-18. Selectivity of SlyD for Ni(II) vs. Fe(II). The change in absorbance at 315 nm upon

titration of apo SlyD (25 M) with Ni(II) (crosses) is shown. A similar titration curve is obtained when

adding Ni(II) to a SlyD sample (25 M) pre-incubated with 250 M Fe(II) (diamonds), indicating that

Ni(II) coordination to the protein is unperturbed in the presence of excess Fe(II).

Mn(II) binding to SlyD.

Considering the large number of carboxylate moieties and His residues present in SlyD, Mn(II) is

another metal ion that could be coordinated by this protein. The addition of Mn(II) to SlyD did

not produce any change in the absorption profile of the protein (data not shown), however

equilibrium dialysis indicated that a very small fraction of the protein was bound to the metal, i.e.

0.2 equivalents of metal bound (Table 3-3). Therefore to determine whether manganese could

affect the Ni(II)-binding activity of SlyD, a Ni(II) titration was carried out using a SlyD sample

that was first incubated with 8 molar excess Mn(II). Overlay of the results from this titration

with a control experiment conducted with nickel alone yielded similar binding curves indicating

that Ni(II) can effectively compete with Mn(II) (data not shown). The selectivity of the protein

for Mn(II) vs. Ni(II) was evaluated by a separate method, in which the protein was allowed to

0

0.05

0.1

0.15

0.2

0.25

0.3

0 2 4 6 8 10

ΔA

bso

rban

ce a

t3

15

nm

Equivalents of Ni(II)

86

equilibrate with a metal solution containing 8 molar excesses of Ni(II) as well as Mn(II).

Analysis of the metal content of this sample revealed SlyD exclusively bound to nickel, further

corroborating the selective binding of nickel over manganese ions (Table 3-3).

In vivo metal selectivity of SlyD.

(A) Susceptibility of E. coli to external metal concentrations.

Given that SlyD can coordinate several different types of metal ions in vitro with considerable

affinity, it is feasible that SlyD plays a role in maintaining optimal metal levels in E. coli by

functioning as a general metal buffer/detoxifier. To test this hypothesis, the growth rates of wild

type and ΔslyD strains were examined by measuring optical density at different metal

concentrations in both LB and minimal media. During these investigations we noted that when

using minimal media, metal concentrations > 0.7 mM could not be investigated due to significant

amounts of precipitation. Furthermore, the maximum external metal concentrations that could be

used without severely inhibiting the growth of either bacterial strain were much lower in minimal

media compared to LB media, signifying that the components in the rich media influence the

available metal concentrations (Table 3-1). Nevertheless, no significant differences in the

growth rates were observed for the ΔslyD strain compared to wild type at any of the metal

concentrations tested in either LB or minimal media under aerobic conditions (Figure 3-19 and

data not shown) suggesting that SlyD is not essential to confer metal resistance against cobalt,

nickel, copper or zinc in E. coli under aerobic growth conditions.

87

Figure 3-19. Sensitivity of E. coli to transition metals. Representative data set for the effect on the

growth rate of a wildtype and a ΔslyD strain upon addition of various metal concentrations detected via

OD600. These experiments were conducted in minimal media supplemented with the following amounts

of metal: (A) 10 M Ni(II), (B) 30M Ni(II), (C) 50 M Cu(II), (D) 250 M Cu(II), (E) 150 M Zn(II),

(F) 500 M Zn(II). The black lines correspond to the wildtype and the coloured lines indicate the growth

detected for the ΔslyD strain. The solid lines and the dashed lines represent growth detected in untreated

and treated cultures, respectively.

0

0.3

0.6

0.9

1.2

0 2 4 6 8 10

OD

600

Time (h)

0

0.3

0.6

0.9

1.2

0 2 4 6 8 10

OD

600

Time (h)

(A) (B)

0

0.4

0.8

1.2

1.6

0 2 4 6 8 10

OD

600

Time (h)

0

0.4

0.8

1.2

1.6

0 2 4 6 8 10

OD

600

Time (h)

(C) (D)

0

0.5

1

1.5

2

2.5

0 2 4 6 8 10

OD

600

Time (h)

0

0.5

1

1.5

2

2.5

0 2 4 6 8 10

OD

600

Time (h)

(E) (F)

88

(B) Transcriptional response of metal transporters.

To further assess the potential role of SlyD in metal homeostasis, we examined the transcription

of several metal transporters in E. coli using quantitative PCR (qRT-PCR). The basis for

monitoring mRNA levels of transporters stems from the fact that the expression of these proteins

are tightly regulated by metal sensing transcription factors that gauge the concentration of a

particular metal with extreme selectivity (43). A comparison of the mRNA levels of the

transporters between wild type strain and ∆slyD strains should enable us to determine the impact

of SlyD on a given metal homeostasis pathway.

Thus far, nickel is the only metal ion connected to the in vivo function of SlyD (21). To test

whether deletion of SlyD could lead to perturbation of the nickel ion balance in E. coli, the

expression levels of the ABC type nickel importer NikABCDE and the nickel exporter RcnA

were evaluated by monitoring the mRNA levels of nikA and rcnA, respectively. While the

expression of rcnA was induced when cells were treated with 10 M Ni(II) as expected, we

observed no difference between transcript levels in ∆slyD compared to wild type cells in

aerobically-grown bacteria (Figure 3-20). We next examined the transcript levels of nikA and

rcnA in anaerobically grown E. coli with or without 10 M Ni(II). Similar to the observations

made under aerobic conditions, while a substantial induction in rcnA expression was observed

upon metal addition, the change in mRNA levels in the wild type and ΔslyD strains were

comparable, indicating that SlyD does not affect the nickel exporting pathway (Figure 3-21). In

contrast, a ~3 fold difference in the nikA mRNA level was detected in the ΔslyD strain when

compared to wild type without nickel supplementations of the media (Figure 3-21). As

anticipated, upon addition of 10 M Ni(II) a ~10 fold and a ~8 fold reduction in the expression

level of nikA was observed for the wild type and ΔslyD strains respectively (Figure 3-21). When

considering the mRNA levels of the nickel-treated cells the wild type contains ~2 fold more nikA

transcript than the deletion strain. Therefore, these diminished nikA transcript levels in the ΔslyD

strain indicate an impact on the nickel uptake pathway upon deletion of this gene. To test

whether SlyD contributes to nickel homeostasis specifically even in the presence of other metals,

nikA and rcnA transcript levels were monitored in anaerobically-grown bacteria treated for 30

min with 10 M Ni(II) as well as 400 M Zn(II), a concentration high enough to elicit a response

in zinc homeostasis (see below). The transcript levels for the two transporters in these growths

followed a similar trend to that observed when bacteria were treated with nickel alone, thereby

89

indicating that SlyD is able to specifically contribute to the maintenance of nickel ions even in

the presence of excess zinc in vivo (Figure 3-22).

Figure 3-20. Expression profiles of zinc, copper and nickel transporters in aerobic growth. Relative

mRNA levels detected upon addition of 10 M Zn(Ni) (Panel A), 400 M Zn (II) (Panel B-C) and 150 M

Cu(II) (Panel D-E). The bars denoted –Zn(II)/Cu(II) and Ni(II) correspond to bacteria cultured in minimal

media without added metal. The mRNA detected for each transcript in the wildtype and ΔslyD strain under

identical growth conditions are shown in solid and clear bars, respectively. The reported relative mRNA levels

were normalized with respect to holB which encodes a DNA gyrase. A p-value > 0.05 was calculated for the

transcript levels of the wildtype and ΔslyD strain under each growth condition, indicating no significant

difference between the two strains.

Figure 3-21. Expression of Ni(II) transporters in response to Ni(II). Change in mRNA levels from

anaerobically-cultured bacteria upon treatment with 10 M Ni(II). The bars denoted -Ni(II) indicated the

transcripts detected for bacteria grown without extra Ni(II) added to the media. The amount of mRNA

observed in wildtype and ΔslyD strains are shown in black and clear bars respectively. The observed

transcripts levels are relative to that of the DNA gyrase, holB. The p-value of < 0.05 indicates the values are

significantly different and p-values > 0.05 are considered not to be significantly different.

0

0.5

1

1.5

2

2.5

3

-Zn(II) +Zn(II)

0

4

8

12

16

20

-Cu(II) +Cu(II)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

-Zn(II) +Zn(II)

0

0.5

1

1.5

2

2.5

-Ni(II) +Ni(II)

rcnA znuA zntA copA

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

-Cu(II) +Cu(II)

cusF

Re

lati

ve

mR

NA

le

ve

ls

(A) (B) (C) (D) (E)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

-Ni(II) +Ni(II)

p < 0.05

p < 0.05

0

5

10

15

20

25

30

35

-Ni(II) +Ni(II)

Rela

tive

mR

NA

leve

ls

nikA rcnA

p > 0.05

p > 0.05

90

Figure 3-22. Expression the Ni(II) transporters in the presence of excess Zn(II). Relative change in

mRNA levels of nikA and rcnA detected upon addition of 10 M Ni(II) and 400 M Zn(II). Bacteria

were grown in media optimized for hydrogenase expression under anaerobic conditions. The bars

denoted –Ni(II) correspond to bacteria cultured in media without added metal. The mRNA detected for

each transcript in the wildtype and ΔslyD strain under identical growth conditions are shown in solid and

clear bars respectively. The reported relative mRNA levels were normalized with respect to holB which

encodes a DNA gyrase. The p-values < 0.05 indicate the data sets are significantly different whereas p-

values > 0.05 are considered to be not significantly different.

To determine whether SlyD can influence zinc homeostasis transcription of two zinc-responsive

genes were monitored upon addition of 400 M ZnSO4. The two transcripts monitored

correspond to znuA, which encodes a subunit of the major inducible high-affinity Zn(II) uptake

ABC transporter system, and zntA, which encodes a Zn(II) exporter from the P1-type ATPase

family (44). The concentration of zinc was chosen based on the toxicity experiments, which

revealed that addition of 400 M zinc results in slower growth (as opposed a lethal concentration

leading to no growth). Furthermore, several previous investigations assessing the genome-wide

transcriptional response of E. coli to zinc have demonstrated that [Zn(II)] >100 M is sufficient

to observe a change in the transcript levels of the above-mentioned transporters, albeit under

different media compositions (45, 46). While the addition of zinc led to an increase in zntA

transcription and a decrease in znuA transcription as expected we observed no significant

difference between wild type and ∆slyD strains for these transcripts (Figure 3-20). Similarly, no

difference was detected between wild type and ΔslyD in the absence of oxygen (Figure 3-23).

0

1.5

3

4.5

6

7.5

9

-Ni(II) +Ni(II)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

-Ni(II) +Ni(II)

Re

lative

mR

NA

le

ve

ls

nikA rcnA

p > 0.05

p < 0.05

p > 0.05

p < 0.05

91

Figure 3-23. Expression profiles of zinc and copper transporter in anaerobic growth. Relative

mRNA levels detected upon addition of 400 M Zn(II) (Panels A-B) and 150 M Cu(II) (Panel C-D).

The bars denoted –Zn(II)/Cu(II) correspond to bacteria cultured in minimal media without added metal.

The mRNA detected for each transcript in the wildtype and ΔslyD strain under identical growth

conditions are shown in solid and clear bars respectively. The reported relative mRNA levels were

normalized with respect to holB which encodes a DNA gyrase. A p-value > 0.05 was calculated for

wildtype and ΔslyD strain under each growth conditions indicating no significant difference between the

two strains.

We next tested the involvement of SlyD in copper homeostasis, by examining copA and cusF

transcripts under high Cu(II) levels by addition of 150 M Cu(II) to the media in the presence of

molecular oxygen. Similar to zinc, this particular copper concentration was also selected based

on our toxicity studies that indicated 150 M is sufficient to reduce the growth rate of the

bacteria. In addition, previous investigations on copper toxicity in E. coli by Outten et al.

revealed that CuSO4 above 100 M is sufficient to cause extreme stress levels (47). At these

high copper concentrations the primary copper exporter CopA as well as the CusCFBA system

that exports copper across both membranes to the extracellular space are necessary to render

copper tolerance to the bacteria (47). While the expression of both exporters increase in the

presence of copper, copA transcripts are induced to a higher level compared to cusF in the

presence of oxygen (Figure 3-20). In contrast, in anaerobically grown bacteria a more dramatic

increase in mRNA levels was detected for cusF relative to copA (Figure 3-23), confirming the

differential expression of the two exporters in response to molecular oxygen observed previously

(47). Although the expected responses were detected upon treatment with copper, the fold

0.0

0.3

0.6

0.9

1.2

1.5

-Zn(II) +Zn(II)

0

20

40

60

80

-Cu(II) +Cu(II)

0.0

1.5

3.0

4.5

6.0

7.5

9.0

-Zn(II) +Zn(II)

znuA zntA copA cusFR

ela

tive

mR

NA

le

ve

ls

0

0.3

0.6

0.9

1.2

1.5

1.8

-Cu(II) +Cu(II)

(A) (B) (C) (D)

92

increase in mRNA levels were similar in both the wildtype and ΔslyD strains indicating that

deletion of SlyD did not lead to perturbation of the copper homeostasis in E. coli (Figure 3-20

and 3-23).

Discussion

An understanding of how proteins detect and coordinate the correct metal ion is an important

issue when considering proteins involved in metal homeostasis. In this investigation, metal-

binding activity and metal-selectivity of SlyD, a protein involved in maintaining the balance of

nickel ions in E. coli, was examined (48). Among the metals tested, SlyD was capable of

binding Zn(II), Cu(I), and Co(II) in vitro whereas binding of Mn(II) or Fe(II) to the protein was

not detected. Given the diverse array of potential metal-binding residues present in the MBD of

SlyD, it is not surprising that the protein can coordinate many of the first row transition metals in

addition to Ni(II), which was described previously (10). In addition to the non-discriminating

primary structure, the lack of specificity is likely also due to the highly malleable conformation

of the MBD of the protein, supported by the fact that the structure of this region cannot be

resolved from NMR measurements even upon addition of 1 equivalent of Ni(II) (5).

Zn(II) ions bind to SlyD in a progressive and overlapping fashion similar to that observed for

Ni(II), indicating that zinc binding is similarly non-cooperative. Competition titrations indicate

that SlyD has five zinc-binding sites and can chelate this metal with an apparent KD of 10-10

M.

The CD spectroscopy, together with charge state distribution detected by mass spectrometry is

indicative of small but distinctive changes in the protein structure. Although Ni(II) also leads to

structural changes in SlyD, the overall shape of the CD spectrum for Zn(II)-bound SlyD is

different. In addition to a lower metal-binding capacity when compared to Ni(II), the amount of

secondary structure formed appears to be less when binding zinc ions. Therefore it is possible

that the two metal ions are sequestered utilizing distinct metal sites in SlyD that result in

different protein folds. This interpretation is supported by the competition titrations in which the

addition of Ni(II) together with Zn(II) only resulted in a decrease in the intensity of nickel

LMCT band, as opposed to complete elimination of the band, at least at the metal concentrations

tested. The ability of zinc to partially replace nickel and vice versa is additional evidence for

some “Ni-specific” as well as “Zn-specific” sites. Interestingly, when considering nickel and

zinc ions, the order in which the metal is added to the protein appears to govern the final

93

metallation state of the protein. Therefore, it appears that binding of the first metal may lock the

protein in to a specific conformation such that only a few sites are then accessible for metal

exchange reactions, further strengthening the distinct folds adapted by SlyD for Ni(II) vs Zn(II).

Changes in protein structure, also termed allostery, is one of the mechanisms widely used by

proteins to overcome the thermodynamically dictated metal specificity (2, 43). In this event, the

binding of the “correct metal” is exploited to drive the changes in the

secondary/tertiary/quaternary structure of the protein and or dynamics (i.e. protein-protein

interactions) that ultimately results in a biological function. In the case of SlyD, the difference

in the fold of the protein upon Ni(II) vs. Zn(II) binding may allow the protein to discriminate

between the two metal ions and produce the desired biological function of SlyD in vivo only

when the correct metal is bound.

Table 3-4. KD of SlyD-Metal complexes.§

Metal KMe (M)

Co(II) 4 ± 1 x 10-9 a

Ni(II) 10-10 b

Cu(I) 1.5 ± 0.3 x 10-17 c

Zn(II) 1.0 ± 0.1 x 10-10 d

§Data were treated assuming binding of a single metal ion to obtain a macroscopic KD.

aDetermined via competition

with Fura-2. bEstimated from competition with PAR, Fura-2 and Zn-SlyD.

cDetermined by competition with Bcs.

dCalculated from competition experiments with PAR.

