DEFINING THE CATALYTIC AND KINETIC MECHANISM AND NATURAL FUNCTION OF THE HIGHLY CONSERVED ACYL-AMP HYDROLASE, HINT1
A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA BY
SANAA BARDAWEEL
IN PARTIAL FULFILMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTORE OF PHILOSOPHY
DR. CARSTON R. WAGNER, ADVISER
DECEMBER 2010
© Sanaa Bardaweel, 2010
ACKNOWLEDGEMENTS
I would like to gratefully and sincerely thank my thesis adviser, Dr. Carston R.
Wagner, for his guidance, understanding and patience. His mentorship was paramount in providing a well rounded experience consistent with my long-term career goals. He encouraged me to not only grow as an experimentalist but also as an instructor and an independent researcher. In every sense, none of this work would have been possible without him.
I gratefully acknowledge Dr. Michael Sadowsky for his advice, supervision, and
crucial contributions, which made him a backbone of this research and so to this thesis. I
am also deeply thankful to Dr. Patrick Hanna, for the valuable advice and discussions that
inspired me through out the work. Many thanks must also go to my thesis committee
members, Dr. Chengguo Xing and Dr. Howard Towle, for their valuable input and
suggestions that improved my dissertation.
I would like to make a special reference to Dr. Yusuf Abul-Hajj for his unflinching encouragement and support in various ways. His support and guidance during the very first steps in my PhD have had a remarkable influence on my entire progress in the program.
I am grateful to former members of Dr. Wagner’s laboratory, Dr. Brahma Ghosh,
Dr. Brandie Brummer, Dr. Yan Jia, Dr. Qing Li, Dr. Brian White and Dr. Xin Zhou for
their help and great friendship. I also thank current members of the lab, Dr. Adrian
Fegan, Dr. Matthew Cuellar, Dr. Sidath Kumarapperuma, Amit Ganger and Ryan Holton
for the numerous and productive discussions that helped me improve my knowledge
during my graduate study.
i
I owe my loving thanks to my mother who raised me with a love of science and supported me in all my life. Her endless faith inspired me to become the person who I am today. A special thank-you goes to my wonderful sons, Hashem, Yamen and Othman who made my graduate life a lot less stressful. I hope that I may serve as an example for them as my mother has for me.
My final, and most heartfelt, acknowledgment must go to my husband Rizq. His support, encouragement and unwavering love have turned my journey through graduate school into a pleasure. For never once complaining through all difficult times and challenges, for taking all responsibility for the care of our family, for sharing all the good and bad times with endless support, for all that, this dissertation is dedicated to him.
ii
ABSTRACT
Histidine triad nucleotide binding proteins (Hints) are members of the histidine triad (HIT) protein superfamily of nucleotidyl transferases and hydrolyases. It has been recently demonstrated that Hints are efficient phosphoramidases and therefore activators of potent antiviral and anticancer pronucleotides. In spite of their high evolutionary conservation among all kingdoms of life, and the several regulatory functions in which
Hints have been implicated, a clear connection between their observed function and their catalytic efficiency has not been elucidated. To gain a comprehensive understanding of the essential role of these ubiquitous enzymes, our laboratory has devoted a considerable effort toward the delineation of the principles governing Hints catalysis and cellular function. Such understanding will provide an unprecedented ability to assess the role of these highly conserved, but functionally unknown enzymes.
Since Hints are found in both prokaryotes and eukaryotes, we have attempted to understand their function, mechanism, and structural determinants in prokaryotes, under the assumption that their role may be at least partially conserved among members of the tree of life. Recently, we have demonstrated by E. coli gene disruption studies that the bacterial Hint enzyme is necessary for growth under high salt conditions, and when alanine is a carbon and nitrogen source. Through a combination of phenotypic screening and complementation experiments with wild-type and ecHinT knock-out E. coli strains, we have shown that catalytically-active ecHinT is required for growth on D-alanine. In addition, using Hint-inhibitors and active-site mutants, we have demonstrated that expression of catalytically-active ecHinT is essential for the activity of the enzyme D- amino acid dehydrogenase (DadA) (equivalent to D-amino acid oxidase in eukaryotes), a
iii necessary component of the D-amino acids metabolic pathway. These results are considered as the first report in literature that shows a successful connection between a discovered Hint-related phenotype and the catalytic activity of Hint.
Previously, we have demonstrated that lysyl-AMP generated by LysRS is a substrate for both human and E. coli Hints. In addition, we have shown that the ability of Hint to hydrolyze lysyl-AMP depends on its enzymatic activity. Here, we demonstrate that the molecular determinants governing this regulation appear to reside in the C- terminus region of Hint. Interestingly, the ecHinT-DadA interaction appears also to be governed by both ecHinT-activity and the C-terminus loop.
We have also expanded our scope to look at possible toxicity of D-alanine in E. coli strains lacking dadA or hinT . Our results demonstrate that E. coli mutants lacking dadA or hinT are highly susceptible to D-alanine toxicity and that the catalytic activity of
Hint is an essential requirement to protect E. coli from the observed toxicity of D-alanine.
Based on careful analysis of the combined results from the ecHinT-LysRS and ecHinT-
DadA potential interactions, and comprehensive understanding of the D-alanine metabolic pathway in bacteria, we proposed a possible regulatory mechanism of Hint,
LysRS and DadA on global protein translational processes to prevent D-amino acids toxicity in E. coli .
iv
Table of Contents
Acknowledgements ...... i
Abstract ...... iii
Table of Contents ...... v
List of Tables ...... xi
List of Figures ...... xiii
List of Schemes ...... xvi
List of Abbreviations ...... xvii
Chapter One: Introduction: Histidine Triad Nucleotide Binding Proteins ...... 1
I. Discovery and Classification of HIT Superfamily...... 2
II. Phosphoramidase Activity of Hints ...... 6
III. Structural Properties of Hints ...... 12
IV. Structural Studies of Hints ...... 13
V. Role of Hint Homodimerization ...... 14
VI. Possible Biological Functions of Hints ...... 18
VII. Prokaryotic Hint ...... 20
VIII. The Interaction of Hints with Aminoacyl tRNA Synthetases ...... 23
IX. Current Research ...... 27
Chapter Two: Probing the Impact of the ecHinT C-Terminal Domain on Structure
and Catalysis...... 28
v
I. Introduction ...... 29
II. Experimental Procedures ...... 32
A. Crystallization and X-ray Data Collection ...... 32
B. Structure Determination and Refinement ...... 32
C. Site-directed Mutagenesis ...... 35
D. Expression and Purification of Recombinant Proteins ...... 35
E. Circular Dichroism Spectroscopy ...... 37
F. Size-exclusion Chromatography ...... 37
G. Steady-state Kinetics ...... 38
H. Pre-steady-state Kinetics ...... 38
I. pH-dependence of Steady-state Kinetics ...... 39
J. Lysyl-AMP-dependent Adenylation of ecHinT by LysU ...... 39
K. PDB Accession Numbers ...... 39
III. Results and Discussion ...... 40
A. Structural Properties of ecHinT ...... 40
B. Ligand Binding ...... 48
C. Design of C-terminus Mutants and Protein Purification...... 54
D. Secondary Structure Analysis by Circular Dichroism (CD) Spectroscopy ...... 54
E. Thermal Stability ...... 55
F. Size-exclusion Chromatography ...... 60
G. Effect on Catalysis ...... 60
H. Lysyl-AMP-dependent Adenylation of ecHinT by LysU ...... 71
vi
IV. Summary and Concluding Remarks ...... 74
Chapter Three: HinT, a Histidine Triad Nucleotide Binding Protein, is Essential for
Alanine Metabolism in Escherichia coli ...... 76
I. Introduction ...... 77
II. Experimental Procedures ...... 80
A. Bacterial Strains, Media, and Growth Conditions ...... 80
B. PCR Verification of Mutants ...... 80
C. Verification of hinT Deletion Mutant by Loss of Activity ...... 80
D. Phenotype Analysis Using Biolog™ GN2-MicroPlates ...... 83
E. Carbon Source Utilization Assays ...... 83
F. Induction of D-amino Acid Dehydrogenase Activity and Enzyme Assays ...... 84
G. Preparation of Membrane Fractions...... 84
H. Reverse Transcription-PCR ...... 85
I. Site-directed Mutagenesis ...... 85
J. Phenotype Complementation Studies ...... 86
K. General Synthetic Procedures and Materials ...... 86
L. Synthesis of 2’,3’-Isopropylidine-5’-O-(4-Chlorophenoxy)Carbonyl Guanosine
(Guanosine -5’-Carbonate Acetonide) ...... 87
M. Synthesis of 2’,3’-Isopropylidine-5’-O-[(3-Indolyl)-1-Ethyl]Carbamoyl
Guanosine (Guanosine -5’-Carbamate Acetonide) ...... 87
vii
N. Synthesis of 5’-O-[(3-Indolyl)-1-Ethyl]Carbamoyl Guanosine (Guanosine-5’-
Tryptamine Carbamate, TpGc) ...... 88
III. Results and Discussion ...... 89
A. Operon Structure of Genes Associated with hinT ...... 89
B. Phenotype of ecHinT ...... 89
C. D-Amino Acid Dehydrogenase Transcription and Activity Testing ...... 93
D. Phenotype Rescue by ecHinT ...... 98
E. Design, Synthesis and Characterization of ecHinT Inhibitor ...... 100
Chapter Four: Synthesis and Evaluation of Potential Inhibitors of Human and
Escherichia coli Histidine Triad Nucleotide Binding Proteins ...... 111
I. Introduction ...... 112
II. Experimental Procedures ...... 113
A. General Procedure for synthesis of 2’,3’-Isopropylidine-5’-(indole-3- propionyl)
nucleosides: (Adenosine Acetonide-5’-Indole-3-Propionate (1) ...... 113
B. General Procedure for Synthesis of 5’-(indole-3- propionyl) nucleosides
(Adenosine 5’-Indole-3-Propionate(2)...... 114
C. General Procedure for Synthesis of 5’-O-(4-chlorophenoxy) carbonyl 3’-azido-3’-
deoxy thymidine (AZT-5’-carbonate (3) ...... 115
D. General Procedure for Synthesis of 5’-O-[(3-indolyl)-1-ethyl]carbamoyl 3’-azido-
3’-deoxy thymidine (AZT-5’-carbamate (4) ...... 115
viii
E. General Procedure for Synthesis of 5’-O-[(3-indolyl)-1-ethyl]carbamoyl 3’-
amino-3’-deoxy thymidine (AZT-3’-NH2-5’-carbamate(5) ...... 116
F. General Procedure for Synthesis of 2’,3’-Isopropylidine-5’-O-(4-
chlorophenoxy)carbonyl nucleosides (Adenosine-5’-carbonate acetonide (6) ....116
G. General Procedure for Synthesis of 2’,3’-Isopropylidine-5’-O-[(3-indolyl)-1-
ethyl]carbamoyl nucleosides (Adenosine 5’-tryptamine-carbamate acetonide (7)117
H. General Procedure for Synthesis of 5’-O-[(3-indolyl)-1-ethyl]carbamoyl
nucleosides (Adenosine 5’-tryptamine-carbamate (8) ...... 118
I. Phosphoramidase Assay...... 118
J. Phenotype Testing ...... 119
III. Results and Discussion ...... 119
A. Inhibitors’ Synthesis and Evaluation ...... 119
B. Phenotype Testing ...... 130
IV. Summary and Concluding Remarks ...... 130
Chapter Five: Possible Physiological Role of The Histidine Triad Nucleotide
Binding Protein in Escherichia coli : Regulation of D-alanine Detoxification ...... 133
I. Introduction ...... 134
II. Experimental Procedures ...... 138
A. Bacterial Strains, Media, and Growth Conditions ...... 138
B. Growth Inhibitory Effects of D-amino Acids on Escherichia coli ...... 138
C. Growth Kinetics ...... 139
ix
D. D-alanine Toxicity in Presence of ecHinT Inhibitors ...... 139
E. Mammalian Cell Lines and Cell Culturing ...... 140
F. Cell Viability Assay ...... 140
G. D-alanine Cytotoxicity in MCF-7, MDA-MB-231 and MIA PaCa-2 Cell Lines 141
H. D-alanine Cytotoxicity in HPB-MLT and Raji Cell Lines ...... 141
III. Results ...... 142
A. D-amino Acids Toxicity in E. coli ...... 142
B. Effect of ecHinT Inhibition on E. coli Sensitivity to D-alanine ...... 148
C. D-alanine Toxicity in Mammalian Cell Lines ...... 152
IV. Discussion ...... 152
BIBLIOGRAPHY ...... 162
x
List of Tables
Chapter One
Table 1. Hydrolysis rates of substrates by human Hint and E. coli Hint ...... 9
Table 2. Sequence alignment for ecHinT, human and rabbit Hint ...... 22
Chapter Two
Table 1. Data collection and refinement statistics for echinT and its H101A mutant GMP
complexes ...... 33
Table 2. Mutagenic forward primers for the C-terminus mutants ...... 36
Table 3. Steady-state and pre-steady-state kinetic parameters for ∆114-119 and ∆117-
119 C-terminus deletion mutants ...... 65
Table 4. Steady-state kinetic parameters for L114A, H116A, K117A, and L119A C-
terminus alanine mutants ...... 66
Chapter Three
Table1. Bacterial strains used in this study ...... 81
Table 2. Primers used for sequence verification PCR reaction ...... 82
Chapter Four
Table 1. Inhibition constants determined in HEPES buffer (pH7.2) at 25ºC ...... 123
xi
Chapter Five
Table 1. Summary of IC 50 values ...... 147
Table 2. Summary of the ecHinT-LysRS and ecHinT-DadA studies ...... 160
xii
List of Figures
Chapter One
Figure 1. Reactions catalyzed by HIT superfamily ...... 3
Figure 2. P-N bond hydrolysis of adenosine phosphoramidates by Hint ...... 8
Figure 3. X-Ray Crystallographic structure of hHint1 active-site bound AMP (PDB:
1KPF) ...... 10
Figure 4. a) X-ray crystallographic structure of hHint1 with bound AMPCP (PDB:
1AV5) ...... 15
Figure 4. b) Structure of the hHint1 C-terminus loop ...... 16
Figure 5. Rabbit Hint1 guanosine monophosphate binding pocket ...... 17
Figure 6. Comparison of the C-terminal domain for ecHinT (PDB code 3N1S)
(green/cyan) and hHint (PDB code 1av5) ...... 24
Figure 7. Catalytic processes involving lysyl-tRNA synthetase and Hints ...... 25
Figure 8. Proposed catalytic mechanism for lysyl-adenylate hydrolysis by Hints ...... 26
Chapter Two
Figure 1. Difference electron density map for ecHinT ...... 42
Figure 2. ecHinT structure superimposed on the structure of hHint1 ...... 44
Figure 3. ecHinT C-terminus structure ...... 46
Figure 4. Environment of GMP binding ...... 49
Figure 5. a) Environment around GMP in domain A showing the disposition of the
HxHxHxx motif of wild-type ecHinT compared with that of the H101A mutant...... 52
xiii
Figure 5. b) View of binding pocket in ecHinT showing GMP phosphate interactions ....53
Figure 6. Secondary structure analysis ...... 56
Figure 7. Thermal stability studies...... 58
Figure 8. Size-exclusion chromatograms from the Superdex G 200 column ...... 61
Figure 9. pH-dependence rate profile for wild-type ecHinT, ∆114-119 and ∆117-119 C-
terminus deletion mutants ...... 67
Figure 10. Stopped flow trace of ecHinT exhibited a biphasic profile, a burst phase
followed by a linear phase ...... 70
Figure 11. Adenylation of wild-type ecHinT and ∆114-119 mutant by ec LysU ...... 72
Figure 12. Time-dependence of ecHinT and ∆114-119 mutant adenylation by ec LysU .73
Chapter Three
Figure 1. Sequence verification PCR ...... 90
Figure 2. Phosphoramidase assay to detect ecHinT activity in E. coli cell-free lysates ....91
Figure 3. Summary of metabolic fingerprints ...... 92
Figure 4. Bacterial growth curves of wild-type E. coli BW25113 and ∆hinT mutant in M9
medium in the presence of either 20 mM glucose or D,L-alanine...... 94
Figure 5. Alanine transport and metabolism in E. coli ...... 95
Figure 6. RT-PCR was performed with equivalent amounts of mRNA obtained from
wild-type E. coli BW25113, ∆hinT, and ∆ycfL mutants using dadA primers ...... 97
Figure 7. ecHinT structural and activity requirement for phenotype rescue ...... 99
Figure 8. Inhibition of phenotype rescue by ecHinT inhibitor ...... 102
xiv
Figure 9. Superposition of the eight independent monomers observed in the structure of the ecHinT-GMP complex ...... 107
Figure 10. Phenotype rescue by the alanine scan mutants ...... 108
Chapter Four
Figure 1. Structures of a) Tryptamine 5’-adenosine phosphoramidate and its b) ester and
c) carbamate analogues ...... 122
Figure 2. All compounds exhibited a non-competitive inhibition profile ...... 126
Figure 3. ecHinT inhibition resulted in impaired phenotype ...... 132
Chapter Five
Figure 1. D-amino acids IC 50 determination in E. coli strains: wild-type BW25113, ∆hinT and ∆dadA ...... 143
Figure 2. Sensitivity of E. coli strains, wild-type BW25113, ∆hinT and ∆dadA , to L- alanine treatment ...... 146
Figure 3. Growth kinetic curves of three E. coli strains: wild-type BW25113, ∆hinT and
∆dadA ...... 149
Figure 4. D-alanine and D-lysine Synergism ...... 150
Figure 5. Effect of ecHinT inhibition on E. coli sensitivity to D-alanine ...... 151
Figure 6. D-alanine toxicity in mammalian cell lines ...... 153
Figure 7. The structure of peptidoglycan ...... 156
Figure 8. D,L-alanine metabolism and possible ecHinT-LysRS regulation ...... 161
xv
List of Schemes
Chapter Two
Scheme 1. Proposed kinetic mechanism for both phosphoramidate and acyl-AMP hydrolysis by Hints ...... 69
Chapter Three
Scheme 1. General synthetic scheme of TpGc ...... 101
Chapter Four
Scheme 1. General synthetic scheme of compounds 1 and 2 ...... 121
Scheme 2. General synthetic scheme of compounds 3 and 4 ...... 124
Scheme 3. General synthetic scheme of compound 5...... 125
xvi
List of Abbreviations
AARS Aminoacyl tRNA synthetase
ADP Adenosine diphosphate
AIPA Adenosine indole propionic acid
Ala Alanine
AMP-lysine AMP-N-ε-(N-α -acetyl lysine methyl ester) 5'-phosphoramidate
AMP Adenosine 5'-monophosphate
AMPCP Adenosine 5'-( α,β-methylene) diphosphate
AMP-NH2 Adenosine 5'-monophosphpramidate
AP 3A Diadenosine p1,p3-triphosphate
AP 4A Diadenosine p1,p4-tetraphosphate
Arg Arginine
Asp Aspartic acid
ATP Adenosine 5'-triphosphate
AZT 3'-Azido-3'-deoxythymidine
CD Circular dichroism
Cdk7 Cyclin dependent kinase 7 cDNA Complementary DNA
CNS Central nervous system
DadA D-amino acid dehydrogenase
DCPIP 2,6-dichlorophenol-indolphenol
DMBA 7,12-dimethylbenzanthracene
DMEM Dulbecco's modified eagle medium
xvii
DNA 2'-Deoxyribonucleic acid
DTT Dithiothreitol
E. coli Escherichia coli ec LysU E. coli lysyl tRNA synthetase ec DHFR E. coli dihydrofolate reductase ec Hin T E. coli histidine triad nucleotide-binding protein
EDC Ethyl dimethylaminopropyl carbodiimide
EDTA Ethylenediaminetetracetic acid
ESI-HRMS electrospray ionization high-resolution mass spectrometry
FAD Flavin adenine dinucleotide
Fhit Fragile histidine triad
GalT Galactose-1-phosphate uridylyltransferase
Glu Glutamic acid
Gly Glycine
GMP Guanosine 5'-monophosphate
GMP-Lysine GMP-N-ε-(N-α-acetyl lysine methyl ester ) 5'-phosphoramidate
GTP Guanosine 5'-triphosphate
h Hour
1H NMR Proton nuclear magnetic resonance
HEPES 4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid
HI-FBS Heat-inactivated fetal bovine serum
HINT Histidine triad nucleotide binding protein
HIT Histidine Triad
xviii hHint1 Human histidine triad nucleotide-binding protein1
His Histidine
HPLC High performance liquid chromatography
IPTG Isopropyl -β-D-thiogalactopyranoside
IC 50 Inhibitory concentration
Kan Kanamycine
kDa Kilodalton
Ki Inhibition constant
LB Luria bertani medium
LC Liquid chromatography
ESI-MS/MS Electrospray ionization tandem mass spectrometry
Lys Lysine
LysRS Lysyl-tRNA synthetase
M Molar
MALDI-TOF Matrix assisted laser desorption ionization-time of flight
Min Minute
MITF Microphthalmia-associated transcription factor
mg milligram
ml milliliter
mM millimolar
MOPS 3-(N-morpholino) propanesulfonic acid
MRE Mean residue ellipticies
mRNA Messenger RNA
xix
MTX Methotrexate
Mw Molecular weight
NADPH β-Nicotinamide adenine dinucleotide 2'-phosphate-reduced form
nm Nanometer
NMP Nucleotide monophosphate
NMR Nuclear magnetic resonance
OD Optical density
ORF Open reading frame
PAGE Polyacrylamide gel electrophoresis
PCR Polymerase chain reaction
PBS Phosphate-buffered saline
PEG Polyethylene glycol
31 pNMR Phosphorus nuclear magnetic resonance
Pi Inorganic phosphate
PM Phenotype microarray
PMSF Phenylmethanesulfonyl fluoride
P-N bond Phosphorus -Nitrogen bond
Ppase Yeast inorganic pyrophosphatase
Ppi Inorganic pyrophosphate
PVDF Polyvinylidene fluoride qRT-PCR Quantitative real time polymerase chain reaction
RNA Ribonucleic acid
xx
RNase H Ribonuclease H rpm Revolutions per minute
RPMI Roswell park memorial institute medium
SD Standard deviation
SDS Sodium dodecyl sulfate
SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis
SEC Size exclusion chromatography
SLS Static light scattering
TB Terrific broth
TFIIH Transcription factor II H
TLC Thin layer chromatography
TMP Trimethylphosphate
TpAd Tryptamine adenosine phosphoramidate
TpGc Tryptamine guanosine carbamate
Tris Tris(hydroxymethyl)aminomethane tRNA Transfer RNA
Trp Tryptophan
UDP Uridyl diphosphate
USF2 Upstream simulatory factor 2
UV Ultraviolet
Val Valine vol Volume
WT Wild type
xxi
X-gal 5-Bromo-4-chloro-3-indolyl-β-D-galactopyranoside
µM micromolar
µl microliter
xxii
Chapter One – Introduction
Histidine Triad Nucleotide Binding Proteins
1
I. Discovery and Classification of HIT Superfamily
HIT proteins were initially recognized and classified based on their virtual
sequence similarity regardless of their functions. Generally, the HIT superfamily can be
divided into five branches 1; GalT, Aprataxin 2, DcpS/DCS-1, Fhit and Hint. 1; 3; 4
Although members of each branch carry out a distinct enzymatic activity (Fig. 1), the absolutely conserved motif, His-X-His-X-His-XX (X is a hydrophobic residue), can be found in all members of the superfamily. 2; 5
The histidine triad nucleotide binding proteins (Hint) are the most ancient branch and are considered to be the ancestor of the HIT superfamily. Search results from the
Basic Local Alignment Search Tool (BLAST) program analyses indicate that Hints are found in all kingdoms of life. 6 While prokaryotes typically carry only one Hint gene,
eukaryotes generally contain two to three different Hint genes. 6 For example, three human Hint genes have been identified (hHint1, hHint2 and hHint3), while Drosophila have two genes resulting from alternative splicing (Chou, T. F. and Wagner, C. R. unpublished communication ). Although mammalian Hints share high sequence identity
(93%), eukaryotic Hints, generally, show larger variations within their core amino acid sequences. Remarkably, several predicted proteins from bacteria and archaea have been identified that have greater identity to mammalian Hints than do eukaryotic counterparts. 7
Hints are homodimeric proteins that have been shown to function as purine
phosphoramidases. 8; 9; 10; 11 Brenner and coworkers were the first to describe the
phosphoramidase activity of Hint1, by demonstrating that rabbit Hint1 and yeast Hnt
2
Figure 1. Reactions catalyzed by the HIT superfamily.
O O GalT N N Gal-1-P O O N O O O N O Glc-O P O P O O Gal-O P O P O O O- O - O - O - Glc-1-P OH OH OH OH
NH 2 NH 2 N N N N
O O N N Aprataxin O N N DNA-O P O P O O HO P O O O - O - O -
OH OH DNA-PO 3 OH OH
Me O Me O N N NH NH N N O O N NH2 O N NH 2 DcpS HO P O O3P O P O P O O O O- O - O - OH OH OH OH PPi NH2 NH2 N N N N
O O N N Fhit O N N AMP-O P O P O O HO P O O O - O - O - OH OH ADP OH OH
NH 2 NH2 N N N N
O N N O N N H Hints Alkyl Group N P O O HO P O O O- O- OH OH OH OH Alkyl Amine
NH 2 NH2 N N N N O O N N Hints O N N Alkyl Group O P O O HO P O O - O O - OH OH Alkyl acid OH OH
3 convert adenosine 5’-monophosphoramidate (AMP-NH 2) to AMP and ammonia and adenosine lysine phosphoramidate to AMP and lysine (Fig. 1). 9
The evolutionarily conserved fragile histidine triad (Fhit) protein is composed of
147 amino acids and represents the second branch of the histidine triad (HIT) protein superfamily. 2 The Fhit branch is found only in animals and fungi. 2 Like Hint1, Fhit is a homodimer with two identical purine mononucleotide binding sites. 12 Fhit, like all
members of the HIT superfamily, possesses nucleotide-binding and nucleotide-
hydrolyzing activity. Fhit catalyzes the hydrolysis of Ap nA (n = 3 or 4), producing AMP as one of the two mononucleotide products (Fig. 1). 12
The human Fhit gene is disrupted in many tumors resulting in impaired or missing expression of the encoded Fhit protein. 13 Loss of Fhit expression was reported to represent an early event during neoplasia and was correlated with tumor progression and detrimental prognosis. 14; 15 Consistent with its tumor-suppressor function,
expression of Fhit in multiple Fhit-negative tumor cells resulted in inhibition of tumor
growth in Fhit knock-out mice. 16 The molecular basis of the tumor-suppression activity of Fhit appears to be mediated by a tyrosine phosphorylation-dependent modulation of the Akt-survival pathway. 17; 18 Site-directed mutagenesis of the nucleophilic histidine 96 of the HIT motif to asparagine generates a catalytically inactive mutant Fhit protein.
However, the enzymatically dead protein still shows tumor-suppressive activity, suggesting that the hydrolysis of diadenosine polyphosphates is not involved in the tumor-suppressive function of Fhit. 19; 20
Galactose-1-phosphate uridylyltransferase (GalT) is the second enzyme in the
Leloir pathway of galactose utilization. 21; 22 The GalT branch consists of nucleoside
4 monophosphate transferases, including galactose-1-phosphate uridylyltransferase, diadenosine tetraphosphate phosphorylase, and adenylyl sulfate-phosphate adenylytransferase. 2
Though GalT and Hint are both dimers, the GalT monomer is more than twice the size of the Hint dimer. Except for the slightly altered HIT motif, members of the GalT branch share little overall sequence similarity with Hint or Fhit branch members.
Nevertheless, the three-dimensional structure of the monomer GalT is similar to that of the Hint or Fhit homodimer. 2; 7 GalT also has the same folding structure in the HIT region of the protein and mode of nucleotide binding as the latter two proteins. 2; 7
The fourth branch is Aprataxin, a zinc finger containing member of the HIT
superfamily that is mutated in ataxia-oculomotor apraxia. 1; 2; 23 Human Aprataxin is
expressed in two splice forms 23 ; the minor splice form that encodes for a Hint domain and an apparent C-terminal zinc finger, and the major splice form that encodes for the N- terminus domain. 23; 24 The Hint domain of Aprataxin, which shows 31% amino acid identity with rabbit and human Hint, is believed to contain a dimerization interface as well as a phosphoramidase catalytic domain. 2 In addition to the phosphoramidase activity, members of this branch were shown to have Ap 4A hydrolase activity and
DNA/RNA binding properties. 1 Recently, the physiological substrates for Aprataxin
have been proposed to be abortive adenylated DNA ligation intermediates (Fig. 1). 25
The last branch of the HIT superfamily is the scavenger mRNA decapping enzyme, DcpS/DCS-1, a 7-methyl-GpppG hydrolase (Fig. 1). 3; 4 Although DcpS shares no obvious sequence homology with other HIT proteins, except for the conserved HIT
5 motif, its 80 kDa size determined by gel filtration suggests that it too has the potential to form a homodimer. 26
Human DcpS was cloned and purified as an enzyme that hydrolyzes compounds
such as 7meGpppG and small capped oligoribonucleotides and has been proposed to
function in the hydrolysis of short oligomers that remain after 3’ to 5’-exonucleolytic
degradation of mRNA (Fig. 1). 3 However, the specificity for methylated guanosine cap demonstrates that DcpS is distinct from the previously reported Fhit, which can efficiently hydrolyze both methylated and unmethylated structures. 27; 28
The central histidine of the HIT motif in DcpS (His-277) is involved in its pyrophosphatase activity as it catalyzes the formation of a covalent nucleotidyl phosphohistidyl intermediate (Fig. 1). 3 Consistent with the significance of the central
histidine in the pyrophosphatase activity, substitution mutagenesis of the central histidine
in DcpS (H277N) demonstrates that DcpS also contains a functional HIT motif. 3
II. Phosphoramidase Activity of Hints
Rabbit Hint was initially purified, as an abundant cytosolic protein, based on its
nucleotide binding properties using adenosine agarose affinity columns. 29 Hint
phosphoramidase activity was first reported by Brenner and coworkers when they
screened a series of compounds as Hint substrates and found that Hint hydrolyzes the
natural product adenosine-5’-monophosphoramidate (AMPNH 2) in an active-site dependent manner at second order rates exceeding 1×10 6 M-1 s-1. 9 Wagner and coworkers later demonstrated by 31 P NMR that human Hint1 and a previously uncharacterized E. coli ORF, ycfF (renamed ecHinT), were purine nucleoside phosphoramidases. 8
6
A sensitive and continuous fluorescence assay to measure the phosphoramidase activity of Hint was developed by Wagner and coworkers. 11 Using this assay, the general outline of human and E. coli Hint phosphoramidase substrate specificities was determined. 11 A series of substrates linking the naturally fluorogenic indole derivatives to nucleoside 5’-monophosphates was synthesized and their steady-state kinetic parameters of hydrolysis by Hint were evaluated (Fig. 2). The substrate specificity studies revealed a preference for purines over pyrimidines and unhindered over sterically arylated amines. Consistent with the observed hydrogen bonding between the 2’-OH group of adenosine monophosphate and the active-site residue (Fig. 3), maintenance of an electrophilic or hydrogen bonding group at the ribose 2’-position appears to be an essential requirement for a Hint substrate (Fig. 3). 11 In addition, Acyl nucleoside
monophosphate was found to be an excellent substrate with kcat /K m values ranging from
10 6 to 10 7 M-1s-1 (Table 1). 11
Despite strong evidence that Hints are phosphoramidases, the identity of the natural phosphoramidate substrates has been a mystery. With the exception of enzyme active-site lysine-adenylates that are intermediates for enzymes, such as DNA and RNA ligases 30 , and the observation that adenylyl sulphate transferase, in the presence of high
31 concentrations of ammonia, could generate AMP-NH 2 , common naturally occurring macromolecular or small molecule purine based or pyrimidine based phosphoramidates have not been reported. Recently, both bacterial and human Hint1 have been shown to hydrolyze acyl-AMP generated by lysyl tRNA synthetase, which suggests that acyl-AMP generated by aminoacyl tRNA synthetases may be a natural substrate of Hints. 32
7
Figure 2. P-N bond hydrolysis of adenosine phosphoramidates by Hint.
NH2 H NH2 N N N N N H O N N N H O N N N P O O HO P O O OH HINT NH OH + 2 OH OH OH OH
8
Table 1. Hydrolysis rates of the most efficiently hydrolyzed Hint1.
NH2 NH O 2 N N N N N NH O O N N O N N O N N H NH2 HN O P O H N P O O P O O O- N O HN O- HN O- OH OH OH OH OH OH GMP-Tryptamine AIPA AMP-Tryptamine
-1 -1 -1 3 kc at (s ) Km(µ M) kcat /K m (M s )x 10
hHint1 ecHinT hHint1 ecHinT hHint1 ecHinT AMP-Tryptamine 2.3±0.1 4.5±0.1 0.13±0.02 5.2±0.2 17000±2000 870±50 GMP-Tryptamine 2.3±0.1 4.0±0.4 0.21±0.02 6.0±1 11000±1000 700±200 AIPA 1.98±0.02 4.0±0.1 0.04±0.002 4.0±0.4 53000±500 1000±200
9
Figure 3. X-ray crystallographic structure of hHint1 active-site bound AMP (PDB:
1KPF).
10
In vitro studies indicate that human Hint1 can bind various nucleotides, including
33; 34 AMP, ADP, and the diadenosine polyphosphates Ap 3A and Ap 4A. Rabbit Hint1 also binds several purine nucleosides and nucleoside 5’-phosphates. 29 In vitro, the human and rabbit Hint1 proteins do not hydrolyse dinucleoside polyphosphates (Ap nA) or ATP 5; 7 , but they do hydrolyse ADP. 33; 35
Although the HIT superfamily is comprised of at least five distinct subfamilies,
extensive investigations of their catalytic and kinetic mechanisms of action have only
been carried out on GalT and Fhit. In both cases, catalysis proceeds with the formation
of a histidine-NMP intermediate and inversion of the phosphorous configuration,
followed by transfer to either water (Fhit) or galactose (GalT). For GalT, uridinylation of
the nucleophilic histidine by UDP-glucose and subsequent deuridylylation by galactose-
1- phosphate were found to be rapid with rates of 281 s -1 (40ºC) and 166 s -1 (40ºC),
36 respectively. Since neither one of these steps was found to coincide with the k cat value
(62 s -1 at 40ºC), it was proposed that either the product release or a conformation change may be the rate limiting. 36 Most importantly, structural analysis of the GalT active-site
reveals that two of the conserved histidines and a close cysteine formed a coordination
complex with an atom of Fe. 37
While conversion of Ap 3A to ADP and AMP by Fhit proceeds through a double
displacement mechanism, less is known about the specific catalytic steps. 38 When the putative nucleophilic histidine was replaced by glycine, free histidine was found to rescue the enzymatic activity, in addition, AMP-imidazole and AMP N-methylimidazole were found to be substrates. 39 Results of steady-state kinetic and mutagenesis studies have implied that one member of the catalytic triad, His-98, is probably responsible for the
11 donation of a proton to the leaving group, as well as the substrate binding. 40 This is in marked contrast to GalT, in which the corresponding histidines are coordinated to Fe.
The mechanism of Hint catalysis has been proposed to proceed through a two-step mechanism that is analogous to Fhit catalysis. In the first step, the active-site nucleophilic His-112 is thought to form a covalent Hint-AMP intermediate followed by release of nucleoside monophosphate after water hydrolysis of the intermediate. The proposed mechanism was supported by site-directed mutagenesis studies, in which activity was abolished by mutations of the nucleophilic histidine. Zhou et al have provided the first elucidated kinetic study of hHint1 by pre-steady-state and steady-state kinetic analyses, pH-vs-rate analysis and viscometric studies (Zhou, X. and Wagner, C.
R. unpublished data), thus setting a foundation for a comprehensive understanding of the rules governing Hint1 catalysis in particular and the HIT superfamily in general.
III. Structural Properties of Hints
Crystallographic data have been reported for six human Hint1 and five rabbit Hint proteins as either the wild-type apo or AMP- or GMP-bound complexes. 5; 6; 10; 33 Two persistent errors have been reported in the Hint protein scientific literature. When first isolated, the sequence of bovine Hint was annotated as a PKC inhibitor-1. 41; 42 However,
subsequent studies have demonstrated that, while Hint binds to PKC, it is a modulator of
PKC function and not a PKC inhibitor. 5; 29; 43 It was also reported that “bovine PKCI-1”
dried onto nitrocellulose filters binds Zn +2 . 42 This observation was supported when the
active-site of Hint1 was examined and shown to contain four histidines, which were
proposed to bind to Zn +2 . 44 In 1996, a crystal structure of the “zinc form” of Hint was
published that had no zinc electron density and no change in structure from the non-zinc
12 form of the protein. 5 Brenner and coworkers reported that metal binding at the
nucleotide binding site in Hint would be competitive with binding to the ribose ring of
the nucleoside or nucleotide. 7
IV. Structural Studies of Hints
X-ray crystallography studies have demonstrated that hHint1 exists as a homodimeric protein with α + β overall folding topology (Fig. 4a). Each monomer contains a five-stranded antiparallel sheet and two helices. The two monomers are brought together in the dimer to form a 10-stranded antiparallel sheet, such that the central helix from one monomer packs against the central helix from the other monomer.5
Residues in the nucleotide binding site are not involved in the dimerization interface and are positioned 25Å apart on the opposite ends of each monomer (Fig. 4a). The carboxy terminal amino acids of both monomers wrap around one another allowing the carboxylate group of Gly-126 to form a salt bridge with Arg-119 from the other monomer. In addition, the C-terminus of each monomer resides in close proximity to the catalytic residues from the opposite monomer and does form a range of contacts with the nucleotide binding site (Fig. 4b).
In the crystal structures of the complexes, Hint-GMP, Hint-AMP and Hint-8Br-
AMP, two identical nucleotide-binding sites were found in the homodimeric structure of
Hint. 7 A set of conserved hydrophobic residues composes the binding site for the purine base, while conserved nonpolar and polar residues form the binding site for the ribose.
Conserved polar residues, including His-110 and His-112 from the HIT motif, make up the binding site for the α-phosphate (Fig. 5). 7 Sequences, from 17 HIT proteins, within
the nucleotide binding region have been compared and only six residues were found to be
13 identical in every HIT protein (Phe-19, His-51, Leu-53, His-110, His-112, His-114). Five of the six absolutely conserved residues make direct contact with the nucleotide. 7
V. Role of Hint Homodimerization
To study the importance of homodimerization on the catalytic activity of Hint1,
Wagner and coworkers designed the monomer version of Hint1 by destabilizing the
dimerization interface. Replacement of Val97 of hHint1 with Asp, Glu, or Arg resulted
in monomeric mutants of Hint1, which were characterized by a combination of size-
exclusion chromatography, static light scattering, and chemically induced dimerization
studies. Significant perturbations of the active-site residues were not detected by
molecular dynamics simulations of the monomeric hHint1. The combined kinetic and
structural results demonstrate that, for monomeric hHint1, the efficiency (k cat /K m) of
acylated-AMP hydrolysis, but not maximal catalytic turnover (k cat ), is dependent on homodimerization. 45
14
Figure 4. a) X-ray crystallographic structure of hHint1 with bound AMPCP (PDB:
1AV5). hHint1 exists as a homodimeric protein with α + β overall folding topology.
Residues in the nucleotide binding site are not involved in the dimerization interface and
are positioned 25Å apart on the opposite ends of each monomer.
15 b) Structure of the hHint1 C-terminus loop. The C-terminus of each monomer resides in close proximity to the catalytic residues from the opposite monomer. A tryptophan residue (Trp-123) in the C-terminus of one monomer (blue) is in close proximity to the active-site of the other monomer (yellow).
16
Figure 5. Rabbit Hint1 guanosine monophosphate binding pocket: the signature histidine triad residues are mainly responsible for stabilizing the binding to the α-phosphate,
whereas the base moiety of the nucleoside is sandwiched between two phenylalanines
and an isoleucine.
17
VI. Possible Biological Functions of Hints
During the last decade, evidence has begun to accumulate that Hints are involved
in a wide array of biological processes. Two-hybrid screening experiments revealed that
Hint1 interacts with Cdk7, the catalytic subunit of the cyclin dependent kinase activation
complex Cdk7-cyclin H-MAT1. 46 Analogous interactions have been found between
HNT and Kin28, the yeast orthologs of Hint and Cdk7. 9 Nevertheless, Hint1 mouse knock-out studies indicate that Hint1 is not required for Cdk7 function. 47 hHint1 has also been shown to directly interact with human Pontin and Reptin in the TCF-β-catenin transcription complex. 48 Further investigations with the knock-out mice have shown that
at 2~3 years of age, both heterozygous and homozygous mice were been found to have an
increased susceptibility to the induction of ovarian and mammary tumors by the
carcinogen 7, 12-dimethylbenzanthracene (DMBA) and to the production of spontaneous
tumors. 49 Up regulation of Hint1 and the significantly reduced in vivo tumorigenicity of
5-aza-dC-treated human non-small cell lung cancer cell line NCI-H522 demonstrated that hHint1 might be a tumor-suppressor. 50 The recent observation that hHint1 is involved in the modulation of apoptosis, independent of its enzymatic activity, suggests a possible mechanism for its tumor-suppressor activity. 51 Consistent with its potential tumor- suppressive function, over-expressed Hint1 was found to inhibit cell growth, as well as activator protein-1 activity in the human colon cancer cell line SW480. 52 In addition, it has been suggested that the tumor-suppressing ability of Hint1 is related to a potential role in enhancing DNA repair processes associated with the histone variant, H2AX and the gene, ataxia telangiectasia mutated (ATM). 53
18
Because Hint is widely expressed in the brain, including the frontal cortex, Hint1 has been suggested to have a physiological function in the CNS. 54 Studies with knock-
out mice have revealed a decreased ability for spontaneous locomotion and
supersensitivity to amphetamine, suggesting that Hint1 is involved in the regulation of
postsynaptic dopamine transmission. 55 Recent studies have provided evidence that control of the activity of morphine on the function of the µ-opiod receptor is associated with zinc-dependent binding of Hint1 to the N-terminal cysteine rich domain of specific
GTPase activating proteins of receptor-activated GazGTP subunits (RGSZ’s). 56 Further analysis has revealed that the Hint1-RGSZ complex functions as an adaptor that facilitates control of the downstream activity of PKC γ by regulating its association with the µ-opiod in a zinc-dependent manner; thus preventing desensitization of the µ-opiod receptor. 56
Like Hint1, human Hint2 is a purine nucleotide phosphoramidase. 57 hHint2 has
61% sequence similarity to hHint1 and is thought to be a homodimer. 57 hHint2 is found
exclusively in mitochondria and has been shown to be a mitochondrial apoptotic
sensitizer, which is down-regulated in hepatocellular carcinomas. 57 It also appears to be
involved in the regulation of calcium-independent steroidogenesis. 58 hHint2 silencing in
H295R cells resulted in a marked reduction of the steroidogenic response. The duration
of the mitochondrial calcium signal induced by angiotensin II was also reduced upon
hHint2 silencing, but not affected after its overexpression, suggesting that under basal
conditions, hHint2 is optimally expressed to regulate steroidogenesis. 58
Little is known about Hint3. Wagner and coworkers have discovered that hHint3,
which is only 28% sequence identical to hHint1, is likely to be a distinct branch of the
19
HIT superfamily, since it is a multimeric oligomer and strongly prefers acyl-adenylates over purine nucleotide phosphoramidates as substrates. 59 Although of unknown physiological consequence, a natural monomeric polymorph has been identified. 59 In contrast to hHint1, when ectotopically expressed, hHint3 appears in both the cytoplasm and nucleus. 59
VII. Prokaryotic Hint
E. coli Hint (ecHinT) was first discovered by Wagner and coworkers when an
open reading frame, designated ycfF , at the 16-min position (161090–1161467) on the E.
coli genetic map was found to have 47% amino acid sequence identity to rabbit and
human Hint1. 8 Moreover, after cloning, purifying and characterizing the activity of the
purified protein in vitro , it was shown that the hinT gene is indeed a nucleoside
monophosphoramidase. Based on these finding, ycfF was renamed as hinT . 8 Further characterization demonstrated that bacterial Hint is homodimeric and capable of hydrolyzing adenosine and guanosine 5’-phosphoramidate monoesters significantly faster than mammalian Hint. 8 Analysis of the lysates from a constructed hinT knock-out strain of E. coli demonstrated that all of the cellular nucleoside phosphoramidase activity is due to ecHinT and that it is the only purine phosphoramidase expressed by E. coli . 8
An observed association of ecHinT with growth under high salt conditions was
the first suggestion of a possible physiological role of Hint in bacteria. 8 Interestingly,
the reported phenotype appeared not to be connected to the phosphoramidase activity of
ecHinT, since the expression of either wild-type ecHinT or the catalytically-impaired
mutant, H101A, appeared to rescue the knock-out strain from the cation-dependent
phenotype. 8
20
Despite their ubiquity and high sequence similarity to eukaryote Hint1 (i.e, ecHinT is 48% indentical to hHint1), only a few studies of prokaryote Hints have been reported. ecHinT is alleged to form stable protein-protein interactions with six species: a putative oxidoreductase and formate dehydrogenase (b1501), the heat shock protein 70
(Hsp70), the β-subunit of DNA polymerase III ( dnaN ), a membrane-bound lytic murein transglycosylase D ( dniR ), ET-Tu elongation factor ( tufA ), and a putative synthetase
(yjhH ). 60 In addition, Mycoplasma HinT has been shown to interact with two
membrane proteins (P60 and P80). 61; 62 However, the physiological and biochemical
importance of these interactions remains unresolved.
Sequence analysis of both prokaryote and eukaryote Hints revealed that a
significant amount of sequence diversity between species resides in the C-terminus.
Although ecHinT and hHint1 share nearly 50% sequence similarity, the C-termini are not
sequence similar nor of the same length. Alignment of the human (126 residues), rabbit
(126 residues) and E. coli Hint (119 residues) sequences (Table 2) reveals that there is a
12-residue deletion in the N-terminus of ecHinT compared to the mammalian sequence
(residues 5-16), a single residue insertion at residue 79 of ecHinT, and a four-residue extension of the C-terminus of ecHinT compared to mammalian sequences (Table 2). To probe the impact of the C-terminus of Hints, Wagner and coworkers constructed two chimeric proteins in which the C-terminal loops were switched between hHint1 and ecHinT. The Human/ec chimera, which contains the C-terminus of ecHinT, exhibited
nearly identical specificity constants (k cat /K m) to those found for ecHinT, whereas the specificity constants of the ecHinT/Hs chimera, which contains the C-terminus of hHint1, were found to approximate those for hHint1. The conclusion was that substrate
21
Table 2. Sequence alignment for ecHinT, human and rabbit Hint.
1 9 19 29 39 ec _HINT MAEE------TIFS KIIRREIPSD IVYQDDLVTA FRDISPQAPT HILIIPNILI 1 11 21 31 41 51 human_HINT madeiakaqv arpggdtifg kiirkeipak iifeddrcla fhdispqapt hflvipkkhi rabbit_HINT madeiakaqv arpggdtifg kiirkeipak iifeddqcla fhdispqapt hflvipkkhi
49 59 69 79 89 99 ec_HINT PTVNDVSAEH EQALGRMITV AAKIAEQEGI AEDGYRLIMN TNRHGGQEVY HIHMHLLGGR 61 71 81 91 100 110 human_HINT sqisvaeddd esllghlmiv gkkcaadlgl –nkgyrmvvn egsdggqsvy hvhlhvlggr rabbit_HINT sqisaaedad esllghlmiv gkkcaadlgl –kkgyrmvvn egsdggqsvy hvhlhvlggr
109 119 ec _HINT PLGPMLAHKGL 120 126 human_HINT qmhwppg---- rabbit_HINT qmnwppg----
22 specificity could be transferable by C-terminal loop exchange between hHint1 and ecHinT. 63
Recently, the first comprehensive high-resolution crystal structures of the full
length N-terminal and C-terminal residues of wild-type and the H101A mutant of ecHinT
were reported. 64 As described in Chapter Two, the structure of ecHinT shows the general α and β type fold observed for other Hint proteins 5; 6; 10; 33 and is characterized by five antiparallel β-strands and a central helix (residues 56-76). The homodimer is
formed by the interactions of the central helix and the joining of β-strands into a 10- stranded antiparallel sheet. The C-terminal residues of both monomers fold across one another, but unlike the human Hint1 structures, the terminal residues do not form a salt bridge as observed between domains A (Gly126) and domain B (Arg119) of the human
Hint1 structure (Fig. 6). 33
VIII. The Interaction of Hints with Aminoacyl tRNA Synthetases
Recently, Wagner reported the first evidence of a connection between Hint1 proteins and LysRS. 32 A series of catalytic radiolabeling, mutagenesis, and kinetic experiments was conducted with purified LysRSs and Hints from human and E. coli. The results of these studies have confirmed that lysyl-AMP generated by LysRS is a natural substrate for ecHinT and hHint1 (Fig. 7). 32 The proposed mechanism for the lysyl-AMP
hydrolysis by Hint proceeds through the formation of an adenylated-Hint intermediate
followed by water hydrolysis of the intermediate (Fig. 8). 32 Site-directed mutagenesis studies of the active-site histidine triad abolished Hint labeling when either His-101 of ecHinT or His-112 of hHint1 was replaced by either alanine or glycine. 32
23
Figure 6. Comparison of the C-terminal domain for ecHinT (PDB code 3N1S)
(green/cyan) and hHint (PDB code 1av5) (yellow/red) highlighting the hydrogen bond between G126 (red) and R119 (yellow) in hHint1.
24
Figure 7. Catalytic processes involving lysyl-tRNA synthetase and Hints.
25
Figure 8. Proposed catalytic mechanism for lysyl-adenylate hydrolysis by Hints.
26
Consistent with pyrophosphate being an inhibitor for aminoacyl tRNA synthetase, incubations in the presence of pyrophosphatase resulted in enhanced formation of Hint-
AMP. 32 However, the rationale for the interaction of Hints and LysRS is not apparent, since the over-expression of Hints appears not to inhibit protein translation or to have any other deleterious effects on bacteria.
IX. Current Research
The research described in the current thesis elucidates the contributions to the characterization of ecHinT from the aspect of function, mechanism, and structural determinants. The current work provides the first strong evidence of a biological function of the highly conserved Hint in E. coli . In Chapter 2, the impact of the C-terminus as a
structural determinant of ecHinT activity and function is described. In Chapter 3, our
discovery of a bacterial phenotype for ecHinT that is dependent on the enzymatic
catalytic activity is described. Structural and activity requirements of ecHinT for the
discovered phenotype are elaborated in this chapter. Chapter 4 describes the synthesis
and evaluation of novel Hint inhibitors. The prepared inhibitors are considered as the first
cell permeable Hint inhibitors reported in the literature. Finally, in Chapter 5, the
importance of ecHinT on D-amino acids detoxification in E. coli is described.