In addition to zinc, SlyD can bind up to 4 Cu(I) ions with an apparent affinity of ~10-17

M, the

tightest detected for all the metals tested. Unlike Ni(II) and Zn(II), which are able to bind to the

protein concurrently, Cu(I) can reach its full capacity of 4 metal ions per SlyD monomer even

when the protein is pre-loaded with Ni(II) or Zn(II). These results are not surprising considering

that the average KD measured for Cu(I) is seven orders of magnitude tighter than that observed

for zinc and nickel ions. Thus, the specificity of SlyD towards Cu(I) appears to be dictated by the

relative thermodynamic parameters (i.e. KMe) measured. In addition to the extremely high

affinity, copper is further distinguished by the fact that binding of this metal to SlyD appears to

be partially cooperative, which may be another factor that enables the protein to bind Cu(I)

selectively over nickel or zinc. SlyD can also bind to cobalt ions with an average stoichiometry

of ~4 and from competition experiments an apparent KD of ~ 10-9

M was detected. However,

when considering the selectivity of the protein with respect to Co(II) the trend is opposite to that

94

observed for Cu(I). For instance, addition of Zn(II) to a cobalt-saturated sample yields SlyD that

is solely chelated by Zn(II) and similarly when protein is presented with equimolar solutions of

nickel and cobalt only Ni(II)-bound SlyD can be isolated. Therefore, in agreement with the

measured KMe, the selectivity data clarify that in the presence of either nickel or zinc the much

weaker Co(II)-SlyD complexes are not formed. Based on these in vitro metal-binding studies

conducted on SlyD, the preference for metal ions can be ordered as follows, Mn(II), Fe(II) <

Co(II) < Ni(II), Zn(II) << Cu(I). This series resembles the natural order of stabilities for

biologically essential divalent metals, commonly termed the Irving-Williams series (49) which

predicts that copper, zinc and nickel will form the tightest complexes, followed by cobalt, ferrous

iron and manganese and finally calcium and magnesium forming the weakest complexes. From

these results it is evident that the strength of the metal binding is dictated by the properties of the

metals rather than the protein and that SlyD is sufficiently flexible to be able to accommodate the

coordination preference of each of these metal ions.

It is well known that due to their flexible nature, many proteins can accommodate a variety of

different metals in their metal-binding site(s). Several examples for this trend can be found by

just considering some Ni(II) homeostatic proteins. For example UreE, a chaperone dedicated to

shuttling Ni(II) for urease biosynthesis can bind Co(II), Cu(II) as well as Zn(II). Not only do

these metals bind to the protein they can effectively compete with binding of Ni(II) ions (50).

Through extensive spectroscopic investigations it was shown that although UreE can bind

several transition metals, the geometry and the coordination sphere utilized by the protein to

coordinate each metal is distinct (50). Therefore, only Ni(II) binding is expected to elicit the

correct fold of UreE, which subsequently enables the protein to interact with the partner proteins

involved in urease biosynthesis. NikR from E. coli can chelate several other divalent metal ions

in vitro (i.e.. Co(II), Cu(II), Zn(II), Cd(II)) in addition to Ni(II). Similarly to SlyD, the affinity of

NikR for the first row metals tested also followed the Irving-Williams series (37). Spectroscopic

studies revealed that like UreE, NikR utilizes a unique geometry and coordination sphere for

each metal (51), chemical and thermal denaturation data along with in vitro DNA-binding assays

provided evidence for a Ni(II)-selective enhancement in the stability and activity of the protein

(37, 52). Therefore, binding of Ni(II) alone is considered to provide a favourable protein

conformation that allows NikR to achieve its biological function, as has been verified in vivo

(37, 51). Additional examples are found when considering the proteins involved in maintaining

95

nickel homeostasis in Helicobacter pylori, such as the putative Ni(II) storage protein, Hpn. For

this small histidine-rich protein, even though an in vitro preference of Cu(II) > Ni(II) > Zn(II)

was observed a preferential effect of nickel was detected during in vivo metal toxicity studies

(53).

Although conclusive in vitro evidence for a Ni(II)-specific conformation of the protein such as

that described for NikR and UreE is currently lacking for SlyD, the specificity of the protein can

be inferred from in vivo studies. The Ni(II) homeostasis pathway in E. coli is complex, and

there is evidence that there are two distinct pools of nickel that are maintained by two different

transcription factors, RcnR and NikR (54). One of the nickel pools is connected to the exporter,

RcnA, which is de-repressed by the transcription factor RcnR in response to elevated Ni(II) and

this response is observed both aerobically and anaerobically (54). While a specific Ni(II) uptake

system is not known to be expressed under aerobic conditions, nickel can still gain access to the

cytoplasm through non-specific metal importers such CorA that are unaffected by the presence of

oxygen (55). However, a specific nickel uptake system, the Nik transporter, is expressed under

anoxic conditions when the demand for nickel is higher due to the expression of the [NiFe]

hydrogenase enzymes (56). As a result a second nickel pool is established to provide nickel for

enzyme maturation and is connected to the activity NikR (55, 56). The role of SlyD in nickel

homeostasis was analyzed via monitoring the transcript levels of the nickel importer nikA and the

exporter rcnA. The results obtained for these two importers provide several insights into nickel

homeostasis in E. coli. Firstly, deletion of slyD reduced the nickel capacity of the cytoplasm

such that nikA transcription was down regulated in the absence of the protein. Secondly, deletion

of slyD did not impact the exporter expression, indicating that the protein does contribute to the

nickel exporting pathway. Furthermore the in vivo data suggest that SlyD contributes only to the

nickel pool maintained by the Nik importer. The nickel brought into the cytoplasm through the

Nik importer is considered to be preferentially routed to [NiFe]-hydrogenase maturation through

the action of metallochaperones. Therefore, these findings corroborate the nickel-related

function of SlyD in enzyme maturation, and strengthen the assignment of this protein as a nickel

metallochaperone.

Results from the toxicity studies clearly indicate that although the protein is capable of

coordinating several types of transition metals, SlyD does not function as a general metal

detoxifier like some other nickel storage factors such as hpn or hpn-like proteins (53, 57). In

96

addition, deletion of SlyD did not lead to a significant change in the transcription profiles of the

copper or zinc transporter suggesting that although the protein could not preferentially coordinate

nickel in vitro, its activity is specific for Ni(II) in vivo. Therefore SlyD is yet another protein

that can be included in the growing repertoire of proteins for which thermodynamically dictated

affinities in vitro cannot be extrapolated to in vivo function.

The mis-matched metal selectivity of SlyD in vitro versus in vivo raises the question as to how

SlyD is able to overcome the thermodynamically-dictated metal specificity within the cell under

normal growth conditions. It is now clear that the effective intracellular concentrations of metal

ions are tightly controlled within a cell. In E. coli, copper homeostasis is regulated by the copper

sensor CueR that has an affinity for copper in the zeptomolar range (10-21

M) (58). This

estimated affinity of the regulator implies that all cellular copper is tightly bound and buffered

within the cytosol. A similar scenario is observed for zinc as well, where the Zn(II) sensors Zur

and ZntR in E.coli, bind this metal with Kd of 10-15

M (23). This tight regulation of the

cytoplasmic metal concentrations suggests virtually no free metal ions of Cu(I) or Zn(II), thus

SlyD may not be able to gain access to these metal ions unless the protein can interact with a

metal importer directly. As Cu(I) and Zn(II) are the only metals that can disrupt the Ni(II)

binding capacity of SlyD the limited availability of these metals will likely leave the protein free

to bind Ni(II) and function in nickel homeostasis in E. coli. This concept is supported by the fact

that when excess Zn(II) was added along with Ni(II), the nikA mRNA levels were similar to that

observed for Ni(II) alone. Thus, the data imply even under toxic metal conditions when zinc

homeostasis is disrupted, SlyD is capable of retaining its Ni(II)-specific function in vivo. The

fact that the deletion of slyD did not change the transcript levels of the copper and zinc

transporters under toxic metal levels is additional evidence that supports the Ni(II)-specific role

of SlyD in E. coli.

Similar to copper and zinc, nickel concentrations are thought to be tightly buffered to less than

one free nickel ion within the E. coli by the previously mentioned RcnR and NikR (54). Then

how is SlyD able to bind nickel in vivo and contribute to the nickel homeostasis? NikR has two

nickel-binding sites, with affinities of 10-12

and 10-7

M (55). Nickel binding to both sites allows

the regulator to bind to DNA with a tighter affinity than when only a single site is occupied by

nickel, indicating two levels of gene regulation at different metal concentrations. If one were to

follow the above argument presented for Cu(I) and Zn(II) ion regulation, then SlyD that chelates

97

nickel with ~10-10

M affinity would not be able to compete with NikR for the high affinity metal

sites. However, it is possible that the protein is able to sequester enough metal for storage and/or

transfer before the low affinity site of NikR is filled and transcription of the nickel importer is

completely turned off. On the other hand, there may be no such competition for nickel if SlyD

can directly interact with the Nik transporter. Nickel transported via the Nik system is directed

to the biosynthesis of the [NiFe]-hydrogenase as deletion of the importer abrogates the enzyme

activity (56). In addition it has been shown that there is no competition between NikR and nickel

transported for metallocentre assembly suggesting that metallochaperones sequester nickel for

hydrogenase maturation at the transporter (54, 56). Given that SlyD is known to form a protein-

protein complex with the nickel chaperone HypB (21, 59), SlyD may also gain access to nickel

via ligand exchange reactions at the transporter. An interaction of SlyD or the SlyD-HypB

complex with the Nik transporter would circumvent the thermodynamically dictated metal

affinities of the proteins, and ensure the population of the metallochaperones with nickel alone

through kinetic means by relying on protein-protein interactions. Further experiments involving

cross-linking and co-localization are necessary to determine whether such complexes are indeed

formed in vivo.

In addition to thermodynamic gradients several strategies are utilized by cells to overcome the

challenges in populating each metallo-protein with the correct metal. For instance, specificity is

achieved through first and second coordination spheres of the protein, metal-induced allostery as

well as limiting the amounts of free metal available for those that are high in the stability series

(i.e. copper and zinc). Utilizing metallochaperones for metal delivery and compartmentalization

are yet more ways to overcome thermodynamically dictated preference for metals by proteins.

This study exemplifies the need for these additional mechanisms and highlights the crucial role

of cellular strategies in maintaining the relative abundance of a particular metal that enable a

protein to selectively bind its cognate metal within the cell.

98

References:

1. Robinson, N. J. (2007) A more discerning zinc exporter, Nat. Chem. Biol. 3, 692-693.

2. Waldron, K. J., and Robinson, N. J. (2009) How do bacterial cells ensure that metalloproteins get

the correct metal?, Nat. Rev. Microbiol. 7, 25-35.

3. Leach, M. R., and Zamble, D. B. (2007) Metallocenter assembly of the hydrogenase enzymes,

Curr. Opin. Chem. Biol. 11, 159-165.

4. Zhang, J. W., Leach, M. R., and Zamble, D. B. (2007) The peptidyl-prolyl isomerase activity of

SlyD is not required for maturation of Escherichia coli hydrogenase, J. Bacteriol. 189, 7942-

7944.

5. Martino, L., He, Y., Hands-Taylor, K. L., Valentine, E. R., Kelly, G., Giancola, C., and Conte, M.

R. (2009) The interaction of the Escherichia coli protein SlyD with nickel ions illuminates the

mechanism of regulation of its peptidyl-prolyl isomerase activity, FEBS J. 276, 4529-4544.

6. Weininger, U., Haupt, C., Schweimer, K., Graubner, W., Kovermann, M., Bruser, T., Scholz, C.,

Schaarschmidt, P., Zoldak, G., Schmid, F. X., and Balbach, J. (2009) NMR solution structure of

SlyD from Escherichia coli: spatial separation of prolyl isomerase and chaperone function, J.

Mol. Biol. 387, 295-305.

7. Knappe, T. A., Eckert, B., Schaarschmidt, P., Scholz, C., and Schmid, F. X. (2007) Insertion of a

chaperone domain converts FKBP12 into a powerful catalyst of protein folding, J. Mol. Biol. 368,

1458-1468.

8. Scholz, C., Eckert, B., Hagn, F., Schaarschmidt, P., Balbach, J., and Schmid, F. X. (2006) SlyD

proteins from different species exhibit high prolyl isomerase and chaperone activities,

Biochemistry 45, 20-33.

9. Erdmann, F., and Fischer, G. (2007) The nickel-regulated peptidyl prolyl cis/trans isomerase

SlyD, in Metal Ions in Life Sciences (Sigel, A., Sigel, H., and Sigel, R. K. O., Eds.), pp 501-518,

John Wiley and Sons.

10. Kaluarachchi, H., Sutherland, D. E., Young, A., Pickering, I. J., Stillman, M. J., and Zamble, D.

B. (2009) The Ni(II)-Binding Properties of the Metallochaperone SlyD, J. Am. Chem. Soc. 131,

18489-18500.

11. Li, Y., and Zamble, D. B. (2009) Nickel homeostasis and nickel regulation: an overview, Chem.

Rev. 109, 4617-4643.

12. Wulfing, C., Lombardero, J., and Pluckthun, A. (1994) An Escherichia coli protein consisting of

a domain homologous to FK506-binding proteins (FKBP) and a new metal binding motif, J. Biol.

Chem. 269, 2895-2901.

13. Finney, L. A., and O'Halloran, T. V. (2003) Transition metal speciation in the cell: Insights from

the chemistry of metal ion receptors, Science 300, 931-936.

14. Salgado, M. T., Bacher, K. L., and Stillman, M. J. (2007) J. Biol. Inorg. Chem. 12, 294.

15. Zimmermann, M., Clarke, O., Gulbis, J. M., Keizer, D. W., Jarvis, R. S., Cobbett, C. S., Hinds,

M. G., Xiao, Z. G., and Wedd, A. G. (2009) Metal Binding Affinities of Arabidopsis Zinc and

Copper Transporters: Selectivities Match the Relative, but Not the Absolute, Affinities of their

Amino-Terminal Domains, Biochemistry 48, 11640-11654.

16. Atanassova, A., Lam, R., and Zamble, D. B. (2004) Anal. Biochem. 335, 103.

17. Martell, A. E., and Smith, R. M. (2003) NIST Standard Reference Database 46, version 8.0

National Institute of Standards and Technology, Rockville, M.D.

99

18. Fahrni, C. J., and O'Halloran, T. V. (1999) Aqueous coordination chemistry of quinoline-based

fluorescence probes for the biological chemistry of zinc, J. Am. Chem. Soc. 121, 11448-11458.

19. Golynskiy, M. V., Gunderson, W. A., Hendrich, M. P., and Cohen, S. M. (2006) Metal binding

studies and EPR spectroscopy of the manganese transport regulator MntR, Biochemistry 45,

15359-15372.

20. Baba, T., Ara, T., Hasegawa, M., Takai, Y., Okumura, Y., Baba, M., Datsenko, K. A., Tomita,

M., Wanner, B. L., and Mori, H. (2006) Construction of Escherichia coli K-12 in-frame, single-

gene knockout mutants: the Keio collection, Mol. Syst. Biol. 2, 2006 0008.

21. Zhang, J. W., Butland, G., Greenblatt, J. F., Emili, A., and Zamble, D. B. (2005) A role for SlyD

in the Escherichia coli hydrogenase biosynthetic pathway, J. Biol. Chem. 280, 4360-4366.

22. Graham, A. I., Hunt, S., Stokes, S. L., Bramall, N., Bunch, J., Cox, A. G., McLeod, C. W., and

Poole, R. K. (2009) Severe Zinc Depletion of Escherichia coli: ROLES FOR HIGH AFFINITY

ZINC BINDING BY ZinT, ZINC TRANSPORT AND ZINC-INDEPENDENT PROTEINS, J.

Biol. Chem. 284, 18377-18389.

23. Outten, C. E., and O'Halloran, T. V. (2001) Femtomolar sensitivity of metalloregulatory proteins

controlling zinc homeostasis, Science 292, 2488-2492.

24. Ngu, T. T., and Stillman, M. J. (2009) Metal-binding mechanisms in metallothioneins, Dalton

Trans. 5425-5433.

25. Kaltashov, I. A., and Abzalimov, R. R. (2008) J. Am. Soc. Mass Spectrom. 19, 1239.

26. Tanaka, M., Funahash.S, and Shirai, K. (1968) Kinetics of Ligand Substitution Reaction of

Zinc(2)-4-(2-Pyridylazo)Resorcinol Complex with (Ethylene Glycol)Bis(2-Aminoethyl Ether)-

N,N,N',N'-Tetraacetic Acid, Inorg. Chem. 7, 573-&.

27. McNulty, T. J., and Taylor, C. W. (1999) Extracellular heavy-metal ions stimulate Ca2+

mobilization in hepatocytes, Biochem. J. 339, 555-561.

28. Hottenrott, S., Schumann, T., Pluckthun, A., Fischer, G., and Rahfeld, J. U. (1997) J. Biol. Chem.

272, 15697.

29. Perczel, A., Hollosi, M., and Fasman, G. D. (1996) Circular Dichroism and the Conformational

Analysis of Biomolecules (Fasman, G. D. Ed.), pp 285-380, Plenum Press, New York.

30. Woody, R. W., Dunker, A. K., and Fasman, G. D. (1996) Circular Dichroism and the

Conformational Analysis of Biomolecules (Fasman, G. D. Ed.), pp 285-380, Plenum Press, New

York.