27
Chapter Two
Probing the Impact of the ecHinT C-Terminal Domain on Structure and Catalysis
28
I. Introduction
Histidine triad nucleotide binding protein (Hint) belongs to a ubiquitous
superfamily consisting primarily of nucleoside phosphoramidates and acyl-AMP
hydrolases, dinucleotide hydrolases and nucleotidylyl transferases. The histidine triad
(HIT) superfamily has a characteristic C-terminal active-site motif, HXHXHXX, where
X is a hydrophobic residue. 63 Based on their enzymatic function, sequence composition, and structural similarity, HIT proteins have been classified into five branches; fragile HIT
(Fhit), Hint, galactose-1-phosphate uridyl transferase (GalT) 2, Aprataxin 1, and
DcpS/DCS-1. 3; 4
Hint is considered as the ancestor of the HIT protein superfamily and is highly conserved from bacteria to humans. While prokaryote genomes, including a wide array of both Gram-negative and Gram-positive bacteria, typically encode one Hint gene, eukaryotes generally express multiple forms of Hint. Although E. coli gene disruption studies have suggested that the bacterial enzyme is necessary for growth under high-salt conditions 8, the cellular function of Hint and the rationale for its evolutionary conservation in bacteria have remained a mystery.
E. coli Hint (ecHinT) is alleged to form stable potential protein-protein interactions with six species: a putative oxidoreductase and formate dehydrogenase
(b1501), the heat shock protein 70 (Hsp70), the β-subunit of DNA polymerase III ( dnaN ), a membrane-bound lytic murein transglycosylase D ( dniR ), ET-Tu elongation factor
(tufA ), and a putative synthetase ( yjhH ). 60 In addition, Mycoplasma HinT has been shown to interact with two membrane proteins (P60 and P80). 601; 62 However, the
physiological and biochemical importance of these interactions has remained unresolved.
29
In the last decade, evidence has begun to accumulate that Hints are involved in a wide array of biological processes. Mouse Hint1 gene knock-out studies have demonstrated that Hint1 acts as a tumor suppressor. 49; 55 The recent observation that human Hint (hHint1) is involved in the modulation of apoptosis, independent of its enzymatic activity, suggested a possible mechanism for its tumor-suppressor activity. 51
Consistent with its potential tumor-suppressing function, over-expressed Hint1 was found to inhibit cell growth and activator protein-1 activity in the human colon cancer cell line
SW480. 52 Hint1 has also been implicated as a modulator of central nervous system sensitivity to amphetamine. 65 Hint2, which is highly homologous to Hint1, is found in
mitochondria and appears to function as a regulator of apoptosis. 57 Recently, protein- protein interaction studies have found that the transcription factor, TFIIH, complexes of
MITF or USF2, and complexes of lysyl-tRNA synthetase (LysRS) are associated with human Hint1 (16, 17). 34; 66
Crystallographic data for six human Hint1 and five rabbit Hints as the wild-type
(wt) apo and inhibitor-bound complexes have been reported. 5; 6; 10; 33 More recently, a number of Hint-like protein structures from pathogenic organisms have been deposited in the Protein Data Bank (PDB) by several Structural Genomics Consortia. These data reveal that the typical length of the Hint is about 120 residues and show that the protein functions as a homodimer, which is characterized by an α + β fold that contains a five-
stranded antiparallel sheet and two helices. The monomers come together to form a
homodimer with a 10-stranded antiparallel sheet that makes extensive contacts between a
helix and the carboxy-terminal amino acids of one protomer and the corresponding
residues in the other protomer. 5 The histidine signature sequence of the superfamily is
30
HXHXHXX near the nucleoside phosphate binding site. There is a fourth histidine involved in the binding pocket that is not in the contiguous sequence. Structural data reported for mammalian species reveal an incomplete trace of the full-length monomers, as electron density was not interpretable for the first N-terminal 13 residues in the human or rabbit Hint structures.
Sequence analysis of both prokaryote and eukaryote Hints revealed that the greatest sequence diversity between species resides in the C-terminus. Although ecHinT and hHint1 share a nearly 50% sequence similarity, the C-termini are neither similar in sequence nor of the same length. Alignment of the human (126 residues), rabbit (126 residues) and E. coli Hint (119 residues) sequences reveals that there is a 12-residue deletion in the N-terminus of ecHinT compared to the mammalian sequence (residues 5-
16), a single-residue insertion at residue 79 of ecHinT, and a four-residue extension of the
C-terminus of ecHinT compared to mammalian sequences.
In a previous work, we have shown that the C-terminal motif of Hint1 proteins appears to be responsible for mediating phosphoramidate and lysyl-AMP substrate specificity. 11 In our effort to understand the mechanism of ecHinT catalysis, we report the first structural characterization of a bacterial Hint protein. Our data reveal the presence of four unique homodimers in the asymmetric unit of the monoclinic crystal lattice of ecHinT with an observation of conformational flexibility in terminal loop regions. The reported crystal complexes with GMP are the first structural data reported for a bacterial Hint protein. In addition, through deletion mutagenesis, steady-state and pre-steady-state kinetics, and LysRS-generated lysyl-AMP hydrolysis studies, we have probed the importance of the C-terminus of ecHinT in catalysis.
31
II. Experimental Procedures
A. Crystallization and X-ray Data Collection (Data collection and analysis were
carried out by Dr. Vivian Cody).
Recombinant wild-type ecHinT and its H101A mutant were cloned, isolated, and
purified as described previously. 8 Crystals were grown using microbatch under paraffin oil at 20 oC from enzyme incubated with GMP prior to crystallization. Protein droplets
contained 40% polyethylene glycol (PEG) 20K, 0.1 M Mg acetate, and 0.1 M Na acetate
(pH 5.0) and the sample buffer contained 20 mM Tris (pH 7.0), 1 mM EDTA, 10%
glycerol, and 2 mM GMP. The protein concentration was 7.8 mg/ml for wild-type
ecHinT and 6.7 mg/ml for the H101A mutant protein. Crystals grew over several days’
time and were cryopreserved in PEG 400 at 16%. Data were collected on beamline 11-1
at the Stanford Synchrotron Radiation Laboratory (SSRL) using the remote access
facility. 67; 68; 69 Crystals of both complexes were monoclinic, belonged to space group
P2 1 and diffracted to 1.3 Ǻ and 1.8 Ǻ resolutions for the wild-type and H101A mutant
complexes, respectively. Data were processed with HKL2000 in Denzo and scaled with
SCALEPACK for the wild-type structure and with Mosflm and SCALA for the H101A
mutant. 70 The diffraction statistics for these structures are shown in Table 1.
B. Structure Determination and Refinement
These structures were solved by molecular replacement using the coordinates of human Hint (PDB code 1av5) 5 in the program Molrep. 71 Inspection of the resulting difference electron density map was performed with the program COOT running on a
Mac G5 workstation and revealed density for the complex. 72 The sequence changes from the search model were made during the map-fitting process. To monitor
32
Table 1. Data collection and refinement statistics for ecHinT and its H101A mutant
GMP complexes.
Data collection ec hinT wt GMP H101A ec hinT GMP PDB number 3N1S 3N1T Space group P2 1 P2 1 Cell dimensions ( Ǻ) 75.18 65.30 99.12 49.33 64.66 74.75 β = 109.75 β = 109.04 Beamline SSRL 11-1 SSRL 11-1 Resolution ( Ǻ) 1.30 (1.45) 1.70 (1.76) Wavelength ( Ǻ) 0.975 0.975 Rmerge 0.165 0.080 a,b Rsym (%) 0.115 0.108 Completeness (%) a 95.2 (77.7) 86.4(25.8) Observed reflections 222,096 140,655 Unique reflections 149,986 41,405 I/ σ(I) 5.3 9.5 Multiplicity a 3.4 (2.5) 3.4 (2.1)
Refinement and model quality
Resolution range ( Ǻ) 37.96 – 1.45 35.33-1.72 No. of reflections 149,986 39,308 R-factor c 17.4 19.4 d Rfree -factor 20.3 25.2 Total protein atoms 9071 3973
Total water atoms 1068 317 Average B-factor ( Ǻ2) 20.4 24.8 Luzzati error 0.163 0.212 Rms deviation from ideal Bond lengths ( Ǻ) 0.01 0.02 Bond angles ( o) 1.52 2.03 Ramachandran plot Residues in most favored regions (%) 97.3 97.8 Residues in additional allowed regions (%) 2.6 2.0 Residues in generously allowed regions (%) 0.1 0.2 Residues in disallowed regions (%) 0.0 0.0
a The values in parentheses refer to data in the highest resolution shell. b Rsym = ΣhΣi|I h,i - | / ΣhΣi|I h,i |, where is the mean intensity of a set of equivalent reflections. c R-factor = Σ|F obs – Fcalc | / ΣFobs , where F obs and F calc are observed and calculated structure factor amplitudes. d Rfree -factor was calculated for R-factor for a random 5% subset of all reflections.
33 the refinement, a random subset of all reflections was set aside for the calculation of R free
(5%). The final cycles of refinement were carried out using the program Refmac5 from the CCP4 suite of programs. 73 The Ramachandran conformational parameters from the
last cycle of refinement, as generated by PROCHECK 74 , showed that more than 95% of
the residues have the most favored conformation and only 10 residues are located in the
disallowed regions of the protein. These residues are near the terminal residues that had
poor electron density.
Analysis of the data showed that the wild-type ecHinT crystallized with four
homodimers in the asymmetric unit of the monoclinic lattice, while the H101A mutant
complex crystallized with two homodimers in the asymmetric unit. Non-crystallographic
symmetry was not used during the refinement, and each monomer was refined
independently. There are 119 residues in each ecHinT monomer, compared with 126
residues in the human protein. The final cycle of refinement for ecHinT revealed
residues 1-117 in chain A, residues 2-119 in chain B, residues 1-116 in chain E, residues
2-119 in chain F, residues 2-116 in chain I, residues 1-119 in chain J, and residues 2-116
in chains M and N. There were eight copies of GMP. Twenty-eight copies of ethylene
diol from the buffer, as well as 1113 water molecules, were located in the structure. This
is the first report on the complete trace of the residues in a Hint protein, as previous
structures reported no interpretable electron density for the first 14 N-terminal residues
and for the last 5 C-terminal residues. In the ecHinT H101A mutant protein, the final
refinement revealed residues 5-112 for chain A, residues 5-116 for chain B, residues 5-
114 for chain E, and residues 4-114 for chain F. Four copies of GMP and 317 water
molecules were observed in the electron density.
34
C. Site-directed Mutagenesis
C-terminus deletion and alanine scan variants were generated from E. coli hinT -
pSGA02 expression vector harboring the gene encoding wild-type ecHinT using the
Quick-Change mutagenesis kit (Stratagene) in accordance with the manufacturer’s
protocol. Mutagenic primers are listed in Table 2. Encoding sequences of the C-
terminus mutants were confirmed by DNA automated sequencing (Biomedical Genomic
Center).
D. Expression and Purification of Recombinant Proteins
E. coli hinT -disrupted strain BB2 was used for the expression of ecHinT proteins
to avoid possible wild-type ecHinT contaminants, as previously described. 8 BB2 cells
containing the respective plasmids were grown at 37ºC in terrific broth (TB) medium
containing 100 µg/ml ampicillin and 50 µg/ml chloramphenicol to an OD 600 of 0.6.
Expression was induced by adding isopropyl-β-D-thiogalactopyranoside to a final concentration of 0.5 mM, and the culture was incubated for a further 10 h. Cells were harvested by centrifugation for 15 min at 6000 g, and the pellet was resuspended in buffer
A [20 mM Tris (pH 7.0), 1 mM EDTA, and 1 mM DTT]. Cell lysate was obtained by
15-s sonication (nine times) in lysis buffer [1 mg/ml lysozyme, 20 mM Tris (pH 7.00), 1 mM EDTA, 1 mM DTT, and protease inhibitor]. Lysates were centrifuged at 25,000 g at
4ºC for 30 min. ecHinT proteins were purified by an AMP-agarose affinity column
(Sigma) followed by a PD-10 desalting column (GE Healthcare). The purified proteins were exchanged with buffer A and concentrated with an Amicon stirred cell with a YM-
10 membrane (Millipore).
35
Table 2. Mutagenic primers for the C-terminus mutants.
Protein Primer
117-119 GGACCAATGCTGGCGCATTAAGCGCGCTCGAGGGTACCC
114-119 GGCCGTCCGCTGGGACCAATGTAAGCGCGCTCGAGGGTA
l114A GACCAATGGCGGCGCATAAAGGTCTGTAA
H116A TGCTGGCGGCTAAAGGTC
K117A TGGCGCATGCAGGTCTGTAA
L119A ATAAAGGTGCGTAAGCGCGTC
36
Homogeneity was analyzed by SDS-PAGE and gel filtration chromatography. Protein concentrations were determined with the Bradford protein assay (Bio-Rad). The E. coli
BS68 strain harboring the pBAS39 (derived from pET3a) plasmid for expressing ec LysU as a C-terminal His6-tag fusion protein was a gift from Dr. Paul Schimmel (The Scripps
Research Institute). 75 LysU protein was purified by Ni +2 agarose binding according to a previously published procedure. 76
E. Circular Dichroism Spectroscopy
Circular dichroism (CD) experiments were performed on a Jasco J710 spectropolarimeter equipped with a temperature-controlled water bath. Proteins at a concentration of 0.1 mg/ml in buffer A were analyzed in a quartz cuvette with a path length of 1 mm under N 2. The spectra of the wild-type and the mutants were determined in the far-UV region (190–260 nm). The spectra were accumulated and averaged over nine scans, and the buffer background was subtracted from the protein spectra. The measured ellipticities were converted into mean residue ellipticities, MRE, using Jasco software. To determine the thermal denaturation of proteins, CD unfolding measurements were obtained at 222 nm with a temperature gradient of 1°C/min between
10°C and 80°C.
F. Size-exclusion Chromatography
Size-exclusion chromatography was conducted to determine apparent molecular weights (Mw) using a Superdex G 200 column (GE Healthcare) at a flow rate of 0.5 ml/min. Protein samples (25 µM) were prepared in buffer A with 5% (vol/vol) glycerol.
Proteins were eluted with P500 buffer [0.5 M NaCl, 50 mM potassium phosphate, and 1 mM EDTA, (pH 7.0)] and the retention time of proteins was monitored by absorbance at
37
280 nm using an in-line UV detector (Beckman Gold 168). The molecular standards used were blue dextran (200 kDa), albumin (66 kDa), DHFR2-1DDG (36 kDa), carbonic anhydrase (29 kDa), cytochrome c (12.4 kDa), and aprotinin (6.5 kDa) (Sigma).
G. Steady-state Kinetics
Turnover rates were measured by following the hydrolysis of either tryptamine
5’-adenosine phosphoramidate (TpAd) or adenosine 5’-indole-3-propionic adenylate
(AIPA) as fluorogenic substrates, by the ecHinT proteins, as previously described. 11
Excitation wavelength was set at 280 nm; fluorescence emission was measured at 360 nm, and all the kinetic assays were performed in duplicate at 25ºC. The rate of hydrolysis of the substrates was determined by measuring the increase in fluorescence intensity upon the addition of the enzyme over the course of the reaction. The Michaelis-
-1 Menten constants, k cat (s ) and K m (µM) were determined by a nonlinear regression analysis of the initial velocity versus concentration using the JMP IN 7 software (SAS
Institute, Inc., Cary, NC, USA).
H. Pre-steady-state Kinetics
Pre-steady-state kinetic experiments were performed on a fluorescence two- syringe stopped-flow apparatus (model SX.18MV, Applied Photophysics). The reaction rates of the ecHinT adenylation were monitored at 25°C with either AIPA or TpAd in
HEPES buffer [20 mM, and 1 mM MgCl 2, (pH 7.2)]. The sample was excited at 280 nm
and the fluorescence emission at 340 nm was monitored with a cutoff filter of 320 nm.
The time-course curves were fitted, using the JMP IN 7 software with the equation: P(t) =
-kobst A0 – A(e ) + k hydro •t, where A0 is the fluorescence intensity at time zero, A is the
amplitude, k hydro is the hydrolysis rate of the intermediate, k obs is the pseudo-first-order
38 rate constant for the adenylation step, and t is time. The results represent the average of five experiments.
I. pH-dependence of Steady-state Kinetics
The pH-dependence of the steady-state turnover rates was determined over a pH range of 5.8-9.0 using phosphate buffer (0.1M, pka=7.2). Hydrolysis rates were determined at 25ºC and k cat values were plotted versus the pH.
J. Lysyl-AMP-dependent Adenylation of ecHinT by LysU
E. coli LysU (2 µM) was incubated with [ α-32 P]ATP (0.2 µM, 800 Ci/mmol, MP
Biomedicals) in buffer B [10 µl, 25 mM Tris-HCl (pH 7.8), 100 mM NaCl, 2 mM MgCl 2,
1 mM dithiothreitol, 113 µM Lysine, 0.02 unit/µl, inorganic pyrophosphate, and protease
inhibitor tablet (Roche Applied Science)] at 25°C for 1 min, then ecHinT proteins (5 µM)
were added and incubated for 10 min. The reaction was terminated by the addition of
SDS sample buffer (4x, 5 µl; Invitrogen). The reaction mixture was boiled for 3 min, and
the proteins were separated by SDS-PAGE and electroblotted onto a polyvinylidene
difluoride membrane. Labeled proteins were visualized by subjecting dried
polyvinylidene difluoride membranes to autoradiography with a storage phosphor screen
for 12 h, followed by scanning with a Storm 840 PhosphorImager.
K. PDB ACCESSION NUMBERS
Coordinates for wild-type ecHinT and H101A ecHinT have been deposited in the
PDB with accession codes 3N1S and 3N1T respectively.
39
III. Results and Discussion
A. Structural Properties of ecHinT
This is the first report on the crystal structures of full-length Hint from E. coli (a
119 residue homodimeric protein) and its H101A mutant, both in complex with GMP.
There are four unique homodimers in the asymmetric unit of the monoclinic crystal lattice for the wild-type ecHinT-GMP complex and two unique homodimers in the asymmetric unit for the H101A mutant complex. In all cases, GMP is bound in both monomeric domains, unlike the example of the hHint1 complex, which revealed that only one ligand was bound to the homodimer. 5; 33
The quality of the electron density maps for the ecHinT GMP complex was such that nearly all the N-terminal and C-terminal residues could be traced for the four independent homodimeric pairs of ecHinT in the asymmetric unit of the crystal lattice
(Fig. 1). As structural data reported for human or rabbit Hints revealed no interpretable electron density for the first 14 N-terminal residues, our structural analysis of ecHinT is the first report to show the complete trace of the N-terminal domain of a Hint protein.
Analysis of electron density for both the N-terminal and the C-terminal residues of ecHinT-GMP complex indicates that the loops in some monomers can adopt more than one conformation. The largest variations are observed in the N-terminal and C-terminal regions and for the loop regions opposite the guanine ring of GMP. The observation of conformational flexibility in the terminal loop regions could explain the presence of multiple homodimers in the asymmetric unit of these structures.
The structure of ecHinT shows the general α+β− type folds observed for other Hint proteins 5; 6; 10; 33 and is characterized by five antiparallel β− strands and a central helix
40
(residues 56-76). The homodimer is formed by the interactions of the central helix and the joining of β− strands into a 10 −stranded antiparallel sheet. β− sheet interactions across
the homodimeric interface are made by residues 83-89 of both domains (Fig. 2a), while
on the opposite surface, the helices are nearly perpendicular to each other, crossing Gly63
(Fig. 2b), similar to that of the hHint1 structure (Gly75). 5; 33 The central helix of ecHinT is longer than that of the human Hint1 and is displaced relative to it (Fig. 2b).
The C-terminal residues of both monomers fold across one another; however, unlike the human Hint1 structure, the terminal residues do not form a salt bridge, as observed between domains A (Gly126) and domain B (Arg119) of the hHint1 structure
(Fig. 3). 33 Rather, the C-terminal residues in ecHinT extend beyond those of the human
Hint1 sequence (Fig. 3). Additionally, the N-terminus of ecHinT reveals an extended loop region with little tertiary structure among the eight monomers in the asymmetric unit of the wild-type structure. Also shown (Fig. 3b) are the residues of ecHinT removed in the ∆114-119 mutant, which reduces the surface interface of the homodimer and make it more like that of the hHint1 structure.
41
Figure 1. Difference electron density map for a) ecHinT showing the GMP binding
pocket. 2Fo-Fc (1 σ) (blue); Fo-Fc (3 σ) (green).
42 b) ecHinT H101A mutant with GMP.
43
Figure 2. ecHinT structure superimposed on the structure of hHint1 a) View of ecHinT
domains A (green) and B (cyan) superimposed on the structure of hHint (PDB code 1av5)
(yellow/red) illustrating the beta sheet structure and the interleaving of the C-terminal
residues of each domain.
44 b) View of the opposite side of the A-B homodimer showing the helical dimer interface crossing at Gly63 (space fill) and the binding of GMP compared with hHint. The B domains were used to fit these structures. Drawn with PyMol .
45
Figure 3. ecHinT C-terminus structure a) Comparison of C-terminal domain for ecHinT
(green/cyan) and hHint (PDB code 1av5) (yellow/red) highlighting the hydrogen bond
between G126 (red) and R119 (yellow) in hHint1.
46 b) Surface representation of residues removed for the ecHinT ∆114-119 mutant. Drawn
with PyMol.
47
B. Ligand Binding
Hint protein superfamily is characterized by the conserved histidine triad (His-X-
His-X-His-XX) sequence motif where X is a hydrophobic residue. Structural data reveal the proximity of a fourth His that can be considered part of the histidine triad catalytic site (Fig. 4a). These data show that the substrate GMP or AMP binds in a cleft near the
His triad binding site, with the phosphate oxygen atoms involved in hydrogen −bonding
contacts with the side chain functional groups of His101, His103, and Asn88, and the
backbone functional groups of Glu96 and Val97, in addition to solvent. Interactions of
the GMP indicate that the nucleoside ring occupies an open cleft region that is in contact
with the C-terminus of other domains which pack against this region. Furthermore, the
C-terminal residues of ecHinT extend farther than the C-terminal Gly126 of human
Hint1, which forms a salt bridge with Arg119 of the other domain (Fig. 3). This
extended structure in ecHinT permits a closer involvement of the C-terminus with the
nucleoside binding site.
The structure of hHint1 (PDB code 1av5) was shown to bind only one ligand in
the homodimer. When domain B is used to superimpose the two structures, which both
have a bound ligand, there is a good fit of the two B domains (Fig. 4b); however, there is
a displacement in the positions of the secondary structural features of domain A,
compared to the fit of domain B (Fig. 4a). Part of the shift may be caused by a collapse
of the binding pocket, as the superposition of the structure (PDB code 1kpe), with a
ligand bound in each domain, shows a similar pattern of shifts in domain A that is
intermediate between that of ecHinT and that of hHint1 (PDB code 1av5).
48
Figure 4. Environment of GMP binding a) in domain A to signature His (39, 99, 101,
103 in ecHinT) (green) and hHint (yellow).
49 b) Similar view with ecHinT domain B (cyan) and hHint with Ap2 in domain B (red).
Drawn with PyMol.
50
Comparison of the ecHinT H101A mutant-GMP complex with the wild-type structure reveals that the effect of changing His to Ala at this position permits a small shift in the phosphate group of GMP such that the contact distance from the phosphate oxygen to His101, as observed in the wild-type ecHinT structure (3.03/3.33 Ǻ for domains A and B, respectively), is shorter for the mutant model (2.35/2.91 Ǻ). The structure of the transition-state analogue ADV from human Hint1 (PDB code 1kpe) 5 shows the formation of a covalent link of the transition state analogue with His110
(2.52 Ǻ). The equivalent contact in the ecHinT-GMP complex is 3.43 Ǻ. These data
support the diminished catalytic activity observed for the ecHinT H101A mutant, as the
active-site nucleophile is no longer available (Fig. 5).
51
Figure 5. a) Environment around GMP in domain A showing the disposition of the
HxHxHxx motif of wild-type ecHinT (green) compared with that of the H101A mutant
(violet).
52 b) View of binding pocket in ecHinT showing GMP phosphate interactions. Drawn with
PyMol.
53
C. Design of C-terminus Mutants and Protein Purification
The significance of the C-terminus of Hint1 proteins in mediating their phosphoramidate and acyl-AMP hydrolyase activity was studied previously. 63 Studies with chimeric ecHinT, in which the E. coli C-terminus was replaced with the hHint1 C-
terminus, and chimeric hHint1, in which the hHint1 C-terminus was replaced with the
ecHinT C-terminus, revealed that the substrate specificity of each enzyme could be
transferred to the other by a simple C-terminal switching. 63 To gain further insights into
the critical role of the C-terminal loop in governing the catalysis and substrate specificity
of Hint1 proteins, we designed four alanine mutants of the C-terminus and two sequential
deletion mutants, that shortened the length of the C-terminus by three ( ∆117-119) and six
(∆114-119) amino acids. Plasmids encoding the DNA sequences for the mutants were generated, expressed in an E. coli strain which lacks the ability to endogenously express wild-type ecHinT, and purified using affinity chromatography. The purification of ecHinT wild-type using an AMP-agarose column was reported previously. 8 The same
protocol was followed to purify the mutants with minor modifications.
Compared to wild-type and ∆117-119 mutant, the purified yield of ∆114-119 mutant, using the AMP-agarose column, was significantly reduced, consistent with our earlier finding that the C-terminal loop plays an important role in ligand binding. 63
D. Secondary Structure Analysis by Circular Dichroism (CD) Spectroscopy
To investigate the robustness of the secondary structure of ecHinT to the sequential deletion of 3 and 6 residues from the C-terminal loop, far-UV CD spectra of the wild-type and the deletion mutants were determined. Relative to the wild-type, the
CD spectra of the recombinant C-terminus deletion proteins show that the helical
54 structure of the mutants has likely been increased. However, since The CD spectra of all three proteins shared a minimum at 222 nm at 25ºC, it is likely that shortening the C- terminal loop by 3 or 6 amino acid residues did not lead to major changes in the secondary structure (Fig. 6a). The secondary structures of the four alanine mutants were also studied. Similar to wild-type, all four mutants shared a minimum point at 222 nm at
25ºC (Fig. 6b), suggesting that the overall secondary structure was not affected by any of the alanine mutations in the C-terminus loop. Interestingly, although the H116A mutant shared the same minimum point at 222 nm, analysis of its CD spectrum revealed the presence of more β-sheets in the secondary structure.
E. Thermal Stability
To determine the effect of C-teminus deletion on the thermal stability of ecHinT,
the temperature dependence of the CD spectra was measured. The denaturation curves
showed midpoints at 55 ± 0.4ºC and 49 ± 0.90ºC for ∆117-119 and ∆114-119 ecHinT, respectively. Wild-type ecHinT has a Tm value of 61 ± 0.5ºC under the same experimental conditions. Based on these results, it is apparent that the C-terminal loop has a significant impact on ecHinT stability (Fig. 7a). Nevertheless, given that the C- terminus of hHint1 is three amino acids shorter than the ecHinT C-terminus, it is likely that protein stability is also influenced by other structural motifs. Thermal stability studies of the four alanine mutants did not reveal any significant changes in Tm values of the alanine mutants relative to wild-type ecHinT (Fig. 7b).
55
Figure 6. Secondary structure analysis of a) ∆114-119 and ∆117-119 C-terminus deletion
mutants relative to wild-type ecHinT.
56
b) L114A, H116A, K117A, and L119A C-terminus alanine mutants relative to wild-type ecHinT.
57
Figure 7. Thermal stability studies of a) ∆114-119 and ∆117-119 C-terminus deletion mutants relative to wild-type ecHinT.
58 b) L114A, H116A, K117A, and L119A C-terminus alanine mutants relative to wild-type ecHinT.
59
F. Size-exclusion Chromatography
To determine the dimerization state of the proteins under native conditions, samples of wild-type ecHinT, ∆114-119, ∆117-119 and the four alanine mutants were analyzed on a Superdex G 200 column with in-line UV detector. Wild-type ecHinT eluted with a retention time of 36.5 min corresponding to the theoretical mass (26 kDa) for the dimer (Fig. 8). In good agreement with wild-type ecHinT, only homodimeric protein was observed for all mutants with retention times of 36 min ( ∆114-119), 36.5 min
(∆117-119) (Fig. 8a), and 36 ± 0.5 min for all the four alanine mutants (Fig. 8b).
G. Effect on Catalysis
Previously, it has been demonstrated that Hint1 catalysis proceeds by the
formation of an active-site His-NMP intermediate, followed by intermediate hydrolysis.
32 Evidence in support of this hypothesis for ecHinT has arisen from several observations. First, the ecHinT active-site residue, His101, has been shown to be essential for catalytic activity. 8 Second, His101-dependent intermediate formation was directly
observed during the ecHinT and hHint1 hydrolysis of LysRS-generated lysyl-AMP. 32
Recently, we have shown that the C-terminus of Hint1 is a significant contributing factor in determining the phosphoramidate substrate specificity. 63 Given the proximity of the
C-terminus to the active-site of ecHinT and our observation that it can adopt multiple conformations ( vide supra ), we chose to investigate the role of the C-terminus in catalysis
by comparing the catalytic efficiencies of the deletion mutants, ∆117-119 and ∆114-119.
In addition, from previous studies of engineered monomeric hHint1 45 and C-terminal chimeras of hHint1 and ecHinT 63 , we suspected that the C-terminus may play a role in the hydrolysis of lysyl-AMP generated by E. coli LysRS.
60
Figure 8. Size-exclusion chromatograms from the Superdex G 200 column equipped
with UV (absorbance of 280 nm): a) ∆114-119 (pink) and ∆117-119 (blue) C-terminus deletion mutants compared to wild-type ecHinT (green).
61 b) L114A, H116A, K117A, and L119A C-terminus alanine mutants compared to wild- type ecHinT.
30 30 Det 168-280nm Det 168-280nm Det 168-280nm Det 168-280nm Det 168-280nm Det 168-280nm WildType-1 114-1 116-1 117-1 119-1
20 20
10 10 m A U m A U
0 0
-10 -10 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 37.5 40.0 Minutes
62 c) An overlay of all the C-terminus deletion and the C-terminus alanine mutants with wild-type ecHinT.
30 30 Det 168-280nm Det 168-280nm Det 168-280nm Det 168-280nm Det 168-280nm Det 168-280nm Det 168-280nm 117 114 114-1 116-1 117-1 119-1 WildType-1
25 25
20 20
15 15
10 10 m A U m A U
5 5
0 0
-5 -5
-10 -10 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 Minutes
63
To characterize the effect of C-terminus deletion on ecHinT catalysis, the steady- state and pre-steady-state kinetic parameters were determined with a model substrate phosphoramidate substrate, TpAd 11 , and a model acyl-AMP substrate 45 , AIPA. When
first assessed by steady-state kinetics, ∆117-119 and ∆114-119 mutants exhibited an approximately 9-fold reduction in catalytic efficiency (i.e. k cat /K m values) compared to
wild-type ecHinT for the hydrolysis of both substrates. The basis for this perturbation
differed for each mutant. For ∆114-119, the reduction in k cat /K m was largely due to the 5-
fold reduction in k cat , with only a modest 2-fold increase in K m. In contrast, a larger 5-
fold increase in the K m and only a modest 2-fold decrease in k cat were observed for the
∆117-119 mutant (Table 3). Catalytic efficiencies in the steady-state step, with the two model substrates, for the four alanine mutants were also evaluated (Table 4). All alanine mutants exhibited 3-5 fold reduced catalytic efficiencies relative to the wild-type. Except for the H116A mutant, the reduction in k cat /K m was mainly due to an increase in K m values.
To verify that the observed reduction in the steady-state parameters for the
deletion mutants was not because of a shift in the pH-versus-rate profile relative to wild-
type ecHinT, k cat and K m values were determined for the hydrolysis of TpAd at variable
pH points. Wild-type ecHinT and the two deletion mutants exhibited bell-shaped pH
profiles from pH 5.8 to 9.0, with pKa values of 6.9 and 8.5 and pH optima of 7.8 (Fig. 9).
64
Table 3. Steady-state and pre-steady-state kinetic parameters for ∆114-119 and ∆117-
119 C-terminus deletion mutants compared to wild-type ecHinT, using AIPA or TpAd as substrate, in HEPES buffer (pH 7.2, 20 mM HEPES, 1 mM MgCl 2) at 25°C.
NH2 NH 2 N N N N O N N O H O N N N P O O O- HN O P O O HN - O OH OH OH OH
AIPA TpAd
Rate constant WT 114-119 117-119 WT 114-119 117-119
-1 k2 (s ) 135±5 35±2 39±3 116±6 22±1 33±2 adenyl Km (µM) 7.5±2.0 18±2 16±1 11.0±2 23 ±3 21±4 adenyl 6 -1 -1 k2/K m (×10 s M ) 18.0±1 1.9±0.5 2.4±0.4 10.5±0.2 0.96±0.10 1.6±0.3 -1 kcat (s ) 4.0±0.1 0.79±0.10 1.9±0.2 4.5±0.1 0.88±0.10 2.3±0.8 Km (µM) 4.2±0.4 13±1 15±2 5.2±0.2 9.2±0.8 23.8±2.4 3 -1 -1 kcat /K m (×10 s M ) 952±200 59±4 123±15 870±50 96±18 97±24
65
Table 4. Steady-state kinetic parameters for L114A, H116A, K117A, and L119A C- terminus alanine mutants compared to wild-type ecHinT, using AIPA or TpAd as substrate, in HEPES buffer (pH 7.2, 20 mM HEPES, 1 mM MgCl 2) at 25°C.
TpAd AIPA
-1 -1 Protein Kcat (s ) Km(µM) Kcat /K m Kcat (s ) Km(µM) Kcat /K m (×10 3 s-1M-1) (×10 3 s-1M-1) WT ecHinT 4.5±0.1 5.2±0.2 870 ± 50 4±0.1 4.2±0.4 952 ± 200
∆114-119 0.88±0.1 9.2±0.8 96 ± 18 0.79±0.1 13±1 59 ± 4
∆117-119 2.3±0.8 23.8±2.4 97 ± 24 1.9±0.2 15±2 123 ±15
L114A 2.4±0.6 13.9±2.5 173±11 2.8±0.5 16.6±1.6 169±7
H116A 2.3±0.5 3.2±0.84 719±22 3.0±0.2 4.6±.3 652±20
K117A 3.5±0.3 12.5±1.2 280±13 2.9±0.7 14.7±1.6 197±12
L119A 1.9±0.2 8.3±0.4 229±15 2.2±0.1 9.1±0.8 242±14
66
Figure 9. pH-dependence rate profile for wild-type ecHinT, ∆114-119 and ∆117-119 C- terminus deletion mutants.
67
It had been previously proposed that the kinetic mechanism for both phosphoramidate and acyl-AMP hydrolysis by Hints is likely to proceed through active- site adenylation, followed by the enzyme intermediate hydrolysis (Scheme 1). 32 If the rate of adenylation is faster than intermediate hydrolysis, we hypothesized that one should be able to observe a burst of tryptamine or indolylic acid after treatment with
TpAd or AIPA, respectively. This fast first step (k2) would then be followed by a slower,
possibly rate-limiting, intermediate hydrolysis step (k cat ). Consistent with our hypothesis, transient-state kinetic measurements revealed a burst, followed by a linear rate, corresponding to the observed steady-state turnover (Fig. 10). At pH 7.2, the burst rate
-1 -1 (k 2) for wild-type ecHinT was found to be 116 + 6 s and 135 + 5 s for TpAd and
AIPA, respectively (Table 3). The burst rate values (k 2) for the deletion mutants, ∆117-
119 and ∆114-119, were only modestly reduced by approximately 3.5 to 6-fold, while the
burst Km increased by only 2 to 3-fold. Thus, reduction of the C-terminal length by either 3 or 6 residues resulted in a moderate reduction in the efficiency of the burst step.
Presumably, the burst phase corresponds to either the rate of enzyme adenylation and/or a conformational step. Ongoing studies should clarify the specific nature of the catalytic and kinetic mechanisms of ecHinT and the role of the C-terminus. Nevertheless, neither deletion mutation appears to significantly alter the overall kinetic pathway.
68
Scheme 1. Proposed kinetic mechanism for both phosphoramidate and acyl-AMP hydrolysis by Hints.
k1 k2 kcat E + S E•S AMP-E + P E + AMP k-1 S = TpAd or AIPA P = Trptamine or Indole-3-acetic Acid
69
Figure 10. Stopped flow trace of ecHinT exhibited a biphasic profile, a burst phase followed by a linear phase.
70
H. Lysyl-AMP-dependent Adenylation of ecHinT by LysU
Previously we had demonstrated that lysyl-AMP generated by ec LysU is a native substrate for ecHinT. 32 In situ adenylation of ecHinT can be observed after treatment
with ec LysU and α-32 P-labeled ATP. As a proposed mechanism for the adenylation
reaction, nucleophilic attack at the phosphorous atom of either TpAd or AIPA by active-
site residue, His-101, is believed to be responsible for ecHinT-AMP intermediate
formation. 32 Evidence for this mechanism was provided by the inability of the
catalytically impaired active-site mutant H101A to hydrolyze lysyl-AMP and form the
labeled ecHinT intermediate. 32 To assess the role of the C-terminus on the hydrolysis of
lysyl-AMP generated by LysRS, we examined the ability of the deletion mutants to form
the enzyme-AMP intermediate. As can be seen in Fig. 11, the intensity of wild-type
ecHinT labeling by ec LysU (lane 2) was found to be at least 20-fold higher when
compared to ∆114-119 deletion mutant labeling (lane 4) while labeling of ∆117-119 was not observed. Since the extent of labeling could be observed for the ∆114-119 deletion mutant relative to wild-type ecHinT, the rate of the labeling reaction was determined and compared to the rate observed for wild-type ecHinT (Fig. 12). Although labeling could be observed within 10 s for wild-type ecHinT, the rate of intermediate formation for the
∆114-119 mutant was found to be at least 50-fold slower while again no labeling of the
∆117-119 mutant was observed. Taken together, these results provide insights into the crucial role of the C-terminus on the hydrolysis of lysyl-AMP generated by LysRS.
71
Figure 11. Adenylation of wild-type ecHinT and ∆114-119 mutant by ec LysU. Lane1, 3:
controls, [ α-32 P] ATP (0.2 µM) was incubated in buffer A (10 µl, 25 mM Tris HCl, pH
7.8, 100 mM NaCl, 2 mM MgCl 2, 1 mM dithiothreitol, 113 µM Lysine, 0.02 unit/µl,
Inorganic Pyrophosphate, protease inhibitor tablet) at 25°C for 1 min followed by the addition of ecHinT proteins and incubated for 10 min. Lane 2,4,5: E. coli LysU (2 µM)
was pre-incubated with [ α-32 P] ATP at 25°C for 1 min before the addition of ecHinT proteins.
1.2 1 1
0.8
0.6
0.4
0.2
Normalized P-32 Intensity 0.049 0 ecHint ∆114-119
72
Figure 12. Time-dependence of ecHinT and ∆114-119 mutant adenylation by ec LysU.
[α-32 P] ATP (0.2 µM) was incubated in buffer A in presence of ec LysU(2 µM) for 1 min followed by the addition of either ecHinT (5 µM) or ∆114-119 (16 µM). Proteins were incubated for: 10, 20, 30, and 40 seconds before the reaction was terminated.
73
IV. Summary and Concluding Remarks
In summary, we report the first comprehensive high-resolution crystal structure of the full length N-terminal and C-terminal residues of wild-type and the H101A mutant of the histidine triad nucleotide binding protein from Escherichia coli (ecHinT) complexes with GMP. These data reveal the presence of four unique homodimers in the asymmetric unit of the monoclinic crystal lattice and two unique homodimers in the asymmetric unit of the H101A mutant protein. The presence of multiple homodimers in the asymmetric unit can be explained by the conformational flexibility observed in the terminal loop regions of these structures, although the role of His-101 on the conformations remains unclear. In an effort to probe the impact of the C-terminus loop on the structural and kinetic characteristics of ecHinT, two C-terminus deletion mutants of ecHinT were constructed. Our results show that sequential deletions of 3 and 6 amino acids from the C-terminus did not result in major perturbations of the secondary structure
o of the protein. However, a significant shift by approximately 10 C in the T m value was observed for the C-terminus deletion mutants, suggesting that the C-terminus is important for maintaining ecHinT stability. Kinetic studies of wild-type ecHinT and the C-terminus deletion mutants revealed that the catalytic efficiency of ecHinT with diffusible model substrates is only modestly influenced by the C-terminal loop. In contrast, the ability of the C-terminus deletion mutants to hydrolyze the native substrate lysyl-AMP generated by LysRS was greatly impaired by the loss of as few as three amino acids. One would expect that if ecHinT could only hydrolyze lysyl-AMP after release from LysRS, little difference in the catalytic efficiency in hydrolysis of the model substrates and the enzyme generated substrates would be observed. This should be especially true for the ∆117-119
74 mutant, which was the most catalytically efficient of the two mutants ( vide supra ).
Nevertheless, unlike wild-type ecHinT and the ∆114-119 mutant, enzyme intermediate formation was not observed. Since it is well-recognized that dissociation of amino acid adenylates from amino acyl-tRNA synthetases is highly disfavored 77 , the ability of ecHinT to hydrolyze LysRS generated lysyl-AMP suggests a mechanism that may rely on a direct transfer process.
Recent results of E. coli protein-protein interaction studies demonstrated that the elongation factor (EF-Tu) may bind not only to lysyl, alanyl, and isoleucyl-tRNA synthetase but also ecHinT. 60 These results, combined with our recent finding that lysyl-AMP generated by LysRS is a physiological substrate for Hints 32 , suggest that
Hints might function as regulators of protein translation processes. Whether this
regulation occurs as a pre-transfer editing or scavenging of inadvertent aminoacyl-
adenylates, it appears to be mediated by the proposed mechanism of direct transfer of
lysyl-AMP. It is also possible that the interaction of Hints with aminoacyl tRNA
synthetases (AARS) facilitate an as yet undetermined non-canonical function of AARS.
Ongoing studies attempting to unravel the physiological function of Hints in E. coli
should shed light on either hypothesis .
75
Chapter Three
HinT, a Histidine Triad Nucleotide Binding Protein, is Essential for
Alanine Metabolism in Escherichia coli
76
I. Introduction
Histidine triad nucleotide binding proteins (Hints) are conserved from bacteria to humans. 6 While prokaryotes typically encode only one Hint gene, eukaryotes generally contain two to three different Hint genes. 2 The physiological role of Hints in mammalian cells is just now being delineated. Two hybrid screening experiments revealed that Hint1 interacts with Cdk7, the catalytic subunit of the cyclin dependent kinase activation complex Cdk7-cyclin H-MAT1. 46 However, Hint1 mouse knock-out
studies indicated that Hint1 is not required for Cdk7 function. 47 The human Hint1
(hHint1) has also been shown to directly interact with human Pontin and Reptin in the
TCF-β-catenin transcription complex 48 , and it has been reported that hHint1 might be a
tumor suppressor. 49; 50 In addition, recent studies have shown that hHint1 is involved
in the regulation of postsynaptic dopamine transmission. 55 Recently, we reported that
Hints hydrolyze lysine-AMP generated by LysRS in eukaryotes, thus suggesting that
Hints have a specific role in regulating LysRS. 32
E. coli Hint (encoded by hinT ) has high sequence similarity to human Hint1
(ecHinT is 48% identical to hHint1 at the amino acid level) and has been shown to form stable potential protein-protein interactions with six proteins: a putative oxidoreductase and formate dehydrogenase (b1501), the heat shock protein 70 (Hsp70), the β-subunit of
DNA polymerase III ( dnaN ), a membrane-bound lytic murein transglycosylase D ( dniR ),
ET-Tu elongation factor ( tufA ), and a putative synthetase ( yjhH ). 60 In addition, the
HinT from Mycoplasma has been shown to interact with two membrane proteins (P60 and P80). 61; 62 However, despite these determined interactions, and the diversity of
functions that Hints have been shown to be involved in, from regulators of tumor
promotion to amphetamine sensitivity, the physiological and biochemical importance of 77 the relationship of Hints and these proteins remain unresolved. Moreover, the foundational reason for their wide-spread conservation across all three kingdoms of life remains enigmatic.
Crystal structure studies of hHint1 and, recently, ecHinT have revealed that both proteins are homodimers containing an active-site with four conserved histidines. 5; 64
While similar, close inspection of the two structures revealed little sequence similarity between their C-termini. In contrast to hHint1, the longer C-terminus of ecHinT was found to adopt eight different conformations in the unit cell. 64 Chimeric domain swap mutants, in which the C-termini of both ecHinT and hHint1 have been switched, have demonstrated the importance of the C-termini on model substrate specificity. 63
Moreover, deletion mutagenesis studies have shown that the loss of just three C-terminal side-chains abolishes the ability of ecHinT to hydrolyze LysRS-generated lysine-AMP while having only a modest effect on the catalytic efficiency of the enzyme with model substrates. 64 Although catalytic insights of Hint activity have been garnered from
these studies, a defined biochemical rationale for the function of Hints in general, and
ecHinT in particular, has remained elusive.
Phenotype characterization is an essential approach for understanding structure-
function relationships among a variety of biological systems. While several advanced
and comprehensive technologies have been developed to sequence and identify functions
of genes and their products, as well as assign them to particular metabolic pathways, the
function of many genes in most organisms that have been sequenced to date remains
unknown. For example, although E. coli is considered to be amongst the most genetically-characterized microorganisms, about 30-40% of its genes have unknown
78 function. 78 In an effort to determine the function of many of these “unknown” genes, a
library of single gene knock-out mutants of all nonessential E. coli K-12 genes has been
generated. 79 The metabolic profiles of many of these knock-out mutants have been
characterized using Phenotype MicroArrays (PM) that allow testing of a large number of
cellular phenotypes in 96-well microplates. 80 Based on the same redox chemistry,
Biolog™ phenotypic screening plates have been developed as a simplified universal reporter of metabolism in a single bacterium.
Given the high sequence similarity between hHint1 and ecHinT, we hypothesized that determining the function of Hint in E. coli may reveal a conserved biochemical and
physiological role for Hints in general. In E. coli K-12, in silico analysis indicates that
hinT is located in an apparent operon consisting of the hinT(ycfF)-ycfL-ycfM-ycfN(thiK)- ycfO(nagZ)-ycfP genes. In the study reported here, we investigated the role of hinT in E. coli by using single gene deletion mutants and metabolic analyses. In addition, we investigated the role of ecHinT catalysis and structure on the observed ecHinT phenotype with a combination of site-directed mutagenesis and chemical biological studies. Our results show that ecHinT catalytic activity and the C-terminal domain are required for E. coli to grow on D-alanine as a sole carbon and energy source by the regulation of D- amino acid dehydrogenase activity. Taken together, our results demonstrate that ecHinT plays an essential role in the regulation of the fate of alanine in cellular compartments and thus link for the first time the catalytic activity of a Hint protein with a physiological function.
79
II. Experimental Procedures
A. Bacterial Strains, Media, and Growth Conditions
The bacterial strains used in this study were obtained from the E. coli Genetic
Stock Center at Yale University and are listed in Table 1. All strains were received as
glycerol stocks on filters, sub-cultured twice on Luria–Bertani (LB) agar medium 81 , and incubated at 37°C for 48 h before testing. Liquid cultures were grown aerobically in LB medium supplemented with kanamycin (50 µg/ml) at 37°C, with shaking, for 18 h.
Strains were stored as 10% glycerol stocks at - 80°C.
B. PCR Verification of Mutants
The mutations in gene knock-out mutants were verified using the polymerase chain reaction (PCR) technique. Primers sequences for each PCR reaction are listed in
Table 2. DNA from freshly isolated colonies served as templates for PCR and reactions were initiated following a ‘‘hot start’’ protocol at 95°C. The first PCR reaction was used to confirm the expected size of hinT operon with a single gene deletion and insertion of a kan resistance gene. The second PCR reaction confirmed the junction points of the kan resistance gene in each knock-out mutant. PCR conditions were as follows: 94°C for 2 min followed by 30 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 2 min.
C. Verification of hin T Deletion Mutant by Loss of Activity
The turnover rates of the fluorogenic substrate tryptamine 5’-adenosine phosphoramidate (TpAd) by ecHinT protein in cell-free lysates of wild-type E. coli
BW25113 and the hinT deletion mutant ( ∆hinT ) was determined using a previously
80
Table1. Bacterial strains used in this study
Gene deleted Strain Gene Size (Kb) Gene Product
hinT JW1089 0.360 Purine nucleoside Phosphoramidase ycfL JW1090 0.378 Predicted protein
ycfM JW5157 0.642 Predicted outer membrane lipoprotein ycfN JW1092 0. 825 Thiamine kinase
nagZ JW1093 1.026 Beta-N-glucosaminidase
ycfP JW5158 0.543 Conserved protein
81
Table 2. Primers used for sequence-verification PCR reaction
Mutant Forward primer Reverse primer
hinT CGTCTACGGCACACCGCGTAA GAATATAGTTTCTTCTGCCAC
ycfL TGGCGCATAAAGGTCTGTAA GGCGTAGCGACTCATTTTTGTC
ycfM GGGGGCGCACAAAGTCAGACAA GGGATTATTGCTGCGAATCGGC
ycfN CTGGTCAGGTAAAGGTGCCGTT GACATCCAACATTACTGGACCC
ycfO CCTGGCGGCAGCTATTAATAAA CCGGACTGTTAGAGTCAAAACC
ycfP AAGCGTTCAAAACCCTCGGGTAA CGCCGCCGACAATCACAATC
82 described phosphoramidase assay. 11 The excitation wavelength was 280 nm and
fluorescence emission was measured at 360 nm. All kinetic assays were performed in
duplicate at 25ºC. The substrate hydrolysis rate was determined by measuring
fluorescence increase following addition of 5-10 µl of lysates (8 ng total protein).
D. Phenotype Analysis Using Biolog™ GN2-MicroPlates
Phenotypic analyses of hinT operon mutants containing single gene deletions
were determined using Biolog™ GN2-MicroPlates (Biolog Inc., Hayward, CA). Mutants
were inoculated onto R2A agar plates containing, per liter, 0.5 g casamine acids, 0.5 g of
D-glucose, 0.3 g sodium pyruvate, 0.3 g K 2HPO 4, 0.05 g MgSO 4·7H 2O, 0.5 g proteose
peptone, 0.5 g soluble starch, 0.5 g yeast extract, and 15 g agar and incubated overnight
at 37°C. Cells were swabbed from the surface of the agar plate, suspended in GN/GP-IF
inoculating fluid (Biolog Inc., Hayward, CA) to a final OD 600nm of 0.7, and inoculated
into the GN2-MicroPlate (150 µl per well). The microplates were incubated at 30°C for
24 h, and absorbance values at 540 nm and 630 nm were determined using an EL808
Ultra Microplate reader (Bio-Tek Instruments Inc., Winooski, VT).
E. Carbon Source Utilization Assays
Overnight cultures of wild-type and mutant E. coli strains were grown at 37°C,
with shaking at 200 rpm, in LB medium supplemented with kanamycin (50 µg/ml) as
needed. Cell pellets were obtained by centrifugation at 6000 g at 4°C for 10 min. Pellets
were washed twice and re-suspended in minimal medium containing (per liter): 6.8 g
Na 2HPO 4, 3.0 g KH 2PO 4, 0.5 g NaCl, 1.0 g NH 4Cl, 0.001 g CaC1 2, 0.001 g MgSO 4, and
0.004 g thiamine (pH 7.0). Carbon sources, D,L-alanine, glycerol, or glucose, were
added, after autoclaving the medium, to a final concentration of 20 mM when needed.