31. Beltramini, M., Bubacco, L., Salvato, B., Casella, L., Gullotti, M., and Garofani, S. (1992) The

aromatic circular dichroism spectrum as a probe for conformational changes in the active site

environment of hemocyanins, Biochim. Biophys. Acta. 1120, 24-32.

32. Rensing, C., and Grass, G. (2003) Escherichia coli mechanisms of copper homeostasis in a

changing environment, FEMS Microbiol. Rev. 27, 197-213.

33. Cobine, P. A., George, G. N., Jones, C. E., Wickramasinghe, W. A., Solioz, M., and Dameron, C.

T. (2002) Copper transfer from the Cu(I) chaperone, CopZ, to the repressor, Zn(II)CopY: Metal

coordination environments and protein interactions, Biochemistry 41, 5822-5829.

34. Badarau, A., Firbank, S. J., McCarthy, A. A., Banfield, M. J., and Dennison, C. Visualizing the

Metal-Binding Versatility of Copper Trafficking Sites, Biochemistry 49, 7798-7810.

35. Xiao, Z., Brose, J., Schimo, S., Ackland, S. M., La Fontaine, S., and Wedd, A. G. Unification of

the Copper(I) Binding Affinities of the Metallo-chaperones Atx1, Atox1, and Related Proteins:

DETECTION PROBES AND AFFINITY STANDARDS, J. Biol. Chem. 286, 11047-11055.

100

36. Iwig, J. S., Leitch, S., Herbst, R. W., Maroney, M. J., and Chivers, P. T. (2008) Ni(II) and Co(II)

sensing by Escherichia coli RcnR, J. Am. Chem. Soc. 130, 7592-7606.

37. Wang, S. C., Dias, A. V., Bloom, S. L., and Zamble, D. B. (2004) Selectivity of metal binding

and metal-induced stability of Escherichia coli NikR, Biochemistry 43, 10018-10028.

38. Maret, W., and Vallee, B. L. (1993) Cobalt as Probe and Label of Proteins, Metallobiochem. 226,

52-71.

39. Adrait, A., Jacquamet, L., Le Pape, L., de Peredo, A. G., Aberdam, D., Hazemann, J. L., Latour,

J. M., and Michaud-Soret, I. (1999) Spectroscopic and saturation magnetization properties of the

manganese- and cobalt-substituted Fur (ferric uptake regulation) protein from Escherichia coli,

Biochemistry 38, 6248-6260.

40. Kwan, C. Y., and Putney, J. W. (1990) Uptake and Intracellular Sequestration of Divalent-

Cations in Resting and Methacholine-Stimulated Mouse Lacrimal Acinar-Cells - Dissociation by

Sr-2+ and Ba-2+ of Agonist-Stimulated Divalent-Cation Entry from the Refilling of the Agonist-

Sensitive Intracellular Pool, J. Biol. Chem. 265, 678-684.

41. Grynkiewicz, G., Poenie, M., and Tsien, R. Y. (1985) A new generation of Ca2+ indicators with

greatly improved fluorescence properties, J. Biol. Chem. 260, 3440-3450.

42. Morleo, A., Bonomi, F., Iametti, S., Huang, V. W., and Kurtz, D. M., Jr. Iron-nucleated folding of

a metalloprotein in high urea: resolution of metal binding and protein folding events,

Biochemistry 49, 6627-6634.

43. Giedroc, D. P., and Arunkumar, A. I. (2007) Metal sensor proteins: nature's metalloregulated

allosteric switches, Dalton Trans. 3107-3120.

44. Waldron, K. J., Rutherford, J. C., Ford, D., and Robinson, N. J. (2009) Metalloproteins and metal

sensing, Nature 460, 823-830.

45. Lee, L. J., Barrett, J. A., and Poole, R. K. (2005) Genome-wide transcriptional response of

chemostat-cultured Escherichia coli to zinc, J. Bacteriol. 187, 1124-1134.

46. Yamamoto, K., and Ishihama, A. (2005) Transcriptional response of Escherichia coli to external

zinc, J. Bacteriol. 187, 6333-6340.

47. Outten, F. W., Huffman, D. L., Hale, J. A., and O'Halloran, T. V. (2001) The independent cue

and cus systems confer copper tolerance during aerobic and anaerobic growth in Escherichia coli,

J. Biol. Chem. 276, 30670-30677.

48. Kaluarachchi, H., Chung, K. C. C., and Zamble, D. B. (2010) Microbial nickel proteins, Nat.

Prod. Rep. 27, 681-694.

49. Irving, H., and Williams, R. J. P. (1948) Order of Stability of Metal Complexes, Nature 162, 746-

747.

50. Colpas, G. J., Brayman, T. G., McCracken, J., Pressler, M. A., Babcock, G. T., Ming, L. J.,

Colangelo, C. M., Scott, R. A., and Hausinger, R. P. (1998) Spectroscopic characterization of

metal binding by Klebsiella aerogenes UreE urease accessory protein, J. Biol. Inorg. Chem. 3,

150-160.

51. Leitch, S., Bradley, M. J., Rowe, J. L., Chivers, P. T., and Maroney, M. J. (2007) Nickel-specific

response in the transcriptional regulator, Escherichia coli NikR, J. Am. Chem. Soc. 129, 5085-

5095.

52. Bloom, S. L., and Zamble, D. B. (2004) Metal-selective DNA-binding response of Escherichia

coli NikR, Biochemistry 43, 10029-10038.

101

53. Ge, R., Zhang, Y., Sun, X., Watt, R. M., He, Q. Y., Huang, J. D., Wilcox, D. E., and Sun, H.

(2006) Thermodynamic and kinetic aspects of metal binding to the histidine-rich protein, Hpn, J.

Am. Chem. Soc. 128, 11330-11331.

54. Iwig, J. S., Rowe, J. L., and Chivers, P. T. (2006) Nickel homeostasis in Escherichia coli - the

rcnR-rcnA efflux pathway and its linkage to NikR function, Mol. Microbiol. 62, 252-262.

55. Li, Y., and Zamble, D. B. (2009) Nickel homeostasis and nickel regulation: an overview, Chem.

Rev. 109, 4617-4643.

56. Rowe, J. L., Starnes, G. L., and Chivers, P. T. (2005) Complex transcriptional control links

NikABCDE-dependent nickel transport with hydrogenase expression in Escherichia coli, J.

Bacteriol. 187, 6317-6323.

57. Zeng, Y. B., Yang, N., and Sun, H. Metal-binding properties of an hpn-like histidine-rich protein,

Chemistry 17, 5852-5860.

58. Changela, A., Chen, K., Xue, Y., Holschen, J., Outten, C. E., O'Halloran, T. V., and Mondragon,

A. (2003) Molecular basis of metal-ion selectivity and zeptomolar sensitivity by CueR, Science

301, 1383-1387.

59. Leach, M. R., Zhang, J. W., and Zamble, D. B. (2007) The role of complex formation between

the Escherichia coli hydrogenase accessory factors HypB and SlyD, J. Biol. Chem. 282, 16177-

16186.

102

Chapter 4 Significance of the HypB-SlyD complex in [NiFe]-hydrogenase

biosynthesis3

4.1 Introduction

Accounting for nearly one-third of all proteins in nature, metalloproteins catalyze some of the

most complex and important processes that are essential for all cellular life (1). Incorporation of

the inorganic cofactor into these metalloproteins is a challenging process due to the prevalence of

buried metallocenters and the complexity of many of these metal-containing active sites. This

process is further complicated by the fact that assembly of these proteins needs to be achieved in

the context of exceedingly low concentrations of free metal ions within a cell. Several

mechanisms have been adapted by nature to meet these demands, one of which is utilization of

proteins dedicated to handling a specific metal ion during the biosynthesis process (2). These

proteins, termed metallochaperones, not only ensure the delivery of the correct metal ion to a

specific target even in an environment of low metal availability but also protect other biological

components of the cell from the detrimental effects of metal ions (3). Furthermore, these helper

proteins are one of the key factors that enable a cell to overcome the inherent thermodynamic

metal affinities of individual factors, via transfer of a specific metal through transient protein-

protein interactions (1).

The hydrogenase enzymes found in many bacteria and archaea catalyze the oxidation of

hydrogen gas and/or the reverse reaction (4). Central to microbial energy metabolism as well as

fuel cell applications, this chemically challenging reaction occurs at an intricate organometallic

metal centre consisting of unique diatomic ligands such as carbon monoxide and cyanide (5). In

the case of the [NiFe]-hydrogenases, the gas processing occurs at the dinuclear metal site that is

buried within the large subunit of a heterodimeric protein. Much of the current knowledge about

the assembly of this sophisticated metal cluster is based on investigations conducted on the

Escherichia coli hydrogenase 3 iso-enzyme and presumably occurs in two steps. The initial iron

incorporation is performed by HypCDEF proteins that are responsible for synthesis of the

diatomic ligands and delivery of the decorated iron centre to the large subunit (4). The

3The in vivo experiments presented in this chapter were designed and performed by J. W. Zhang.

103

subsequent nickel insertion step is carried out by the three metallochaperone proteins HypA,

HypB and SlyD (4, 6).

Figure 4-1. Architecture of E. coli SlyD and HypB. (A) SlyD consists of an N-terminal FKBP-type

PPIase domain, an IF (insertion in the flap) chaperone domain and a C-terminal metal-binding domain

that is rich in potential metal-binding residues (7) A loop in the chaperone domain (residue 107-111) is

essential for the SlyD-HypB complex formation (8). (B) The first seven residues of HypB (CXXCGC)

encompass the high-affinity Ni(II) binding-motif (KDNi(II)

1.2 x 10-13

M) while the residues C166, H167

and C198 located in the GTPase domain contribute to the low-affinity metal-binding site that can bind

either Ni(II) or Zn(II) with micromolar affinity (9, 10). HypB interacts with SlyD via the the linker

region (8).

HypA consists of a N-terminal metal-binding site that can coordinate Ni(II) with micromolar

affinity (6). The protein also contains a structural zinc site at the C-terminus typical of protein

domains mediating protein-protein interactions (6). As such, HypA is anticipated to serve as a

scaffold for the nickel insertion complex during enzyme maturation (11). HypB has an essential

GTPase activity that is required for hydrogenase maturation and nickel insertion in vivo (12-14).

E. coli HypB has two metal-binding sites, one at the N-terminus of the protein that binds nickel

with picomolar affinity and a second metal-binding site in the GTPase domain (G-domain) that

can bind nickel or zinc with low-micromolar affinity (10) (Figure 4-1). Mutation of the residues

contributing to either of the metal sites results in a hydrogenase-deficient phenotype, indicating

the importance of both metal sites in HypB for hydrogenase biogenesis (9). In addition to these

two metal-binding sites, some HypB homologues such as that from B. japonicum contain poly-

histidine stretches that are capable of binding nickel with high capacity but low affinity (15).

This polyhistidine stretch is absent in E. coli HypB and consequently the nickel storage role is

thought to be carried out by SlyD. SlyD is a multi-domain protein harbouring both protein

folding (16, 17) and metal-binding activities (18, 19) (Figure 4-1). In support of the suggested

1 196

Peptidyl-Prolyl Isomerase

Domain

Metal-Binding

Domain146

IF- Chaperone Domain

76 120

(A)

C2XXC5GC7

GTPase Domain

2 8 77 290

Linker

Region

High-Affinity

Ni(II) site

C166H167C198

Low-Affinity

Metal-Binding

site

(B)V107EITA111

104

nickel storage function, in vitro characterization of SlyD revealed that the protein has

considerable capacity for nickel (7 potential nickel-binding sites) and binds the metal with an

affinity in the nanomolar range (19). The unique C-terminal metal-binding domain (MBD) of

SlyD features a large concentration of potential metal-binding residues and is essential for this

nickel chelation in vitro (19). Corroborating the importance of the metal-binding activity of

SlyD, the MBD is essential for optimal hydrogenase biogenesis as well as nickel accumulation in

E. coli (8).

SlyD was initially recognized as an accessory factor of the hydrogenase maturation pathway

when the protein was found in a complex with HypB, in multiprotein pull-down assays designed

to discover additional components of the [NiFe]-hydrogenase maturation pathway (20).

Subsequent in vitro work has established that these two proteins form a heterodimer through the

proline-containing linker region of HypB located between the high-affinity N-terminal metal-

binding site and the G-domain, and a loop in the IF domain of SlyD (8) (Figure 4-1). This

protein-protein interaction is important for the hydrogenase-related function of SlyD in vivo, as

disruption of this interaction through mutagenesis results in decreased hydrogenase activity, to a

similar extent as that observed in the ∆slyD strain (8). However, the consequences of this

interaction and how it enhances hydrogenase biogenesis are not clearly understood yet. The aim

of this study was to gain a better understanding of how the SlyD and HypB interaction modulates

the metal-binding and GTPase activities of HypB. The results presented provide evidence for

the nickel storage function of SlyD and show that these stored nickel ions can be used as a source

to metallate HypB, even under competitive conditions. This investigation also revealed that the

HypB-SlyD interaction results in faster metal release from the N-terminal high-affinity Ni(II) site

of HypB, as has been observed before (8), and this effect is enhanced by the nature of the

nucleotide bound to HypB, suggesting a gated metal release process. Furthermore, we find that

the GTPase activity of HypB is enhanced in the presence of SlyD, which can rescue the

enzymatic activity that is inhibited by binding of metal to the G-domain. The implications of the

altered metal-binding and hydrolysis functions that emerge as a result of protein complex

formation on [NiFe]-hydrogenase metallocentre assembly are discussed.

105

4. 2 Materials and Methods

Materials. All metal salts and the chelators, EDTA, EGTA, glycine, were purchased from

Sigma-Aldrich as 99.9% pure salts unless otherwise stated. The working solutions of the

chelators were adjusted to pH 7.5 prior to use and the concentrations of the metal stock solutions

were established by ICP-AES. GDP, GTP and the GTP analog, guanosine 5′-[β,γ-

imido]triphosphate (GMP-PNP) were purchased from Sigma and used as supplied unless

otherwise noted. Buffers for all assays were treated with Chelex-100 (Bio-Rad) whereas buffers

for MS-ESI experiments were treated with Chelex-100 that was converted to the NH4+ form from

the Na+ form according to the manufacturer’s protocol. Tryptone, yeast extracts and glycerol

were purchased from BioShop Canada Inc. All other chemicals were purchased from Sigma-

Aldrich unless otherwise noted.

Growth conditions. The MC4100 strain and the ΔslyD strain in the same background were

grown anaerobically for 5 hrs in sealed flasks of buffered tryptone-yeast extract-potassium

phosphate (TYEP) media supplemented with 0.8% glycerol, 15 mM sodium formate, 1 M

sodium molybdate and 1 M sodium selenite, following inoculation with 1% (v/v) of an

aerobically grown overnight culture. Radioactive 63

Ni(II) (Perkin Elmer) was included in the

growth media at a final concentration of 0.25 M. Following the 5 h growth, cells were washed

thoroughly with chilled 50 mM potassium phosphate pH 7.0 to remove any excess nickel. The

cells were then grown anaerobically for an additional 16 h at 37 oC in buffered TYEP media or

media that was supplemented with dimethylglyoxime (DMG). At the start of each new growth

step, chloroamphenicol (34 g/mL) was added to the ΔslyD bacterial cultures.

Cellular nickel accumulation. After the final growth step, cells were harvested and extensively

washed with 50 mM potassium phosphate pH 7.0. Cell extracts were prepared by sonication in

the same buffer followed by centrifugation at 18,000 rpm for 20 min. Radioactivity was

measured by scintillation counting in 3 mL of UltimaGold scintillation fluid with a Tri-Carb

2100TR (Packard Instruments) liquid scintillation analyzer.

Hydrogenase activity assay. For analysis of hydrogenase activity in the cell cultures, cell

extracts were prepared according to the method described by Zhang et al. (20). The hydrogenase

activity of all three isozymes in E. coli was measured by hydrogen-dependent reduction of

benzyl viologen according to the procedure of Ballantine and Boxer (21).

106

Protein expression and purification. SlyD wildtype protein and SlyD variants were over

expressed and purified according to the methods previously described (19, 20). Only fully

reduced SlyD was used for the subsequent experiments as determined via an N-ethyl maleimide

modification assay (19). HypB was expressed and purified according to the method used by

Leach et. al, and supplementation of the growth media with 1 mM NiSO4 is expected to yield

purified HypB with close to stoichiometric amount of Ni(II) bound (10). The amount of metal

bound to HypB was determined by treating HypB with excess p-hydroxymecuribenzoic in the

presence of 4-(2-pyridylazo)resorcinol (PAR) followed by detection at 500 nm. The resulting

Ni(II)-PAR signal was compared to a standard curve prepared under the same conditions to

determine the fraction of metal bound and the protein was routinely found to contain > 0.95

equivalents of Ni(II). To assess the free thiol content of purified HypB, tris(2-

carboxyethyl)phosphine (TCEP) was removed by PD10 gel-filtration columns (GE Healthcare)

and the protein was mixed with 5,5’-dithio-bis(2-nitrobenzoic acid) (DTNB) in the presence of 6

M guanidinium hydrochloride and compared against β-mercaptoethanol as a standard. The

HypB mutant C2,5,7A was expressed and purified in a similar manner to the wildtype protein

except the growth media was not supplemented with additional nickel Ni(II) (10). Analysis of the

metal content of the purified HypB mutant according to the method described above indicated

the presence of only apo protein.

Sample preparation for electrospray-ionization mass spectrometry (ESI-MS). All sample

preparations were conducted in an anaerobic chamber. Apo HypB samples were prepared by

incubating the protein with 10 mM EDTA and 5 mM TCEP for 3 days at 4 oC. The sample was

then buffer exchanged via PD-10 columns equilibrated with 10 mM ammonium acetate, pH 7.5

(MS buffer) and the fractions containing the protein were collected and concentrated to a volume

of ~100 L. This sample was further buffer exchanged four times with 500 L aliquots of MS

buffer via a 10K MWCO nanosep centrifugal device (PALL lifesciences) to remove salt. Holo

HypB and apo SlyD samples were prepared by incubating the proteins with 1 mM TCEP

overnight at 4 oC and buffer exchanging into 10 mM ammonium acetate, pH 7.5 in a similar

manner to the apo HypB samples. Ni(II)-bound SlyD samples were prepared by addition of

appropriate amount of nickel with subsequent equilibration of the sample overnight at 4 oC. To

ensure reproducibility all experiments mentioned below were conducted at least in triplicate.

107

Metal transfer using ESI-MS. HypB and SlyD protein concentrations of 100 M each were used

for these reactions unless otherwise noted. After mixing all of the components necessary for the

reaction, samples were left at room temperature in an anaerobic glovebox. To monitor the

metallation state of the protein mixture at a given time point, an aliquot of the mixture was

removed and diluted 10-fold using MS buffer into septum-capped vials to reduce exposure to air

and analyzed via ESI-MS immediately. When metal chelators were included in the reaction

mixture, a 10 molar excess of the competitor relative to the HypB protein concentration was

added to the mixture.

Equilibrium dialysis. Using microchambers of DIALYZER (Harvard Apparatus, Inc.), a 100 M

sample of apo HypB (250 L) was dialyzed against a 100 M SlyD sample loaded with 1

equivalent of Ni(II) (250 L). The chambers were separated by a 1 MWCO membrane, and the

samples were allowed to equilibrate at room temperature in an anaerobic glove box. For ESI-MS

analysis at a given time point, solutions from each of the chambers were diluted 10-fold into

septum-capped vials using MS buffer.

Metal-release assays using ESI-MS.

a) Metal release from HypB in the presence of SlyD variants. Holo HypB and apo SlyD at 100

M each were used for these reactions unless otherwise noted. An EGTA concentration of 1

mM also used. After equilibrating the samples for the indicated amount of time at room

temperature in an anaerobic glovebox, they were diluted 10-fold into septum-capped vials for

analysis via ESI-MS. Control experiments with HypB and chelator alone were carried out under

identical experimental conditions.

b) Metal release from HypB using different chelators. To initiate the reactions, proteins were

mixed at a 1:2 ratio with 50 M holo HypB and 100 M apo SlyD, into buffer containing 250

M of the chelator under investigation. At the indicated times, aliquots of these samples that

were allowed to equilibrate at room temperature in an anaerobic glove box were diluted 10-fold

into septum-capped vials and analyzed by ESI-MS immediately.

c) Metal release from HypB in the presence of nucleotide. GDP (purchased as the Na+ form

from Sigma-Aldrich) was buffer exchanged using a C18 Vydac analytical column (GRACE)

using MS Buffer, prior to use in ESI-MS experiments. GMP-PNP was used as purchased

108

(obtained as Li+ form from Sigma-Aldrich). Prior to conducting the assays, holo HypB samples

were incubated with the respective nucleotide (5 molar excess relative to HypB) at room

temperature in an anaerobic glove box for 30 min. Reactions were initiated by mixing holo

HypB bound to nucleotide (100 M) and apo SlyD (100 M) with buffer containing 1 mM

EDTA. An aliquot of the sample was transferred to a septum-capped vial at various time points

and diluted 10-fold prior to analysis by ESI-MS immediately.

ESI-MS instrumental parameters, data acquisition and processing. The mass spectra were

acquired on a AB/Sciex QStarXL mass spectrometer equipped with an ion spray source in the

positive ion mode and a Hot Source-Induced Desolvation (HSID) interface (Ionics Mass

Spectrometry Group Inc.). Ions were scanned from m/z 1300-3000 with accumulation of 1 s per

spectrum with no interscan time delay and averaged over a 2-4 min period. The instrument

parameters were as follows: ion source temperature 200 oC, ion source gas 50.0 psi; curtain gas

50.0 psi; ion spray voltage 5000.0 V; declustering potential 60.0 V; focusing potential 60 V;

collision gas 3.0; MCP (detector) 2200.0 V. The resulting spectra were deconvoluted without

further manipulation using the Bayesian protein reconstruction program over an appropriate mass

range using a step mass of 1 Da, signal/noise ratio of 20 and the minimum intensity detected set

to 1%.

Metal-release assays using electronic absorption spectroscopy: Holo HypB samples (100 M)

in 10 mM ammonium acetate, pH 7.5, and 1 mM MgCl2 were incubated with 500 M nucleotide

for 30 min at room temperature prior to the start of the reaction. SlyD (100 M) was rapidly

mixed with holo HypB at 1:1 ratio into buffer containing 1 mM EDTA and loaded into an

anaerobic cuvette in the anaerobic glovebox. All samples were allowed to equilibrate for ~1 min

prior to collection of data. Spectra were recorded every 15 min for 2 h and the amount of Ni(II)

bound to HypB was determined by the intensity of the signal at 320 nm. The fraction of holo

HypB was calculated using the following ratio (A320/Amax) where Amax is the absorption detected

at 320 nm at the first time point. Apo HypB does not have a detectable signal at 320 nm.

GTPase kinetic assays and HPLC analysis. To measure the GTPase activity, protein samples

were buffer exchanged into 20 mM Tris, pH 7.5, 100 mM NaCl, 5 mM MgCl2 and 1 mM TCEP.

The following reaction conditions were used for al1 GTP hydrolysis measurements that were

conducted under aerobic conditions: A final concentration of 1 M HypB, 50 M SlyD and 450

109

M GTP. When SlyD was included in the reaction mixture, the two proteins were allowed to

equilibrate at room temperature for 30 mins. Assays were initiated by rapid mixing of the

protein solutions with GTP followed by incubation of the samples at 37 oC under aerobic

conditions. After 10 min, the reactions were quenched by adding 1 M perchloric acid and after

10 min, neutralized with 4 M sodium acetate, pH 4.0. Precipitated protein and salt was removed

by centrifuging at 18000 rpm for 5 min and 25 L of the resulting supernatant was diluted into

Milli-Q water to a total volume of 100 L for analysis. Samples were injected on a Dionex

BioLC equipped with an A50 autosampler with an injection volume set to 50 L. GDP and GTP

were resolved on a C18 Vydac analytical column (GRACE) with an isocratic mobile phase

consisting of 100 mM sodium phosphate buffer, pH 6.6, 10 mM tetrabutylammonium bromide,

4% acetonitrile. Nucleotides were detected via the absorbance at 254 nm and under these

conditions, GDP eluted at ~7.8 min and GTP eluted at ~9.5 min. Integration of the peak area

using PeakNet Client v 6.4 (Dionex Inc.) relative to control samples (with no enzyme present)

treated in an identical faction allowed for quantification of moles GDP produced L-1

min-1

,

from which kcat was then determined. For analysis of GTPase activity as a result of metal

coordination to the site at the G-domain, 10 M purified HypB was incubated with either 10

molar excess of Ni(II) or 1.5 molar excess of Zn(II) respectively at room temperature for 1 h.

The GTPase activities of these samples were measured in an identical manner as described above

by subsequent dilutions of these samples to 1 M.

4.3 Results

SlyD functions as a nickel storage protein in E. coli. The ability of SlyD to coordinate

multiple nickel ions, coupled with the fact that deletion of this gene results in reduced nickel

content in E. coli when grown under anaerobic conditions, has led to the designation of SlyD as a

nickel reservoir (19, 20). These stored nickel ions on SlyD are expected to be utilized for

[NiFe]-hydrogenase metallocentre assembly when bacteria are faced with nickel-limiting

conditions. If this hypothesis is true, then a ΔslyD strain should be deficient in hydrogenase

activity compared to the wildtype strain, especially when bacteria are faced with Ni(II)

starvation. Cell cultures were first exposed to Ni(II) under anaerobic conditions to allow E. coli

to accumulate nickel and potentially store Ni(II) in SlyD. Then, hydrogenase activity was

assessed following washing of the cells and further growth for 16 h anaerobically in media with

110

or without dimethyl glyoxime (DMG). Supplementation with DMG is expected to mimic nickel-

limiting conditions because this chelator binds nickel with high affinity (22, 23). Deletion of

slyD results in 80% reduction in hydrogenase activity whereas the relative activity decreases

even further to 11% of the wildtype when grown in nickel-limited media (Table 4-1). Addition

of DMG decreases the hydrogenase activity in each strain by 54% and 74% for the wildtype and

ΔslyD strains, respectively (Table 4-1). This decrease in activity for the deletion strain under

limiting nickel conditions is evidence that SlyD contributes to hydrogenase maturation by

functioning as a nickel reservoir.

Table 4-1. Total hydrogenase activity.

- DMG + DMG

% of enzyme activity in +

DMG relative to - DMG

Wildtype 1.00 0.46 ± 0.09 46%

∆slyD 0.20 ± 0.03 0.05 ± 0.02 26%

% enzyme activity in ΔslyD

relative to WT 20% 11%

Hydrogenase activity in cultures that were first grown with 0.25 M

63Ni, followed by washing and further

anaerobic growth for 16 h. Total hydrogenase activity was measured using the benzyl-viologen assay. The data are

normalized to the amount of hydrogenase activity detected (measured as units/mg) in the wildtype strain grown

without addition of DMG. The errors reported are standard deviation from at least three independent replicates.

Table 4-2. Cytoplasmic 63

Ni(II) levels.

- DMG + DMG

% of Ni(II) in + DMG

relative to - DMG

Wildtype 1.00 0.70 ± 0.08 70%

∆slyD 0.45 ± 0.15 0.05 ± 0.02 12%

% Ni(II) in ΔslyD

relative to WT 45% 8%

Amount of radioactive nickel remaining in bacteria that were cultured for 5 hrs in media containing 0.25 M

63Ni

anaerobically, followed by washing and further anaerobic growth for 16 h. The data are normalized to the amount

of 63

Ni detected (measured as cpm/mg) in the wildtype strain grown without addition of DMG. The errors reported

are from at least three independent replicates.

To determine whether the decrease in hydrogenase activity is in fact due to less nickel retention

by the ΔslyD strain, the nickel content was also analyzed for the two bacterial strains grown

under different media compositions. As expected (20), bacterial strains with lesions in the slyD

gene only accumulate 45% of Ni(II) levels observed in the wildtype strain when grown in media

without DMG. A more dramatic decrease in the nickel content is observed when the bacteria are

111

grown in media supplemented with DMG, with only 8% nickel accumulation in the ΔslyD strain

relative to the wildtype strain grown under identical conditions (Table 4-2). Similarly, when

considering the impact of limiting available nickel by treatment with DMG the effects are more

pronounced for the ΔslyD strain as the cytoplasmic nickel content dropped by 88% whereas only

a 30% decrease is observed in the wildtype strain (Table 4-2). Therefore, the data indicate that

under nickel-limiting conditions, deletion of slyD, results in a 5.8-fold decrease of the nickel

quota within the cell, further supporting the hypothesis that SlyD functions as a nickel storage

protein.

Metallation of apo-HypB using SlyD as a Ni(II) source. Metal stored under abundant Ni(II)

conditions is typically utilized for enzyme biosynthesis when bacteria are faced with metal

limiting conditions (11). Therefore, it is possible that Ni(II) stored in SlyD is mobilized for

assembly of the [NiFe]-hydrogenase metallocentre. The most likely partner for metallation by

SlyD is HypB, because a HypB-SlyD protein complex is observed in vivo (20). To test this

hypothesis, metal transfer experiments were conducted in vitro. These transfer reactions were

initiated by mixing 100 M SlyD containing 1 equivalent of Ni(II) with 100 M apo HypB

under anaerobic conditions. HypB and SlyD are expected to form a heterodimer with a KD of

~10 M (8), therefore protein concentrations of 100 M each were chosen to ensure > 70%

complex formation. It is estimated that there are approximately 25000 molecules of HypB and

10000 molecules of SlyD expressed per cell of bacteria, which roughly translates to 52 M

HypB and 17 M SlyD assuming a cell volume of 10-15

L (7, 24). Therefore, the concentrations

utilized in these in vitro assays are within the range of physiological concentrations of the two

proteins.

ESI-MS was used for analysis of the metal transfer reactions because it detects all the proteins

present in solution simultaneously and it reveals the metallation state of each of the proteins.

Representative spectra for a metal transfer assay are shown in Figure 4-2. The peak intensities

from the reconstructed spectra were used to calculate the fractions of holo HypB assuming that

both apo and holo HypB species in the reaction mixture are ionized similarly in the ESI source

and that the signal detected for each species in the gas phase is proportional to their respective

concentration in the solution phase. Ni(II)-bound HypB is not detected after a 15 min incubation

period (data not shown), but a gradual increase in the amount of the Ni(II)-bound HypB species

112

with a concomitant decrease in the amount of holo SlyD species is observed when the protein

mixture is allowed to equilibrate for longer (Figure 4-2 and Figure 4-3A). Although HypB has

two metal-binding sites only a single Ni(II) per HypB monomer is detected in these spectra

signifying metal loading into a single metal site. Considering the affinities of the Ni(II) sites of

HypB (~pM and ~M) (10) and those of SlyD (~nM) (19) it is likely that the high-affinity metal

site at the N-terminus of HypB is being loaded via this metal transfer process, rather than the

weaker site in the G-domain. To verify this explanation, metal transfer from holo SlyD to the

HypB mutant C2,5,7A was assessed. This HypB mutant contains an intact G-domain metal site

but mutating the three cysteines to alanines disrupts the high-affinity nickel-binding activity (10).

Ni(II) transfer between holo SlyD and the HypB mutant was not observed (data not shown),

indicating that SlyD only transfers metal to the N-terminal high-affinity site. Nevertheless, these

data denote that SlyD can function as a Ni(II) donor to HypB, albeit with slow metal exchange

rates. To ensure that incomplete metallation of HypB observed after 4 h is not due to

compromised integrity of HypB, a control experiment was conducted with an apo HypB sample

that was incubated with 1 molar equivalent of metal and analyzed under identical conditions at

various time points. The mass spectra indicate that fully metallated HypB is obtained in less than

30 min incubation and remains fully metal bound over the course of 4 h, suggesting that partial

metallation is a SlyD-dependent effect and not due to compromised integrity of the HypB protein

(data not shown).

In addition to the expected apo HypB peak at 31432 Da, an additional peak is observed at 31460

Da in the reconstructed spectrum for apo HypB (Figure 4-1). Only the apo-protein is susceptible

to this modification (+28 Da) since addition of 1 equivalent of Ni(II) to the protein yields a

single peak corresponding to the Ni-bound HypB (Figure 4-3), suggesting that these peaks result

from modification of apo protein during the ionization process. Exposure to reactive oxygen

species during the ionization process can cause oxidative modifications of amino acid side chains

that result in an increase in the protein mass (25, 26). Typically, lowering the voltage on the ion

source is expected to reduce the oxidation of the protein (25), however lowering of the voltage

did not reduce the oxidation of HypB (data not shown). A further decrease in the applied voltage

was precluded by the significant decrease in the signal due to insufficient ionization energy. The

N-terminal region containing the metal-binding site is a likely candidate for this modification

because binding of Ni(II) is able to protect the protein from oxidation.

113

Figure 4-2. Metal transfer detection via ESI-MS. Representative reconstructed mass spectra

corresponding to SlyD (left panels) and HypB (right panels). Mass spectra corresponding to SlyD bound

to 1 equivalent of Ni(II) and apo HypB before mixing (top panels). Metallation state of the proteins after

equilibrating for 4 h (middle panels) and the same reaction conducted in the presence of a 10 molar excess

EGTA (bottom panels). Numbers heading the dotted lines represent the number of Ni(II) ions bound to

the protein. The (*) in HypB spectra corresponds to fully reduced apo HypB whereas (**) indicates apo

HypB with the +28 Da modification.

In order to assess whether metallation of HypB is due to direct metal transfer achieved through

heterodimer formation and not through dissociation of the Ni(II) from SlyD followed by

subsequent binding by HypB, SlyD bound to 1 equivalent of metal was dialyzed against an apo

HypB sample. Metallation of apo HypB is still observed, indicating that a fraction of the Ni(II)-

HypB formed is due to this diffusion-mediated process (Figure 4-3A). However, the metallation

rate is considerably reduced compared to that achieved when the two proteins are incubated

together. These results suggest that although diffusion-dependent metal transfer does occur,

Ni(II) transfer is significantly enhanced due to protein-protein complex formation.