83
F. Induction of D-amino Acid Dehydrogenase Activity and Enzyme Assays
Induction of D-amino acid dehydrogenase in wild-type and mutant strains was achieved by adding 100 mM D,L-alanine to minimal medium containing 10% (vol/vol) glycerol as the sole carbon source. Cultures were grown at 37°C for 24 h prior to collection of cell pellets. D-amino acid dehydrogenase activity was measured by following the reduction of 2,6-dichlorophenol-indophenol (DCPIP) at 600 nm as previously described. 82 Reaction mixtures contained 50 mM potassium phosphate
buffer (pH 7.0), 10 mM KCN, 0.5 mM phenazine methosulfate, 10 mM DCPIP, and 100
mM of D,L-alanine. The reduction of absorbance at 600 nm was measured for 5 min
using Varian Cary 50 UV-visible spectrophotometer (Palo Alto. CA, USA). The amount
of DCPIP reduced was determined using a molar extinction coefficient of 21,500. All
assays were produced in triplicate, and rates were determined from the first 2 min of
activity. Activity is expressed as nmoles DCPIP reduced/min/mg protein, and rates
between replicates varied by < 5%.
G. Preparation of Membrane Fractions
Wild-type and mutant strains were grown to late exponential phase (OD 600nm =
0.9-1.1) in minimal medium containing 100 mM D,L-alanine and 10% (vol/vol) glycerol,
harvested by centrifugation at 10,000g, and washed once in 0.85% NaCl. The pellets
were stored at -80°C. Cell pellets were thawed, suspended in MOPS Buffer (10 mM
MOPS (pH 7.5), 20 mM MgCl 2, and 10% glycerol) and passed twice through a French
Pressure Cell at 11,000 psi. Cell lysates were centrifuged for 10 min at 14,500g and the
resulting supernatant was centrifuged for 80 min at 37,000g in a Beckman
84
Ultracentrifuge. The membrane fractions were obtained as previously described. 83
Aliquots of membrane fractions were stored at -80°C and used only once after thawing.
H. Reverse Transcription-PCR
Total RNA was isolated from E. coli strains using Qiagen RNeasy minikit
(Qiagen, Valencia, CA) according to the manufacturer’s protocol. RNA concentration was determined spectrophotometrically at 260 nm and RNA quality was confirmed by electrophoresis of 1 µg of total RNA from each preparation on a 1.2 % denatured agarose
gel. The following primers were used for RT-PCR analysis: LP
5’GTTTCGATAACCGCATTCGT3’ and RP 5’CCGGTATTCAGCCACAGATT3’. The
RT-PCR conditions were adapted from Qiagen OneStep RT-PCR Kit handbook and
were as follows: 50°C for 30 min, 95°C for 15 min followed by 30 cycles of 94°C for 0.5
min, 50°C for 0.5 min, 72°C for 0.5 min followed by a single 10-min cycle at 72°C for
extension.
I. Site-directed Mutagenesis
All hinT mutants were generated from E. coli hinT -pSGA02, the expression
vector harboring the gene encoding wild-type hinT , by using the Quick-Change
mutagenesis kit (Stratagene, Santa Clara, CA) following the manufacturer’s protocol.
Mutagenesis of ecHinT and hHint1 to produce ec /Hs and Hs/ ec chimeric proteins was
described previously. 63
85
J. Phenotype Complementation Studies
The preparation of competent cells was achieved using a Z-Competent E. coli
Transformation Kit and Buffer Set (Zymo Research Corporation, Orange, CA). The
∆hinT mutant was transformed with 20 ng of plasmid DNA encoding either the wild-type
hinT , H101A, H101G , ∆114-119, ∆117-119, L114A, H116A, K117A, L119A, Hs/ ec chimera or ec /Hs chimera. Cultures were grown in LB medium, induced with IPTG (500
µM) when the culture reached OD 600nm = 0.4-0.7, and incubated for an additional 10 h.
Cell pellets were obtained by centrifugation at 6000g for 10 min at 4ºC, washed twice in
M9 minimal medium 81 , and re-suspended in 200 µl of the same medium. A 10-20 µl
aliquot of the cell suspension was inoculated into M9 medium supplemented with 20 mM
D,L-alanine or glucose, with ampicillin (100 µg/ml) and kanamycin (50 µg/ml), to an
initial OD 600nm = 0.06 - 0.08. Cells were incubated at 37ºC and OD 600nm was measured after for 36 - 40 h of growth.
K. General Synthetic Procedures and Materials (Compounds’ synthesis and characterization were carried out by Dr. Brahma Ghosh)
All reagents were purchased from commercial vendors and used without further purification. Anhydrous pyridine was used from previously unopened SureSeal or
AcrosSeal bottles. Analytical thin layer chromatorgraphy (TLC) was performed on 0.25 mm precoated Merck silica gel (SiO 2) 60 F254. Column chromatography was performed on Purasil 60A silica gel, 230–400 mesh (Whatman). 1H and 31 P NMR were recorded on a Varian Mercury-300 or a Varian MR 400 spectrometer. Chemical shifts are reported in ppm relative to residual dimethyl sulfoxide (DMSO-d6) peak. High-resolution mass spectrometry (HRMS) data were obtained on a Biotof II (Bruker) ESI-MS spectrometer.
86
L. Synthesis of 2’,3’-Isopropylidine-5’-O-(4-Chlorophenoxy)Carbonyl Guanosine
(Guanosine -5’-Carbonate Acetonide, 1)
Guanosine-2’,3’- acetonide (0.231 mmol) was dissolved in anhydrous pyridine (7 ml) in a 2-neck round-bottomed flask equipped with an Ar. inlet and a microsyringe. The stirred solution was chilled over a dry ice-acetone bath and 4-chlorophenyl chloroformate
(0.278 mmol, 1.20 eq) was added dropwise under Ar. The cooling bath was removed and the solution was slowly allowed to return to room temperature. Stirring was continued until thin layer chromatographic and ESI-MS analyses showed that all of the starting material had disappeared (after approximately 4 h). Pyridine was evaporated in vacuo by
co-distillation with n-heptane and the crude solid obtained was purified by
chromatography on SiO 2 gel to isolate the guanosine-5’-carbonate as a while solid with a
1 90% yield. The H NMR spectrum was (DMSO-d6): 1.30 (s, 3H, -CH 3), 1.56 (s, 3H, -
CH 3), 4.28- 4.58 (m, 3H, C2’+C3’+ C4’-H), 5.22 (s, 2H, 5’-CH 2), 6.22 (s, 1H, C1’-H),
6.58 (s, 2H, NH 2), 7.22 (d, 2H, -Ph), 7.50 (d, 2H, -Ph), 7.86 (s, 1H, C8-H), 10.74 (s, 1H,
N1-H) ppm.
M. Synthesis of 2’,3’-Isopropylidine-5’-O-[(3-Indolyl)-1-Ethyl]Carbamoyl
Guanosine (Guanosine -5’-Carbamate Acetonide, 2)
A mixture of guanosine-5’-carbonate (0.255 mmol) and tryptamine (0.640 mmol,
2.5 eq) in anhydrous THF (10 ml) was heated to 60ºC (oil bath) under Ar. and stirred at that temperature until TLC showed complete consumption of the starting material (4 h).
The solvent was removed by rotary evaporation and the residue chromatographed on
SiO 2 gel using CH 2Cl 2:MeOH:H 2O (5:2:0.25) to give the title carbamate as white
87 amorphous solid (yield 90%). The 1H NMR spectrum was (DMSO-d6): 1.32 (s, 3H, -
CH 3), 1.58 (s, 3H, -CH 3), 2.80 (t, 2H, -CH 2-), 3.24 (q, 2H, -CH2 NHCO), 4.00- 4.30 (m,
3H, 5’-CH 2+ C4’-H), 5.10 (d, 1H, 5’-C3’-H), 5.24 (d, 1H, C2’-H), 6.00 (s, 1H, C1’-H),
6.58 (s, 2H, NH 2), 6.96 (t, 1H, indole), 7.12 (t, 1H, -indole), 7.18 (s, 1H, indole), 7.36 (d,
1H, indole), 7.40 (t, 1H, -NH), 7.50 (d, 1H, indole), 7.84 (s, 1H, C8-H), 10.72 (s, 1H, N1-
H), 10.82 (s, 1H, NH-indole) ppm.
N. Synthesis of 5’-O-[(3-Indolyl)-1-Ethyl]Carbamoyl Guanosine (Guanosine-5’-
Tryptamine Carbamate, TpGc)
The above 2’, 3’acetonide-protected guanosine carbamate (100 mg) was dissolved in TFA:H 2O (4:1, 5 ml) and stirred at room temperature until TLC and ESI-MS showed no starting material left (60 min). The mixture was stripped to dryness and the residue purified by SiO 2 gel chromatography using CH 2Cl 2:MeOH:H 2O (5:3:0.5) as eluant. The
relevant UV-active fractions were collected and concentrated to afford the final
compound as a white solid (yield 40%). The 1H NMR spectrum was (DMSO-d6): 2.38 (t,
2H, -CH 2-), 3.17 (q, 2H, -CH2NHCO), 4.23- 4.33 (m, 2H, 5’-CH 2), 4.56 (t, 1H, C4’-H)
5.14 (d, 1H, C3’-H), 5.39 (d, 1H, C2’-H), 6.10 (s, 1H, C1’-H), 6.38 (s, 2H, NH 2), 6.96 (t,
1H, indole), 7.15 (t, 1H, -indole), 7.21 (s, 1H, indole), 7.32 (d, 1H, indole), 7.40 (t, 1H, -
NH), 7.48 (d, 1H, indole), 7.80 (s, 1H, C8-H), 10.65 (s, 1H, N1-H), 10.78 (s, 1H, NH- indole) ppm [OHs exchanging with H 2O in NMR solvent] .
88
III. Results and Discussion
A. Operon Structure of Genes Associated with hinT
In E. coli K-12 strains, the hinT gene is located adjacent to ycfL , ycfM , ycfN (thiK ), ycfO (nagZ ), and ycfP . PCR analyses were performed to determine if the deletion/insertion mutations occurred in frame without affecting the operon structure.
Results in Figure 1 show that the operon structure is still intact after deletion of each gene and insertion of a kan resistant gene. Further evidence of hinT deletion was obtained by measuring the phosphoramidase activity in cell-free lysates. Figure 2 shows that the hinT deletion mutant ( ∆hinT ) did not have appreciable phosphoramidase activity that is associated with the expression of hinT in wild-type E. coli BW25113.
B. Phenotype of ecHinT
Despite the fact that Hint is present in E. coli and other prokaryotes, the function of this conserved group of proteins in bacteria has remained largely unknown. To ascertain the role of ecHinT in microbial metabolism, we obtained a series of mutants from the Keio collection containing deletions in the hinT (ycfF ), ycfL , ycfM , ycfN (thiK ), ycfO (nagZ ), and ycfP genes. Biolog™ plates were used for phenotypic screening to estimate the differential metabolic potential between wild-type E. coli BW25113 and the hinT deletion mutant ( ∆hinT ). Results from Biolog™ substrate utilization studies (Fig. 3) indicated that metabolic profiles of wild-type E. coli BW25113 and the ∆hinT mutant
were highly similar, with the exception of the utilization of D- and L-alanine.
89
Figure 1. Sequence verification PCR (representative gel)
MW 1 2 3 4 marker
Lane1: hinT (annealing Temp 38ºC)
Lane2: hinT (annealing Temp 40ºC)
Lane3: hinT (annealing Temp 43ºC)
Lane4: hinT (annealing Temp 43ºC)
Expected fragment size= 1.15 hinT size (0.36Kb) + Kan gen (0.795)
90
Figure 2. Phosphoramidase assay to detect ecHinT activity in E. coli cell-free lysates.
91
Figure 3. Summary of metabolic fingerprints. Dataset of the GN2 plate assay of wild- type E. coli BW25113 and ∆hinT mutant strains was processed as described in experimental procedures. The overlay represents the difference between the ability of the two strains to use 95 carbon sources. This is a representative overlay of three independent experiments.
92
The initial screening results were verified and confirmed by culturing wild-type and mutant strains in minimal media containing D,L-alanine (Fig. 4). These results were highly reproducible, even when a starvation phase was introduced prior to culturing.
To further define the role of genes in the apparent hinT operon in alanine
metabolism, E. coli strains from the Keio collection with verified deletions in ycfL
(JW1090), ycfM (JW5157), ycfN (JW1092), ycfO (JW1093), and ycfP (JW5158) were
evaluated for carbon source utilization. Compared to the wild-type BW25113 strain,
none of the tested mutants showed significant difference in alanine catabolism when
tested using Biolog™ plates and test tube cultures.
C. D-Amino Acid Dehydrogenase Transcription and Activity Testing
Inspection of the metabolic fingerprints of wild-type BW25113 and the ∆hinT mutant revealed that while the ∆hinT mutant failed to grow on both D and L isomers of
alanine, it did grow on serine and glycine. Based on these results, and what is known
about the alanine catabolic pathway in E. coli , we hypothesized that the ∆hinT mutant
was neither deficient in the amino acid transporter system for D,L-alanine nor in alanine
racemase, but most likely in D-amino acid dehydrogenase activity encoded by dadA (Fig.
5). The inducible dadA encodes for the broad specificity bacterial D-amino-acid
dehydrogenase. DadA catalyzes the oxidation of D-amino acids, including alanine, into
their corresponding ketoacid. 84 The bacterial D-amino-acid dehydrogenase is a
heterodimer complex formed by a 45-kDa and a 55-kDa subunit. 85 To avoid overproduction of reactive oxygen species that could damage cellular components, D- amino acid oxidation in bacteria is not coupled to O 2 reduction and the two electrons from the substrate are received by FAD in the small subunit before being transferred to a
93
Figure 4. Bacterial growth curves of wild-type E. coli BW25113 and ∆hinT mutant in
M9 medium in the presence of either 20 mM glucose or D,L-alanine. Measurements were done in duplicate and variants were less than 5% between two independent cultures.
94
Figure 5. Alanine transport and metabolism in E. coli.
95 sulfur-iron center in the large subunit. 85 Bacterial D-amino-acid dehydrogenase appears to have two main functions in bacteria. Initially, it allows growth of bacteria using D- amino acids as the sole carbon, nitrogen, and energy source. 86; 87 Additionally, it prevents the accumulation of D-amino acids in cellular compartments, as some D-amino acid analogues have specific inhibitory effects on bacterial growth. 88; 89 In bacteria, D-
amino acids are used to stabilize the peptidoglycan structure 90 , and have recently been shown to regulate biofilm disorganization 91 and are thought to play a role in quorum sensing.
The inability of the ∆hinT mutant to grow on D,L-alanine can be explained by either failure to induce transcription of the D-amino-acid dehydrogenase or the presence of a defective D-amino-acid dehydrogenase that is incapable of metabolizing D-alanine.
To test these hypotheses, RT-PCR was performed using primers for dadA and an
equivalent amounts of mRNA obtained from wild-type BW25113 or the ∆hinT mutant
grown in the presence of D,L-alanine. Results in Figure 6 show that equivalent amounts
of dadA transcript were produced by either the wild-type strain or the ∆hinT mutant. To
determine if the inability of the ∆hinT mutant to grow on D-alanine was due to a lack of
DadA enzyme activity, membrane fractions were isolated and analyzed for D-amino acid
dehydrogenase activity. Results showed that the ∆hinT mutant possessed 38-fold less D-
amino acid dehydrogenase activity (< 5 nmol/min/mg) relative to the wild-type
BW25113 strain (190 nmol/min/mg).
96
Figure 6. RT-PCR was performed with equivalent amounts of mRNA (100 ng) obtained from wild-type E. coli BW25113, ∆hinT, and ∆ycfL mutants using dadA primers under the described conditions. 1 µg of each PCR reaction was loaded on 1% agarose gel.
97
In this study, we report the first evidence of a regulatory relationship between the evolutionary conserved enzymes; DadA and ecHinT. Since the transcripts encoding for
DadA were produced in the ∆hinT mutant, our results suggested that D-amino acid dehydrogenase activity in E. coli is likely modulated by ecHinT in a post-translational
manner.
D. Phenotype Rescue by ecHinT
Transformation of a plasmid encoding the wild-type hinT into the ∆hinT mutant
(JW1089) allowed rescue of the mutant for growth on D,L-alanine to a level of growth
comparable to wild-type BW25113 grown under the same conditions (Fig. 7). This
indicated that other mutations in strain JW1089 were not involved in the inability of the
∆hinT mutant to grow on D,L-alanine.
A mutational approach was used to fully examine the relationship between ecHinT structure and activity, and DadA activity. Two active-site mutants were prepared and characterized as described previously. 8 ecHinT His-101 residue, which was
previously suggested to be the key catalytic residue for enzymatic activity 8, was
exchanged by site-directed mutagenesis into either alanine (H101A) or glycine (H101G).
Enzymatic activities of these mutants were previously quantified by 31 P-NMR using
adenosine-5´-monophosphoramidate (AMP-NH 2) as a substrate and the catalytic efficiency was found to be 3,0000-fold lower than that of wild-type ecHinT. 8 Plasmids
encoding the catalytically impaired mutants of ecHinT, His-101A or His-101G, failed to
complement the ∆hinT mutant for growth on D,L-alanine. This indicated that the catalytically active ecHinT is needed for DadA activity (Fig. 7).
98
Figure 7. ecHinT structural and activity requirement for phenotype rescue. Each column represents the growth of the ∆hinT mutant that was transformed with a plasmid containing either hinT , a mutant hinT , ec /Hs chimera, Hs/ ec chimera, or hHint1 , except for the first column of wild-type E. coli BW25113. OD 600 values were determined 48 h
after inoculation into M9 medium in presence of 20 mM D,L-alanine.
99
E. Design, Synthesis and Characterization of ecHinT Inhibitor
Tryptamine guanosine carbamate (TpGc) was prepared according to synthetic
Scheme 1. The rationale for designing such an inhibitor was based on the structural activity relationship of Hint substrates that was reported previously 11 , which found that the phosphoramidate substrate tryptamine 5’-guanosine monophosphate (TpGd) (Fig. 8a)
-1 5 - was an excellent substrate with a kcat value of 4.0 s and a k cat /K m value of 7.0 x 10 M
1s-1. 11 It was hypothesized that substitution of the hydrolysable phosphoramidate moiety with a carbamate linker (TpGc) would result in an inhibitor of ecHinT and not a substrate
(Fig. 8b). In addition, the incorporation of guanosine would insure that potential off- target inhibition of Tryptophan tRNA synthetase would be minimal.
Using the previously described spectrophotometric phosphoramidase assay 11 , the
ecHinT Ki value for TpGc was determined. Although designed to be a competitive
inhibitor, TpGc was found to be a non-competitive inhibitor of ecHinT with a Ki value of
42 µM (Fig. 8c). Consistent with our previous finding, that ecHinT catalytic activity is
required for the phenotype, when compared to the growth of ∆hinT E. coli strain on D,L-
alanine, cultures of wild-type BW25113 grown in the presence of D,L-alanine and 100
µM of TpGc exhibited a similarly diminished growth capacity (Fig. 8d).
100
Scheme 1. General synthetic scheme of TpGc.
O O
N N NH Cl O NH N N NH a N N NH HO 2 O O 2 O O
O O O O
1
b
O O N N O NH O NH NH NH c N N N NH2 N NH2 N O N O H O H O O O OH OH TpGc 2
a) p-Cl phenyl chloroformate, pyridine, RT, 4h, 65% b) tryptamine, anhyd THF, 60°C, 4h, 90% c) TFA:H2O (4:1), RT, 60 min, 40%
101
Figure 8. Inhibition of phenotype rescue by ecHinT inhibitor.
a) The structure of tryptamine 5’-guanosine monophosphate (TpGd).
O
N NH O H N N NH N P O O 2 HN O- OH OH
TpGd
102 b) The structure of tryptamine guanosine carbamate (TpGc).
O
N NH O
N N NH N O O 2 HN H
OH OH
TpGc
103 c) TpGc exhibits a non-competitive inhibition profile.
104 d) Wild-type E. coli BW25113 was grown in M9 medium supplemented with either 20 mM glucose (blue bar) or 20 mM D,L-alanine (brown bar) in the presence or absence of
100 µM TpGc. OD 600 values were determined after 48 h.
105
To probe structural elements of ecHinT that are determinant for the discovered
phenotype, non-active-site deletion mutations that reside in the C-terminus region of
ecHinT were prepared and characterized as described previously. 64 Recently, we
investigated the C-terminus region of ecHinT protein and found that this loop
accommodates several conformational forms in the 1.3°A crystal structure (Fig. 9). 64
We designed, and fully characterized, two deletion mutants of ecHinT C-terminus in which three or six amino acids were deleted from the 11 amino acids C-terminal loop.
The two deletion mutants were equivalent in their secondary structure to wild-type ecHinT and preserved the ability to form homodimers. The catalytic efficiencies of the deletion mutants were within one order of magnitude lower than that for the wild-type ecHinT. Despite the partial activity possessed by these mutants, both mutants failed to rescue ∆hinT when growing on D,L-alanine (Fig. 7).
To further evaluate the impact of the C-terminal loop on the observed phenotype and in an effort to determine if the phenotype rescue is dependent on specific residues in the C-terminus of ecHinT, four alanine mutants of the C-terminal loop, L114A, H116A,
K117A, and L119A were designed and characterized as previously described in Chapter
Two. All of the alanine mutants retained partial activities and were structurally equivalent to wild-type ecHinT when studied by CD spectroscopy. The ability each of these mutants to rescue ∆hinT when growing on D,L-alanine was studied as shown in Figure 10. All the alanine mutants were able to rescue the phenotype, suggesting that the length rather than the sequence of the C-terminus is what determines the ability of ecHinT to function as a regulator of DadA activity.
106
Figure 9. Superposition of the eight independent monomers observed in the structure of
the ecHinT-GMP complex.
107
Figure 10. phenotype rescue by the alanine scan mutants. Each column represents the
growth of the ∆hinT mutant that was transformed with a plasmid containing wild-type ecHint, L114A, H116A, K116A, or L119A mutant except for the first column of wild- type E. coli BW25113. OD 600 values were determined 48 h after inoculation into M9
medium in presence of 20 mM D,L-alanine.
108
Although both the N and C-terminal loops of ecHinT can adopt more than one
conformation in the ecHinT-GMP complex structure 64 , the overall fold of the structure is similar to hHint1. 64 To investigate the possibility of phenotype rescue by the human version of Hint protein (hHint1), which is 48% homologous to ecHinT, the E. coli ∆hinT mutant was transformed with a plasmid encoding hHint1 and the ability to grow on D,L- alanine was examined. The hHint1 failed to rescue the growth phenotype suggesting that certain structural characteristics, in addition to catalytic activity, govern the regulation of
DadA activity by ecHinT. Based on our finding that the C-terminus loop of ecHinT plays a significant role in the observed relationship between ecHinT and DadA activity, the ability of two chimeric proteins to rescue ∆hinT cells was studied. A hHint1-ecHinT
chimera (Hs/ ec ), in which the C-terminus of hHint1 was replaced with ecHinT C-
terminus, was unable to rescue the D-alanine growth defect exhibited by E. coli ∆hinT cells. However, and in contrast, a ecHinT-hHint1 chimera ( ec /Hs), in which the C- terminus of ecHinT was replaced with the hHint1 C-terminus, had only a slightly reduced ability to complement the phenotype when compared to wild-type ecHinT (Fig. 7).
Taken together, these results suggest that other inherent structural requirements, in addition to the C-terminus, are critical to the ecHinT-mediated DadA activation.
Ongoing studies should shed light on the post-translational regulatory relationship between ecHinT and DadA and whether this regulation influences the localization, dimerization and/or degradation of DadA.
In summary, by using a series of mutagenic and chemical biological studies, ecHinT has been shown to be necessary for the growth of E. coli on D-alanine by regulating the catalytic activity of a key metabolizing enzyme, DadA. The role of ecHinT
109 has been shown to be dependent on the catalytic activity of ecHinT, components of its C- terminus domain and potentially unidentified structural factors. This marks the first example of a direct relationship between the catalytic activity of a Hint protein and a physiological function. Interestingly, the characteristics of ecHinT that appear necessary for the discovered phenotype are also required for the hydrolysis of LysU-generated lysyl-AMP by ecHinT. 32; 64 Whether this observation is a coincidence or evidence of a link between the ecHinT regulation of DadA and its interaction with LysRS remains to be determined.
110
Chapter Four
Synthesis and Evaluation of Potential Inhibitors of Human and Escherichia coli Histidine Triad Nucleotide Binding Proteins
111
I. Introduction
Histidine triad nucleotide binding protein (Hint) is named after the highly
conserved motif related to the sequence His-X-His-X-His-XX in the active-site, where X
is a hydrophobic amino acid. Hint is considered to be the ancestor of the histidine triad
protein superfamily and can be found in both prokaryotes and eukaryotes. 6 While
eukaryotes generally express multiple forms of Hint (Hint1, Hint2 and Hint3), prokaryote
genomes, including Gram-negative and Gram-positive bacteria, typically encode one
Hint gene. 2
Recently, evidence has begun to emerge that Hint1 may function as a tumor suppressor through multiple molecular mechanisms involving triggering apoptosis 51 ,
regulating gene transcription and cell cycle modulation. 34; 66 Hint1 deleted mice studies
have revealed that an increased susceptibility to the carcinogenic DMBA induced ovarian
and mammary tumors, as well as an increased occurrence of spontaneous tumors. 49
Eukaryotic Hint1 has been shown to associate with, and suppress the β-catenin Wnt signaling pathway transcriptional activity by direct interactions with Reptin and Pontin. 48
Hint1 has also been associated with transcription factors such as, TFIIH 46 , MITF 34; 92 ,
and USF2. 66 In addition, the interactions between MITF and USF2 appear to be mediated by LysRS. 34; 66; 93 Recently, we have demonstrated that lysyl-adenylate (lysyl-
AMP) generated by lysyl-tRNA synthetases (LysRS) is also a substrate for Hints. 32
Despite such a wide variety of possible biological functions in eukaryotes, the cellular function and rationale for the evolutionary conservation of Hint in bacteria has remained a mystery. Moreover, none of the previously proposed functions has been successfully linked to the enzymatic activity of Hints. Recently, we have shown that the
112 expression of the catalytically active Hint is essential for the activity of the enzyme D- amino acid dehydrogenase (DadA) in Escherichia coli . 94 Consequently, we have chosen to synthesize novel Hint1 inhibitors that can be used as chemical probes to study the connection between the newly discovered Hint1-associated phenotype in E. coli and its enzymatic activity.