Mass (Da)

20700 20900 21100 21300

Mass (Da)

31300 31400 31500 31600

Rel

ativ

e In

tens

ity

0 1 2 3

**

**

**

*

1

*

*

Before

Mixing

After

4 h

After 4h

with EGTA

SlyD HypB

114

Figure 4-3. Ni(II) transfer from SlyD to HypB detected via ESI-MS. (A) Equilibrating SlyD (100

M) bound to 1 equiv. of Ni(II) with apo HypB (100 M) leads to a gradual increase in the amount of

Ni(II)-bound HypB with time (diamonds). Equilibrating the two proteins separated by a membrane

results in significantly slower metal transfer rates (circles). (B) Fraction of Ni(II)-bound HypB obtained

with time upon equilibrating 1 equiv. of Ni(II)-bound SlyD (100 M) with apo HypB (100 M) in the

presence of 1 mM EGTA (squares). A lower yield of Ni(II)-bound HypB is obtained when 100 M Ni(II)

is added to a solution of apo HypB (100 M) and 1 mM EGTA (triangles), suggesting protein-mediated

metal transfer. Error bars represent standard deviation of at least 3 independent trials.

Figure 4-4. Analysis of holo HypB by ESI-MS. Representative spectra of HypB bound to 1 equivalent

of Ni(II) (denoted as holo HypB) after leaving the protein at room temperature for 25 h. The m/z spectra

and the reconstructed spectra are shown in the left and right panels, respectively. The dotted-lines denote

the apo-HypB species corresponding to masses of 31432 Da and 31460 Da.

0

0.1

0.2

0.3

0.4

0.5

0 1 2 3 4

Fra

ctio

n o

f N

i(II)-

bo

un

d H

yp

B

Time (h)

0

0.1

0.2

0.3

0.4

0.5

0 1 2 3 4

Fra

ctio

n o

f N

i(II)-

bo

un

d H

yp

B

Time (h)

(A) (B)

Mass (Da)

31300 31400 31500 31600

Holo HypB

Rel

ativ

e In

tens

ity

m/z

1600 2000 2400 2800

115

To validate whether SlyD can still function as a source of nickel for HypB in a competitive

environment, metal transfer from SlyD to HypB was monitored in the presence of two metal

chelators, EGTA and glycine. EGTA forms a 1:1 complex with Ni(II) with an apparent KD ~10-

11 M at pH 7.5 whereas two glycine molecules are required to form a complex with Ni(II) and

binds this metal with an affinity in the low micromolar range (27). Transfer of nickel from SlyD

to the HypB high-affinity metal site is still observed and remains unchanged in the presence of

10 molar excess (i.e. 1 mM) of the weak competitor glycine (data not shown). In the case of the

stronger competitor EGTA, the fraction of Ni(II)-bound HypB is diminished compared to that

observed without the competitor (Figure 4-2 and 4-3B). However, addition of 1 equivalent of

nickel without SlyD to apo HypB solution containing 1 mM EGTA results in further diminished

fraction of Ni(II)-HypB obtained (Figure 4-3B). Therefore, these data are consistent with the

model that SlyD is not only able to provide Ni(II) to HypB, but it is able to do so in a strongly

competitive environment.

SlyD modulates the HypB high-affinity Ni(II)-binding site. The reason why fully-saturated

HypB could not be obtained in the metal-transfer assays even after long equilibration periods that

were sufficient to remove all the nickel bound to SlyD was puzzling. Earlier investigations by

Leach et al. demonstrated that SlyD can modulate the N-terminal nickel site of HypB, likely

through the formation of a protein-protein complex (8). In order to determine if partial

metallation is a result of this modulation of the HypB Ni(II) site by SlyD, a protein mixture

containing equimolar apo SlyD and holo HypB was analyzed by ESI-MS. Upon addition of apo

SlyD to holo HypB samples, a loss of Ni(II) ions from the high-affinity site of HypB is observed

with increasing incubation time and at 25 h only half of HypB remains bound to nickel (Figure 4-

4). At any of the time points analyzed no metal was bound to SlyD, suggesting that metal loss

was not due to chelation by SlyD. To ensure that metal loss was due to the presence of SlyD,

control experiments were conducted in which a sample of HypB alone was analyzed under

identical experimental conditions. No metal loss was detected from HypB even after allowing the

sample to remain at room temperature for a 25 h period under anaerobic conditions (Figure 4-4).

Therefore, the data indicate that an interaction with SlyD leads to a dramatic change in the high-

affinity Ni(II) site of HypB such that the nickel-HypB protein complex does not survive the

ionization process during the MS analysis.

116

Figure 4-5. SlyD modulates the high-affinity metal site of HypB. Representative reconstructed mass

spectra corresponding to apo SlyD (left panel) and holo HypB (right panel) before mixing (top panels)

and after allowing the two proteins to equilibrate for 25 h (bottom panels). Numbers heading the dotted

lines represent the number of Ni(II) ions bound to the protein. Partial loss of metal occurs from HypB as

indicated by the appearance of apo-peaks (* and **), but this fraction of lost metal is not coordinated by

apo SlyD as the mass of the protein remains unchanged. The (*) in HypB spectra corresponds to fully

reduced apo HypB whereas (**) is apo HypB containing the +28 Da modification.

Features of SlyD important for increased metal-release rates. In order to establish which

attributes of SlyD are necessary to enhance the metal release rates from HypB, several SlyD

mutants were analyzed. The metal release experiments were conducted in the presence of 1 mM

EGTA (chelator added as an acceptor of Ni(II)) with 100 M each of HypB and SlyD, followed

by ESI-MS analysis. As mentioned above, when compared to HypB alone, inclusion of SlyD in

the reaction leads to acceleration of Ni(II) release from HypB. Addition of SlyD1-146, a truncated

version of SlyD lacking the metal-binding domain, does not modulate the Ni(II) site of HypB as

the metal release rate was similar to that of HypB alone (Figure 4-6). To verify that protein-

protein complex formation is necessary for this accelerated metal release, a SlyD mutant that is

unable to form a heterodimer with HypB was used. This SlyD mutant, referred to as the ΔIF

SlyD, was described previously and has a deletion in the IF-chaperone domain corresponding to

Mass (Da)

20700 20900 21100 21300

Mass (Da)

31300 31400 31500 31600

Rel

ativ

e In

tens

ity0 1 2 3 4

**

*

1

Before

Mixing

After

25 h

SlyD HypB

117

residues 107-111, which abrogates complex formation (8). Even after allowing the proteins to

equilibrate for 6.25 hours, 92% of HypB retained bound Ni(II) indicating that the SlyD-HypB

complex formation is essential for modulation of the HypB Ni(II) site (Figure 4-6). When a

variant of SlyD corresponding to the metal-binding domain of SlyD was utilized (i.e. SlyD146-196

referred to as the SlyD-MBD) a metal-release profile comparable to that of ΔIF SlyD was

observed (Figure 4-6). Cross-linking experiments involving the SlyD-MBD and WT HypB

failed to produce a cross-linked product suggesting that the IF-domain is essential for the

complex formation (data not shown). Taken together the results obtained for the three mutants

indicate that the IF domain that is essential for complex formation, as well as the metal-binding

domain are both necessary to accelerate release of the Ni(II) from the N-terminal metal-site of

HypB. To determine whether Cys residues in SlyD are necessary for modulation of the metal-

binding activity of HypB by SlyD, a variant devoid of Cys residues designated as the triple

mutant (triple mutant contains the following mutations and deletion: C167,168,184,185A and

Δ193-196) was utilized. Although the triple mutant has reduced affinity as well as reduced

capacity for Ni(II) binding (19), the metal release profile was not affected when compared to the

WT SlyD protein (Figure 4-6). Thus, the results indicate that the Cys residues on SlyD are not

required for affecting the high-affinity Ni(II) site of HypB.

Thiol-based acceptor enhances the metal release from HypB. If nickel flows from the SlyD-

HypB complex to the hydrogenase enzyme precursor, it is likely to be uni-directional in vivo.

This assumption is based on the fact that once the Ni(II) is coordinated by the Cys residues of the

precursor [NiFe]-Hydrogenase large subunit, this subunit is expected to undergo proteolytic

cleavage at the C-terminus with subsequent folding to yield a mature enzyme, thereby trapping

the metal ion at the active site. To mimic this thermodynamic nickel sink, metal release from

HypB by SlyD was re-evaluated in the presence of EDTA, a chelator with high affinity for Ni(II)

(apparent KD of 10-15

M at pH 7.5) (27). As expected, inclusion of EDTA leads to faster metal

release. Only 64% of HypB remains bound to Ni(II) after a 30 min reaction compared to 72%

with EGTA (Figure 4-7). Similar to the above mentioned metal-release assays, nickel-bound

SlyD species are not detected indicating that the nickel is not transferred to SlyD but sequestered

by the chelator once released from the HypB metal site. To observe whether addition of an

acceptor containing thiol moieties could affect the metal release rate, a seven residue peptide

(termed B7) was used. The basis for addition of chelator containing thiol groups is that it would

118

be a closer mimic to the [NiFe]-hydrogenase precursor as the coordination sphere of the enzyme

consists of Cys residues (11). This B7-peptide corresponds to the minimal sequence of the high-

affinity nickel-binding site of HypB. Similar to the full-length protein, the peptide has high

affinity for nickel ions, and coordinates this metal with a KD of ~10-13

M. (Douglas, C., Zamble,

D. B., manuscript in preparation). Although the peptide binds Ni(II) with lower affinity than

EDTA, the metal release from HypB is greatest in the presence of this peptide as only 28% of

HypB remains bound to Ni(II) after a 30 minute incubation, suggesting that the thiol residues

favour this metal release process (Figure 4-7). For all chelators used, control experiments with

50 M HypB and 250 M chelator alone were conducted to ensure the enhanced metal-release

rates observed were in fact due to the presence of SlyD (Figure 4-7).

Figure 4-6. Monitoring metal release from HypB to EGTA in the presence of SlyD variants via

ESI-MS. Addition of 1 mM EGTA to holo HypB (100 M) results in Ni(II) loss from the high-affinity

site (empty diamonds). Addition of 100 M wildtype (filled squares) or the triple mutant SlyD (filled

diamonds) leads to an increase in the rate of metal release to EGTA. Addition of MBD-SlyD (empty

circles), ΔIF SlyD (empty squares) or SlyD1-146 (empty triangles) at a concentration of 100 M each does

not affect the rate of Ni(II) release from HypB to EGTA. Error bars represents the standard deviation of

at least 3 independent trials.

0.4

0.5

0.6

0.7

0.8

0.9

1

0 1 2 3 4 5 6 7

Fra

ctio

n o

f N

i(II)-

bo

un

d H

yp

B

Time (h)

119

Figure 4-7. Metal release from HypB using different acceptors. Fraction of holo HypB remaining

after incubating HypB (50 M) with 250 M chelator (filled bars) for 30 mins compared to fraction of

holo HypB remaining after including apo SlyD (100 M, empty bars) in the reaction mixture. The trend

in metal release rate does not correlate with the KD of the chelators for Ni(II) (EDTA < B7-peptide <

EGTA).

Effect of SlyD on GxP bound form of HypB. In addition to the metal-binding activity, the

GTPase function of HypB is also essential for the biogenesis of the [NiFe] hydrogenase enzyme

(9, 28). HypB can bind GDP with micro-molar affinity in vitro (26, 27). GTP hydrolysis is

routinely used as a molecular switch to alter the conformation of proteins (29), thus it is possible

that binding of nucleotide could affect the high-affinity Ni(II) site of HypB. In order to

investigate this possibility, Ni(II) release from the N-terminal metal site of HypB to EDTA was

carried out in the presence of GDP or GMP-PNP, a non-hydrolysable analogue of GTP. The

addition of 500 M GDP to 100 M HypB leads to ~90 % of the protein bound to GDP as

detected by ESI-MS (Figure 4-8). However, in agreement with previous data, GMP-PNP

appears to have lower affinity for HypB when compared to GDP and only results in ~60% GTP

bound form (Figure 4-8) (28). No change in metal release profiles could be detected when HypB

alone was incubated with either form of the nucleotide indicating that the affinity of the Ni(II)

site is not affected when GMP-PNP or GDP is bound by HypB (Figure 4-9). In contrast, upon

addition of apo SlyD to the protein mixture, an accelerated loss of Ni(II) was detected for the

0

0.2

0.4

0.6

0.8

1

B7 peptide EGTA EDTA

Frac

tion

of N

i(II)

boun

d H

ypB

120

GDP-bound form of HypB, whereas no change in metal release profile was observed for GMP-

PNP-bound HypB when compared to that of HypB without any nucleotide (Figure 4-9).

Figure 4-8. GxP binding to HypB via ESI MS. A representative reconstructed mass spectrum for

HypB bound to 1 equivalent of Ni(II) at the high-affinity site, denoted as holo HypB (top panel).

Incubation of 100 M holo HypB with 500 M GDP (middle panel) or the GTP analog, GMP-PNP

(bottom panel) followed by 10-fold dilution prior to MS analysis results in the formation of holo HypB-

nucleotide complexes. The relative abundance of 90% for GDP-HypB and 60% for GMP-PNP-HypB

was calculated based on the peak intensities of each species.

Mass (Da)

31400 31600 31800 32000

Rel

ativ

e In

tens

ity

Holo HypB

holo HypB + GDP

holo HypB

holo HypB

holo HypB + GMP-PNP

121

Figure 4-9. SlyD together with GDP enhances the nickel release rate from HypB. Amount of holo

HypB (100 M) remaining upon equilibration of the protein in 1 mM EDTA for 15 mins. (filled bars).

Including SlyD (100 M) in the HypB, EDTA reaction mixture results in a larger apo HypB fraction

under identical equilibration conditions (unfilled bars). An additional acceleration in the metal release

rate is observed when GDP is added to the HypB/SlyD/EDTA mixture, whereas no change in metal

release rate observed when a GTP analog (GMP-PNP) is added to the mixture. These metal release

assays were analyzed by using ESI-MS.

As we could not detect fully GMP-PNP loaded HypB through mass spectrometry, the observed

nucleotide-dependent metal release activity of HypB was further examined by using electronic

absorption spectroscopy. This analysis method allowed for the inclusion of 1 mM MgCl2 in the

protein buffer (MS is not tolerant to such high salt concentrations), which would facilitate

binding of the nucleotide to the protein. Ni(II) binding to the Cys residues at the N-terminal

metal site of HypB results in a ligand to metal charge transfer (LMCT) band at 320 nm (320

7300 M-1

cm-1

) (10). Therefore, the loss of metal from HypB can be monitored by the decrease

in the intensity of this LMCT band. Taking advantage of the knowledge that SlyD addition to

HypB does not lead to Ni(II) transfer to SlyD (discussed above), the decrease in LMCT should

then correlate with Ni(II) loss from HypB as EDTA chelates Ni(II). In these metal release assays

the change at 320 nm was monitored over a 2 h time period upon mixing holo HypB and apo

SlyD (100 M each) in the presence of 1 mM EDTA. To evaluate the effect of the nucleotide,

holo HypB was incubated with 500 M nucleotide for 30 min at room temperature prior to

addition of SlyD and EDTA. The resulting metal release data are presented in Figure 4-10. In

agreement with the ESI-MS data, an increase in SlyD-dependent metal release is observed that is

further accelerated when HypB is bound to GDP. By fitting the metal release profiles to a 1st

0

0.2

0.4

0.6

0.8

1

No Nucleotide + GTP analog + GDP

Fra

ctio

n o

f N

i(II)

bo

und

Hyp

B

122

order decay event a t1/2 ~ 24 h and 73 min was obtained for HypB and for the HypB-SlyD

mixture, respectively. The half-life for the GMP-PNP loaded HypB in the presence of SlyD

remains unaffected (t1/2 ~ 75 min) whereas the half-life of the GDP bound form decreases to t1/2 ~

33 min. Therefore, the data provide clear evidence that SlyD-dependent modulation of the Ni(II)

site of HypB is further enhanced when HypB is bound to GDP.

Figure 4-10. Metal release from HypB to EDTA monitored by electronic absorption spectroscopy.

Change in the fraction of 100 M Ni(II)-bound HypB in the presence of 1 mM EDTA (empty symbols)

monitored by decreasing absorption of the LMCT band at 320 nm. Metal release from HypB (empty

squares) is not affected upon addition of GDP (empty circles) or GTP analog (unfilled diamonds) to

HypB. Including 100 M SlyD in the reaction mixture (filled symbols) results in an increase in the rate

of metal release. Metal release from HypB bound to GTP analog is not effected (filled diamonds)

whereas SlyD further accelerates the metal release process when HypB is bound to GDP (filled circles).