Hint1 has been found to be a nucleoside phosphoramidase and acyl-AMP hydrolase. 6 Probing the Hint1 base recognition site, substrate specificity for both purine and pyrimidine phosphoramidates was determined.11 Structure–activity relationship studies revealed that Hint1 prefers purine over pyrimidine phosphoramidates. In addition, based on kinetic and crystallographic studies, the 2- and 3-hydroxyl groups of the ribose ring were shown to be required for maximal phosphoramidase efficiency. 11
II. Experimental Procedures
(Compounds’ synthesis and characterization were carried out by Dr. Brahma Ghosh).
A. General Procedure for synthesis of 2’,3’-Isopropylidine-5’-(indole-3- propionyl)
nucleosides: (Adenosine Acetonide-5’-Indole-3-Propionate (1)): [M+ Na] 501.0283
A mixture of the appropriate 2’,3’-isopropylidine nucleoside (0.231 mmol) ,
indole-3-propionic acid (0.254 mmol, 1.1 eq), DCC (0.254 mmol, 1.1 eq), and DMAP
(0.254 mmol, 1.1 eq) was suspended in CH 2Cl 2 (10 ml) and a few drops of DMSO were
added to assist in solubility. The mixture was stirred at RT and after 24 h an additional
equivalent of DCC and DMAP was added. After 36 h, the mixture was diluted with
CH 2Cl 2 (50 ml) and washed with H 2O (3 X 10 ml). The organic layer was collected and concentrated. The residue obtained was purified by SiO 2 gel chromatography using
113
1 CH 2Cl 2:MeOH:H 2O (5:2:0.25) as eluant. H- DMSO-d6: 1.32 (s, 3H, -CH3), 1.54 (s, 3H,
-CH3), 2.54 (s, 1H), 2.64 (m, 2H, -CH2-), 2.90 (t, 2H, -CH2-), 4.12- 4.27 (ddd, 2H, 5’-
CH2), 4.35 (qn, 1H), 4.99 (m, 1H), 5.43 (dd, 1H), 6.18 (d, 1H, C1’-H), 6.95 (t, 1H, indole), 7.05 (t, 1H, indole), 7.08 (s, 1H, indole), 7.20 (d, 1H, indole), 7.34 (br s, 2H, -
NH2), 7.47 (d, 1H, indole), 8.16 (s, 1H, C2-H), 8.29 (s, 1H, C8-H), 10.77 (s, 1H, -NH indole) ppm. 13 C- DMSO-d6: 173.42, 156.59, 152.79, 148.73, 140.08, 136.90, 127.12,
121.84, 121.16, 118.41, 117.93, 114.21, 113.35, 111.09, 90.83, 84.94, 84.32, 81.48,
63.76, 34.76, 26.22, 24.40, 26.22, 24.39, 20.66 ppm.
B. General Procedure for Synthesis of 5’-(indole-3- propionyl) nucleosides
(Adenosine 5’-Indole-3-Propionate(2)) : [M+ 1] : 439.0332
The 2’, 3’acetonide-protected nucleoside 5’-ester (100 mg) was dissolved in
TFA:H 2O (4:1, 5 ml) and stirred at RT for 2 h. The mixture was evaporated to dryness and the residue purified by SiO 2 gel chromatography using CH 2Cl 2:MeOH:H 2O (5:3:0.5)
as eluant. The relevant UV-active fractions were collected and concentrated to afford the
final compounds as white solids. 1H- DMSO-d6: 2.72 (m, 2H, -CH2-), 3.00 (t, 2H, -CH2-
), 4.18- 4.42 (ddd, 2H, 5’-CH2), 4.30 (m, 2H), 4.65 (t, 1H), 6.01 (d, 1H, C1’-H), 7.00 (t,
1H, indole), 7.15 (t, 1H, indole), 7.20 (s, 1H, indole), 7.37 (d, 1H, indole), 7.57 (d, 1H,
indole), 8.20 (br s, 2H, -NH2), 8.38 (s, 1H, C2-H), 8.58 (s, 1H, C8-H), 10.82 (s, 1H, -NH
indole) ppm. 13 C- DMSO-d6: 172.48, 159.81, 150.01, 148.92, 140.74, 136.20, 126.84,
122.34, 120.97, 119.15, 118.27, 118.78, 11.95, 111.37, 87.89, 81.66, 73.14, 70.19, 63.82,
34.31, 20.16 ppm.
114
C. General Procedure for Synthesis of 5’-O-(4-chlorophenoxy) carbonyl 3’-azido-3’- deoxy thymidine (AZT-5’-carbonate (3) ): (M+1) 422.0207
AZT (0.231 mmol) was dissolved in anhydrous pyridine (7 ml) in a 2-neck round- bottomed flask equipped with an Ar. inlet and a microsyringe. The stirred solution was chilled using a dry ice-acetone bath and 4-chlorophenyl chloroformate (0.278 mmol, 1.20 eq) was added dropwise to it under Ar. The cooling bath was then removed and the solution slowly allowed to return to RT and stirred for an additional 3h. Pyridine was evaporated in vacuo by co-distillation with n-heptane. The crude solid obtained was
1 purified by chromatography on SiO 2 gel to afford the title carbonate as a white solid. H-
DMSO-d6: 1.76 (s, 3H, -CH3), 4.14 (m, 1H, C’-H), 4.4- 4.6 (m, 3H, 5’-CH2 + C4’-H),
6.18 (t, 1H, C1’-H), 7.24(d, 2H, -Ph), 7.50 (d, 2H, -Ph), 7.56 (s, 1H, C6-H), 11.38 (s, 1H,
N3-H) ppm.
D. General Procedure for Synthesis of 5’-O-[(3-indolyl)-1-ethyl]carbamoyl 3’-azido-
3’-deoxy thymidine (AZT-5’-carbamate (4)): (M-1] 452.1490
A mixture of the foregoing AZT-5’-carbonate (0.255 mmol) and tryptamine
(0.640 mmol, 2.5 eq) in anhydrous THF (10 ml) was heated to 60ºC (oil bath) under Ar.
Stirring was continued at that temperature for 2.5 h at which point TLC showed complete
disappearance of starting material. The solvent was removed by rotary evaporation and
the residue chromatographed on SiO 2 gel using CH 2Cl 2:MeOH:H 2O (5:2:0.25) to give the
desired carbamate as a white solid. 1H- DMSO-d6: 1.78 (s, 3H, -CH3), 2.30 (m, 1H), 2.46
(q, 1H), 2.84 (t, 2H, -CH2-), 3.30 (q, 2H, -CH2NH-), 4.12- 4.26 (ddd, 2H, 5’-CH2), 4.44
(m, 1H), 6.13 (t, 1H, C1’-H), 6.97 (t, 1H, indole), 7.05 (t, 1H, -indole), 7.14 (s, 1H,
115 indole), 7.33 (d, 1H, indole), 7.45 (t, 2H, -NH + C6-H), 7.51 (d, 1H, indole), 10.80 (s,
1H, NH-indole) ppm.
E. General Procedure for Synthesis of 5’-O-[(3-indolyl)-1-ethyl]carbamoyl 3’-amino-
3’-deoxy thymidine (AZT-3’-NH2-5’-carbamate(5)) : [M+1]: 428.2349
To a solution of the above AZT-5’-carbonate in MeOH (5 ml) was carefully added 10% Pd/C (10% by weight of the starting material). The reaction vessel was evacuated, filled with H 2, and stirred at RT. After 16h the reaction was deemed was complete by ESI-MS analysis. The mixture was diluted 15 ml of MeOH and the catalyst filtered off. The filtrate was concentrated under vacuum and the residue chromatographed
1 on a SiO 2 gel column using CH 2Cl 2:MeOH:H 2O (5:3:0.5) to elute the product. H-
DMSO-d6: 1.76 (s, 3H, -CH3), 1.98 (m, 1H), 2.14 (m, 1H), 2.81 (t, 2H, -CH2-), 3.26 (q,
2H, -CH2NH-), 3.36 (m, 1H), 3.69 (m, 1H), 4.08- 4.23 (ddd, 2H, 5’-CH2), 4.44 (m, 1H)
6.13 (t, 1H, C1’-H), 6.95 (t, 1H, indole), 7.04 (t, 1H, -indole), 7.12 (s, 1H, indole), 7.30
(d, 1H, indole), 7.36 (t, 1H, -NHCO), 7.42 (s, 1H, C6-H), 7.48 (d, 1H, indole), 10.79 (s,
1H, NH-indole) ppm. 13 C- DMSO-d6: 163.94, 156.28, 150.61, 136.42, 127.39, 122.81,
121.09, 118.41, 111.75, 111.56, 109.81, 83.67, 79.40, 64.40, 52.10, 41.39, 25.71, 12.36
ppm.
F. General Procedure for Synthesis of 2’,3’-Isopropylidine-5’-O-(4-
chlorophenoxy)carbonyl nucleosides (Adenosine-5’-carbonate acetonide (6)): (M+1)
462.1246
Adenosine-2’,3’- acetonide (0.231 mmol) was dissolved in anhydrous pyridine (7
ml) in a 2-neck round-bottomed flask equipped with an Ar. inlet and a microsyringe. The
116 stirred solution was chilled over a dry ice-acetone bath and 4-chlorophenyl chloroformate
(0.278 mmol, 1.20 eq) was added dropwise to it under Ar. The cooling bath was then removed and the solution slowly allowed to return to RT. Stirring was continued until
TLC showed all of the starting material had disappeared (4 h, monitored by TLC and
ESI-MS). Pyridine was evaporated in vacuo by co-distillation with n-heptane and the crude solid obtained was purified by chromatography on SiO 2 gel to afford the title carbonated as a white solid. 1H- DMSO-d6: 1.30 (s, 3H, -CH3), 1.58 (s, 3H, -CH3), 4.3-
4.6 (m, 3H, 5’-CH2 + C4’-H), 5.16 (d, 1H, C3’), 5.42 (d, 1H, C2’), 6.21 (s, 1H, C1’-H),
7.20 (d, 2H, -Ph), 7.40 (s, 2H, -NH2), 7.50 (d, 2H, -Ph), 8.18 (s, 1H, C2-H), 8.36 (s, 1H,
C8-H) ppm.
G. General Procedure for Synthesis of 2’,3’-Isopropylidine-5’-O-[(3-indolyl)-1- ethyl]carbamoyl nucleosides (Adenosine 5’-tryptamine-carbamate acetonide (7)):
(M+1) 494.2723
A mixture of the foregoing nucleoside-5’-carbonate (0.255 mmol) and tryptamine
(0.640 mmol, 2.5 eq) in anhydrous THF (10 mL) was heated to 60ºC (oil bath) under Ar. and stirred at that temperature until TLC showed complete consumption of the starting material (12 h). The solvent was removed by rotary evaporation and the residue chromatographed on SiO 2 gel using CH 2Cl 2:MeOH:H 2O (5:2:0.25) to give the appropriate carbamate and white amorphous solids. 1H- DMSO-d6: 1.34 (s, 3H, -CH3),
1.58 (s, 3H, -CH3), 2.80 (t, 2H, -CH2-), 3.20 (q, 2H, -CH2 NHCO), 4.0- 4.40 (m, 3H, 5’-
CH2 + C4’-H), 5.00 (d, 1H, C3’), 5.42 (d, 1H, C2’), 6.18 (s, 1H, C1’-H), 6.90- 7.54 (7H,
indole + -NH2), 8.18 (s, 1H, C2-H), 8.35 (s, 1H, C8-H), 10.81 (s, 1H, -NH indole) ppm.
117
H. General Procedure for Synthesis of 5’-O-[(3-indolyl)-1-ethyl]carbamoyl nucleosides (Adenosine 5’-tryptamine-carbamate (8)): (M+1) 454.0683
The 2’, 3’acetonide-protected nucleoside carbamate (100 mg) was dissolved in
TFA:H 2O (4:1, 5 ml) and stirred at RT until TLC and ESI-MS showed no starting material left (30 min). The mixture was stripped to dryness and the residue purified by
SiO 2 gel chromatography using CH 2Cl 2:MeOH:H 2O (5:3:0.5) as eluant. The relevant UV- active fractions were collected and concentrated to afford the final compounds as white solids. 1H- DMSO-d6: 2.80 (t, 2H, -CH2-), 3.24 (q, 2H, -CH2 NHCO) 4.00 (m, 2H, C4’-
H, + -OH) 4.25 (m, 2H, 5’-CH2), 4.65 (m, 1H, -OH), 5.38 (d, 1H, C3’), 5.52 (d, 1H,
C2’), 5.90 (s, 1H, C1’-H), 6.90- 7.58 (7H, indole + -NH2), 8.16 (s, 1H, C2-H), 8.38 (s,
1H, C8-H), 10.80 (s, 1H, -NH indole) ppm.
I. Phosphoramidase Assay
In the absence of inhibitors, turnover rates were measured by following the
hydrolysis of the fluorogenic substrate tryptamine 5’-adenosine phosphoramidate (TpAd)
by ecHinT and hHint1 proteins as previously described. 11 Excitation wavelength was set
at 280 nm, fluorescence emission was measured at 360 nm and all the kinetic assays were
performed in triplicate at 25ºC. The rate of hydrolysis of the substrate was determined by
measuring the increase of fluorescence intensity upon the addition of the protein over two
-1 minutes. The Michaelis-Menten constants, k cat (s ) and K m (µM), were determined by a nonlinear regression analysis of the initial velocity versus concentration using JMP IN 7 software. In the presence of inhibitors, the proteins were pre-incubated with the tested inhibitor for 30 s, at 25ºC, before the addition of the substrate. The reaction was followed under the same condition as of in the absence of inhibitors.
118
J. Phenotype Testing
Bacterial strain of wild-type Keio (BW25113) was obtained from the Genomic
Center at Yale University. After growing cultures of wild-type in LB media at 37°C
overnight, cultures were centrifuged at 6000g for 10 min at 4ºC. Cells were re-suspended
and washed twice with M9 medium before inoculation. Cells were grown under
starvation in M9 medium for 24 h before testing. To test the ability of the wild-type strain
to grow on D,L-alanine in the presence and absence of the most potent inhibitors, starved
cells were transferred into M9 medium supplemented with either 22 mM D,L-alanine or
10 mM glucose. Cultures were grown at 37°C for 40 h before OD 600 was measured.
III. Results and Discussion
A. Inhibitors’ Synthesis and Evaluation
Based on the previous findings and knowledge of the general outline of substrate specificity requirements, compounds 1 and 2 (Fig. 1) were designed and synthesized following Scheme 1. Both compounds were titrated into the phosphoramidase assay using the fluorogenic substrate tryptamine 5’-adenosine phosphoramidate (TpAd) (Fig.
1a) as described previously 11 and in the experimental procedures section.
While Hints did not show any appreciable esterase activity against compound 1
and 2, kinetic analysis revealed a significant and dose-dependent decrease in the apparent
Vmax of Hint1 in the presence of either compound 1 or 2 (Fig. 2). In contrast, the K m value for TpAd was unaltered by the presence of the inhibitor. Thus, the mechanism of inhibition was clearly of a noncompetitive type, suggesting that the inhibitor and TpAd
119 substrate do not share a common binding site on the enzyme. Ki values were estimated from the Dixon plot and are summarized in Table 1.
A second series of carbamate analogues was prepared following Schemes 2 and 3.
As shown in Table 1, compound 3, a thymidine carbamate analogue in which the 3- hydroxyl group of the ribose ring was replaced with a primary amine group, inhibited
Hint with a Ki value of 122 µM for ecHinT and 139 µM for hHint1. Replacement of the amine group with an azido moiety at the 3’ position, compound 4, resulted in an inhibitor with a Ki value of 565 µM for ecHinT and 715 µM for hHint1. The 5-fold enhancement in equilibrium binding conferred by the amine group at the 3’ position is most likely explained by the formation of favorable hydrogen bonding interactions between the inhibitor and its binding site in the protein structure. Adenosine carbamate, compound 5, and guanosine carbamate 94 , compound 6, were the most potent inhibitors prepared in this
study. As shown in Table 1, Ki values for the adenosine carbamate analogue ranged
between 73 µM for ecHinT and 103 µM for hHint1 while the Ki values for the guanosine carbamate analogue ranged between 42 µM for ecHinT and 34 µM for hHint1. All compounds exhibited non-competitive inhibition profiles (Fig. 2).
120
Scheme 1. General synthetic scheme of compounds 1 and 2.
H H N O N O R1 R R1 HO 1 O O a n O b n O O O O O O OH OH 2 1
a) indole-3-propionic (n =1) or -butyric (n =2 ) acid, DCC, DMAP, CHCl (0.1% DMSO), RT, 16-48h, 2 2 b) TFA:H2O (4:1), RT, 60 min, 40%
121
Figure 1. Structures of a) Tryptamine 5’-adenosine phosphoramidate and its b) ester and
c) carbamate analogues.
a) NH2 N N H O N N N P O O HN O- OH OH
TpAd
b) H O N R1 n O O
R3 R2
1 n= 1 2 n= 2
c) H N O R1 N O O H
R3 R2
122
Table 1. Inhibition constants determined in HEPES buffer (pH7.2) at 25ºC.
Compound R1 R2 R3 ecHinT hHint Ki (µM) a Ki (µM) a
1 Adenine OH OH 375 (±13) 400 (±27)
2 Adenine OH OH 393 (±17) 457 (±32)
3 Thymine H NH 2 122 (±10) 139 (±17)
4 Thymine H N3 565 (±23) 715 (±35) 5 Adenine OH OH 73 (±4) 103 (±18)
6 Guanine OH OH 42 (±6) b 34 (±4) a Values are means of three experiments, standard deviation is given in parentheses. b Value was taken from reference 94.
123
Scheme 2. General synthetic scheme of compounds 3 and 4.
O O H O NH Cl NH N NH O b O N O a N O N O HO O O N O O O H O N N N3 3 4 3 3
H O c N NH O N O N O H O 5 NH2
a) p-Cl phenyl chloroformate, pyridine, quant. b) tryptamine, anhyd THF, 60ºC, 64% c) 10%Pd-C, H2, MeOH, 72%
124
Scheme 3. General synthetic scheme of compound 5.
125
Figure 2. All compounds exhibited a non-competitive inhibition profile.
a) A representative inhibition profile exhibited by the ester analogue, compound 1.
126 b) A representative inhibition profile exhibited by compound 3.
127 c) A representative inhibition profile exhibited by compound 4.
128 d) A representative inhibition profile exhibited by compound 5.
129
B. Phenotype Testing
Recently, we have been able to connect ecHinT to an established phenotype in E. coli . Using active-site mutants, we demonstrated that the catalytically active ecHinT is required for E. coli to utilize D-alanine as a carbon source. To assess the ability of our compounds to inhibit ecHinT in vivo , we tested the ability of E. coli BW25113 strain to grow on D-alanine in the presence of 100 µM of either compound 3 or 5. Cultures were grown in M9 media, supplemented with either 22 mM D,L-alanine or 10 mM glucose, at
37ºC for 40 h before OD 600 was measured. Consistent with our previous finding, that the
catalytic activity of ecHinT is required for E. coli to grow on D-alanine, inhibition of
ecHinT in BW25113 resulted in its impaired growth when D-alanine is the sole carbon
source (Fig. 3). In addition to the cell permeability properties of compounds 3 and 5, no
toxicity was observed with either inhibitor when glucose is the sole carbon source in the
growth medium.
IV. Summary and Concluding Remarks
In conclusion, we have designed and synthesized the first generation of cell-
permeable Hint inhibitors. Previously, we have shown that ∆hinT E. coli, a mutant strain in which hinT was deleted, exhibited impaired growth when D-alanine is the sole carbon source. 94 Using the developed inhibitors as chemical probes, this work has
demonstrated that inhibition of ecHinT resulted in similar impaired growth observed for
∆hinT in the presence of D-alanine as a sole carbon source. Guanosine carbamate 94 , one
of the best inhibitors reported in this study, may serve as structural template for the
design and development of more potent inhibitors with enhanced inhibition constants.
Ongoing crystallographic studies of the inhibitor-bound form of Hint should provide a
130 structural foundation for understanding the mechanism of inhibition and for the design of selective Hint inhibitors.
131
Figure 3. ecHinT inhibition resulted in impaired phenotype.
*Measurements were carried out in duplicates and the standard deviation was ± 2%.
132
Chapter 5
Possible Physiological Role of The Histidine Triad Nucleotide Binding Protein in Escherichia coli : Regulation of D-alanine Detoxification
133
I. Introduction
Amino acids are best known as the building blocks of proteins, which themselves form the biological machinery of all cells. The chemical side chains of amino acids can be arranged clockwise or counterclockwise forming D or L amino acids. In all kingdoms of life, cells predominantly use the overwhelmingly “left-handed” form of amino acids,
L-amino acids, as building blocks in protein synthesis. Nevertheless, significant quantities of D-amino acids can be produced in most bacteria. D-Ala and D-Glu are the most widely produced amino acids in bacteria. 95 The two amino acids are incorporated
into peptidoglycan in the cell wall structure. 90 Peptidoglycan is a strong and elastic
heteropolymer that is synthesized and modified by penicillin binding proteins. 90 The
chemistry and structure of the peptidoglycan is dynamic and depends on factors that
regulate its D-amino acid composition and architecture. 95; 96
Recently, it has been shown that D-amino acids are released, by a diverse number
of bacterial species, in the stationary phase as agents that can control cell wall assembly
and modification. 97 The regulatory role of D-amino acids appears to be mediated by
signaling metabolic slowing in cell wall and cytoplasmic compartments when resources
become scarce. 97 Supporting the proposed signaling role of D-amino acids in bacteria, it
has been found that a mixture of D-amino acids produced in bacteria prevents biofilm
formation and signals for their disassembly in certain bacterial populations. 91 Indeed, D- tyrosine, D-leucine, D-tryptophan, and D-methionine were shown to function as active inhibitory factors of biofilm formation. 91 Not only did these D-amino acids prevent biofilm formation, but they also disrupted existing biofilm. 91 Interestingly, the inhibitory
role of D-amino acids on biofilm formation is believed to be mediated by the release of
134 amyloid fibers that linked cells in the biofilm together. 91 Although a growing evidence has emerged that bacteria have taken advantage of amino acid chirality to sense and respond to specific environmental conditions, it is still controversial whether D-amino acids have a potential role in signaling between individual bacteria.
Since bacteria are the major producers of D-amino acids in nature, it is therefore not surprising that bacteria possess the biochemical machinery to synthesize and degrade such components. In bacteria, neutral D-amino acids can be oxidized to the corresponding α-keto acids by a flavoenzyme named D-amino acid dehydrogenase
(DadA). 84; 85; 98 The enzyme has a broad specificity towards D-amino acids, of which D- alanine is the best substrate. 84; 85; 98 DadA is a membrane bound protein that is directly linked to a respiratory chain. 86 To protect bacteria from the harmful reactive oxygen species that could damage cellular components, D-amino acid oxidation is not coupled to
86; 87 O2 reduction in bacteria. E. coli DadA is a heterodimer formed by a small subunit and a large subunit. The small subunit contains FAD that receives two electrons from the substrate and transfers them to a sulfur-iron center located in the large subunit. 86; 87
DadA was shown to play two main roles in bacteria. Initially, it allows bacteria to grow using D-amino acids as the sole carbon, nitrogen, and energy source. 86; 87 Second, it
prevents local accumulation of D-amino acids as some D-amino acid analogues have
been shown to exhibit specific inhibitory effects on certain bacterial processes, such as
germination. 88; 89
It has been shown that the expression of DadA is induced by the presence of either D-alanine or L-alanine in the growth medium. 99 A catabolic alanine racemase
converts the D- into the L-stereoisomer of alanine. However, in mutants lacking alanine
135 racemase, D-alanine does not induce transcription from the dad operon, suggesting that
L-alanine is an internal inducer of the operon. 100 In addition, it has been shown that the
induction of DadA is regulated by catabolite repression and therefore the induced and
non-induced levels of DadA are significantly lower in cells grown in the presence of
glucose. 82; 100; 101 The extent of catabolite repression depends upon the concentration of
c-AMP (3'-5 cyclic adenosine monophosphate) inside the cells. Binding of c-AMP to
CAP (Catabolite Activator Protein) causes it to undergo a transition which allows it to
bind to the DNA in the promotor region and thereby permits transcription. 99
In Chapter 3, we have demonstrated, by E. coli gene disruption studies, that the
ecHinT enzyme is necessary for growth under conditions when alanine is the sole carbon
and nitrogen source. 94 Further investigations, with active-site mutants and ecHinT
inhibitors, revealed a unique connection between the discovered phenotype and the
catalytic activity of ecHinT. 94 In addition, we have demonstrated that the expression of
catalytically-active ecHinT is essential for the activity of the enzyme D-alanine
dehydrogenase. 94
Histidine triad nucleotide binding proteins (Hints) are ubiquitous enzyme
members of the histidine triad (HIT) protein superfamily of nucleotidyl transferases and
hydrolases.2 The Hint branch represents the most ancient branch of this superfamily, and can be found in Archaea, Bacteria, and Eukaryotae. Despite their wide-spread and high conservation among all kingdoms of life, a rationale for the occurrence of Hint remains elusive. In particular, previous to our work, the role of Hint1 in the life cycle of prokaryotes in general, and bacteria, in particular, has not been investigated.
136
In this study, we demonstrate that, in addition to its role in the regulation of D- amino acid metabolism, ecHinT is also involved in D-alanine detoxification in E. coli .