GTPase activity of HypB. Similar to most other small guanine nucleotide-binding proteins, the

intrinsic GTP hydrolysis rate of E. coli HypB is very slow (kcat = 0.17 min-1

) (28, 30). However

for some of these proteins, an interaction with a partner protein increases its enzymatic activity

(31, 32). To test whether SlyD has this effect on HypB, GTP hydrolysis rates under saturation

conditions were monitored with or without SlyD. HypB with Ni(II) bound at the N-terminal

metal site catalyzes the hydrolysis of GTP with a kcat of 0.27 ± 0.06 min-1

, a rate that is slightly

higher but comparable to that has been observed before (28, 30). A moderate three-fold increase

in the catalytic rate is observed when HypB is pre-incubated with SlyD (Table 4-3) and control

0

0.2

0.4

0.6

0.8

1

0 15 30 45 60 75 90 105 120

Fra

ction o

f N

i(II)-

bou

nd H

ypB

Time (min)

123

experiments with SlyD alone were conducted to ensure that the increase in GTP hydrolysis is

from HypB only.

Table 4-3. GTP hydrolysis rates of HypB under saturating conditions.

Metal

bound at the

G-domain

HypBWT

kcat (min-1

)

HypBWT +

SlyDWT

kcat (min-1

)

HypBWT +

SlyD1-146

kcat (min-1

)

HypBC2,5,7A

kcat (min-1

)

-- 0.27 ± 0.06 0.82 ± 0.04 0.36 ± 0.08 0.25 ± 0.09

Ni(II) 0.13 ± 0.03 0.87 ± 0.01 0.12 ± 0.03 0.12 ± 0.02

Zn(II) 0.07 ± 0.04 0.91 ± 0.09 0.06± 0.04 0.04 ± 0.01 Assays were performed using 1 M HypB protein and 450 M GTP and when required SlyD was included in the

reaction at a final concentration of 50 M. HypB samples with SlyD and or ligand were incubated for 1 hr at room

temperature prior to assay. The kcat are averages and standard deviations from more than three independent

experiments. To load the G-domain metal site, protein was incubated with 10 molar excess Ni(II) or 1.5 molar

excess Zn(II).

E. coli HypB contains a second metal-binding site within the G-domain of the protein that can

bind Zn(II) or Ni(II) with a KD of 1 M and 12 M, respectively (10). To assess whether SlyD

could still affect GTPase activity when HypB is metallated at both metal sites, HypB with either

Ni(II) or Zn(II) bound to the low-affinity site was incubated with SlyD, and the hydrolysis rates

of these protein mixtures were measured. HypB activity was reduced when the G-domain metal

site was metallated with either Zn(II) or Ni(II) as has been shown previously (30). Inclusion of

SlyD in the reaction mixture results in complete recovery of enzymatic activity with hydrolysis

rates comparable to SlyD/HypB mixture in which HypB is only bound to Ni(II) at the N-terminal

metal site (Table 4-3). Given that SlyD can bind multiple Ni(II) and or Zn(II) ions with an

affinity in the nanomolar range ((19) and Kaluarachchi, H., Zamble D. B., unpublished data), the

observed recovery could be a result of metal loss from the G-domain metal due to chelation by

SlyD. To test this possibility, GTP hydrolysis of HypB with metal bound at the G-domain was

measured in the presence of SlyD1-146. This SlyD variant lacking the MB domain is unable to

bind either Ni(II) or Zn(II) (19) and Kaluarachchi, H., Zamble, D. B., unpublished data). Unlike

the WT protein, addition of the SlyD1-146 mutant did not lead to the moderate increase in GTPase

activity, suggesting that the MBD of SlyD is necessary for this activity. In addition, the mutant

failed to rescue the inhibition of enzymatic activity when the G-domain metal site was occupied

with either nickel or zinc (Table 4-3). To verify that SlyD did indeed demetallate the low-

affinity metal site of HypB, a holo HypB sample (10 M) was incubated with 0.8 equivalents of

Zn(II) for 30 min, followed by incubation of this sample with apo SlyD (10 M) which was

124

analyzed via ESI-MS. The resulting mass spectra shows that this metal transfer process is ~95%

complete within 15 min (Figure 4-11) signifying that the experimental conditions used are

sufficient to remove all the metal from the G-domain.

Figure 4-11. SlyD removes Zn(II) from the G-domain of HypB. Representative reconstructed mass

spectra corresponding to holo HypB containing 0.8 equivalents of Zn(II) (top right panel) and apo SlyD

(top left panel) prior to mixing. Reconstructed spectra of the proteins after equilibrating the solution for

15 min at room temperature (bottom panels). Quantitation of the Zn(II) peak by area indicate that >90%

of the Zn(II) has been removed from the HypB and is subsequently coordinated by SlyD.

Analysis of this process using SlyD1-146 does not reveal any metal loss from HypB even after

allowing the proteins to incubate for 2 h (Figure 4-12), indicating that the MBD of SlyD is

necessary for this process. Given that the affinity for Ni(II) to this G-domain site is 10-fold

lower than that for Zn(II), this chelation by SlyD will also remove Ni(II) thus leading to the

observed recovery of the GTPase activity (Table 3). Therefore, the MS data supports a model in

which SlyD removes the metal bound at the G-domain of HypB thus leading to recovery of the

GTPase activity of HypB.

Mass (Da)

20700 20900 21100 21300

Mass (Da)

31300 31400 31500 31600

Rel

ativ

e In

tens

ity

0 1 2 3 +Ni(II) +Ni(II) & Zn(II)

125

Figure 4-12. SlyD1-146 cannot competitively remove Zn(II) bound to the G-domain of HypB.

Representative reconstructed mass spectra corresponding to holo HypB containing 0.8 equivalents of

Zn(II) (top right panel) and apo SlyD1-146 (top left panel) prior to mixing. Reconstructed spectra of the

proteins after equilibrating the solution for 2 h at room temperature (bottom panels). No change in the

mass spectra are detected indicating Zn(II) remains bound to HypB.

To assess whether high-affinity nickel site modulates the enzymatic activity of HypB, the

GTPase activity of the C2,5,7A HypB mutant was analyzed. While this mutant protein contains

an intact G-domain metal site, it cannot bind Ni(II) at the N-terminus because the three Cys

residues coordinating the Ni(II) have been mutated to Ala residues (10). The kcat measured for

the apo protein and for protein with either Zn(II) or Ni(II) bound at the G-domain was analogous

to that of the wildtype HypB indicating that the N-terminal nickel-site does not alter the GTPase

activity of the protein (Table 4-3).

Mass (Da)

20700 20900 21100 21300

Mass (Da)

31300 31400 31500 31600

Rel

ativ

e In

tens

ity

Apo +Zn(II) +Ni(II) +Ni(II) & Zn(II)

126

4.4 Discussion

The lack of a poly-histidine tail in E. coli HypB compared to its homologs prompted a search for

proteins that could functionally replace this nickel storage activity. This search led to the

discovery of SlyD as a protein that contributes to the maturation of the [NiFe]-hydrogenase

enzymes in this bacteria (20). The C-terminal region of SlyD corresponds to an unusual metal-

binding region that is rich in both the number and diversity of available metal-binding residues,

which are proposed to confer nickel storage capability to SlyD. These stored metal ions are then

anticipated to serve as a source of nickel when bacteria are faced with metal scarce

environments. Although analysis of SlyD in vitro and in vivo thus far supports this model (8, 19,

20), the experiments described herein provide direct evidence to validate this hypothesis. When

comparing bacteria cultured with or without the nickel chelator, DMG, the cytoplasmic nickel

content of the ΔslyD strain is significantly depleted in comparison to that of the wildtype strain

indicating that SlyD is a major determinant of nickel storage in E. coli. Similarly, when

considering the amount of active hydrogenase produced, the inhibitory effect of the chelator is

more pronounced in the ΔslyD strain, indicating bacteria without SlyD are highly-dependent on

exogenous nickel for [NiFe]-hydrogenase assembly. Therefore these results provide

physiological evidence that SlyD acts as a nickel reservoir in E. coli and this storage capacity is

useful for enzyme biosynthesis when nickel is scarce.

Given that SlyD was originally found to contribute to hydrogenase biosynthesis through its

interaction with HypB (20) and that HypB has two metal-binding sites required for hydrogenase

biosynthesis (9, 10), it was feasible that HypB could function as an acceptor of nickel from SlyD.

The results from the metal transfer assays reveal that SlyD can metallate the high-affinity nickel

site of HypB even under competitive conditions albeit with slow metal transfer rates. These data

imply that stored Ni(II) ions may be mobilized from SlyD to precursor hydrogenase enzyme, via

HypB. This observed transfer within a SlyD-HypB complex suggests that the critical nickel can

be supplied to HypB without relying on diffusion, thus avoiding competition with the abundant

metal-chelating molecules present in the cytoplasm. The slow metallation of HypB suggests that

HypB and SlyD alone are not sufficient to achieve ideal metal transfer conditions. Additional

factors that could contribute to an acceleration of this process may include interactions with other

protein partners such as HypA. It is also possible that docking of the Ni(II)-insertion protein

complex onto the large subunit of the precursor enzyme may enhance this process. Nevertheless,

127

these results together with the in vivo data provide evidence that E. coli can use SlyD to supply

Ni(II) for enzyme maturation under nickel limiting conditions.

Although SlyD may act as a nickel source, delivery of nickel by HypB is considered to be the

privileged pathway because hydrogenase maturation is completely abolished in the absence of

HypB (8, 33, 34) whereas deletion of SlyD only results in decreased hydrogenase activity in vivo

(8). Furthermore, by adding excess Ni(II) to the media, hydrogenase activity in the ΔslyD strain

can be fully restored whereas only partial recovery is detected for ΔhypB (8, 33, 34). The ideal

location of the nickel site (N-terminally located Ni(II) site that is readily accessible for metal

exchange) coupled to the fact that HypB can bind nickel with high affinity (1.2 x 10-13

M, (10))

would enable this protein to compete for nickel and act as a Ni-chaperone that funnels sufficient

metal ions for the hydrogenase enzyme biogenesis. However, if metal transfer from HypB to the

precursor enzyme was to rely on diffusion, the extremely high affinity of HypB would translate

to a very slow metal release rate from HypB (~19 h, assuming a kon only limited by diffusion),

which is likely irrelevant considering the short life cycle (i.e. minutes) of E. coli. Therefore, it is

not surprising that an additional protein factor such as SlyD is necessary to make this nickel

available for hydrogenase biosynthesis on a physiologically relevant time-frame. The results

presented here and by Leach et al. (8) reveal that the interaction of SlyD with HypB leads to a

dramatic increase in the rate of nickel release from the N-terminal metal site. A survey of SlyD

variants indicates that while the Cys residues on SlyD are not necessary for the modulation of

metal-binding activity of HypB, a successful heterodimer formation as well as the metal-binding

domain of SlyD are an absolute must for activation of metal release as has been reported

previously (8). Investigation of different chelators also revealed that the metal release process

can be enhanced by addition of acceptors with thiol moieties. This implies that Ni(II) release

from HypB can be further catalyzed upon docking of HypB onto the precursor enzyme, which

would place the Ni(II) ion in close proximity to the enzyme active site consisting of Cys

residues. Use of thiol exchange reactions is a common feature among the copper

metallochaperones such as Atox1 and Ccs2 (35), whether these types of reactions are also used

by HypB for metal transfer is an issue that remains to be verified in the future.

The requirement for an NTPase during the assembly of enzyme metallocentres appears to be an

emerging theme (6). This concept is exemplified by HypB, which contains a C-terminal GTPase

domain (i.e. G-domain) that is conserved among the HypB homologues (6). GTP binding and

128

hydrolysis are mandatory for assembly of the enzyme active-site because mutations that

abrogates GTPase activity in vitro abolishes enzyme activity in vivo (28). Although it is known

that this enzymatic function of HypB is essential, research to date has not clearly established as

to how energy derived from the hydrolysis step is utilized during the assembly of the [NiFe]-

centre of the enzyme. However, answers to this key issue can be obtained from the results

presented in this report.

First, the SlyD-mediated decrease in affinity of the N-terminal nickel site is enhanced when

HypB is bound to GDP, but GDP alone does not modulate the high-affinity site, suggesting that a

more favourable interaction is achieved between SlyD and GDP-bound HypB. Currently the

only other known GDP-specific effect on E. coli HypB is that the presence of nucleotide

enhances dimerization of HypB (30). However, it is unlikely that the observed change is a result

of this altered oligomeric state of HypB as we would expect to observe a similar enhancement

for GTP as well, since dimerization is not dependent on the identity of the nucleotide (30). A

similar enhancement in an interaction upon nucleotide binding has been observed between the

GTPase chaperone MeaB and the B12-dependent isomerase, methylmalonyl-CoA mutase

(MCM). The affinity of the protein complex of the cobalt-dependent MCM and MeaB was

found to increase by ~15 fold when MeaB is bound to a GTP when compared to the GDP-bound

form of MeaB (32). Evidence for nucleotide-dependent protein-protein interactions is also

observed during assembly of the urease active site, where mutations that decrease nucleotide

binding to the GTPase UreG prevents docking of the accessory protein complex UreGDF onto

the precursor urease enzyme (36). The results presented here also suggest that while the N-

terminal nickel-site does not contribute to enzymatic activity of the protein, GTP hydrolysis can

affect the metal-transfer from the N-terminal site of HypB. However, this link is only achieved

when SlyD is present because nucleotide-dependent increase in metal release is not observed for

HypB alone. This gating mechanism by SlyD is likely a useful mechanism in vivo, which would

ensure that HypB does not release the Ni(II) routed for the [NiFe]-hydrogenase enzyme

prematurely. Furthermore, these observations supports a model where GTP hydrolysis precedes

the Ni(II) insertion step to the precursor enzyme as GDP enhances the metal-release from HypB-

SlyD complex.

Second, in addition to modulating the high-affinity metal-binding site on HypB, SlyD is able to

moderately enhance the GTPase activity of HypB. Although the effect is minor this is the first

129

protein factor discovered so far that can positively contribute to the enzymatic activity of HypB.

This enhancement of the low intrinsic hydrolysis rate upon binding to a partner protein gives us

insight into how protein-protein interactions can accelerate biological processes and is not

uncommon among NTPases. For example, HydF (the GTPase involved in metal insertion into

the[FeFe]-hydrogenase) enzymatic activity increases by 50% in the presence of the other

accessory protein HydE or HydG (31). The aforementioned MeaB provides another example, as

its GTP hydrolysis rate increases ~100 fold in the presence of the MCM enzyme (32). It is

noteworthy that while many of these partner proteins significantly affect the hydrolysis rates, the

effect of SlyD is very minor. This indicates that SlyD alone is not sufficient to fulfill the role of

the GTP-activating protein (GAP) and additional factors such as HypA or even the precursor

enzyme may be required to enhance the GTPase activity of HypB.

Third, the G-domain metal site of HypB is demetallated in the presence of SlyD. As reported

here and by Cai, et al. (30), metal binding to this G-domain site is linked to the GTPase activity

and binding of either Zn(II) or Ni(II) inhibits GTPase turnover (30). The G-domain metal site

and/or the link it provides to GTP hydrolysis is essential for hydrogenase maturation because

mutating any of the residues that coordinate the metal ion abolishes in vivo hydrogenase activity

(9). This metal-induced inhibition is circumvented by the presence of SlyD because SlyD

competitively binds these metal ions with a much higher affinity than HypB. This metal removal

by SlyD leading to recovery of the GTPase activity of HypB is further supported by the fact that

SlyD1-146, which cannot bind metal with an appreciable affinity (19), is unable to reverse this

metal-dependent inhibition. By combining the in vitro data presented here and our current

knowledge of HypB, the in vivo significance of this G-domain metal site can now be inferred. It

is likely that this metal site functions as a regulatory switch for the GTPase function of HypB,

such that metal loading to the G-domain site will prevent the protein from needlessly

hydrolyzing GTP until a successful Ni(II)-insertion complex is formed with the other partner

proteins. Successful protein-protein complex formation and subsequent metal extraction by

SlyD then allows HypB to begin GTP hydrolysis, which results in successful metal transfer to

the precursor enzyme.

In summary, the results presented here provide evidence that the role of SlyD in hydrogenase

maturation goes beyond that of a reservoir of nickel ions needed during hydrogenase

biosynthesis under nickel-limiting conditions. Although SlyD is not absolutely essential for the

130

Ni(II) insertion process, as are HypA or HypB, the results support a model as to how E. coli

utilize this protein to optimize the biogenesis of the [NiFe]-hydrogenase (Figure 4-12).

Additionally we report a link between GTP hydrolysis and the Ni(II)-transfer step from HypB

that is only established in the presence of SlyD. These discoveries give us insight as to how the

protein-protein interactions as well as the GTPase function of HypB can be exploited to ensure

the timely delivery of the metal ion. Furthermore, these protein-protein mediated effects

highlight the importance of the coordinated actions of multiple protein factors to overcome the

thermodynamic and kinetic barriers of inserting the nickel into the partially assembled

hydrogenase. It is likely that this process is more complicated than presented here, as the effect

of HypA was not assessed in these experiments, although this protein is known to be necessary

for enzyme maturation (19). Nevertheless, the results presented here serves as a starting point to

elucidate the exact mechanism of nickel insertion and the interplay of the metal-binding and

enzymatic activities of the proteins in the nickel insertion complex HypA-HypB-SlyD.

131

Scheme 4-12. Proposed model for Ni(II) insertion to the hydrogenase precursor under limited (a)

and abundant (b) exogenous nickel availability. The initial two steps involve metal transfer within the

HypB-SlyD complex. Under nickel-limiting conditions, Ni(II) stored on SlyD is transferred to apo HypB

(1a → 2a). Under abundant nickel conditions HypB is metallated at the high-affinity site as well as the

G-domain site and SlyD competitively removes Ni(II) from the G-domain, readying HypB for GTP

hydrolysis (1b → 2b). The HypB/SlyD complex associates with HypA, which anchors this Ni(II)-

insertion complex to the large subunit of the precursor enzyme (denoted as H2ase). The third step involves

GTP hydrolysis by HypB and the catalytic rate is now enhanced (denoted by change in colour of HypB)

due to the formation of the multi-protein complex including SlyD and the absence of metal coordinated at

the G-domain site. Upon GTP hydrolysis, the GDP-bound HypB in association with SlyD and/or HypA

enables facile transfer of the Ni(II) from the N-terminal site of HypB. Following dissociation of the

nickel insertion complex, the final step involves recognition of the Ni(II)-bound hydrogenase enzyme by

the HycI protease, which leads to C-terminal cleavage and subsequent folding into a mature enzyme.

HypA

HypB

GDP

SlyD

Ni2+ Ni2+

Ni2+ Ni2+ +

SlyD

Ni2+ Ni2+

Ni2+ Ni2+

HypB

GTP

HypA

(1a)

H2ase

HypA

SlyD

Ni2+

Ni2+ Ni2+

HypB

GTP

Ni2+

Fe2+

HypB

GTP

SlyD

Ni2+ Ni2+

+

SlyD

Ni2+ Ni2+

HypB

GTP

Ni2+

Ni2+

HypB

GTP

Ni2+

Ni2+

HypA

GTP

GDP + Pi

H2ase

Fe2+

HypA

SlyD

Ni2+

Ni2+ Ni2+HypB

Ni2+

GDP

(4)

Ni2+

(2a)

(3)

SlyD

Ni2+

Ni2+ Ni2+

Ni2+

H2ase

Ni2+

HycI

(5)

Ni(II) limited conditions Abundant Ni(II)

(1b)

(2b)

132

References

1. Waldron, K. J., and Robinson, N. J. (2009) How do bacterial cells ensure that

metalloproteins get the correct metal?, Nat. Rev. Microbiol. 7, 25-35.

2. Kuchar, J., and Hausinger, R. P. (2004) Biosynthesis of metal sites, Chem. Rev. 104, 509-

525.

3. Waldron, K. J., Rutherford, J. C., Ford, D., and Robinson, N. J. (2009) Metalloproteins

and metal sensing, Nature 460, 823-830.

4. Forzi, L., and Sawers, R. G. (2007) Maturation of [NiFe]-hydrogenases in Escherichia

coli, Biometals 20, 565-578.

5. Ragsdale, S. W. (2009) Nickel-based Enzyme Systems, J. Biol. Chem. 284, 18571-18575.

6. Kaluarachchi, H., Chan Chung, K. C., and Zamble, D. B. (2010) Microbial nickel

proteins, Nat. Prod. Rep. 27, 681-694.

7. Roof, W. D., Fang, H. Q., Young, K. D., Sun, J., and Young, R. (1997) Mutational

analysis of slyD, an Escherichia coli gene encoding a protein of the FKBP immunophilin

family, Mol. Microbiol. 25, 1031-1046.

8. Leach, M. R., Zhang, J. W., and Zamble, D. B. (2007) The role of complex formation

between the Escherichia coli hydrogenase accessory factors HypB and SlyD, J. Biol.

Chem. 282, 16177-16186.

9. Dias, A. V., Mulvihill, C. M., Leach, M. R., Pickering, I. J., George, G. N., and Zamble,

D. B. (2008) Structural and biological analysis of the metal sites of Escherichia coli

hydrogenase accessory protein HypB, Biochemistry 47, 11981-11991.

10. Leach, M. R., Sandal, S., Sun, H., and Zamble, D. B. (2005) Metal binding activity of the

Escherichia coli hydrogenase maturation factor HypB, Biochemistry 44, 12229-12238.

11. Li, Y., and Zamble, D. B. (2009) Nickel homeostasis and nickel regulation: an overview,

Chem. Rev. 109, 4617-4643.

12. Olson, J. W., and Maier, R. J. (2000) Dual roles of Bradyrhizobium japonicum nickelin

protein in nickel storage and GTP-dependent Ni mobilization, J. Bacteriol. 182, 1702-

1705.

13. Maier, T., Lottspeich, F., and Bock, A. (1995) GTP hydrolysis by HypB is essential for

nickel insertion into hydrogenases of Escherichia coli, Eur. J. Biochem. 230, 133-138.

14. Mehta, N., Benoit, S., and Maier, R. J. (2003) Roles of conserved nucleotide-binding

domains in accessory proteins, HypB and UreG, in the maturation of nickel-enzymes

required for efficient Helicobacter pylori colonization, Microb. Pathogen. 35, 229-234.

15. Olson, J. W., Fu, C., and Maier, R. J. (1997) The HypB protein from Bradyrhizobium

japonicum can store nickel and is required for the nickel-dependent transcriptional

regulation of hydrogenase, Mol. Microbiol. 24, 119-128.

16. Scholz, C., Eckert, B., Hagn, F., Schaarschmidt, P., Balbach, J., and Schmid, F. X. (2006)

SlyD proteins from different species exhibit high prolyl isomerase and chaperone

activities, Biochemistry 45, 20-33.

17. Hottenrott, S., Schumann, T., Pluckthun, A., Fischer, G., and Rahfeld, J. U. (1997) The

Escherichia coli SlyD is a metal ion-regulated peptidyl-prolyl cis/trans-isomerase, J. Biol.

Chem. 272, 15697-15701.

18. Hottenrott, S., Schumann, T., Pluckthun, A., Fischer, G., and Rahfeld, J. U. (1997) The

Escherichia coli SlyD is a metal ion-regulated peptidyl-prolyl cis/trans-isomerase, J. Biol.

Chem. 272, 15697-15701.

133

19. Kaluarachchi, H., Sutherland, D. E., Young, A., Pickering, I. J., Stillman, M. J., and

Zamble, D. B. (2009) The Ni(II)-binding properties of the metallochaperone SlyD, J. Am.

Chem. Soc. 131, 18489-18500.

20. Zhang, J. W., Butland, G., Greenblatt, J. F., Emili, A., and Zamble, D. B. (2005) A role

for SlyD in the Escherichia coli hydrogenase biosynthetic pathway, J. Biol. Chem. 280,

4360-4366.

21. Ballantine, S. P., and Boxer, D. H. (1985) Nickel-Containing Hydrogenase Isoenzymes

from Anaerobically Grown Escherichia-Coli K-12, J. Bacteriol. 163, 454-459.

22. Benanti, E. L., and Chivers, P. T. (2009) An intact urease assembly pathway is required

to compete with NikR for nickel ions in Helicobacter pylori, J. Bacteriol. 191, 2405-

2408.

23. Seshadri, S., Benoit, S. L., and Maier, R. J. (2007) Roles of His-rich hpn and hpn-like

proteins in Helicobacter pylori nickel physiology, J. Bacteriol. 189, 4120-4126.

24. Maier, T., Jacobi, A., Sauter, M., and Bock, A. (1993) The product of the hypB gene,

which is required for nickel incorporation into hydrogenases, is a novel guanine

nucleotide-binding protein, J. Bacteriol. 175, 630-635.

25. Boys, B. L., Kuprowski, M. C., Noel, J. J., and Konermann, L. (2009) Protein oxidative

modifications during electrospray ionization: solution phase electrochemistry or corona

discharge-induced radical attack?, Anal. Chem. 81, 4027-4034.

26. Takamoto, K., and Chance, M. R. (2006) Radiolytic protein footprinting with mass

spectrometry to probe the structure of macromolecular complexes, Annu. Rev. Biophys.

Biomol. Struct. 35, 251-276.

27. Martell, A. E., and Smith, R. M. (2003) NIST Standard Reference Database 46, version

8.0, National Institue of Standards and Technology, Rockville, MD.

28. Maier, T., Lottspeich, F., and Bock, A. (1995) GTP hydrolysis by HypB is essential for

nickel insertion into hydrogenases of Escherichia coli, FEBS J. 230, 133-138.

29. Caldon, C. E., Yoong, P., and March, P. E. (2001) Evolution of a molecular switch:

universal bacterial GTPases regulate ribosome function, Mol. Microbiol. 41, 289-297.

30. Cai, F., Ngu, T. T., Kaluarachchi, H., and Zamble, D. B. Relationship between the

GTPase, metal-binding, and dimerization activities of E. coli HypB, J. Biol. Inorg. Chem.

In Press.

31. Broderick, J. B., Shepard, E. M., McGlynn, S. E., Bueling, A. L., Grady-Smith, C. S.,

George, S. J., Winslow, M. A., Cramer, S. P., and Peters, J. W. (2010) Synthesis of the

2Fe subcluster of the [FeFe]-hydrogenase H cluster on the HydF scaffold, Proc. Nat.

Acad. Sci. U.S.A. 107, 10448-10453.

32. Padovani, D., Labunska, T., and Banerjee, R. (2006) Energetics of interaction between

the G-protein chaperone, MeaB, and B-12-dependent methylmalonyl-CoA mutase, J.

Biol. Chem. 281, 17838-17844.

33. Maier, T., Jacobi, A., Sauter, M., and Bock, A. (1993) The Product of the Hypb Gene,

Which Is Required for Nickel Incorporation into Hydrogenases, Is a Novel Guanine

Nucleotide-Binding Protein, J. Bacteriol. 175, 630-635.

34. Olson, J. W., Mehta, N. S., and Maier, R. J. (2001) Requirement of nickel metabolism

proteins HypA and HypB for full activity of both hydrogenase and urease in Helicobacter

pylori, Mol. Microbiol. 40, 270-270.

35. Robinson, N. J., and Winge, D. R. Copper metallochaperones, Annu. Rev. Biochem. 79,

537-562.

36. Moncrief, M. B., and Hausinger, R. P. (1997) Characterization of UreG, identification of

a UreD-UreF-UreG complex, and evidence suggesting that a nucleotide-binding site in

134

UreG is required for in vivo metallocenter assembly of Klebsiella aerogenes urease, J.

Bacteriol. 179, 4081-4086.

135

Chapter 5 Summary and Perspective

Enzymes that use nickel are considered to be relics from an era pre-dating the generation of

dioxygen by photosynthesis into the atmosphere, a time when nickel was more plentiful (1).

Nevertheless nickel still serves as an essential catalytic cofactor in many enzymes found in

eubacteria, archaebacteria, algaea, fungi and plants today (2). Currently there are eight known

nickel enzymes that catalyze a plethora of reactions resulting in the use and/or production of

gases such as carbon monoxide, methane, hydrogen and oxygen (3). In addition to the enzymes

that employ nickel, organisms with a requirement for nickel also express a variety of proteins

that contribute to the transport, distribution, regulation as well as assembly of the nickel active

sites. These proteins ensure that proper balance of metal is maintained, as excess nickel, like

many other transition metals can be detrimental to an organism. Although these proteins

involved in nickel homeostasis are essential to the survival of an organism, not much is known

about the mechanisms of action of these individual components, how they interrelate within the

pathways, or how metal selectivity is achieved. This dissertation focuses on understanding the

molecular details of one such component in E. coli nickel homeostasis, SlyD and provides some

insight into the rules and requirements that govern intracellular transition metal homeostasis at a

molecular level.

5.1 Ni(II)-binding properties of SlyD

The biosynthesis of the intricate metallocentre in [NiFe]-hydrogenase requires the coordinated

action of several accessory proteins that in E. coli are encoded by hypA-F (2). SlyD was recently

identified as an additional accessory protein involved in this maturation pathway, specifically

contributing to the nickel delivery step to the enzyme (4). SlyD posses a very unique C-terminal

sequence that is rich in both the number and types of metal-binding ligands. A vital role for this

metal-binding domain was shown as deletion of this domain compromises the activity of SlyD in

hydrogenase maturation as well as nickel accumulation in E. coli (5). To understand the

molecular details of this in vivo function, the nickel-binding activity of SlyD was investigated by

using a combination of various spectroscopic methods and mass spectrometry. The study

demonstrated that SlyD binds multiple nickel ions with a capacity for coordinating up to seven

136

nickel ions. The analysis of non-covalent metal-protein complexes via ESI-MS revealed that

SlyD coordinates nickel in a non-cooperative manner with affinity in the nanomolar range. In

addition, the ESI-MS revealed that the protein exists in a mixture of metalloforms that is

dependent on the availability of nickel ions in solution. Structural analysis via CD spectroscopy

suggested that this metallochaperone undergoes small but distinct changes in the structure upon

metal binding and that these nickel-binding sites are predominantly assembled through β-turn

formation. Although the C-terminal metal-binding domain is primarily responsible for metal

chelation, metal binding also perturbs the structure of the N-terminal domains. An investigation

of the nickel sites by X-ray absorption spectroscopy (XAS) showed that SlyD binds nickel ions

by using several different geometries and coordination numbers that range from 4-6 coordinate.

Also, the XAS data suggest that Cys and His residues as well as carboxylate moieties are utilized

by the protein to coordinate the nickel ions. Finally, the characterization of SlyD mutants

demonstrated that the cysteine residues in the C-terminal domain confer tighter affinity as well as

increased binding capacity to SlyD but are not essential for Ni(II) binding. However, in

agreement with the vital role of the metal-binding domain for hydrogenase activity in vivo,

deletion of this region completely abolishes the ability of the protein to coordinate nickel ions.

Therefore, this investigation provided a detailed analysis delineating the molecular basis of

nickel binding by SlyD in vitro and provides support for the proposed nickel-storage role of SlyD

in nickel homeostasis (3).

5.2 Metal specificity of SlyD

Although SlyD can bind nickel and is an important factor in nickel homeostasis in E. coli,

previous investigations demonstrated that this protein can also coordinate several other types of

first row transition metals (6). This observation raised the question of whether SlyD could

preferentially coordinate nickel in the presence of other transition metal ions. In addition,

because SlyD can bind these other metals, it was feasible that the protein contributes to metal

homeostasis pathways other than nickel. In order to answer these questions, the in vitro metal-

binding activity of SlyD to several biologically relevant transition metals: Mn(II), Fe(II), Co(II),

Cu(I) and Zn(II) was examined. Using a combination of mass spectrometry and spectroscopic

methods, it was established that while binding to manganese or ferrous ions was not detected the

protein could coordinate multiple Co(II), Zn(II) as well as Cu(I) signifying the promiscuous

137

nature of this protein. SlyD binds Co(II) with an average stoichiometry of 4 and exhibits

nanomolar affinity for this ion (average KD of 4±1 x 10-9

M) whereas it can coordinate up to 5

zinc ions with an average KD of 1.0±0.4 x 10-10

M. Cu(I) complexes of SlyD have the greatest

stability as the protein binds 4 cuprous ions with an average KD of 1.5 ± 0.3 x 10-17

M. Analysis

of the metal-binding mechanism via ESI-MS revealed that similar to nickel, zinc and cobalt bind

to the protein utilizing a non-cooperative mechanism while Cu(I) binding to the protein followed

a partially cooperative metallation pathway. Through competition titrations the ability of SlyD

to preferentially coordinate nickel in the presence of a second metal ion was investigated. These

metal selectivity studies revealed that the protein has the following preference for metal Mn(II),

Fe(II) < Co(II) < Ni(II), Zn(II) << Cu(I), which correlates with the individually measured

thermodynamic affinities of the protein for each metal. This order of preference also matches the

universal order of preferences for divalent metal ions, termed the Irving-Williams series (1).

Therefore, the in vitro data indicate that SlyD is a flexible protein that can accommodate a

variety of divalent metal ions, and this flexible nature leads to imperfect selection for nickel.