Indeed, we discovered that E. coli mutants lacking dadA or hinT genes are highly susceptible to D-alanine toxicity and that the enzymatic activity of ecHinT is essential to protect E. coli from the D-alanine toxicity.
Recently, the biochemical connection between ecHinT and Lysyl tRNA synthetase has been elucidated. 32 Supported by the results of several radiolabeling and mutagenic experiments, it has been proposed that Hint may function in part to regulate the catalytic activity of LysRS. 32 Interestingly, the catalytic activity of Hint has been
shown to be an essential requirement for ability of Hint to hydrolyze lysyl-AMP
generated by LysRS. 32 The proposed function of Hint provided a possible new
mechanism that regulates protein translation by either pre-transfer editing or scavenging
highly reactive inappropriately released aminoacyl-adenylates. Given the similarity
between the molecular determinants that govern both ecHinT-LysRS and ecHinT-DadA
interactions, we hypothesize that the ability of ecHinT to activate DadA is governed by
the ability of aminoacyl tRNA synthetases to adenylate ecHinT to generate the ecHinT-
AMP intermediate. Once the intermediate formed, DadA becomes functionally active to
catalyze D-alanine degradation and prevent its accumulation in the intracellular
compartments. Failure to activate the metabolic pathway of D-alanine catabolism in
bacteria would probably lead to mis-incorporation of D-alanine in protein structures,
generating functionless proteins and thus toxicity.
137
II. Experimental Procedures
A. Bacterial Strains, Media, and Growth Conditions
All bacterial strains used in this study were obtained from the E. coli Genetic
Resource Center at Yale University. All strains were received as glycerol stocks on
filters, sub-cultured twice, on Luria–Bertani (LB)-agar medium and incubated at 37°C for
48 h before testing. The pH of the medium was adjusted to 7.0 with 1.0 M sodium
hydroxide, and the medium was sterilized at 121°C and 15 psi for 20 min. Liquid cultures
were grown aerobically in LB medium supplemented with kanamycin (kan, 50 µg/µl) at
37°C, with shaking, for 18 h. Strains were stored as 10% glycerol stocks at - 80°C. The
mutations in gene knock-out strains were verified using the Polymerase Chain Reaction
(PCR) technique as described previously in chapter 3.
B. Growth Inhibitory Effects of D-amino Acids on Escherichia coli
IC 50 defined as the D-amino acid concentration that produces 50% of the maximal effect was determined for three E. coli strains, wild-type BW25113, ∆hinT and
∆dadA . A starter culture of 1 µL of each strain was added to 5 ml of LB medium and
allowed to grow overnight at 37ºC while shaking at 220 rpm. The following day, 10 µL
of the starter culture was inoculated into 10 ml LB medium and was allowed to shake at
37ºC. Stock solutions of 1 M of the D-amino acids were prepared in phosphate buffered
saline (pH 7.00), and pH was adjusted using 1 N NaOH and 1 N HCl. The stock solutions
were filter-sterilized and diluted in LB medium to various concentrations ranging
between 0-300 mM. Fresh cultures of each strain were supplemented with the desired D-
amino acid concentration. Optical densities at 600 nm, OD 600 , measurements were
138 obtained, using Varian Cary 50 UV-visible spectrophotometer (Palo Alto, CA, USA), after 24 h incubation at 37ºC with shaking. The concentration of D-amino acid inhibiting bacterial growth by 50% (IC 50 ) was estimated by plotting the millimolar concentration of
D-amino acid versus OD 600 and the concentration that causes 50% growth inhibition was determined.
C. Growth Kinetics
Growth curves of E. coli strains metabolizing D-alanine were constructed by
monitoring OD 600 of cultures growing in LB medium in presence of 65 mM of D-alanine.
The test flasks were shaken at 37ºC and samples were drawn at each time point for OD 600 readings until the stationary phase was reached.
D. D-alanine Toxicity in Presence of ecHinT Inhibitors
The synthesis and evaluation of ecHinT inhibitors were described in Chapter 3 and 4. Cultures of E. coli wild-type BW25113 strain in LB medium were prepared as
described previously in the experimental procedures. Stock solutions of 10 mM of the
TpGc inhibitor were prepared in molecular grade DMSO and diluted with distilled water.
100 µM of the inhibitor was added to each culture, with or without D-alanine, with the
control culture having no inhibitor. Cultures were incubated at 37ºC with shaking.
Absorbance readings at 600 nm were obtained after 24 hours.
139
E. Mammalian Cell Lines and Cell Culturing
The cytotoxicity effects of D-alanine, in mammalian cells, were tested in the
following five human cancer cell lines: the human breast cancer cell lines MCF-7 and
MDA-MB-231, the human pancreatic carcinoma cell line MIA PaCa-2, the T-cell
leukemia cell line HPB-MLT and the human Burkitt's lymphoma cell line Raji. Cell lines
were cultured in either high glucose Dulbecco’s Modified Eagle Medium (DMEM)
(Invitrogen, Carlsbad, CA) or Roswell Park Memorial Institute Medium (RPMI)
(Invitrogen, Carlsbad, CA) containing 10% Heat Inactivated Fetal Bovine Serum (HI-
FBS), 2 mM L-glutamine, 50 U/ml penicillin and 50 µg/ml streptomycin. Cell lines were
maintained at 37°C in a 5% CO 2 atmosphere with 95% humidity. Cells were passaged
weekly, and the culture medium was changed twice a week. According to their growth
profiles, the optimal plating densities were determined. To ensure exponential growth
throughout the experimental period and to ensure a linear relationship between
absorbance at 492 nm and cell number when analyzed by the MTS assay, the following
densities were used for each cell line: 5×10 3 cells/well for the MCF-7, MDA-MB-231
and MIA PaCa-2 cell lines, and 15×10 3 cells/well for the HPB-MLT and Raji cell lines.
F. Cell Viability Assay
Cultured cells were counted using trypan blue exclusion method and seeded to 96- well plates at the desired densities. D-alanine and L-alanine stock solutions at a concentration of 1 M were prepared in phosphate buffer and further diluted in medium to produce the following concentrations: 50, 100, 150, 200 and 300 mM. D-alanine or L- alanine at various concentrations was added to each well in triplicate. Treated cells were incubated in a 37°C 5% CO 2 incubator for 24 h. In order to assess cell viability after
140 incubation with various concentrations of D-alanine or L-alanine, MTS assays were carried out using CellTiter 96 Aqueous One solution cell proliferation assay according to the manufacturer’s protocol. The absorbance at 492 nm was read on a Tecan Plate Reader
(Tecan Group Ltd., Mannedorf, Switzerland).
G. D-alanine Cytotoxicity in MCF-7, MDA-MB-231 and MIA PaCa-2 Cell Lines
Cells were detached with the Non-Enzymatic Cell Dissociation Complex (Sigma
Chemical Co.) to make single-cell suspension, and viable cells were counted and diluted with DMEM medium to give a final density of 5×10 3 cells/well. 100 µL per well of cell
suspension was seeded in 96-well plates and incubated to allow for cell attachment. After
24 h, the cells were treated with either D-alanine or L-alanine. Each concentration of D-
alanine or L-alanine was added to each well in triplicate. The plates were incubated for
24 h at 37°C 5% CO 2. Cell viability was measured using MTS assay. Control wells without D-alanine or L-alanine treatment were prepared under the same conditions.
H. D-alanine Cytotoxicity in HPB-MLT and Raji Cell Lines
Viable cells were counted and diluted with RPMI medium to give a final density of 15×10 3 cells/well. 100 µL per well of cell suspension was seeded in 96-well plates.
After 24 h, the cells were treated with either D-alanine or L-alanine. Each concentration
of D-alanine or L-alanine was added to each well in triplicate. The plates were incubated
for 24 h at 37°C 5% CO 2. Cell viability was measured using MTS assay. Control wells without D-alanine or L-alanine treatment were prepared under the same conditions.
141
III. Results
A. D-amino Acids Toxicity in E. coli
The concentration of D-amino acid that inhibits the growth of half of an inoculum of E. coli (IC 50 ) was estimated for three E. coli strains (Table 1). As shown in Figure 1
(A-E), wild-type BW25113 appears to be the most resistant strain to D-alanine treatment, with an IC 50 value that is 2.6-fold and 3.8-fold higher than the IC 50 value of ∆hinT and
∆dadA strains respectively (Table 1). Compared to D-alanine, none of the tested strains show preferential sensitivity to treatment with other tested D-amino acids. All measurements were carried out in duplicates with SD less than 5%. To confirm that the toxicity observed with D-alanine is specific to the D form and not to the L form of alanine, growth of the three strains, under conditions where LB medium was supplemented with various concentrations of L-alanine, was studied. Consistent with the literature reports, our results show that the D-isomer, but not the L-isomer, of alanine was associated with the observed toxicity in E. coli (Fig. 2). The toxicity of the D-isomers of several amino acids for Gram-positive and Gram-negative bacteria has been clearly established. 102; 103; 104 In addition, high concentrations of glycine were reported to
inhibit bacterial growth by replacing both D- and L-alanine residues in peptidoglycan
subunits 104 thereby impairing transpeptidization within the cell wall and leading to
subsequent lysis of bacteria.
142
Figure 1. D-amino acids IC 50 determination in E. coli strains: wild-type BW25113,
∆hinT and ∆dadA were grown in LB medium at 37°C for 24 h in presence of various concentrations of D-amino acids.
a) D-alanine IC 50 .
2
1.5
600 1 wild-type OD ∆dadA ∆hinT 0.5
0 0 50 100 150 200 250 300
[D-alanine] mM
143 b) D-serine IC 50 .
2
1.5
1 wild-type 600
OD ∆hinT
0.5 ∆dadA
0 0 50 100 150 200 250 300 [D-serine] mM
c) D-proline IC 50 .
2
1.5
1 wild-type 600 ∆hinT OD 0.5 ∆dadA
0 0 50 100 150 200 250 300 [D-proline] mM
144 d) D-lysine IC 50 .
2
1.5
600 1 OD wild-type ∆hinT 0.5 ∆dadA
0
0 50 100 150 200 250 300 D-[lysine ] mM
e) Glycine IC 50 .
2
1.5
1 wild-type
600 ∆hinT ∆dadA OD 0.5
0
0 50 100 150 200 250 300
[Glycine] mM
145
Figure 2. Sensitivity of E. coli strains: wild-type BW25113, ∆hinT and ∆dadA , to L- alanine treatment.
2
1.5
1 wild-type
600 ∆hinT
OD 0.5 ∆dadA
0 0 50 100 150 200 250 300
[L-alanine] mM
146
Table 1. Summary of IC 50 values in millimolar.
Amino acid wild-type ∆hinT ∆dadA
L-alanine > 200 > 200 > 200
D-alanine 230 ± 15 90 ± 5 60 ± 5
D-serine 170 ± 10 170 ± 10 170 ± 10
D-proline > 300 > 300 > 300
D-lysine 60 ± 5 60 ± 5 60 ± 5
glycine 130 ± 10 130 ± 10 130 ± 10
147
The growth kinetic experiments were conducted to examine the inhibitory effect of D-alanine on the growth rate of D-alanine-metabolizing E. coli stains. Relative to wild-type BW25113 strain, both ∆hinT and ∆dadA strains show reduced growth rate in
the exponential phase when grown in LB medium in the presence of 65 mM of D-alanine
(Fig 3).
The lethal effects of toxic amino acids were shown to be additive in several
studies . 105 To study the potential for synergitic toxicity between different D-amino acids, the IC 50 of D-alanine in presence of 30 and 60 mM D-lysine was estimated. As
illustrated in Figure 4, 3.8-fold decrease in the IC 50 value, in the presence of 60 mM D-
lysine, and 1.5-fold decrease in the IC 50 value, in the presence of 30 mM D-lysine, were
observed. Although specific parallel pathways may be affected by different amino acids,
it is also possible that a common target is altered, which leads to an additive effect in
toxicity.
B. Effect of ecHinT Inhibition on E. coli Sensitivity to D-alanine
Growth of the wild-type BW25113 strain in LB medium supplemented with 150
mM D-alanine in presence of ecHinT inhibitor, TpGc, was studied. Synthesis and
evaluation of the inhibitor used in this study were described earlier in Chapter 3. As
shown in Figure 5, compared to the control culture, where the medium was supplemented
with 150 mM D-alanine with no inhibitor added, the inhibition of ecHinT results in 2.7-
fold reduction in the growth level observed for the wild-type BW25113 strain. Similar
growth levels in control cultures, in the presence or absence of 100 µM of TpGc, suggest
that the inhibitor itself is not responsible for the observed toxicity (Fig. 5).
148
Figure 3. Growth kinetic curves of three E. coli strains: wild-type BW25113, ∆hinT and
∆dadA were grown in LB medium at 37°C in presence of 65 mM D-alanine.
149
Figure 4. D-alanine and D-lysine synergism: wild-type BW25113 strain was grown in
LB medium at 37°C in presence of various concentrations of D-alanine and D-lysine.
2
1.5
600 1 0 mM D-lys OD 30 mM D-lys 0.5 60 mM D -lys
0 0 50 100 150 200 250 300
[D-alanine] mM
150
Figure 5. Effect of ecHinT inhibition on E. coli sensitivity to D-alanine.
151
C. D-alanine Toxicity in Mammalian Cell Lines
MCF-7, MDA-MB-231, MIA PaCa-2, HPB-MLT and Raji cells were incubated with different concentrations of either D-alanine or L-alanine for up to 24 h. The degree of cell viability after treatment was obtained by conducting MTS assay. Optical densities at 492 nm was used as an indicator of cell viability and expressed as a percentage of cell viability. For the MCF-7 and MDA-MB-231 breast cancer cell lines, a dose-dependent reduction in the percentage of viable cells when treated with D-alanine, but not with L- alanine, was observed (Fig. 6a, 6b). However, D-alanine toxicity was not observed for the human pancreatic carcinoma cell line MIA PaCa-2, the T-cell leukemia cell line
HPB-MLT and the human Burkitt’s lymphoma cell line Raji (Fig. 6c, 6d, 6e). In addition to other possible contributing pathways of toxicity, the potential toxicity of D-alanine on
MCF-7, MDA-MB-231 might be explained by the oxidative damage to cells by H 2O2, which is formed upon the oxidation of D-alanine by D-amino acid oxidase. 106
IV. Discussion
In E. coli, D-alanine might be utilized in two main pathways that prevent its accumulation in cellular compartments. First, D-alanine is incorporated in the peptidoglycan, the structural heteropolymer found in bacterial cell walls. The incorporation of D-alanine into peptidoglycan precursors is catalyzed via the sequential action of alanine racemase and several ligases. Although this is a dynamic process that is well-regulated by enzymatic activity, less is known about the factors that regulate alterations in the composition and architecture of the peptidoglycan heteropolymer (Fig.
7).
152
Figure 6. D-alanine toxicity on mammalian cell lines. a) MCF-7 cell line.
153 b) MDA-MB-231 cell line.
c) MIA PaCa-2 cell line.
154 d) Raji cell line.
e) HPB-MLT cell line.
155
Figure 7. The structure of peptidoglycan. Peptidoglycan is built from prefabricated units that contain a peptide chain of three to five amino acids attached to the N-Acetyl- muramic acid. The peptide chain contains a terminal d-alanine-d-alanine unit.
156
The second pathway is a metabolic pathway that involves the conversion of D-
alanine into pyruvate, by DadA; thus serving as an energy precursor for cellular
processes. Under basal conditions, the two pathways are optimized to regulate and
maintain certain levels of intracellular pools of D-alanine for cell wall biosynthesis and
energy utilization. When E. coli is exposed to an environment that is rich in D-alanine,
these pathways probably get activated to protect the cells against D-alanine toxicity that
is induced by exceeding the capacity of the cellular and metabolic machinery.
Although potential toxicity has been linked to D-amino acids, the fundamental
mechanism by which toxicity occurs is not fully understood. In animal studies,
administration of D-amino acids to rats and chicks resulted in growth inhibition of the
animals. 107 In addition, D-amino acids were found to accumulate in certain tissues
resulting in serious damage, e.g., suppression of the synthesis of glutamate oxaloacetate
transaminase, glutamic pyruvic transaminase, and lactate dehydrogenase. 107
Several mechanisms were proposed to explain the toxicity associated with D- amino acids in both eukaryotes and prokaryotes. The proposed pathway of D-amino acids toxicity in eukaryotes appears to be related to the oxidative damage to cells by
H2O2, which is formed upon the oxidation of D-amino acids by D-amino acid oxidase
and/or D-aspartate oxidase. 106 In E. coli , however, D-amino acids oxidation by DadA is not coupled to O 2. Hence, the toxicity of D-amino acids in E. coli could not be explained
by the previous mechanism.
Another pathway has been proposed to explain the toxicity of D-amino acids in
both eukaryotes and prokaryotes. Based on several pieces of evidence, it has been
suggested that D-amino acids toxicity is directly related to the formation of D-aminoacyl
157 tRNA. D-Amino acids are usually prevented from being incorporated into proteins because aminoacyl tRNA synthetases have preferential binding sites for L-amino acids.
108; 109; 110; 111 However, it was observed that Escherichia coli and Bacillus subtilis tyrosyl-tRNA synthetases can potentially catalyze the formation of D-tyrosyl-tRNA Tyr to the same extent found for the L-enantiomer. 108; 109 Interestingly, extracts of E. coli , yeast, rabbit reticulocytes, and rat liver were shown to have an enzyme activity capable of accelerating the hydrolysis of the ester linkage of D-Tyr-tRNA, thus releasing the tRNA from D-tyrosine. 112 In E. coli , this enzyme activity has been assigned to D-Tyrosyl-
tRNA Tyr deacylase which is encoded by the yihZ gene. D-Tyrosyl-tRNA Tyr deacylase has
a broad range of substrate specificity and can also hydrolyze D-aspartyl-tRNA Asp and D-
tryptophanyl-tRNA Trp into the corresponding D-amino acid and free tRNA. An E. coli strain, in which the yihZ gene was disrupted, was found to be more sensitive to D- tyrosine, D-tryptophan, D-aspartate, D-glutamate, and D-serine than the wild-type strain.
112 Consistent with this finding, about 40% of the cellular tRNA Tyr in E. coli lacking D- tyrosyltRNA Tyr deacylase was demonstrated to bear D-tyrosyl-tRNA Tyr when the cells were grown in the presence of 2.4 mM D-tyrosine. 111 Over-expression of tRNA Tyr , tRNA Trp , and tRNA Asp protects the E. coli yihZ –knock–out strain against the toxic effect of D-tyrosine, D-tryptophan, and D-aspartate, respectively. 111
The fate of D-aminoacyl tRNA in E. coli has never been studied previously.
However, Dedkova et al . demonstrated that D-methionine and D-phenylalanine charged
on aminoacyl tRNA were incorporated into proteins in vitro 113 , suggesting that the
toxicity of D-amino acids might arise from the mis-incorporation of D-amino acids in
protein structure resulting in the formation of functionless protein.
158
In previous work, we demonstrated that lysyl-AMP generated by LysRS is a
physiological substrate for Hints. 32 In addition, we showed that the ability of Hint to hydrolyze lysyl-AMP depends on its enzymatic activity. Based on these findings, we proposed that Hints might function as potential regulators of protein translation processes. Interestingly, the molecular determinants governing this regulation appear to be overlapping with the determinants of ecHinT-DadA interaction regulation (Table 2). 94
Based on these observations, we hypothesize that there might be a possible correlation between the ability of aminoacyl tRNA synthetases to adenylate ecHinT and the activation of DadA by ecHinT (Fig. 8). Once DadA is functionally active, the metabolic pathway of D-alanine catabolism is also activated; thus preventing D-alanine accumulation in the intracellular compartment. We propose that accumulation of D- alanine would probably result in charging tRNA molecules with D-alanine instead of the
L-isomers of amino acids. Consequently, D-alanine might be mis-incorporated in protein structures generating mis-folded proteins, which would probably lead to the observed toxicity of D-alanine in E. coli . The ability of Hint to cross-talk to aminoacyl tRNA synthetases and hydrolyze aminoacyl-AMP intermediates suggest that Hint has a role in pre-transfer editing or scavenging the inappropriately released D-aminoacyl-adenylates; thus preventing D-amino acids mis-incorporation in protein structures. Ongoing studies attempting to unravel the Hint interaction network in E. coli should shed light on this hypothesis.
159
Table 2. Summary of the ecHinT-LysRS and ecHinT-DadA studies.
Protein Activity LysRS-Adenylation DadA Activity
ecHinT Yes Yes Yes hHint1 Yes No No ecHinT-H101A No No No ∆114-119 Yes No No ∆117-119 Yes No No ec /Hs Yes Yes Yes Hs/ ec Yes Yes No
160
Figure 8. D,L-alanine metabolism and possible ecHinT-LysRS regulation.
YanJ alanine ACSS CycA Serine/Alanine Transporter Glycine APC transporter
O Alanine O Racemase -O -O
NH3+ NH3+ D-Alanine L-Alanine ‘
H2O dadA-Active hinT LysRS•Lys-AMP Pi ? NH4+ dadA-Inactive hinT-AMP LysRS•Lys ATP
O -O Glycolysis 1 O Pyruvate
161
BIBLIOGRAPHY
1. Kijas, A. W., Harris, J. L., Harris, J. M. & Lavin, M. F. (2006). Aprataxin forms a
discrete branch in the HIT (histidine triad) superfamily of proteins with both
DNA/RNA binding and nucleotide hydrolase activities. Journal of Biological
Chemistry 281, 13939-13948.
2. Brenner, C. (2002). Hint, Fhit, and GalT: Function, structure, evolution, and
mechanism of three branches of the histidine triad superfamily of nucleotide
hydrolases and transferases. Biochemistry 41, 9003-9014.
3. Liu, H. D., Rodgers, N. D., Jiao, X. & Kiledjian, M. (2002). The scavenger
mRNA decapping enzyme DcpS is a member of the HIT family of
pyrophosphatases. EMBO Journal 21, 4699-4708.
4. Kwasnicka, D. A., Krakowiak, A., Thacker, C., Brenner, C., and Vincent, S. R.
(2003). Coordinate expression of NADPH-dependent flavin reductase, Fre-1, and
Hint-related 7meGMP-directed hydrolase, DCS-1. J. Biol. Chem. 278, 39051-
39058.
5. Lima, C. D., Klein, M. G., Weinstein, I. B., and Hendrickson, W. A. (1996).
Three-dimensional structure of human protein kinase C interacting protein 1, a
member of the HIT family of proteins. Proceedings of the National Academy of
Sciences of the United States of America 93, 5357-5362.
6. Brenner, C., Bieganowski, P., Pace, H. C. & Huebner, K. (1999). The histidine
triad superfamily of nucleotide-binding proteins. Journal of Cellular Physiology
181, 179-187.
162
7. Brenner, C., Garrison, P., Gilmour, J., Peisach, D., Ringe, D., Petsko, G. A. &
Lowenstein, J. M. (1997). Crystal structures of HINT demonstrate that histidine
triad proteins are GalT-related nucleotide-binding proteins. Nature Structural
Biology 4, 231-238.
8. Chou, T. F., Bieganowski, P., Shilinski, K., Cheng, J. L., Brenner, C. & Wagner,
C. R. (2005). P-31 NMR and genetic analysis establish hinT as the only
Escherchia coli purine nucleoside phosphoramidase and as essential for growth
under high salt conditions. Journal of Biological Chemistry 280, 15356-15361.
9. Bieganowski, P., Garrison, P. N., Hodawadekar, S. C., Faye, G., Barnes, L. D. &
Brenner, C. (2002). Adenosine monophosphoramidase activity of Hint and Hnt1
supports function of Kin28, Ccl1, and Tfb3. Journal of Biological Chemistry 277,
10852-10860.
10. Krakowiak, A., Pace, H. C., Blackburn, G. M., Adams, M., Mekhalfia, A.,
Kaczmarek, R., Baraniak, J., Stec, W. J. & Brenner, C. (2004). Biochemical,
crystallographic, and mutagenic characterization of hint, the AMP-lysine
hydrolase, with novel substrates and inhibitors. Journal of Biological Chemistry
279, 18711-18716.
11. Chou, T.-F., Baraniak, J., Kaczmarek, R., Zhou, X., Cheng, J., Ghosh, B., and
Wagner, C. R. (2007). Phosphoramidate Pronucleotides: A Comparison of the
Phosphoramidase Substrate Specificity of Human and E. coli Histidine Triad
Nucleotide Binding Proteins (Hint1). Mol. Pharmaceutics 4, 208-217.
12. Pace, H. C., Garrison, P. N., Robinson, A. K., Barnes, L. D., Draganescu, A.,
Rosler, A., Blackburn, G. M., Siprashvili, Z., Croce, C. M., Huebner, K. &
163
Brenner, C. (1998). Genetic, biochemical, and crystallographic characterization of
Fhit-substrate complexes as the active signaling form of Fhit. Proceedings of the
National Academy of Sciences of the United States of America 95, 5484-5489.
13. Ohta, M., Inoue, H., Cotticelli, M. G., Kastury, K., Baffa, R., Palazzo, J.,
Siprashvili, Z., Mori, M., McCue, P., Druck, T., Croce, C. M., and Huebner, K.
(1996). The FHIT gene, spanning the chromosome 3p14.2 fragile site and renal
carcinoma-associated t(3;8) breakpoint, is abnormal in digestive tract cancers.
Cell 84, 587-597.
14. Ishii H, O. K., Furukawa Y. (2003). Alteration of the fragile histidine triad gene
early in carcinogenesis: an update. J. Exp. Ther. Oncol. 3, 291-296.
15. Huebner, K., Croce, CM. (2003). Cancer and the FRA3B/FHIT fragile locus: it's a
HIT. Br. J. Cancer 88, 1501-1506.
16. Pekarsky, Y., Zanesi, N., Palamarchuk, A., Huebner, K., Croce, CM. (2002).
FHIT: from gene discovery to cancer treatment and prevention. Lancet. Oncol 3,
748-754.