To determine whether the in vitro metal-binding activity of these other transition metals

translates into participation of SlyD in other metal homeostasis pathways, several in vivo assays

were conducted. Initial assays tested whether SlyD is necessary to confer metal resistance to

bacteria and the results revealed that the presence of the protein was not necessary for growth of

the bacteria at high metal concentration of cobalt, nickel, copper or zinc. Thus, these

investigations suggested that SlyD does not play the role of a metal detoxifier as originally

anticipated. The second set of in vivo investigations evaluated the transcriptional level of several

metal transporters in E. coli via qRT-PCR. These transporter expression levels are tightly

regulated by metal-sensing transcription factors that gauge the concentration of a particular metal

with extreme selectivity (7). By comparing the transcription profile of the wildtype strain to a

ΔslyD strain, the impact of the slyD deletion in a given metal homeostasis pathway was

determined. The transcripts of the zinc importer znuA and the exporter zntA were monitored to

determine the effects on zinc homeostasis whereas the two major copper exporters, copA and

cusF were used to gauge the impact on copper homeostasis. Under all the conditions tested, the

mRNA levels detected for these four transporters in the ΔslyD strain did not significantly differ

from that of the wildtype strain, indicating that SlyD is not an essential component in

maintaining the balance of either copper or zinc in E. coli. The role of SlyD in nickel

138

homeostasis was analyzed via monitoring the transcript levels of the nickel importer nikA and the

exporter rcnA. The results obtained for these two importers provide several insights into nickel

homeostasis in E. coli. Firstly, deletion of slyD reduced the nickel capacity of the cytoplasm

such that nikA transcription was down regulated in the absence of the protein. Secondly, deletion

of slyD did not impact the exporter expression, indicating that the protein does not contribute to

the nickel exporting pathway. This evidence supports a model in which E. coli maintains two

distinct nickel pools within the cytoplasm, one maintained by the Nik importer and the second

linked to the RcnA exporter (8), and SlyD contributes only to the nickel pool maintained by the

Nik importer. The nickel brought into the cytoplasm through the Nik importer is considered to

be preferentially routed to [NiFe]-hydrogenase maturation through the action of

metallochaperones. Therefore, these findings corroborate the nickel-related function of SlyD in

enzyme maturation, and strengthen the assignment of this protein as a nickel metallochaperone.

Furthermore, this investigation also showed that although SlyD is unable to overcome the large

thermodynamic preference in vitro for Cu(I) and exclude Zn(II) chelation even in the presence of

nickel, it is able to retain it’s Ni(II)-specific activity in vivo. Thus, this investigation serves to

highlight the crucial role of cellular trafficking pathways for maintaining the relative abundance

of a particular metal that enables a protein to selectively bind its cognate metal within the cell.

5.3 Significance of the HypB-SlyD complex in [NiFe]-hydrogenase biosynthesis

Nickel chaperones play a central role in microbial nickel homeostasis by shuttling nickel within

the cytoplasm and protecting the cell against toxicity during this transit. The nickel chaperone

role during the [NiFe]-hydrogenase biosynthesis is fulfilled by the proteins HypA, HypB and

SlyD (9).

The hypothesized nickel storage role of SlyD is based on the fact that the MBD can coordinate a

significant number of nickel ions (10). These stored metal ions are then anticipated to serve as a

source of nickel when bacteria are faced with metal scarce environments. Chapter 4 provides

experimental evidence to validate this hypothesis where the impact of deletion of slyD on the

overall nickel content and the hydrogenase activity under nickel limiting conditions was

investigated. Nickel limiting growth conditions were achieved through addition of the metal

chelator dimethylglyoxime, which significantly reduces the cytoplasmic nickel content in ΔslyD

in comparison to that of the wildtype. When hydrogenase activity was assessed, the inhibitory

139

effect of the chelator appeared more pronounced in the ΔslyD strain, indicating bacteria without

SlyD are highly-dependent on exogenous nickel for [NiFe]-hydrogenase assembly. Thus the

results presented provide physiological evidence for the hypothesis that SlyD acts as a nickel

reservoir in E. coli and this storage capacity is useful for enzyme biosynthesis when nickel is

scarce.

In addition to serving as a nickel storage protein, SlyD is known to associate with HypB in vivo

and this heterodimer formation is necessary for optimal hydrogenase activity (5). However, the

significance of this interaction and how it enhances the biogenesis of hydrogenase are not clearly

understood. Our hypothesis was that SlyD contributes to hydrogenase maturation by mobilizing

the stored nickel to the precursor enzyme via HypB. This hypothesis was validated through

metal-transfer experiments which indicated that SlyD can indeed act as a nickel donor to HypB,

even in a competitive environment. This transfer of nickel ions within a SlyD-HypB protein

complex suggests that in vivo, the critical nickel is supplied to HypB without relying on

diffusion and thus avoiding competition with the abundant metal chelating molecules present in

the cytoplasm.

In addition to acting as a nickel source, HypB-SlyD interaction can also stimulate Ni(II)-release

from the N-terminal high affinity Ni(II) site of HypB. A survey of SlyD variants indicates that

successful heterodimer formation as well as the metal-binding domain of SlyD are required for

this modulation of the metal-binding activity of HypB. The significance of this result can be

considered in the context of the role of HypB in vivo. This protein is considered to be the

primary nickel carrier for assembly of the [NiFe]-hydrogenase metallocentre because deletion of

the hypB gene abolishes hydrogenase function in vivo. In addition, the protein has ideal

properties of a metallochaperone, an N-terminally located nickel site with high-affinity for

competitive nickel-binding and subsequent easy metal transfer that could funnel sufficient metal

ions to hydrogenase biogenesis. However, if nickel-transfer were to rely on diffusion, enzyme

maturation would be an extremely slow process. Therefore, it is not surprising that an additional

factor like SlyD is necessary to decrease the affinity of the Ni(II) site of HypB to ensure that the

metal-transfer process can occur in a physiologically relevant time frame. Chapter 4 also reports

several other consequences of the HypB-SlyD interaction. Firstly, addition of chelators with

thiol moieties leads to an increase in metal-release rates from HypB. This suggests that the

nickel transfer from HypB to the precursor enzyme is favoured because the first coordination

140

sphere of the hydrogenase is exclusively composed of thiolates. Secondly, the metal-release

process can be further accelerated when HypB is bound to GDP and the effect of GDP is only

observed in the presence of SlyD. Therefore, these results indicate that GDP-bound

conformation of HypB is necessary to achieve faster metal-transfer, which implies that GTP

hydrolysis likely precedes the nickel insertion step during biosynthesis. Furthermore, SlyD is

necessary to achieve the acceleration of nickel release from GDP-bound HypB suggesting that

protein complex formation provides a link between the transfer of the metal from HypB and the

GTPase activity of the protein.

In addition to modulating the N-terminal Ni(II) site, the affect of SlyD with respect to the

GTPase activity of HypB was also investigated. Interaction with SlyD leads to a ~3 fold increase

in the GTPase activity of HypB, but the effect is more pronounced when HypB contains a metal

at the G-domain metal site. Metal binding to the G-domain can regulate the GTPase activity of

the protein as nickel or zinc binding to this site has an inhibitory effect (11). This inhibitory

effect is circumvented in the presence of SlyD as the protein competitively removes the metal ion

from the G-domain of HypB. Based on the data gathered, we can only speculate about the role

of this G-domain metal site in vivo. It is likely that metal binding to the G-domain is used as a

regulatory component to ensure that GTPase activity is not turned on prematurely, but upon

complex formation, SlyD extracts the metal from the G-domain allowing HypB to enter the

hydrolysis cycle. By combining all the data obtained in this study, a role for SlyD and the SlyD-

HypB interaction in the hydrogenase metallocentre assembly can be derived and is presented in

chapter 4. Although our investigation only involved two of the known partner proteins in the

nickel-insertion step, the insights gained from this study will serve as a framework for building

the molecular and mechanistic details of the nickel insertion step during the hydrogenase

assembly. Furthermore, these protein-protein mediated effects that we discovered, highlight the

importance of the coordinated actions of multiple protein factors in overcoming the

thermodynamic and kinetic barriers to insertion of the nickel ion into the partially assembled

hydrogenase.

5.4 Significance

The utilization of H2(g) for energy is considered to be a crucial feature of very early life on Earth

evident by the presence of hydrogenases in bacteria, archaea, and some eukaryotes (12).

141

Convergent evolution has led to three distinct classes of hydrogenases annotated based on the

combination of metal(s) at the active sites, the [NiFe]-hydrogenase, [FeFe]-hydrogenases and

[Fe]-hydrogenases. These enzymes are able to catalyze the interconversion of molecular H2 and

protons. Consequently they are used to vent excess reducing potential in some anaerobic

microorganisms, while others species use the enzyme to generate a proton motive force that is

necessary for biological functions (13). While hydrogenases have garnered much interest from a

bioinorganic perspective due to the presence of metal-centres decorated with uncommon ligands

such as carbon monoxide and cyanide, interest in hydrogen metabolism and hydrogenases has

grown exponentially over the past decade as these enzymes may have a use in a variety of

biotechnological applications (14).

One of the foremost applications is generation of H2, which is considered to be a sustainable,

cleaner source of energy than fossil fuels. Biological hydrogen production through

microorganism such as algaea and cyanobacteria are at the forefront of this energy production

methods, because H2 production can be coupled to natural sun-light via photosynthesis (15, 16).

Hydrogen evolution can also be achieved through dark fermentation method using bacteria such

as Clostridium sp. that couples the break down of carbohydrate-rich substrates to hydrogen

production. As carbohydrate-rich substrates are naturally found in waste such as food waste or

sewage, this application simultaneously addresses two crucial issues, a supply of energy and

environmental protection (15). Other applications of hydrogenase include the use of enzyme for

bioremediation of toxic heavy metals. Pd(II), Pt(IV) and Ru(III) found in industrial waste can be

converted to the more insoluble metallic forms through reduction using hydrogenases (17).

Hydrogenases are also becoming important from a medical stand point, as these enzymes are

now considered a virulence factor of pathogenic enteric bacteria such as E. coli, Salmonella and

Helicobacter species. These bacteria use the hydrogen produced by the colonic flora within the

animal host as a source of energy, thus proliferating in the anoxic environments within the host

(18). Also, E. coli uses hydrogenases as one of the means to resist the harshly acidic conditions

of the stomach (19). Helicobacter pylori, a lead cause of gastritis and peptic ulcers, requires

[NiFe]-hydrogenases as well as a second nickel-containing enzyme, urease, to successfully

colonize the mucosal surfaces of the human stomach (20). As a human nutritional requirement

for nickel has not been found yet (2), these nickel-dependent enzymes as well as nickel-binding

142

proteins provide an orthogonal pathway that can be targeted by antibiotics without any negative

effects on the host (20, 21).

It is evident that nickel ions as well as nickel-containing enzymes play important roles in all life.

Nature has not only perfected the chemistries performed by these enzymes through the course of

billions of years but has also engineered elaborate systems to ensure the efficient biogenesis of

these enzymes. An in-depth understanding of how this biosynthesis is naturally achieved and

how each protein is optimized to perform their tasks is a pre-requisite for effectively

manipulating hydrogenases or the nickel homeostasis pathway for applications in the future.

This understanding can only come from an investigation of the proteins as well as the interplay

between these proteins involved in these processes.

5.5 Future directions

The research presented in this thesis gives us insight to the molecular details about how the

recently-discovered nickel chaperone protein SlyD binds metal and contributes to nickel

homeostasis in E. coli. The investigations reported here, combined with the pre-existing

knowledge of HypA and HypB, now completes our knowledge about the molecular structure and

metal-binding properties of all three individual accessory protein required for [NiFe]-

hydrogenase biosynthesis. The next step would be to investigate the mechanistic details of metal

transfer and how these shuttling proteins influence or are influenced by each other to attain

successful biogenesis of the hydrogenase enzyme. The HypB-SlyD interaction study

demonstrates the need for having such an integrated system to get an adequate understanding of

exactly how the nickel insertion process is achieved through the concerted action of the

accessory proteins. In addition to SlyD, HypB is known to interact with HypA and a HypA-

HypB-SlyD complex can also be isolated in vivo (9, 22). The consequences of this multi-protein

complex on the metal-binding properties of the proteins and the enzymatic activity of HypB are

essential to understanding why such a protein complex is necessary and how the three proteins

deliver the nickel ion to the metallocentre assembly in vivo. Using this bottom-up approach,

eventually all of the components necessary for assembly of an active [NiFe]-hydrogenase system

in a cell-free system maybe possible, as has been achieved for the nickel-dependent enzyme

urease and the [FeFe]-hydrogenases (23, 24). Although we know that nickel is mobilized via this

protein complex, it is not clearly established where this protein complex specifically acquires

143

nickel. It has been suggested that these proteins gain access to nickel via direct interaction with

the nickel specific importer NikABCDE transporter and thus provide a dedicated source of nickel

for enzyme maturation (25). Which proteins are responsible for this interaction and how the

metal transfer occurs between the transporter and the metallochaperones are questions that

remain to be answered through future investigations.

144

References

1. Waldron, K. J., Rutherford, J. C., Ford, D., and Robinson, N. J. (2009) Metalloproteins

and metal sensing, Nature 460, 823-830.

2. Li, Y., and Zamble, D. B. (2009) Nickel homeostasis and nickel regulation: an overview,

Chem. Rev. 109, 4617-4643.

3. Ragsdale, S. W. (2009) Nickel-based Enzyme Systems, J. Biol. Chem. 284, 18571-18575.

4. Zhang, J. W., Butland, G., Greenblatt, J. F., Emili, A., and Zamble, D. B. (2005) A role

for SlyD in the Escherichia coli hydrogenase biosynthetic pathway, J. Biol. Chem. 280,

4360-4366.

5. Leach, M. R., Zhang, J. W., and Zamble, D. B. (2007) The role of complex formation

between the Escherichia coli hydrogenase accessory factors HypB and SlyD, J. Biol.

Chem. 282, 16177-16186.

6. Wulfing, C., Lombardero, J., and Pluckthun, A. (1994) An Escherichia coli protein

consisting of a domain homologous to FK506-binding proteins (FKBP) and a new metal

binding motif, J. Biol. Chem. 269, 2895-2901.

7. Giedroc, D. P., and Arunkumar, A. I. (2007) Metal sensor proteins: nature's

metalloregulated allosteric switches, Dalton Trans. 3107-3120.

8. Rowe, J. L., Starnes, G. L., and Chivers, P. T. (2005) Complex transcriptional control

links NikABCDE-dependent nickel transport with hydrogenase expression in Escherichia

coli, J. Bacteriol. 187, 6317-6323.

9. Kaluarachchi, H., Chan Chung, K. C., and Zamble, D. B. (2010) Microbial nickel

proteins, Nat. Prod. Rep. 27, 681-694.

10. Kaluarachchi, H., Sutherland, D. E. K., Young, A., Pickering, I. J., Stillman, M. J., and

Zamble, D. B. (2009) The Ni(II)-binding properties of the metallochaperone SlyD, J. Am.

Chem. Soc. 131, 18489-18500.

11. Cai, F., Ngu, T. T., Kaluarachchi, H., and Zamble, D. B. (2011) Relationship between the

GTPase, metal-binding, and dimerization activities of E. coli HypB, J. Biol. Inorg. Chem.

In Press.

12. Vincent, K. A., Parkin, A., and Armstrong, F. A. (2007) Investigating and exploiting the

electrocatalytic properties of hydrogenases, Chem. Rev. 107, 4366-4413.

13. Mulrooney, S. B., and Hausinger, R. P. (2003) Nickel uptake and utilization by

microorganisms, FEMS Microbiol. Rev. 27, 239-261.

14. McGlynn, S. E., Mulder, D. W., Shepard, E. M., Broderick, J. B., and Peters, J. W.

(2009) Hydrogenase cluster biosynthesis: organometallic chemistry nature's way, Dalton

Trans. 4274-4285.

15. Kim, D. H., and Kim, M. S. (2011) Hydrogenases for biological hydrogen production,

Bioresour. Technol. In press.

16. Mertens, R., and Liese, A. (2004) Biotechnological applications of hydrogenases, Curr.

Opin. Biotechnol. 15, 343-348.

17. Vignais, P. M., and Billoud, B. (2007) Occurrence, classification, and biological function

of hydrogenases: an overview, Chem. Rev. 107, 4206-4272.

18. Maier, R. J., Olczak, A., Maier, S., Soni, S., and Gunn, J. (2004) Respiratory hydrogen

use by Salmonella enterica serovar Typhimurium is essential for virulence, Infect.

Immun. 72, 6294-6299.

19. Noguchi, K., Riggins, D. P., Eldahan, K. C., Kitko, R. D., and Slonczewski, J. L.

Hydrogenase-3 contributes to anaerobic acid resistance of Escherichia coli, Plos One 5,

e10132.

145

20. Maier, R. J. (2003) Availability and use of molecular hydrogen as an energy substrate for

Helicobacter species, Microbes and Infection 5, 1159-1163.

21. Maier, R. J. (2005) Use of molecular hydrogen as an energy substrate by human

pathogenic bacteria, Biochem. Soc. Trans. 33, 83-85.

22. Chung, K. C. C., and Zamble, D. B. (2011) The Escherichia coli metal-binding

chaperone SlyD interacts with the large subunit of [NiFe]-hydrogenase 3, FEBS Lett. 585,

291-294.

23. Carter, E. L., Flugga, N., Boer, J. L., Mulrooney, S. B., and Hausinger, R. P. (2009)

Interplay of metal ions and urease, Metallomics 1, 207-221.

24. Kuchenreuther, J. M., George, S. J., Grady-Smith, C. S., Cramer, S. P., and Swartz, J. R.

(2011) Cell-free H-cluster Synthesis and [FeFe] Hydrogenase Activation: All Five CO

and CN- Ligands Derive from Tyrosine, Plos One 6, e20346.

25. Rowe, J. L., Starnes, G. L., and Chivers, P. T. (2005) Complex transcriptional control

links NikABCDE-dependent nickel transport with hydrogenase expression in Escherichia

coli, J. Bacteriol. 187, 6317-6323.