17. Pekarsky, Y., Garrison, P. N., Palamarchuk, A., Zanesi, N., Aqueilan, R. I.,
Huebner, K., Barnes, L. D., Croce, C. M. (2004). Fhit is a physiological target of
the protein kinase Src. Proc Natl Acad Sci USA 101, 3775-3779.
18. Semba S, T. F., Fabbri, M., McCorkell, K. A., Volinia, S., Druck, T., Iliopoulos,
D., Pekarsky, Y., Ishii, H., Garrison, P. N., Barnes, L. D., Croce, C. M. and
Huebner, K. (2006). Fhit modulation of the Akt-survivin pathway in lung cancer
cells: Fhit-tyrosine 114 (Y114) is essentialFhit modulation of the Akt-survivin
pathway. Oncogene 25, 2860-2872.
164
19. Siprashvili, Z., Sozzi, G., Barnes, L. D., McCue, P., Robinson, A. K., Eryomin,
V., Sard, L., Tagliabue, E., Greco, A., Fusetti, L., Schwartz, G., Pierotti, M. A.,
Croce, C. M. & Huebner, K. (1997). Replacement of Fhit in cancer cells
suppresses tumorigenicity. Proceedings of the National Academy of Sciences of
the United States of America 94, 13771-13776.
20. Trapasso, F., Krakowiak, A., Cesari, R., Arkles, J., Yendamuri, S., Ishii, H.,
Vecchione, A., Kuroki, T., Bieganowski, P., Pace, H. C., Huebner, K., Croce, C.
M. & Brenner, C. (2003). Designed FHIT alleles establish that Fhit-induced
apoptosis in cancer cells is limited by substrate binding. Proceedings of the
National Academy of Sciences of the United States of America 100, 1592-1597.
21. Frey, P. A., Wong, L. J., Sheu, K. F. & Yang, S. L. (1982). Galactose-1-
Phosphate Uridylyltransferase - Detection, Isolation, and Characterization of the
Uridylyl Enzyme. Methods in Enzymology 87, 20-36.
22. Frey, P. A. (1996). The Leloir pathway: A mechanistic imperative for three
enzymes to change the stereochemical configuration of a single carbon in
galactose. FASEB Journal 10, 461-470.
23. Moreira, M. C., Barbot, C., Tachi, N., Kozuka, N., Uchida, E., Gibson, T.,
Mendonca, P., Costa, M., Barros, J., Yanagisawa, T., Watanabe, M., Ikeda, Y.,
Aoki, M., Nagata, T., Coutinho, P., Sequeiros, J. & Koenig, M. (2001). The gene
mutated in ataxia-ocular apraxia 1 encodes the new HIT/Zn-finger protein
aprataxin. Nature Genetics 29, 189-193.
24. Date, H., Onodera, O., Tanaka, H., Iwabuchi, K., Uekawa, K., Igarashi, S., Koike,
R., Hiroi, T., Yuasa, T., Awaya, Y., Sakai, T., Takahashi, T., Nagatomo, H.,
165
Sekijima, Y., Kawachi, I., Takiyama, Y., Nishizawa, M., Fukuhara, N., Saito, K.,
Sugano, S. & Tsuji, S. (2001). Early-onset ataxia with ocular motor apraxia and
hypoalbuminemia is caused by mutations in a new HIT superfamily gene. Nature
Genetics 29, 184-188.
25. Ahel, I., Rass, U., El-Khamisy, S. F., Katyal, S., Clements, P. M., McKinnon, P.
J., Caldecott, K. W. & West, S. C. (2006). The neurodegenerative disease protein
aprataxin resolves abortive DNA ligation intermediates. Nature 443, 713-716.
26. Wang, Z. R. & Kiledjian, M. (2001). Functional link between the mammalian
exosome and mRNA decapping. Cell 107, 751-762.
27. Guranowski, A. (2000). Specific and nonspecific enzymes involved in the
catabolism of mononucleoside and dinucleoside polyphosphates. Pharmacology
& Therapeutics 87, 117-139.
28. Guranowski, A., Starzynska, E., Bojarska, E., Stepinski, J. & Darzynkiewicz, E.
(1996). Dinucleoside 5',5'''-P-1,P-3-triphosphate hydrolase from yellow lupin
(Lupinus luteus) seeds: Purification to homogeneity and hydrolysis of mRNA 5'-
cap analogs. Protein Expression and Purification 8, 416-422.
29. Gilmour, J., Liang, N. C. & Lowenstein, J. M. (1997). Isolation, cloning and
characterization of a low-molecular-mass purine nucleoside- and nucleotide-
binding protein. Biochemical Journal 326, 471-477.
30. Shuman, S. & Lima, C. D. (2004). The polynucleotide ligase and RNA capping
enzyme superfamily of covalent nucleotidyltransferases. Current Opinion in
Structural Biology 14, 757-764.
166
31. Fankhauser, H., Schiff, J. A. & Garber, L. J. (1981). Purification and Properties of
Adenylyl Sulfate - Ammonia Adenylyltransferase from Chlorella Catalyzing the
Formation of Adenosine 5'-Phosphoramidate from Adenosine 5'-Phosphosulfate
and Ammonia. Biochemical Journal 195, 545-560.
32. Chou, T. F. & Wagner, C. R. (2007). Lysyl-tRNA synthetase-generated lysyl-
adenylate is a substrate for histidine triad nucleotide binding proteins. Journal of
Biological Chemistry 282, 4719-4727.
33. Lima, C. D., Klein, M. G. & Hendrickson, W. A. (1997). Structure-based analysis
of catalysis and substrate definition in the HIT protein family. Science 278, 286-
290.
34. Lee, Y. N., Nechushtan, H., Figov, N. & Razin, E. (2004). The function of Lysyl-
tRNA synthetase and Ap4A as signaling regulators of MITF activity in Fc epsilon
RI-activated mast cells. Immunity 20, 145-151.
35. Barnes, L. D., Garrison, P. N., Siprashvili, Z., Guranowski, A., Robinson, A. K.,
Ingram, S. W., Croce, C. M., Ohta, M. & Huebner, F. (1996). Fhit, a putative
tumor suppressor in humans, is a dinucleoside 5',5'''-P-1,P-3-triphosphate
hydrolase. Biochemistry 35, 11529-11535.
36. Geeganage, S. & Frey, P. A. (1998). Transient kinetics of formation and reaction
of the uridylyl-enzyme form of galactose-1-P uridylyltransferase and its Q168R-
Variant: Insight into the molecular basis of galactosemia. Biochemistry 37, 14500-
14507.
167
37. Wedekind, J. E., Frey, P. A. & Rayment, I. (1996). The structure of
nucleotidylated Histidine-166 of galactose-L-phosphate uridylyltransferase
provides insight into phosphoryl group transfer. Biochemistry 35, 11560-11569.
38. Abend, A., Garrison, P. N., Barnes, L. D. & Frey, P. A. (1999). Stereochemical
retention of the configuration in the action of Fhit on phosphorus-chiral
substrates. Biochemistry 38, 3668-3676.
39. Huang, K., Arabshahi, A., Wei, Y. & Frey, P. A. (2004). The mechanism of
action of the fragile histidine triad, Fhit: Isolation of a covalent adenylyl enzyme
and chemical rescue of H96G-Fhit. Biochemistry 43, 7637-7642.
40. Huang, K. S., Arabshahi, A. & Frey, P. A. (2005). pH-dependence in the
hydrolytic action of the human fragile histidine triad. European Journal of
Organic Chemistry , 5198-5206.
41. McDonald, J. R. & Walsh, M. P. (1985). Ca-+2-Binding Proteins from Bovine
Brain Including a Potent Inhibitor of Protein Kinase-C. Biochemical Journal 232,
559-567.
42. Pearson, J. D., Dewald, D. B., Mathews, W. R., Mozier, N. M., Zurcherneely, H.
A., Heinrikson, R. L., Morris, M. A., McCubbin, W. D., McDonald, J. R., Fraser,
E. D., Vogel, H. J., Kay, C. M. & Walsh, M. P. (1990). Amino-Acid-Sequence
and Characterization of a Protein Inhibitor of Protein Kinase-C. Journal of
Biological Chemistry 265, 4583-4591.
43. Klein, M. (1997). Studies on the b isoform of protein kinase C and a putative
protein kinase C inhibitor which is a member of the highly conserved histidine
triad protein family. New York, Columbia University, Ph.D. thesis.
168
44. Mozier, N. M., Walsh, M. P. & Pearson, J. D. (1991). Characterization of a Novel
Zinc-Binding Site of Protein-Kinase-C Inhibitor-1. FEBS Letters 279, 14-18.
45. Chou, T. F., Tikh, I. B., Horta, B. A. C., Ghosh, B., De Alencastro, R. B. &
Wagner, C. R. (2007). Engineered monomeric human histidine triad nucleotide-
binding protein 1 hydrolyzes fluorogenic acyl-adenylate and lysyl-tRNA
synthetase-generated lysyl-adenylate. Journal of Biological Chemistry 282,
15137-15147.
46. Korsisaari, N. & Makela, T. P. (2000). Interactions of Cdk7 and Kin28 with
Hint/PKCI-1 and Hnt1 histidine triad proteins. Journal of Biological Chemistry
275, 34837-34840.
47. Korsisaari, N., Rossi, D. J., Luukko, K., Huebner, K., Henkemeyer, M. & Makela,
T. P. (2003). The histidine triad protein Hint is not required for murine
development or Cdk7 function. Molecular and Cellular Biology 23, 3929-3935.
48. Weiske, J. & Huber, O. (2005). The histidine triad protein Hint1 interacts with
Pontin and Reptin and inhibits TCF-beta-catenin-mediated transcription. Journal
of Cell Science 118, 3117-3129.
49. Li, H., Zhang, Y., Su, T., Santella, R. M. & Weinstein, I. B. (2006). Hint1 is a
haplo-insufficient tumor suppressor in mice. Oncogene 25, 713-721.
50. Yuan, B. Z., Jefferson, A. M., Popescu, N. C. & Reynolds, S. H. (2004). Aberrant
gene expression in human non small cell lung carcinoma cells exposed to
demethylating agent 5-aza-2 '-deoxycytidine. Neoplasia 6, 412-419.
169
51. Weiske, J. & Huber, O. (2006). The histidine triad protein Hint1 triggers
apoptosis independent of its enzymatic activity. Journal of Biological Chemistry
281, 27356-27366.
52. Wang, L., Zhang, Y. J., Li, H. Y., Xu, Z. H., Santella, R. M. & Weinstein, I. B.
(2007). Hint1 inhibits growth and activator protein-1 activity in human colon
cancer cells. Cancer Research 67, 4700-4708.
53. Li, H. Y., Balajee, A. S., Su, T., Cen, B., Hei, T. K. & Weinstein, I. B. (2008).
The HINT1 tumor suppressor regulates both gamma-H2AX and ATM in response
to DNA damage. Journal of Cell Biology 183, 253-265.
54. Liu, Q., Puche, A. C. & Wang, J. B. (2008). Distribution and expression of
protein kinase c interactive protein (PKCI/HINT1) in mouse central nervous
system (CNS). Neurochemical Research 33, 1263-1276.
55. Su, T., Suzui, M., Wang, L., Lin, C. S., Xing, W. Q. & Weinstein, I. B. (2003).
Deletion of histidine triad nucleotide-binding protein 1/PKC-interacting protein
growth in mice enhances cell growth and carcinogenesis. Proceedings of the
National Academy of Sciences of the United States of America 100, 7824-7829.
56. Rodriguez-Munoz, M., de la Torre-Madrid, E., Sanchez-Blazquez, P., Wang, J. B.
& Garzon, J. (2008). NMDAR-nNOS generated zinc recruits PKC gamma to the
HINT1-RGS17 complex bound to the C terminus of Mu-opioid receptors.
Cellular Signalling 20, 1855-1864.
57. Martin, J., Magnino, F., Schmidt, K., Piguet, A. C., Lee, J. S., Semela, D., St-
Pierre, M. V., Ziemiecki, A., Cassio, D., Brenner, C., Thorgeirsson, S. S. &
170
Dufour, J. F. (2006). Hint2, a mitochondrial apoptotic sensitizer down-regulated
in hepatocellular carcinoma. Gastroenterology 130, 2179-2188.
58. Lenglet, S., Antigny, F., Vetterli, L., Dufour, J. F. & Rossier, M. F. (2008). Hint2
Is Expressed in the Mitochondria of H295R Cells and Is Involved in
Steroidogenesis. Endocrinology 149, 5461-5469.
59. Chou, T. F., Cheng, J. L., Tikh, I. B. & Wagner, C. R. (2007). Evidence that
human histidine triad nucleotide binding protein 3 (Hint3) is a distinct branch of
the histidine triad (HIT) superfamily. Journal of Molecular Biology 373, 978-989.
60. Butland, G., Peregrin-Alvarez, J. M., Li, J., Yang, W. H., Yang, X. C., Canadien,
V., Starostine, A., Richards, D., Beattie, B., Krogan, N., Davey, M., Parkinson, J.,
Greenblatt, J. & Emili, A. (2005). Interaction network containing conserved and
essential protein complexes in Escherichia coli. Nature 433, 531-537.
61. Hopfe, M., Hoffmann, R. & Henrich, B. (2004). P80, the HinT interacting
membrane protein, is a secreted antigen of Mycoplasma hominis. Bmc
Microbiology 4.
62. Kitzerow, A. & Henrich, B. (2001). The cytosolic HinT protein of Mycoplasma
hominis interacts with two membrane proteins. Molecular Microbiology 41, 279-
287.
63. Chou, T. F., Sham, Y. Y. & Wagner, C. R. (2007). Impact of the C-terminal loop
of histidine triad nucleotide binding protein1 (Hint1) on substrate specificity.
Biochemistry 46, 13074-13079.
171
64. Bardaweel, S., Pace, J., Chou, T.-F., Cody, V., Wagner, C. R. (2010). Probing the
impact of the ecHinT C-terminal domain on structure and catalysis. Journal of
Molecular Biology 404, 627-638.
65. Barbier, E., Zapata, A., Oh, E., Liu, Q., Zhu, F., Undie, A., Shippenberg, T. &
Wang, J. B. (2007). Supersensitivity to amphetamine in protein kinase-C
interacting protein/HINT1 knockout mice. Neuropsychopharmacology 32, 1774-
1782.
66. Lee, Y. N. & Razin, E. (2005). Nonconventional involvement of LysRS in the
molecular mechanism of USF2 transcriptional activity in Fc epsilon RI-activated
mast cells. Molecular and Cellular Biology 25, 8904-8912.
67. McPhillips, T. M., McPhillips, S. E., Chiu, H. J., Cohen, A. E., Deacon, A. M.,
Ellis, P. J., Garman, E., Gonzalez, A., Sauter, N. K., Phizackerley, R. P., Soltis, S.
M. & Kuhn, P. (2002). Blu-Ice and the Distributed Control System: software for
data acquisition and instrument control at macromolecular crystallography
beamlines. Journal of Synchrotron Radiation 9, 401-406.
68. Cohen, A. E., Ellis, P. J., Miller, M. D., Deacon, A. M. & Phizackerley, R. P.
(2002). An automated system to mount cryo-cooled protein crystals on a
synchrotron beamline, using compact sample cassettes and a small-scale robot.
Journal of Applied Crystallography 35, 720-726.
69. Gonzalez, A., Moorhead, P., McPhillips, S. E., Song, J., Sharp, K., Taylor, J. R.,
Adams, P. D., Sauter, N. K. & Soltis, S. M. (2008). Web-Ice: integrated data
collection and analysis for macromolecular crystallography. Journal of Applied
Crystallography 41, 176-184.
172
70. Otwinowski, Z. M., W. (1997). Processing of X-ray Diffraction Data Collected in
Oscillation Mode. In: Methods in Enzymology. Edited by C.W. Carter, Jr. & R.
M. Sweet, 276, Part A, pp 224-225. New York: Academic Press .
71. Vagin, A. & Teplyakov, A. (1997). MOLREP: an automated program for
molecular replacement. Journal of Applied Crystallography 30, 1022-1025.
72. Emsley, P. & Cowtan, K. (2004). Coot: model-building tools for molecular
graphics. Acta Crystallographica Section D-Biological Crystallography 60, 2126-
2132.
73. Collaborative Computational Project, Number 4. (1994). The CCP4 Suite:
Programs for Protein Crystallography. Acta Crystallogr. D50, 760-763.
74. Laskowski, R. A., Macarthur, M. W., Moss, D. S. & Thornton, J. M. (1993).
Procheck - a Program to Check the Stereochemical Quality of Protein Structures.
Journal of Applied Crystallography 26, 283-291.
75. Steer, B. A. & Schimmel, P. (1999). Different adaptations of the same peptide
motif for tRNA functional contacts by closely homologous tRNA synthetases.
Biochemistry 38, 4965-4971.
76. Shiba, K., Stello, T., Motegi, H., Noda, T., MusierForsyth, K. & Schimmel, P.
(1997). Human lysyl-tRNA synthetase accepts nucleotide 73 variants and rescues
Escherichia coli double-defective mutant. Journal of Biological Chemistry 272,
22809-22816.
77. Francklyn, C. S., First, E. A., Perona, J. J. & Hou, Y. M. (2008). Methods for
kinetic and thermodynamic analysis of aminoacyl-tRNA synthetases. Methods 44,
100-118.
173
78. Ito, M., Baba, T. & Mori, H. (2005). Functional analysis of 1440 Escherichia coli
genes using the combination of knock-out library and phenotype microarrays.
Metabolic Engineering 7, 318-327.
79. Baba, T., Ara, T., Hasegawa, M., Takai, Y., Okumura, Y., Baba, M., Datsenko, K.
A., Tomita, M., Wanner, B. L. & Mori, H. (2006). Construction of Escherichia
coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Molecular
Systems Biology 2.
80. Bochner, B. R., Gadzinski, P. & Panomitros, E. (2001). Phenotype MicroArrays
for high-throughput phenotypic testing and assay of gene function. Genome
Research 11, 1246-1255.
81. Sambrook, J. F., E.F. and Maniatis, T. (1989) Molecular Cloning: a laboratory
manual. 2nd ed. N.Y., Cold Spring Harbor Laboratory, Cold Spring Harbor
Laboratory Press, 1659 p. ISBN 0-87969-309-6.
82. Franklin, F. C. H. & Venables, W. A. (1976). Biochemical, Genetic, and
Regulatory Studies of Alanine Catabolism in Escherichia-Coli-K12. Molecular &
General Genetics 149, 229-237.
83. Abrahamson, J. L. A., Baker, L. G., Stephenson, J. T. & Wood, J. M. (1983).
Proline Dehydrogenase from Escherichia-Coli-K12 - Properties of the Membrane-
Associated Enzyme. European Journal of Biochemistry 134, 77-82.
84. Olsiewski, P. J., Kaczorowski, G. J. & Walsh, C. (1980). Purification and
Properties of D-Amino-Acid Dehydrogenase, an Inducible Membrane-Bound
Iron-Sulfur Flavoenzyme from Escherichia-Coli-B. Journal of Biological
Chemistry 255, 4487-4494.
174
85. Pollegioni, L., Piubelli, L., Sacchi, S., Pilone, M. S. & Molla, G. (2007).
Physiological functions of D-amino acid oxidases: from yeast to humans. Cellular
and Molecular Life Sciences 64, 1373-1394.
86. Raunio, R. P. a. J., W. T. (1973). D-alanine oxidase from Escherichia coli:
localization and induction by Lalanine. J. Bacteriol. 115, 560 - 566.
87. Deutch, C. E. (2004). Oxidation of 3,4-dehydro-D-proline and other D-amino acid
analogues by D-alanine dehydrogenase from Escherichia coli . FEMS
Microbiology Letters 238, 383-389.
88. Lobocka, M., Hennig, J., Wild, J. & Klopotowski, T. (1994). Organization and
Expression of the Escherichia-Coli K-12 Dad Operon Encoding the Smaller
Subunit of D-Amino-Acid Dehydrogenase and the Catabolic Alanine Racemase.
Journal of Bacteriology 176, 1500-1510.
89. Yasuda, Y. & Tochikubo, K. (1985). Germination-Initiation and Inhibitory
Activities of L-Alanine and D-Alanine Analogs for Bacillus-Subtilis Spores -
Modification of Methyl-Group of L-Alanine and D-Alanine. Microbiology and
Immunology 29, 229-241.
90. Walsh, C. T. (1989). Enzymes in the D-Alanine Branch of Bacterial-Cell Wall
Peptidoglycan Assembly. Journal of Biological Chemistry 264, 2393-2396.
91. Kolodkin-Gal, I., Romero, D., Cao, S. G., Clardy, J., Kolter, R. & Losick, R. D-
Amino Acids Trigger Biofilm Disassembly. Science 328, 627-629.
92. Razin, E., Zhang, Z. C., Nechushtan, H., Frenkel, S., Lee, Y. N., Arudehandran,
R. & Rivera, J. (1999). Suppression of microphthalmia transcriptional activity by
175
its association with protein kinase C-interacting protein 1 in mast cells. Journal of
Biological Chemistry 274, 34272-34276.
93. Yannay-Cohen, N. & Razin, E. (2006). Translation and transcription: the dual
functionality of LysRS in mast cells. Molecules and Cells 22, 127-132.
94. Bardaweel, S., Ghosh, B., Chou, T. F., Sadowsky, M., Wagner, C. R. (2010).
HINT, a histidine triad nucleotide binding protein, is essential for alanine
metabolism in escherichia coli. Journal of Biological Chemistry, Submitted.
95. Holtje, J. V. (1998). Growth of the stress-bearing and shape-maintaining murein
sacculus of Escherichia coli. Microbiology and Molecular Biology Reviews 62,
181-189.
96. Vollmer, W., Blanot, D. & de Pedro, M. A. (2008). Peptidoglycan structure and
architecture. Fems Microbiology Reviews 32, 149-167.
97. Lam, H., Oh, D. C., Cava, F., Takacs, C. N., Clardy, J., de Pedro, M. A. &
Waldor, M. K. (2009). D-Amino Acids Govern Stationary Phase Cell Wall
Remodeling in Bacteria. Science 325, 1552-1555.
98. Olsiewski, P. J., Kaczorowski, G. J., Walsh, C. T. & Kaback, H. R. (1981).
Reconstitution of Escherichia-Coli Membrane-Vesicles with D-Amino-Acid
Dehydrogenase. Biochemistry 20, 6272-6279.
99. Wild, J. & Obrepalska, B. (1982). Regulation of Expression of the Dada Gene
Encoding D-Amino-Acid Dehydrogenase in Escherichia-Coli - Analysis of Dada-
Lac Fusions and Direction of Dada Transcription. Molecular & General Genetics
186, 405-410.
176
100. Wild, J., Hennig, J., Lobocka, M., Walczak, W. & Klopotowski, T. (1985).
Identification of the Dadx Gene Coding for the Predominant Isozyme of Alanine
Racemase in Escherichia-Coli -K12. Molecular & General Genetics 198, 315-
322.
101. Wild, J. & Klopotowski, T. (1981). D-Amino-Acid Dehydrogenase of
Escherichia-Coli -K12 - Positive Selection of Mutants Defective in Enzyme-
Activity and Localization of the Structural Gene. Molecular & General Genetics
181, 373-378.
102. Fling, M., and Fox, S. W. (1945). Antipodal specificity in the inhibition of growth
of L. arabinosus by amino acids. J. Biol. Chem. 160, 329-336.
103. Yaw, K. E., and Kakavas, J. C. (1952). Studies on the effects of D-amino acids on
Brucella abortus. J. Bacteriol . 63, 263-268.
104. Hammes, W., Schle1fer, K. H. and Kandler, O. (1973). Mode of action of glycine
on the biosynthesis of peptidoglycan. J. Bacteriol . 116, 1029-1053.
105. Beardsley, R. (1962). Amino acid cross resistance in agrobacterium tumefaciens.
J. Bacteriol. 84, 1237-1240.
106. Ercal, N., Luo, X., Matthews, R. H. & Armstrong, D. W. (1996). In vitro study of
the metabolic effects of D-amino acids. Chirality 8, 24-29.
107. Daniello, A., Donofrio, G., Pischetola, M., Daniello, G., Vetere, A., Petrucelli, L.
& Fisher, G. H. (1993). Biological Role of D-Amino-Acid Oxidase and D-
Aspartate Oxidase - Effects of D-Amino Acids. Journal of Biological Chemistry
268, 26941-26949.
177
108. Calender, R., Berg, P. (1966). The catalytic properties of tyrosyl ribonucleic acid
synthetases from Escherichia coli and Bacillus subtilis . Biochemistry 5, 1690-
1695.
109. Calender, R., Berg, P. (1966). Purification and physical characterization of tyrosyl
ribonucleic acid synthetases from Escherichia coli and Bacillus subtilis .
Biochemistry 5, 1681-1690.
110. Soutourina, J., Plateau, P. & Blanquet, S. (2000). Metabolism of D-aminoacyl-
tRNAs in Escherichia coli and Saccharomyces cerevisiae cells. Journal of
Biological Chemistry 275, 32535-32542.
111. Soutourina, O., Soutourina, J., Blanquet, S. and Plateau, P. (2004). Formation of
D-tyrosyl-tRNATyr accounts for the toxicity of D-tyrosine toward Escherichia
coli . J. Biol. Chem. 279, 42560-42565.
112. Ferri-Fioni, M. L., Fromant, M., Bouin, A. P., Aubard, C., Lazennec, C., Plateau,
P. & Blanquet, S. (2006). Identification in archaea of a novel D-Tyr-tRNA(Tyr)
deacylase. Journal of Biological Chemistry 281, 27575-27585.
113. Dedkova, L. M., Fahmi, N. E., Golovine, S. Y. & Hecht, S. M. (2003). Enhanced
D-amino acid incorporation into protein by modified ribosomes. Journal of the
American Chemical Society 125, 6616-6617.
178